BULLET N OF BR] ISH MUSEUM (NATURAL HISTORY) ENTOMOLOGY VOL. II 1951-1952 PRINTED BY ORDER OF THE TRUSTEES OF THE BRITISH MUSEUM (NATURAL HISTORY) LONDON : I9J2 DATES OF PUBLICATION OF THE PARTS No. i. 12 December 1952 No. 2. 21 December 1952 No. 3. 24 November 1952 No. 4. 6 February 1952 No. 5. 21 April 1952 No. 6. 1 8 December 1952 PRINTED IN GREAT BRITAIN AT THE BARTHOLOMEW PRESS DORKING BY ADLARD AND SON, LTD, CONTENTS ENTOMOLOGY VOLUME II No. i. The early literature on Mallophaga. Part II. 1763-75. By THERESA CLAY and G. H. E. HOPKINS (Pis. 1-3.) I No. 2. A revision of the Australian and Tasmanian Gripopterygidae and Nemouridae (Plecoptera). By D. E. KIMMINS 43 No. 3. Sphecidae and Pompilidae (Hymenoptera) collected by Mr. K. M. Guichard in West Africa and Ethiopia. By G. ARNOLD (Pis. 4-5-) 95 No. 4. A collection of Fleas from the bodies of British Birds. With notes on their distribution and host preferences. By MIRIAM ROTHSCHILD 185 No. 5. The Sub-Genus Stegomyia (Diptera : Culicidae) in the Ethio- pian Region. Part I. A preliminary study of the distribu- tion of species occurring in the West African Sub- Region with notes on taxonomy and bionomics. By P. F. MATTINGLY 233 No. 6. A review, and a revision in greater part, of the Cteniscini of the Old World (Hym., Ichneumonidae). By G. j. KERRICH 305 Index to Volume II 4 r* -* 461 CORRIGENDA Pp. 103. The legends to Figs. 5 and 5 (and to PI. 4, fig. i) to read Cerceris longiuscula instead of Cerceris guichardi. P. 203. Below last entry under " Host " add : " Larus argentatus L. Herring Gull. B.M. i ." P. 225. Line 7 from bottom should read " an atypical female, probably of C. borealis, and one male of C.garei (Rothschild, 1948), etc." o. EARLY LITERATURE ON MALLOPHAGA (PART II) THERESA CLAY AND G. H. E. HOPKINS BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ENTOMOLOGY Vol. 2 No. i LONDON : 1951 THE EARLY LITERATURE ON MALLOPHAGA BY THERESA CLAY AND G. H. E. HOPKINS PART II. 1763-1775 Pp. 1-36; Ph. 1-3; 45 Text-figures BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ENTOMOLOGY Vol. 2 No. i LONDON : 1951 THE BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY), instituted in 1949, is issued in five series, corresponding to the Departments of the Museum. Parts appear at irregular intervals as they become ready. Volumes will contain about three or four hundred pages, and will not necessarily be completed within one calendar year. This paper is Vol. 2, No. i of the Entomology series. PRINTED BY ORDER OF THE TRUSTEES OF THE BRITISH MUSEUM Issued November 1951 Price Ten Shillings. THE EARLY LITERATURE ON MALLOPHAGA By THERESA CLAY and G. H. E. HOPKINS PART II, 1763-1775 SYNOPSIS This is the second part of a series of papers in which it is intended to review the species of Mallophaga described between 1758 and 1818. This part deals with the twenty species described by G. A. Scopoli in 1763 and 1772, the single species described by Pontoppidan in 1763, and the three species described by J. C. Fabricius in 1775. IN the first part of this work (Clay and Hopkins, 1950) we remarked on the possibility that specimens in the Nitzsch collection that are not the types of names given by Nitzsch might wrongly be considered as types by writers who do not accept the rules of nomenclature. We were not aware, when we wrote this, that Keler (1941) had listed many of the Halle specimens of species described by Nitzsch in 1818 by reference to older descriptions as the types of the 1818 names. Leaving aside queried instances, the names affected that concern us in our discussion of the early literature are as follows : Trichodectes crassus Nitzsch (p. 129) . The specimens of Pediculus melis in Fabricius's collection are still preserved at Copenhagen and are of necessity the types of crassus as well as of melis. Trichodectes scalaris Nitzsch (p. 130). Since scalaris is merely a nomen novum for Pediculus bovis Linn., the types are the lost specimens of Linne, which we have replaced (1950: 227). The specimens in Nitzsch's collection have no special status. Trichodectes longicornis (p. 130). Although the specimens in the Nitzsch collection at Halle are not types, there is no reason why one of them should not be erected as a neotype, and we intend to take this course in the final instalment of this work. Until this is done they have no special status. No host-records are given in Keler's list mentioned above, but the hosts can be ascertained by reference to his two papers 'Baustoffe zu einer Monographic der Mallophagen ', Parts I and II (1938 and 1939), and (as it happens) in none of the above-mentioned cases is the host of Nitzsch's material definitely different from that of the earlier author on whose description or figure Nitzsch's 1818 names rest their sole claim to validity. We therefore considered whether we could accept Keler's listing of these specimens as types of Nitzsch's names as constituting them neotypes of the earlier names, but we have decided against this. In the instances mentioned above (except in the case of crassus) this course might be possible, but in the case of crassus the original types are still in existence at Copenhagen and in other cases, also, it would be quite impossible to accept Keler's listing. As an instance, Goniodes dis- similis was first described by Denny in 1842 and the specimens listed by Keler (p. 133) would (if they have any status at all) be types of Goniodes dissimilis Nitzsch 1874 (not G. dissimilis Denny 1842, nor G. dissimilis Gurlt 1842) if the species were not the same. We think the principle underlying these statements of fact to be of such 4 THE EARLY LITERATURE ON MALLOPHAGA importance that we find ourselves unable to stretch a point by regarding Keler's listing of specimens as 'Typen' of Nitzsch's 1818 names as constituting them neo- types of the older names on which Nitzsch based them in 1818. We are confirmed in this attitude by the fact that Keler has evidently not studied the early literature at all 'equi (Lin.) ', for instance, does not exist, no species having been described under this name prior to 1842, when Denny described it. We are indebted to the Trustees of the British Museum for permission to publish Figs. 9, 10, 21, 31, 32, 34, drawn by Mr. A. J. E. Terzi, and to Colonel Meinertzhagen for permission to publish Figs. 8, 16-18, 23-25, 29-30, 38-40 drawn by Mr. R. S. Pitcher. Figs. 2, n, 12, 15, 20, 33, 36 were drawn by Miss B. A. Read. The remaining figures were drawn by Miss T. Clay. We are also indebted to Captain W. H. Pollen for the photographs on PI. II, figs. 1-2 ; the other photographs, with the exception of PI. I, fig. 2, were taken by Mr. H. M. Malies. In the measurements given under each species the length of the male genitalia has almost always been taken from a specimen other than the neotype: it has, therefore, not been considered necessary to note this fact in individual instances, as we did in Part I of this work. SCOPOLI 1763 (Entomologia Carniolica. Vindobonae: 381-385) The names in this work are unquestionably in valid form, and the descriptions, though very brief, are usually adequate for the recognition of the genus, which is as much as we can expect from the old descriptions. Scopoli's collections are stated to have been destroyed, either by fire or shipwreck, in 1776. ' Pediculus haematopus (p. 381) The description is undoubtedly that of a Philopterus s.l. and the host-record is 'Habitat in Falcone Palumbrio, & Strige Ulula '. The name would, therefore, refer to a mixture of a Craspedorrhynchus and a Strigi- phihis but for the fact that Scopoli notes that the specimens from the owl were slightly different, thus constituting Falco palumba- rius = Accipiter gentilis (Linn.) the type- host. J. C. Fabricius (1775: 806) unneces- sarily renamed the species Pediculus Strigis (nee P. strigis Pontoppidan, 1763), and Nitzsch (1818: 290) again renamed it Philo- pterus (Docophorus) platyrhynchus. Harrison's erroneous belief that haematopus is preoccu- pied by haematopi Linn., 1758 (which is not a homonym), has caused most later authors to call the species by the name given to it by Nitzsch. Pediculus tinnunculi Latreille ? 1818 (nee Linne 1758) is also a synonym of Craspe- dorrhynchus haematopus (Scopoli) . As pointed 1 Horn, W., and Kahle, J., 1936. Uber entomologische Sammlungen. Ent. Beiheften 3 : 252. FIG. i. Craspedorrhynchus haematopus (Scopoli), male genitalia. X35i. THE EARLY LITERATURE ON MALLOPHAGA 5 out by Hopkins (1949, Ann. Mag. not. Hist. (12) 2: 48), it is based on a drawing copied from one copied by Hooke (in Albin's Natural History of Spiders and Other Curious Insects, 1736) from one of Redi's figures of ' Pollini dell' Astore ', so the host is Accipiter gentilis (Linn.). Neotype male (Fig. i) and neallotype female (Fig. 2) of Craspedorrhynchus haema- topus (Scopoli) from Accipiter g. gentilis (Linn.) from Estonia (Meinertzhagen collec- tion slide No. 1464) ; these specimens agree in characters other than those figured here with the figures published by Merisuo (1945, pi. 2, fig. D; pi. 3, figs. D, d). Neoparatypes: 14 males and 12 females from the same host-form, Estonia. FIG. 2. Craspedorrhynchus haematopus (Scopoli), terminal segments of female abdomen, x 102. The neotypes are automatically also neotypes of Craspedorrhynchus strigis (J. C. Fabricius) 1775 and of Craspedorrhynchus platyrhynchus (Nitzsch) 1818. Pediculus maximus (p. 382) The particulars given, especially the size, can apply only to a Laemobothrion, and this fact has been recognized ever since the genus was erected. The host is Falco Buteo = Buteo buteo (Linn.). J. C. Fabricius (1776: 309) renamed the species Pediculus Buteonis; Nitzsch (1818: 301) proposed the name Liotheum (Laemobothrion) giganteum for maximus Scopoli, buteonis Fabricius, and circi Geoffrey, and we shall deal with Nitzsch's name in such a way as to make it a synonym of Laemobothrion maximum (Scopoli). Eichler has figured the species (1941, fig. 28, and 1942, fig. 4) and in the second of these papers he erects (p. 59) a neotype for the species, the neotype being from a specimen of Buteo vulgaris collected at Agna Man] a, Teneriffe ; Dr. Eichler informs us (in litt.} that this was probably Buteo buteo insularum Floricke. Eichler's neotype of Laemobothrion maximum (Scopoli) is automatically also neo- type of L. buteonis (J. C. Fabricius), but not of L. giganteum (Nitzsch). THE EARLY LITERATURE ON MALLOPHAGA Pediculus coarctatus (p. 382) The host-record is 'Lanio Collur'. Harrison (1916: 12) considered the name to apply to a Menopon s.l. and placed M . fuscocinctum Denny and other names in the synonymy, but this identification of Scopoli's species is quite certainly incorrect, for the mention of a white abdomen with seven conical fuscous spots on each side and with a fuscous apical fascia, coupled with the host-record, is completely diagnostic of a Philopterus. J. C. Fabricius (1798: 570) described a Pediculus lanii, from Lanius collurio Linn., which we consider to be the same species, and Schrank (1803: 187) described a Pediculus collurionis from 'Dorndreher' ( L. collurio}. Schrank's description is independent, but he quotes coarctatus as a synonym and his species is certainly a Philopterus ; we identify this, also, as a synonym of coarctatus. The species of Philopterus parasitizing the small European Passeres are very similar to each other, and in many cases it is difficult (if not impossible) to distinguish the females of two species. Specific differences in the head may be found in the form of the hyaline margin and the shape of the dorsal anterior plate (clypeal signature). The measurements of the head are, in general, unreliable characters ; not only is there considerable variation within a species but, although species may be distinguished from each other by the means of the length or breadth, there is always considerable overlap, making the identification of many specimens by measurements impossible. Again, in most species the proportions of the head show no reliable specific differences ; the cephalic index (breadth : length) tends to remain constant and gives less indica- tion of differences than do the actual measurements of length and breadth. Tables of the C.I. and breadth measurements for three species illustrate these points (Tables 1-4). As in most species with heavily sclerotized plates, there is always considerable variation in the outline of these plates. The exact outline of the prosternal plate, the abdominal tergites, and the sternites of the male and female genital regions cannot, therefore, be used as specific characters. The prosternal plate may vary considerably in outline within one species (Fig. 3), but in some species (P. citrinellae (Schrank), for instance) the posterior part is more heavily sclerotized and pigmented. The female genital plate varies in outline (Fig. 44 shows the out- line in four specimens of P. fringillae taken from the same host individual), but such characters as the curvature of the anterior margin and the ratio of breadth to length may be of specific importance. Comparison of the female genital plate of P. coarctatus with that of P. fringillae shows that the anterior margin always tends to be more rounded in the former than in the latter. The ratio of breadth to length is given in Table 5 ; the variation is caused by the variation in the length of the plate due to the lack of posterior sclerotization in some specimens, the breadth tends to be fairly constant. The chaetotaxy of the thorax and abdomen shows much variation, but some n FIG. 3. Philopterus coarctatus (Scopoli), prosternal plates of 6 females taken from one host individual. X 175. THE EARLY LITERATURE ON MALLOPHAGA FIG. 4 FIG. 5 FIGS. 4-5. Philopterus coarctatus (Scopoli) : 4. Male (setae not shown on legs). 5. Male genitalia. x 196. V FIG. 6 FIG. 7 FIGS. 6-7. Philopterus coarctatus (Scopoli), terminal segments of female abdomen: 6. Ventral view, x 152. 7. Dorsal view, x 107. 8 THE EARLY LITERATURE ON MALLOPHAGA species can be distinguished by the unusually large or small number of setae on cer- tain segments. The male genitalia, as is usual throughout the Ischnocera, give the most reliable specific differences, although in these characters, also, the differences are of small magnitude. Philopterus coarctatus (Scopoli) (Figs. 3-7 ; PI. I, fig. i ; Tables 1-5) is distinguished in the male by the characters of the genitalia, and in the female by the form of the hyaline margin, the prosternal plate, and the proportions of the genital plate. Measurements in mm. Male Female Length Breadth Length Breadth Head o-55 0-50 0-60 o-57 Prothorax . 0-28 0-32 Pterothorax 0-42 0-48 Abdomen . 0-78 0-62 1-02 0-80 Total i-55 1-67 Genitalia . 0-27 Neotype male (Figs. 4-5) and neallotype female (Figs. 3, 6, 7; PI. I, fig. i) of Philopterus coarctatus (Scopoli) from Lanius c. collurio Linn., from Yugoslavia (British Museum (Natural History) collection, slide No. 485). Neoparatypes: 28 males and 51 females from the same host-form from Yugoslavia, Poland, Estonia, Sweden, the British Isles, and the Anglo-Egyptian Sudan. Pediculus ocellatus (p. 382) The unusually good description places it beyond doubt that this species is a Philopterus, as has long been accepted. The hosts, however, are Corvus corax and C. comix, so that the name originally applied to a mixture of Philopterus corvi (Linn.) and a second species which had not then been described. It has commonly been ac- cepted as a principle that the first host mentioned by an author should be regarded as the type-host, but we can only accept this principle when there is no strong reason in favour of its rejection. In the present instance the fact that ocellatus has been accepted since 1818 (at least) as referring to the species found on Corvus corone comix would be a very strong argument against accepting Corvus corax as type-host even if Nitzsch (1818 : 290) had not restricted ocellatus to the species found on C. corone and its sub-species. J. C. Fabricius (1775 : 807) unnecessarily renamed the species as Pediculus Cornicis and dropped the mention of Corvus corax as a host. Nitzsch (1818: 290) retained Scopoli's name, contrary to his usual practice, and the species was subsequently always known as ocellatus (the authorship usually wrongly ascribed to Nitzsch) until Thompson (1935: 214) mistakenly replaced this name by Philopterus corvi (Linn.), with which we have already dealt and which is not the same species. Philopterus ocellatus (Scopoli) is distinguished from P. corvi (Linn.) by having the THE EARLY LITERATURE ON MALLOPHAGA 9 anterior plate (= clypeal signature) heavily sclerotized and pigmented posteriorly, and by the characters of the male genitalia and female genital region. FIG. 8. Philopterus ocellatus (Scopoli), male. Measurements in mm. FIG. 9. Philopterus ocellatus (Scopoli), male genitalia. Male Female Length Breadth Length Breadth Head o-75 0-76 0-80 0-80 Prothorax . 0-47 0-48 Pterothorax 0-68 0-70 Abdomen . i'34 i-ii 1-36 I-I2 Total 2-46 2-50 Genitalia . 0-48 Neotype male (Figs. 8-9, n; PI. I, fig. 2) and neallotype female (Figs. 10, 12) of Philopterus ocellatus (Scopoli) from Corvus corone sardonius Kleinschmidt from ENTOM. 2, I B 10 THE EARLY LITERATURE ON MALLOPHAGA Yugoslavia (British Museum (Natural History) collection, slide No. 484). Neopara- types: 10 males and 9 females from the same host-form from Yugoslavia, Greece, Palestine, and Egypt and 40 males and 30 females from Corvus cor one comix Linn, from the British Isles, Estonia, and Sweden. FIG. 10. Philopterus ocellatus (Scopoli), female. These neotypes are also neotypes of Philopterus cornicis (J. C. Fabricius). There is no such species as Philopterus ocellatus (Nitzsch). Pediculus dolichocephalus (p. 382) There has never been any doubt about the identity of this species, the very brief description agreeing with no oriole-parasite except Ricinus. The original host-record is 'Habitat in Coracia Oriolo' , which Harrison (1916: 66) misconstrued as meaning Coracias and Oriolus, evidently not realizing that in Scopoli's time the oriole was THE EARLY LITERATURE ON MALLOPHAGA n contained in the genus Coracias. Similar errors will be mentioned under Pediculus ardealis and P. troglodytis. J. C. Fabricius (1776 : 310) renamed the species Pediculus Orioli, and Nitzsch (1818 : 302) called it Liotheum (Physostomum) sulphureum, but there is not the slightest FIG. ii FIG. 12 FIGS. 11-12. Philopterus ocellatus (Scopoli), terminal segments of abdomen: n. Male. x8g. 12. Female. x6y. justification for either of these nomina nova. The description given by Fabricius is an abridged version of that of Scopoli, and Nitzsch gives no description and no ' indica- tion ' except a reference to Scopoli, so both these names derive their validity solely from Scopoli 's description. This species is distinguished from other species of Ricinus by the characters of the 12 THE EARLY LITERATURE ON MALLOPHAGA mandibles (Fig. 13), l by the shape of the head and the characteristic colour-pattern of the body (PI. I, fig. 3), the characters of the male genitalia (Fig. 14), and the terminal segments of the female abdomen (Fig. 15). FIG. 13 FIG. 14 FIGS. 13-14. Ricinus dolichocephalus (Scopoli) : 13. Left mandible. X 191. 14. Male genitalia. FIG. 15. Ricinus dolichocephalus (Scopoli), terminal segments of female abdomen, ventral, x 75. Measurements in mm. Male Female Length Breadth Length Breadth Head 0-76 0-68 0-97 0-78 Prothorax . o-57 0-70 Pterothorax o-75 1-03 Abdomen . 0-83 1-16 Total 3-50 4-65 Genitalia . 0-46 Neotype male (Figs. 13-14) and neallotype female (Fig. 15 ; PI. I, fig. 3) of Ricinus dolichocephalus (Scopoli) from Oriolus o. oriolus (Linn.) from NE. Poland (Meinertz- 1 These structures show characters of specific value throughout the genus Ricinus. THE EARLY LITERATURE ON MALLOPHAGA 13 hagen collection, slide No. 4190). Neoparatypes: 3 males and 10 females from same host-form, NE. Poland, Switzerland, and Cyprus. These neotypes are, of necessity, also neotypes of Ricinus orioli (J. C. Fabricius) and of R. sulphur eus (Nitzsch). Pediculus fasciatus (p. 383) The host is Cuculus canorus and the description, unmistakably that of the charac- teristic Cuculiphilus found on this bird, has long caused the name to be correctly ascribed to this louse. J. C. Fabricius (1775 : 807) renamed the species Pediculus Cuculi, his description be- ing copied from that of Scopoli, and Nitzsch (1818: 300) proposed Liotheum (Menopon) phanerostigmaton as a nomen novum for it. The species went under this latter name (and its variant, phanerostigma Giebel) until Harrison (1916: 47) restored Scopoli's name. Uchida (1926: 47) designated Pedi- culus fasciatus Scopoli as type species of Cuculiphilus. Neotype male (Figs. 16-17) and neallotype female (Fig. 18) of Cuculiphilus fasciatus (Scopoli) from Cuculus c. canorus Linn, from NE. Poland (Meinertzhagen collection, slide No. 4211). Neoparatypes: 3 males and 8 females from same host-form, NE. Poland, Ushant (France), and Tanganyika Territory. These neotypes are automatically also neotypes of Cuculiphilus cuculi (J. C. Fabri- cius) and of C. phanerostigmaton (Nitzsch). Neotype of Cuculiphilus phanerostigma (Giebel), a male from Cuculus c. canorus Linn, from Ushant, France (Meinertzhagen collection, slide No. 780), agreeing with the neotype of Cuculiphilus fasciatus (Scopoli). Measurements in mm. FIG. 1 6. Cuculiphilus fasciatus (Scopoli), male. Male Female Length Breadth Length Breadth Head o-35 0-58 o-35 0-63 Prothorax o-37 0-39 Pterothorax o-45 0-48 Abdomen 0-90 o-73 i -02 0-82 Total .. 1-52 1-79 Genitalia 0-48 THE EARLY LITERATURE ON MALLOPHAGA Pediculus auritus (p. 383) The hosts are Picus major and P. martins, i.e. Dryobates major (Linn.) and Dryo- copus martius (Linn.), and the description unquestionably refers to a Penenirmus. In the absence of any indication to the contrary, we regard the first host mentioned by Scopoli as the type-host. Schrank (1803: 188) described a Pediculus Pici from ' Schwarzpecht ' (= Dryocopus martius), placing auritus Scopoli as a synonym, but we cannot accept this as a restriction of auritus to one host, especially as Schrank also placed Pulex picae Redi (i.e. Myrsidea picae (Linn.), from Pica pica) as a synonym. FIG. 17. Cuculiphilus fasciat us (Scopoli), male genitalia. FIG. 18. Cuculiphilus fasciatus (Scopoli), female. Burmeister (1838: 427) and Giebel (1874: 94, pi. 10, fig. 3) described as Docophorus superciliosus an insect that seems to be certainly the same as Penenirmus auritus (Scopoli) and is from Dryobates major (Linn.). Harrison (1916: 88) listed auritus in Philopterus with superciliosus as a synonym. THE EARLY LITERATURE ON MALLOPHAGA 15 Neotype male (Fig. 19 ; PI. I, fig. 4) and neallotype female (Fig. 20) of Penenirmus auritus (Scopoli) from Dryobates major pinetorum (Brehm) from Yugoslavia (British Museum (Natural History) collection, slide No. 498). Neoparatypes: 12 males and 21 females from same host-form, Jugoslavia and NE. Poland; 23 males and 26 females from Dryobates major major (Linn.), Estonia. Measurements in mm. Male Female Length Breadth Length Breadth Head o-53 0-48 o-57 0-50 Prothorax . 0-29 0-30 Pterothorax 0-40 0-46 o-43 0-51 Abdomen . 0-93 0-60 I-I2 0-65 Total 1-72 1-99 Genitalia . 0-28 FIG. 19 FIG. 20 FIGS. 19-20. Penenirmus auritus (Scopoli) : 19. Male genitalia. 20. Terminal segments of female abdomen, x 87. Neotype of Penenirmus superciliosus (Burmeister) a male from Dryobates m. major (Linn.), Estonia (Meinertzhagen collection, slide No. 1568), agreeing with our descrip- tion and figures of the neotype of Penenirmus auritus (Scopoli). Pediculus dentatus (p. 383) The host-record is merely 'in Anate', but as the next species described by Scopoli is from Anas boschas, now known as A. platyrhynchos Linn., we think it only reason- able to assume that dentatus was from the same host, the specific name of which was inadvertently omitted. The description is definitely that of an Anatoecus. Eichler i6 THE EARLY LITERATURE ON MALLOPHAGA (1946: 75) selected Anas platyrhynchos as type-host of this species, but without erecting neotypes. We can find nothing in the subsequent literature that adds anything to our know- ledge of this species ; Harrison (1916 : 13, 93) referred it to Philopterus, placing most FIG. 21 FIG. 22 FIGS. 21-22. Anatoecus dentatus (Scopoli), male. 21. Genitalia. 22. Terminal segments of abdomen, x 97. of the duck-infesting species as synonyms, but pending redescription of the numerous species that have been described from ducks we are unable to suggest which names are synonyms of Anatoecus dentatus (Scopoli). Cummings (1916: 652) was the first author to draw attention to the two distinct forms of genitalia found in male A natoecus those with and those without the so-called 'effractor'. Unfortunately his otherwise excellent account of various species found on the Anatidae is rendered largely useless by his failure to give any indication of the THE EARLY LITERATURE ON MALLOPHAGA 17 host of the majority of the species that he figured. Cummings did not discuss the distribution of the two groups of species, but it has been found that on all the European ducks from which material is available there are two species of Anatoecus, the males of which are separable by the presence or absence of the effractor in addition to other less obvious characters. It has not been possible to assign the females to the different males with certainty, and for this reason no female neallotype nor neoparatypes will be designated for dentatus or other species of Anatoecus with which we shall deal. Cummings made Philopterus icterodes Nitzsch the type-species of Anatoecus and placed it in the group without an effractor ; this species will be dealt with later and Cummings's interpretation will be followed. Cummings did not mention dentatus Scopoli, and in order to fix representative species of the two main groups by the erection of neotypes and publication of figures we have chosen the species from Anas platyrhynchos in which the effractor is present to bear the name dentatus Scopoli. Measurements in mm. Male Length Breadth Head 0-47 0-43 Prothorax . 0-28 Pterothorax o-37 Abdomen . 0-72 0-60 Total 1-45 Genitalia . 0-46 Measurements in mm. of heads of specimens examined Individual No. i 2 3 4 5 6 7 8 9 JO Total length in midline 0-47 0-50 0-47 0-48 o-45 o-45 0-50 0-46 0-47 0-48 Length of hyaline margin 0-16 O'I7 0-16 0-16 0-15 0-15 0-17 0-16 0-15 0-15 Breadth at temples o-43 o-43 0-42 o-43 0-42 0-42 o-43 0-41 0-42 o-43 Breadth at base of hyaline margin .... 0-25 0-26 0-25 0-25 0-25 0-26 0-27 0-24 0-26 0-25 Neotype male (Figs. 21-22; PI. I, fig. 5) of Anatoecus dentatus (Scopoli) from Anas p. platyrhynchos Linn, from NE. Poland (Meinertzhagen collection, slide No. 4176). Neoparatypes : 9 males from same host-form, Poland and England. Pediculus crassicornis (p. 383) The host is Anas Boschas and the louse is undoubtedly an Anaticola. Schrank (1781 : 503) described as Pediculus Anatis a form which is certainly partly crassicornis and which will be discussed later ; his host is 'Anas Boschas, varietas fera ', but there is some evidence of confusion with Anaticola anseris (Linn.). J. C. Fabricius (1798: 571) also described a Pediculus Anatis from Anas Boschas, his description being apparently independent. Nirmus crassicornis 'Olfers' (see Harrison, 1916: 13) does not exist, von Olfers correctly attributing the name to Scopoli, though his host- records indicate the confusion between this species and anseris for which Schrank appears to be responsible. Nitzsch (1818: 292) proposed Philopterus (Lipeurus] ENTOM. 2, I C 18 THE EARLY LITERATURE ON MALLOPHAGA squalidus as a nomen novum for Pediculus anatis Fabricius; this name, also, is a synonym of Anaticola crassicornis (Scopoli). This species is distinguished from Anaticola anseris (Linn.) (see Clay and Hopkins, 1950 : 239) by the characters of the anterior region of the head and the shorter penis (see measurements). FIG. 23 FIG. 24 FIGS. 23-24. Anaticola crassicornis (Scopoli): 23. Male. 24. Male genitalia. X 144 ; a. Tip of paramere from side, x 437. Measurements in mm. Male Female Length Breadth Length Breadth Head 0-67 o-43 0-68 0-46 Prothorax . 0-27 0-30 Pterothorax o-45 0-47 Abdomen . 1-63 0-50 2-IO o-53 Total 2-92 3'4 Genitalia . 0-70 Length of penis (number of specimens in brackets). Anaticola crassicornis Anaticola anseris , 0-13 mm. (i), 0-15 (7), 0-16 (i). 0-27 mm. (3). THE EARLY LITERATURE ON MALLOPHAGA 19 Neotype male (Figs. 23-24) and neallotype female (Fig. 25) of Anaticola crassicornis (Scopoli) from Anas p. platyrhynchos Linn., from NE. Poland (Meinertzhagen collec- tion, slide No. 4242). Neoparatypes: 12 males and 20 females from same host-form, Poland, Hungary, and the British Isles. Pediculus pilosus (p. 384) The host is Fulica atra Linn., but all the description given is that the louse has a red head, a long abdomen, and a particularly hairy anus with the hairs parallel. Although this description is extremely un- satisfactory and Harrison (1916: 17) rejected pilosus as unrecognizable, we think that the description of the anus can only apply (among coot-parasites) to a Pseudomenopon, and the other characters are not inconsistent with this identification. Pseudomenopon could certainly be described as having a red head and its abdomen is moderately elongate, much more so than that of Incidifrons, for instance, which seems the only alternative. The species was redescribed as Menopon tridens by Burmeister (1838 : 440) from material from the same host (Fulica atra) ; Piaget (1880 : 480), wrongly taking Gallinula chloropus as the type-host of tridens, renamed the form from Fulica atra as Menopon tridens var. major (preoccupied by Menopon quadrifascia- tum var. major Piaget 1880 : 441) ; Eichler (1937: 97), noting that the name of Piaget 's variety from the coot was preoccupied, re- FIG. 25. Anaticola crassicornis (Scopoli), female. named it Pseudomenopon thompsoni. Mjoberg (1910: 51) made Menopon tridens 'N.' the type-species of his genus Pseudomenopon. Neotype male and neallotype female of Pseudomenopon pilosum (Scopoli) from Fulica a. atra Linn, from Ireland (Meinertzhagen collection, slide No. 16388). These specimens agree with the figures published by Ferris (1924, Parasitology, 16: 64, fig. 4), although the male drawn by Ferris was from the American coot, Fulica a. americana Gmelin. The stout spine-like seta and the smaller seta below it on each side of the dorsal surface of the prothorax are not shown in fig. 40 (Ferris, 1924), nor are the three setae on each side of the dorsal surface of the metathorax. Neoparatypes : 12 males and 41 females from Fulica a. atra Linn., British Isles. Neotype of Pseudomenopon tridens (Burmeister) a male (Meinertzhagen collection, slide No. 10510) from Fulica a. atra Linn, from Ireland. This specimen agrees with the neotype of P. pilosum (Scopoli). 20 THE EARLY LITERATURE ON MALLOPHAGA The types of Pseudomenopon thompsoni Eichler (Menopon var. major Piaget 1880 : 480 nee 441) are in the British Museum and also agree with the neotypes of P. pilosum (Scopoli). Pediculus colymbinus (p. 384) The host is Colymbus auritus Linn. (= Podiceps auritus (Linn.) of European authors), and Scopoli states that young specimens have an ovate abdomen with elliptical black dorsal spots and that the adult becomes rufous. This description is FIG. 26 FIG. 27. FIGS. 26-27. Aquanirmus colymbinus (Scopoli), male: 26. Genitalia. 27. Terminal segments of abdomen. X223. X 164. FIG. 28 FIG. 28. Aquanirmus colymbinus (Scopoli), terminal segments of female abdomen, ventral, x 208. extremely inadequate, but the black dorsal spots (if not gut-contents) confine us to the Ischnocera and the only genus of Ischnocera reliably reported from the grebes is Aquanirmus. Denny (1842) described a Nirmus fusco-marginatus (a male Aquanirmus) from the same host and a Nirmus podicepis (a female of the same genus) from another THE EARLY LITERATURE ON MALLOPHAGA 21 species of grebe, and later authors have identified what Scopoli considered to be young specimens with N. podicepis Denny. Although thinking the identification of colym- binus as an Aquanirmus far from certain, we see nothing in the description that dis- proves it and there is no object in disturbing the accepted application of Scopoli's name. We cannot accept the assumption that Denny's male and female are conspecific pending a much more careful examination of the forms occurring on different species of grebes than has yet been made, but the lectotype of fusco-marginatus Denny agrees with our male neotype of colymbinus. Measurements in mm. Male Female Length Breadth Length Breadth Head 0-46 o-35 o-53 0-42 Prothorax 0-28 o-33 Pterothorax o-37 o-45 Abdomen 0-87 0-40 1-47 o-55 Total 1-65 2-35 Genitalia o-37 Neotype male (Figs. 26-27 ; PI. II, fig. i) and neallotype female (Fig. 28 ; PL II, fig. 2) of Aquanirmus colymbinus (Scopoli) from Colymbus auritus Linn, from Devon, England (Meinertzhagen collection, slide No. 14736). Neoparatypes: i male and 6 females from the same host-form, England. There are two males and one nymph of Nirmus fusco-marginatus from Podiceps auritus (= Colymbus auritus) in the Denny collection; one of these males (slide No. 349) is hereby selected as lectotype of Aquanirmus fusco-marginatus (Denny). Pediculus ardealis (p. 384) Harrison (1916: n) wrongly considered this name to be a synonym of Ardeicola ardeae (Linn.), doubtless through misreading of the host-record ' Ardea Ciconia' (= Ciconia ciconia). Of the species found on Ciconia ciconia (Linn.), Scopoli's very poor description could only apply to N eophilopterus incompletus (Denny) or to a nymph of Ardeicola ciconiae (Linn.). But, in addition to the description, Scopoli gives two references: to the less hairy figure on Frisch Insect. 5, pi. 4, and to Linne, 1758, p. 613, No. 26. The reference to Frisch is an obvious slip, for plate 4 only con- tains one insect, but the reference to Linne is to Ardeicola ciconiae, and under this species Linne refers to Frisch's plate 6 ; the latter plate does show two insects, one with hairs and the other without, both belonging to the genus Ardeicola. It is, there- fore, certain that Scopoli's insect was the Ardeicola and not the N eophilopterus. Neotype of Ardeicola ardealis (Scopoli) a male from Ciconia c. ciconia (Linn.) from South Africa (British Museum collection, slide No. 430) that agrees with the neotype of Ardeicola ciconiae (Linn.) (see Clay and Hopkins, 1950: 253). Pediculus ovalis (p. 384) The host is Scolopax arquatus and the description states that the louse is smaller than Pediculus humanus, has eight abdominal segments, and a depressed ovate body, 22 THE EARLY LITERATURE ON MALLOPHAGA the antennae are shorter than the head but longer than the thorax, and the colour is rufous-brown. No subsequent author has anything useful to say about the species, and Harrison (1916: 17) discards it as unrecognizable, but we claim that it is recogni- zable with certainty. Admittedly there is no species on the curlew that agrees perfectly with Scopoli's FIG. 29 FIG. 30 FIGS. 29-30. Cummingsiella ovalis (Scopoli) : 29. Male. 30. Male genitalia. description, and in particular none with an 8-segmented abdomen, but there are only three species normally found on this host that could possibly be described as rufous- brown and oval a Cummingsiella, an Austromenopon, and perhaps Saemundssonia humeralis (Denny) . Quite apart from the facts that the Cummingsiella fits the colour- character best and is by far the commonest of the three species (and therefore the most likely to have been observed by Scopoli), the description of the antenna con- vinces us that this is the form that Scopoli described. In the Austromenopon the antennae are concealed and the palps (sometimes mistaken for antennae in early descriptions) certainly do not project far enough beyond the margin of the head to be described as being longer than the thorax, but in the male of the Cummingsiella the antennae are almost as long as the thorax. This identification of Scopoli's species THE EARLY LITERATURE ON MALLOPHAGA 23 involves the relegation of Cummingsiella testudinaria (Denny) to synonymy, but this is in any case inevitable because Docophorus testudinarius Denny and D. biseriatis Denny (the name used on p. 250 in Denny's explanation of his plate i, fig. 6) are not independent names but merely misdeterminations of Children's Nirmus testudinarius FIG. 31. Cummingsiella ovalis (Scopoli), female. and N. biseriatus and therefore invalid. The only other synonym of Cummingsiella ovalis (Scopoli) is Nirmus pseudonirmus Nitzsch. Neotype male (Figs. 29-30) and neallotype female (Fig. 31) of Cummingsiella ovalis (Scopoli) from Numenius a. arquatus (Linn.) from Yugoslavia (British Museum col- lection, slide No. 522). Neoparatypes: 86 males and 99 females from the same host- form, Yugoslavia, Hungary, and the British Isles. Neotype of Cummingsiella pseudonirmus (Nitzsch) a male (British Museum (Natural 2 4 THE EARLY LITERATURE ON MALLOPHAGA History), slide No. 523) from Numenius a. arquatus (Linn.) from Scotland, which agrees with the neotype of C. ovalis (Scopoli). ) Measurements in mm. Male Female Length Breadth Length Breadth Head 0-70 0-72 o-75 o-73 Prothorax . 0-42 o-45 Pterothorax 0-65 0-70 Abdomen . 1-16 1-03 1-50 I'll Total 2-IO 2-56 Genitalia . 0-65 The Denny collection contains one male and two female Cummingsiella labelled Docophorus testudinarius but without host-record. These specimens agree with the neotypes of C. ovalis (Scopoli). Pediculus junceus (p. 384) The host is Tringa vanellus = Vanellus vanellus (Linn.), and the description is obviously that of a Degeeriella s.l. Denny (1842 : 53, 143, pi. 9, fig. 5) made an identifica- tion of this species with which we are in full agreement ; his specimens (3 females) came from the same host, although on p. 143 he added two other hosts. We do not agree with Harrison in considering Pediculus tringae Schrank 1803 (nee O. Fabricius, 1780) to be the present species, and it will be discussed separately. On the other hand, in spite of Schrank's definite statement (1803: 190) that his Pediculus Vanelli, from 'Kybize' (= Vanellus vanellus) is not Scopoli 's insect we can find nothing in his description that supports this statement and suspect that the explanation must be that he misidentified junceus ; our suspicion that this is the case is much strengthened by the fact that we have been unable to find any nirmoid species except junceus on Vanellus vanellus. Measurements in mm. Male Female Length Breadth Length Breadth Head 0-47 0-30 0-47 0-32 Prothorax. O-22 O-22 Pterothorax 0-30 0-32 Abdomen . o-93 0-41 1-03 o-45 Total 1-64 1-82 Genitalia . 0-30 Neotype of Quadraceps junceus (Scopoli) a male (Fig. 32 ; PI. II, fig-3) and neallo- type a female (Fig. 33; PL II, fig. 4) from Vanellus vanellus (Linn.) from Italy (British Museum (Natural History), slide No. 524). Neoparatypes: 154 males and 140 females from the same host-form, Italy and the British Isles. THE EARLY LITERATURE ON MALLOPHAGA 25 Pediculus cuspidatus (p. 385) Denny (1842: 51, 130, pi. 6, fig. 2) redescribed as Nirmus cuspidatus a species that he took to be the same as that of Scopoli, though he thought it necessary to query the FIG. 32 FIG. 33 FIGS. 32-33. Quadraceps junceus (Scopoli) : 32. Male genitalia. 33. Terminal segments of female abdomen, x 125. determination. Scopoli's host-record is 'Ratio aquat.', and Denny's original specimen (no longer in his collection) evidently also came from Rallus aquaticus Linn., because this is the only host mentioned on p. 51. Scopoli's description is not by any means diagnostic, but so far as it goes it fits Rallicola at least as well as can be expected from these old descriptions. No subsequent author seems to have seen the species. Neotype male (Figs. 34-35; PI. II, fig. 5) and neallotype female (Fig. 36; PI. II, fig. 6) of Rallicola cuspidatus (Scopoli) from Rallus a. aquaticus Linn, from Kent, ENTOM. 2, I D 26 THE EARLY LITERATURE ON MALLOPHAGA England (Meinertzhagen collection, slide No. 8332). Neoparatypes: 13 males and 48 females from the same host-form, British Isles. FIG. 35 FIG. 34 FIG. 36 FIGS. 34-36. Rallicola cuspidatus (Scopoli) : 34. Male genitalia. 35. Male antenna, x 173. 36. Terminal segments of female abdomen. X 115. Measurements in mm. Male Female Length Breadth Length Breadth Head o-37 0-30 0-42 o-35 Prothorax . 0-19 O-22 Pterothorax 0-27 0-30 Abdomen . 0-62 o-35 0-85 o-43 Total i-ii 1-44 Genitalia . 0-25 THE EARLY LITERATURE ON MALLOPHAGA 27 Pediculus bidentatus (p. 385) The host is Columba palumbus Linn., and the reference to a hemispherical head rules out all genera of Mallophaga known from European pigeons except Coloceras and Campanulotes. Pediculus bidentatus was constantly placed as a synonym of Campanulotes compar (Burmeister), from Columba lima domestica, until Harrison (1916) reversed this arrangement. Although we find nothing in the original descrip- tion of bidentatus that enables us to decide which of the two genera Scopoli had before him, we think it best to accept this long-standing generic determination of his species. This species is near C. compar (Burmeister) from Columba lima, from which it is distinguished by its greater size (see measurements below). The chaetotaxy and FIG. 37. Campanulotes bidentatus (Scopoli), terminal segments of female abdomen, x 121. general characters are as figured by Keler (1939: 158-160, figs. 89-91) for C. compar except that in fig. 89 a median ventral seta has been omitted on segments VI and VII ; in fig. 90 the latero-dorsal temple spine has been omitted, the dorsal setae on abdominal segments II-III should be about twice as long as shown, and the median dorsal setae on segments V-VI have been omitted ; and in fig. 91 the long ventral seta on the last abdominal segment has been omitted. The genital region of the female of C. bidentatus is shown in Fig. 37. Measurements in mm. Male Female Length Breadth Length Breadth Head o-34 0-47 o-43 o-53 Prothorax . 0-29 o-34 Pterothorax o-37 0-42 Abdomen . 0-61 o-57 0-90 0-66 Total i -20 1-58 Neotype of Campanulotes b. bidentatus (Scopoli) a male (PI. Ill, fig. i) and neallotype a female (Fig. 37) from Columba p. palumbus Linn, from Somerset, England (Meinertz- hagen collection, slide No. 864). Neoparatypes: 63 males and 58 females from the British Isles. ENTOM. 2, i D2 28 THE EARLY LITERATURE ON MALLOPHAGA Comparison of breadth of heads, in mm. Species and No. of specimens Males Breadth in mm. C. bidentatus compar . C. bidentatus bidentatus. Breadth in mm. C. bidentatus compar C bidentatus bidentatus . 0-38 o-59 0-40 0-44 o-45 0-46 0-47 0-48 2 10 7 i 4 5 12 I Females o-43 0-44 o-45 0-46 o-47 0-49 0'50 o-5-r 0-52 0-55 i 4 5 8 2 5 13 9 5 4 Pediculus albiventris (p. 385) The original host-record is ' Motacilla Troglodyte', which Harrison (1916: 87) wrongly took to mean Motacilla and Troglodytes, whereas only one species, the bird now known as Troglodytes troglodytes (Linn.), is mentioned. Clay and Meinertzhagen (1938: 73) showed that Pediculus albiventris (Scopoli) is the same as Docophorus FIG. 38 FIG. 39 FIGS. 38-39. Penenirmus albiventris (Scopoli) : 38. Male. 39. Male genitalia. X 600. THE EARLY LITERATURE ON MALLOPHAGA 29 troglodytis (Waterston), from Troglodytes troglodytes zetlandicus Hartert, and made it the type species of Penenirmus. J. C. Fabricius (1776: 310) renamed the species Pediculus motacillae. Measurements in mm. Male Female Length Breadth Length Breadth Head 0-42 o-37 0-48 0'45 Prothorax . O-2O 0-25 Pterothorax o-33 0-40 Abdomen . o-73 o-45 1-13 0-63 Total i-33 1-85 Neotype of Penenirmus albiventris (Scopoli) a male (Figs. 38-39) and neallotype a female (Fig. 40) from Troglodytes t. troglodytes (Linn.) from Wiltshire, England (Meinertzhagen collection, slide No. 15399). Neoparatypes: 44 males and 52 females from various subspecies of Troglodytes troglodytes from the British Isles. These neotypes are also automatically neo- types of Penenirmus motacillae (J. C. Fabricius) PONTOPPIDAN, 1763 (Den Danske Atlas. 1 Kiobenhavn: 699) Only one of the names of Mallophaga published in this work is new. Pediculus strigis (p. 699, pi. xxx) This species is stated to be new and is des- cribed with a reference to plate xxx b, a figure of an obvious Philopterus s.l. There is no host- record other than that provided by the specific name. Before we go on to discuss the identity of the species it will be as well to consider the sub- sequent history of the name. Scopoli (1772: 124) gives a completely inde- pendent description of a Pediculus strigis which is also clearly a Philopterus s.l. ; there is no host- record, but as he describes a Hippoboscid from Strix bubo we think it nearly certain that this species was also the host of the louse. J. C. Fabri- FIG. 40. Penenirmus albiventris (Scopoli), female. cius (1775 : 806) applied the name Pediculus strigis to Pediculus haematopus Scopoli, Fabricius's nomen novum thus being not only unnecessary but twice preoccupied, to say nothing of the fact that haematopus (and therefore strigis Fabricius) is not from an owl but from a hawk. Miiller (1776 : 185) mentioned the name with a reference to 30 THE EARLY LITERATURE ON MALLOPHAGA 'P.D.A.' (= Pontoppidan's Danske Atlas). O. Fabricius (1780: 216) gave references to Pontoppidan and Miiller and also an independent description of the differences between his material (from Strix nyctea) and Pontoppidan's plate. Harrison (1916: 1 8) did not know strigis Pontoppidan and rejected all the other uses of the name as either unrecognizable or preoccupied. The position may be summarized as follows : P. strigis Pontoppidan is a valid name and easily determinable to the genus. P. strigis Scopoli may or may not be Pontoppidan's species but is certainly con- generic. P. strigis J. C. Fabricius is an unwanted nomen novum for Pediculus (now Craspe- dorrhynchus) haematopus Scopoli. It is twice preoccupied and is not congeneric with the others. P. strigis 'Miiller' does not exist, Miiller merely listing Pontoppidan's species. P. strigis O. Fabricius is partly a reference to Pontoppidan and partly a new species, the name of the latter thrice preoccupied. In the circumstances it seems to us that much the most satisfactory course is to fix strigis Pontoppidan in such a way that it is the same as strigis Scopoli and (if possible) so that its restoration does not upset any well-established name of later date. Not only does Bubo bubo, which we consider to be the host of Scopoli's species, occur in Denmark, but Pontoppidan definitely described another parasite from this host, so that it is very probable that his louse came from this species of bird. Two species of Strigiphilus occur on Bubo bubo, one of which has long been known as 5. heteroceros (Nitzsch) whereas the other had not been named until Eichler (1949 : 14) named it ' Neodocophorus ' hopkinsi, though it is probably a component of 5. cursor (Burmeister) as described by Giebel (1874: 70). The first species, however, cannot retain the name Strigiphilus heteroceros (Nitzsch), because this species was not described until 1861 whereas Grube used the same name (Docophorus heterocerus) for a species found on Strix uralensis liturata Tengmalm ten years earlier (Grube, 1851: 469). Eichler (1949: n) has correctly pointed out this fact and renamed the species with sexually dimorphic antennae found on Bubo bubo as Strigiphilus goniodicerus. Both the species concerned have, therefore, been named and both the names are of equal seniority, so it does not matter to which species we apply the name strigis Pontoppidan (his figure applying fairly well to either), so we have selected hopkinsi to bear Pontoppidan's name. Measurements in mm. Male Length Breadth Head 0-69 0-63 Prothorax . 0-40 Pterothorax 0-56 Abdomen . 0-94 0-84 Total 1-93 Genitalia . 0-47 This species is distinguished from 5. goniodicerus Eichler, from the same host, by the antennae being similar in the two sexes and by the smaller and less complicated THE EARLY LITERATURE ON MALLOPHAGA 31 male genitalia. Although there are two species of females represented in the available material it has not been possible to assign them with certainty to the males, so no neallotype or female neoparatypes have been erected. FIG. 41 FIG. 42 FIGS. 41-42. Strigiphilus strigis (Pontoppidan), male: 41. Terminal segments of abdomen, x 74. 42. Genitalia. x 294. Neotype of Strigiphilus strigis (Pontoppidan) a male (Figs. 41-42, PI. Ill, fig. 2) from Bubo bubo (Linn.) from Russia (Meinertzhagen collection, slide No. 109750;). Neoparatypes: 9 males from the same host-form, Russia and Italy (captive host). Neotype of Strigiphilus strigis (Scopoli) a male (Meinertzhagen collection, slide No. 109756) from Bubo bubo (Linn.) from Russia, that agrees with the neotype of Strigi- philus strigis (Pontoppidan). SCOPOLI, 1772 (Annus V Historico-Naturalis. [PC.] 5. Observationes Zoologicae. Lipsiae: 124-125) The descriptions in this work are very poor. Fortunately they are also very few. Pediculus hirci junioris (p. 124) We only mention this name because we feared at first that it might be the earliest name for a chewing louse of the goat. But we are certain that the description does not refer to a member of the Mallophaga and we think it likely that the insect was one of the Anoplura. Pediculus strigis (p. 124) This name has been dealt with under Pediculus strigis Pontoppidan and a neotype has been erected. Strigiphilus strigis (Scopoli) is both a homonym and a synonym of 5. strigis (Pontoppidan). 32 THE EARLY LITERATURE ON MALLOPHAGA Pediculus ralli (p. 125) There is no host-record other than that provided by the specific name, but the host must be assumed to have been some bird that occurs in Carniolia and was included in the genus Rallus in Scopoli's time. The entire description is that the head of the insect is bidentate and the abdomen glabrous, with pilose and crenate sides. Among parasites of the Rallidae this descrip- tion could only apply to the genus Incidifrons, but as we have not seen this genus from any bird that complies with the conditions we have mentioned as necessary assumptions we are unable to erect a neotype for the species. We think it of the first importance that Incidifrons ralli (Scopoli) should eventually be fixed in such a way that Incidifrons ralli (Denny) becomes a synonym as well as a homonym, thus avoid- ing the confusion that would be caused by the transfer of the name ralli from one species to another. We therefore intend to assume in all future work that the host of Incidifrons ralli (Scopoli), like that of 7. ralli (Denny), was Rallus a. aquaticus Linn., and we most strongly urge other workers to make the same assumption. There are no specimens of 7. ralli in the Denny collection. Pediculus fringillae (p. 125) The entire description is that the head is bidentate and the sides of the abdomen are pilose and rugose, but among parasites of the birds included in Fringilla in Scopoli's time only Philopterus fits this description. There is no host-record other than that provided by the name, so we have chosen as host of our neotype one of the commoner birds included by Scopoli in Fringilla, namely, Fringilla domestica, now known as Passer domesticus Linn. 1 We hope that our action will finally settle the confusion that has arisen (as shown in Part I of this work (Clay and Hopkins, 1950: 270)) through the application of Geoffrey's invalid appellation ' subflavescens ' to the Philop- terus from this host. Fourcroy (1785: 519) gave the name Pediculus passeris to the species described by Geoffroy, and Philopterus passeris (Fourcroy) and P. passeris (Piaget) (together with subflavescens of authors subsequent to Geoffroy) become synonyms of Philopterus fringillae (Scopoli). This species has been discussed above under P. coarctatus (Scopoli) 1763, from which it differs in having a median indentation in the hyaline margin of the head, and in the characters of the male genitalia and female genital plate. Neotype male (Fig. 43 ; PI. Ill, fig. 3 ; Tables I and 3) and neallotype female (Fig. 44 ; Tables 2 and 4-5) of Philopterus fringillae (Scopoli) from Passer d. domesticus (Linn.) FIG. 43. Philopterus frin- gillae (Scopoli), male geni- talia. X3IQ. : This host is not given in the publication under discussion, in which no species of Fringilla are men- tioned, but in Part I of the same work, published in 1769, Scopoli refers to Fringilla domestica on p. 149. THE EARLY LITERATURE ON MALLOPHAGA 33 from Hungary (Meinertzhagen collection, slide No. 80770). Neopamtypes: 47 males and 57 females from the same host-form, Hungary, Estonia, and the British Isles. Measurements in mm. Male Female Length Breadth Length Breadth Head o-55 0-50 0-58 0-56 Prothorax . 0-29 0-32 Pterothorax o-45 0-50 Abdomen . 0-70 0-68 I -02 0-91 Total i-45 1-85 Genitalia . 0-31 FIG. 44. Philopterus fringillae (Scopoli), genital plates of 4 females taken from one host individual, x 102. Pediculus pari palustris (p. 125) 'Cauda quadriseta, ut in P. Pari majoris Linn.' As no differences from Linne's species are mentioned we consider this to be a nomen nudum, but in any case the name clearly does not refer to the Mallophaga but to a mite. J. C. FABRICIUS, 1775 (Systema Entomologiae. Flensburgi et Lipsiae: 804-810) The great majority of the names mentioned in this work have already been dealt with, being either quoted from Linne or entirely unnecessary renamings of Scopoli's species, with descriptions quoted from the latter author. Only the following are genuinely new : Pediculus vulturis (p. 806) The description is quite obviously that of a Laemobothrion, but the host-record is merely 'Habitat in Indiae orientalis vulturibus'. In these circumstances the most reasonable procedure seems to be to attach the name to a Laemobothrion from one of the Indian vultures. We have chosen Pseudogyps bengalensis (Gmelin). This species differs from L. tinnunculi (Linn.) (see Clay and Hopkins, 1950: 228) 34 THE EARLY LITERATURE ON MALLOPHAGA in the larger size, the shape of the head (PI. Ill, figs. 4, 6), in the greater number of prosternal setae 1 and the shape of the sternal plates (Fig. 45), the presence of a line of setae on the lateral margins of the sternal plates, and in details of the male genitalia. No material of Laemobothrion maximum (Scopoli) has been seen from Buteo buteo, but from Eichler's figures (1941 : 363, fig. 28 ; 1942: 59, fig. 4), and examination of specimens (possibly not conspecific with maximum) from other species of Buteo, it seems to differ from vulturis in the smaller size, the smaller number of prosternal setae, the shape of the prosternal plate, and possibly in the details of the male genitalia. Measurements in mm. FIG. 45. Laemoboth- rion vulturis (J. C. Fabri- cius), female sternal plates. X3i. Male Female Length Breadth Length Breadth Head* 1-52 1-70 i-56 i -80 Abdomen . 576 2-48 6-92 3-02 Total 9-24 10-56 Genitalia . 2-90 * The head is liable to distortion in mounted specimens of Laemobothrion and throughout the genus these measurements may show considerable variation in one species. Neotype male (PI. Ill, figs. 4-5) and neallotype female (Fig. 45 ; PI. Ill, fig. 6) of Laemobothrion vulturis (J. C. Fabricius) from Pseudogyps bengalensis (Gmelin) from Deccan, India (Meinertzhagen collection, slide No. 8607). Neoparatypes : i male and 2 females from the same host-form, Deccan and Siam. Pediculus procellariae (p. 808) An elongate, filiform, fuscous species occurring ' in Brasiliae procellariis ' can only be a Halipeurus or a Perineus, the former being the more probable identification because members of this genus are more elongate than those of Perineus (Fabricius seems to have been particularly impressed by this character) and also more fuscous. But in view of the number of species of petrels which occur in Brazilian waters the selection of host must be purely arbitrary. Mr. R. L. Edwards informs us that in a paper he is about to publish he will redescribe Halipeurus procellariae (J. C. Fabricius) from specimens taken from Pterodroma m. macroptera (A. Smith). Halipeurus con- strictiventris (Pessoa and Guimaraes) 1935 will thus become a synonym of H. procellariae. Pediculus diomedeae (p. 808) One of us (Clay, 1940: 300-302) has already discussed this species and erected neotypes from Diomedea m. melanophris Temminck. It is perhaps as well to repeat that it is a Perineus, not a Harrisoniella, has nothing to do with H arrisoniella ferox (Giebel), and that the type species of Harrisoniella is Esthiopterum diomedeae Harrison 1916 (nee J. C. Fabricius 1775) (= Lipeurus ferox Giebel). 1 Examination of further material suggests that this is an unreliable character owing to individual variation. See Part III (in press) for further notes on Laemobothrion vulturis. THE EARLY LITERATURE ON MALLOPHAGA 35 Pediculus hirundinis (p. 810) We have noted above (Clay and Hopkins, 1950 : 26) that this name must be ascribed to Linne and have discussed it under that author. TABLES 1-5, MEASUREMENTS OF PHILOPTERUS SPECIES TABLE I. Breadth (in mm.} of head at temples of males, with number of specimens o-45- 0-47- o-49- 0-51- o-53- o-55- 0-46 0-48 0-50 0-52 o-54 0-56 coarctatus 6 17 4 fringillae 3 15 7 2 citrinellae 3 22 2 TABLE 2. Breadth (in mm.} of head at temples of females, with number of specimens o-47- 0-49- 0-51- 0-53- o-55- 0-57- 0-59- 0-61- 0-48 0-50 0-52 o-54 0-56 0-55 0-60 0-62 coarctatus 3 5 4 18 fringillae 2 6 13 4 5 citrinellae i 3 6 16 4 TABLE 3. Cephalic index of males, with number of specimens 0-90- 0-92- 0-94- 0-96- 0-98- I-OO- I-O2- 0-97 o-93 o-95 0-97 0-99 I-OI 1-03 coarctatus 3 10 7 7 3 fringillae 2 5 10 3 4 citrinellae 4 10 9 3 i TABLE 4. Cephalic index of females, with number of specimens 0-90- 0-92- 0-94- 0-96- 0-98- I-OO- I-O2- 1-04- 0-91 o-93 o-95 0-97 0-99 I-OI 1-03 1-05 coarctatus 3 5 ii 4 6 I fringillae i 4 7 7 7 3 i citrinellae 2 8 ii 7 2 TABLE 5. Ratio of breadth : length of female genital plate, with number of specimens in parentheses 1-38 (i), 1-45 (i). i-53 (2), 1-54 (i). i-55 (i). 1-59 (i), 1-69 (2), 1-75 (i). . 1-79 (i), 1-88 (3), 1-91 (i), 1-94 (2), 2-00 (3), coarctatus fringillae 1-58 (i). 2-10 (i). LIST OF SPECIES The synonymy of the following names has been established. Specific name Present status albiventris Scopoli. Penenirmus albiventris (Scopoli). anatis Fabricius. Anaticola crassicornis (Scopoli). Page 28 THE EARLY LITERATURE ON MALLOPHAGA Specific name ardealis Scopoli. auritus Scopoli. bidentatus Scopoli. biseriatis Denny. buteonis Fabricius. coarctatus Scopoli. collurionis Schrank. colymbinus Scopoli. constrictiventris Pessda & Guimaraes. cornicis Fabricius. crassicornis Scopoli. cuculi Fabricius. cuspidatus Scopoli. dentatus Scopoli. diomedeae Fabricius. dolichocephalus Scopoli. fasciatus Scopoli. fringillae Scopoli. fusco-marginatus Denny. gigantum Nitzsch. haematopus Scopoli. junceus Scopoli. lanii Fabricius. maximus Scopoli. motacillae Fabricius. ocellatus Scopoli. orioli Fabricius. ovalis Scopoli. passeris Fourcroy. passeris Piaget. phanerostigma Giebel. phanerostigmaton Nitzsch. pilosus Scopoli. platyrhynchus Nitzsch. procellariae Fabricius. pseudonirmus Nitzsch. ralli Scopoli. ralli Denny. squalidus Nitzsch. strigis Pontoppidan. strigis Scopoli. strigis Fabricius. subflavescens auctorum. sulphureum Nitzsch. superciliosus Burrneister. testudinarius Denny. thompsoni Eichler. tinnunculi Latreille. tridens Burrneister. tridens var. major Piaget. troglodytis Waterston. vanelli Schrank. vulturis Fabricius. Present status Ardeicola ciconiae (Linn.). Penenirmus auritus (Scopoli). Campanulotes bidentatus (Scopoli). Cummingsiella ovalis (Scopoli). Laemobothrion maximum (Scopoli). Philopterus coarctatus (Scopoli). Philopterus coarctatus (Scopoli). Aquanirmus colymbinus (Scopoli). Halipeurus procellariae (Fabricius) . Philopterus ocellatus (Scopoli). Anaticola crassicornis (Scopoli). Cuculiphilus fasciatus (Scopoli) . Rallicola cuspidatus (Scopoli). Anatoecus dentatus (Scopoli). Perineus diomedeae (Fabricius). Ricinus dolichocephalus (Scopoli). Cuculiphilus fasciatus (Scopoli). Philopterus fringillae (Scopoli). Aquanirmus colymbinus (Scopoli). Laemobothrion maximum (Scopoli). Craspedorrhynchus haematopus (Scopoli). Quadraceps junceus (Scopoli) . Philopterus coarctatus (Scopoli). Laemobothrion maximum (Scopoli). Penenirmus albiventris (Scopoli). Philopterus ocellatus (Scopoli). Ricinus dolichocephalus (Scopoli). Cummingsiella ovalis (Scopoli). Philopterus fringillae (Scopoli). Philopterus fringillae (Scopoli). Cuculiphilus fasciatus (Scopoli) . Cuculiphilus fasciatus (Scopoli). Pseudomenopon pilosum (Scopoli). Craspedorrhynchus haematopus (Scopoli). Halipeurus procellariae (Fabricius). Cummingsiella ovalis (Scopoli). Incidifrons ralli (Scopoli). Incidifrons ralli (Scopoli). Anaticola crassicornis (Scopoli). Strigiphilus strigis (Pontoppidan). Strigiphilus strigis (Pontoppidan). Craspedorrhynchus haematopus (Scopoli). Philopterus fringillae (Scopoli). Ricinus dolichocephalus (Scopoli). Penenirmus auritus (Scopoli). Cummingsiella ovalis (Scopoli). Pseudomenopon pilosum (Scopoli). Craspedorrhynchus haematopus (Scopoli). Pseudomenopon pilosum (Scopoli). Pseudomenopon pilosum (Scopoli). Penenirmus albiventris (Scopoli). Quadraceps junceus (Scopoli) . Laemobothrion vulturis (Fabricius). Page 21 15 2 7 23 5 6 6 20 34 IO 17 13 25 15 34 10 13 32 21 5 4 24 6 5 29 8 13 21 32 32 13 13 19 5 34 23 32 32 18 29 3i 5 32 13 15 24 19 4 19 19 29 24 34 THE EARLY LITERATURE ON MALLOPHAGA 37 REFERENCES Only those papers not referred to in Part I are listed here. CLAY, T. 1940. Anoplura. British Graham Land Exped. Sci. Rep. (Brit. Mus. (Nat. Hist.)), 1: 295-318. & HOPKINS, G. H. E. 1950. The Early Literature on Mallophaga. Pt. I. Bull. Brit. Mus. (Nat. Hist.) Ent. 1 (3) : 223-272. EICHLER, W. 1941. Zur Klassifikation der Lauskerfe. Arch. Naturgesch. B (N.F.), 10: 345-398. 1942. Mallophagen-Synopsis. III. Genus Laemobothrion. Zool. Anz. 137: 52-63. 1946. Mallophagen-Synopsis. VIII. Genus Anatoecus. Tijdschr. Ent. 87: 74-76. 1949. Die Eulenfederlinge. Beitr. Taxon. Zool. 1: 7-22. KELER, S. 1938. Baustoffe zu einer Monographic der Mallophagen. I. Teil. Nova Ada Leop. Carol. (N.F.), 5: 385-467. 1939. Baustoffe zu einer Monographic der Mallophagen. II. Teil. Nova Ada Leop. Carol. (N.F.), 8: 1-254. 1941. Systematisches Verzeichnis der von Nitzsch begriindeten und von Giebel und Taschenberg fortgefuhrten und bereicherten Sammlung von Mallophagen des Zoologischen Instituts der Universitat Halle. Z. Naturw. 95: 123-136. MERISUO, A. K. 1945. Notulae Mallophagologicae. II. Ann. Ent. Fenn. 11: 101-112. WATERSTON, J. 1915. On some Mallophaga in the Kgl. Zool. Museum, Konigsberg. Zool. Jb. (Abt. Syst.}, 39: 17-42. PRESENTED 1 7 DEC 1951 PLATE 1 FIG. i. Philopterus coarctatus (Scopoli), $. FIG. 2. Philopterus ocellatus (Scopoli), . FIG. 3. Ricinus dolichocephalus (Scopoli), . FIG. 4. Penenirmus auritus (Scopoli), <$. FIG. 5. Anatoecus dentatus (Scopoli), $. Bull. B.M. (N.H.} Ent. II, 1 PLATE 1 PLATE 2 FlG. I. Aquanirmus colymbinus (Scopoli), $, FIG. 2. Aquanirmus colymbinus (Scopoli), . FIG. 3. Quadraceps junceus (Scopoli), <$. FIG. 4. Quadraceps junceus (Scopoli), $. FIG. 5. Rallicola cuspidatus (Scopoli), $. FIG. 6. Rallicola cuspidatus (Scopoli), $. f ,. , utj,^ Bull. B.M. (N.H.) Ent. II, 1 PLATE 2 PLATE 3 FIG. i. Campanulotes bidentatus (Scopoli), N.Z. : 119. Newman's description (given below) is good, but owing to the aberrant fore wing, his statement that there are six parallel veins in the fore wing should be read as five. 'Sp. i. Leptoperla Beroe. Fusca, alae opacae, fuscae, versus apicem maculis 50 A REVISION OF THE AUSTRALIAN AND TASMANIAN albidis notatae ; pro- et mesopedes fusci, tibiis medio testaceis : metafemora testacea, apice fusca. (Corp. long. -3 unc. ant. -475 unc. set. caud. -475 unc. alar, dilat. -9 unc.) 'This is a slender and very elegant insect, it differs generically from Isogenus and Perla in the neuration of the fore wings, the exterior portion of which is occupied by six strong parallel longitudinal nervures ; of these the fourth is furcate at the extremity and the fifth unites with the fourth just before its furcation; 1 these longitudinal nervures are intersected by several very delicate transverse nervures : the antennae and caudal setae are very slender, and much longer than the body of the insect ; the terminal segment of the abdomen is furnished below with two leaf-like processes, which curve upwards, passing between the caudal setae and terminating in acute points : the legs are very long and slen- der : the insect is of a dark brown colour, the wings being opaque, dark brown, and the exterior portion of the fore wings regularly spotted with dirty white ; the hind wings are immaculate ; the pro- and mesofemora having a bright testaceous ring; the metafemora are testaceous, with the apex only dark brown, the tibiae are rather paler, and the tarsi nearly black. ' Inhabits Van Diemens Land. There is a single specimen in the cabinet of the Rev. F. W. Hope.' (Now in the Hope Department, University Museum, Oxford.) To this description may be added the following account of the male genitalia. Apical tergites pigmented at bases and sides, sternites with a pair of basal spots. Tenth tergite large, narrowed apically, FIG. 3. Leptoperla beroe Newman, $. Genitalia, a, lateral; b, dorsal; c, ventral. apical margin truncate from the side and carrying a short triangular process. There is a low median dorsal carina, and at the centre of the basal margin is a small quad- rate, membranous area. Supra-anal lobe slender, not projecting much beyond the margin of the tergite. From the side there is a dorsal excision before the apex, giving the appearance of a crochet-hook, and several acute dorsal teeth. Sub-anal plates large, broad, leaf-like, curving upwards, with somewhat twisted apices ; lower basal margins heavily fringed with hairs. Cerci long, at least thirty-seven-segmented. Subgenital plate triangularly produced, pigmented at its centre, giving the im- pression of being narrower. The above genitalic description has been made from a second example in the Hope Department, University Museum, Oxford. 1 Tillyard has shown this to be an aberration ; in the left fore wing all these veins run parallel. GRIPOPTERYGIDAE AND NEMOURIDAE (PLECOPTERA) 51 Leptoperla australica (Enderlein) FIG. 4 Paranotoperla australica Enderlein, 1909, ZooL Anz. 34: 416; 1909, Dtsch. Ent. Z. 1909: 684; 1912, Fauna SW . Aust.: 60; Tillyard, 1921, Trans. Roy. Soc. S. Aust. 45: 270; ? Samal, 1921, Casopis, 18: 66-70. Samal's reference to the occurrence of this species in Tasmania is probably based on a mis-determination and may possibly refer to L. beroe. I have not seen any examples of L. australica from the island and I believe it to be restricted to west and south-west Australia. FIG. 4. Leptoperla australica (Enderlein) $. a, wings ; b, genitalia, ventral. Head brownish-yellow, darker above. Eyes black. Thorax brown, almost rect- angular, somewhat longer than broad, front and hind margins straight, lateral margins straight and diverging slightly from front to back ; no impressed median line, but with anterior and posterior transverse furrows ; anterior angles rounded. Abdomen light brown, in the female paler beneath, subgenital plate brown. Cerci dark brown, longer than abdomen, about thirty segments. Legs brownish-yellow, marked with dark brown at the knees, apices of tibiae, and on the tarsi. Wings light brown, with brown venation, distal cross- veins pale and enclosed in whitish spaces; generally without cross-veins between Cui and wing margin in fore wing. I have seen no males ; Enderlein says that the last tergite in this sex is triangular. His description of the subgenital plate apparently refers to the female. ?. Subgenital plate large, quadrate, angles rounded, projecting somewhat beyond the segment, brown, with pale triangular excised areas, a broad shallow area at apex and a narrow deep one at base. Sub-anal plates triangular, apices curving outward. Length of fore wing $ 9-5-10 mm., $ 7-5-9-5 mm. Length of hind wing $ ?, $ 6-5-8-5 mm. SW. AUSTRALIA: Lunenberg; 22.ix.i9O5, Serpentine, 23~25.ix.i905, Harvey, 27.vii.i9o5 (Hamburg SW. Australia Expedition). WEST AUSTRALIA: Bolgonup, i.x.1922 (R. J. Tillyard). Type in Stettin Museum. Leptoperla tasmanica sp.n. FIG. 5 (In fluid.) Head, thorax, and three terminal segments of abdomen brown, shin- ing, segments one to seven whitish. Antennae longer than fore wing. Pronotum ENTOM. 2, 2 F 52 A REVISION OF THE AUSTRALIAN AND TASMANIAN trapezoidal, anterior margin two-thirds as wide as posterior, a little shorter than width at base. Legs brownish, distinctly banded with dark brown towards apex of femur and base of tibia, apex of tibia and tarsus clouded with darker brown. Wings elongate, fore wing medium fuscous, with pale markings ; all cross-veins, and to a lesser degree FIG. 5. Leptoperla tasmanica sp.n. a, wings, KS an<1 M se P arate at some < e, genitalia $, ventral. from base, about half-way to fork of M. GRIPOPTERYGIDAE AND NEMOURIDAE (PLECOPTERA) 57 cJ. Apical margin of tenth tergite deflexed, and bearing a short process, finger-like from the side, keeled dorsally. Supra-anal lobe upcurved, of medium width at base, tapering to a slender apex from above, upper lateral margins set with triangular teeth, a sinuous hooked projection on the ventral surface directed basally. Cerci brownish, longer than the abdomen, twenty-seven or more segments. Sub-anal plates broad, upcurving, apices truncate and setose, upper angles curving over outwards. Subgenital plate ovate, apex slightly flattened, sometimes slightly excised, extend- ing beyond the bases of the sub-anal plates. ?. Sides of segments eight and nine brownish, dorsum membranous ; tenth tergite brownish, margin produced and rounded. Subgenital plate lightly chitinized, about the width of the segment and scarcely produced, its apex shallowly excised. Sub-anal plates triangular, apices rounded, not produced in long blades. Cerci brownish, about twice as long as the abdomen. Length of fore wing, $ 7 mm., ? 7-75 mm. Length of hind wing, < 6 mm., $ 6-5 mm. AUSTRALIA: Federal Capital Territory, Lee's Springs, xi.i932, 6 <, i $, Murrum- bidgee River, 1.1922, i < (R. J. Tillyard) ; New South Wales, Bathurst, 2,300 ft., io.xi.i884, i < (from McLachlan collection). Type male and one paratype male (from Lee's Springs) as microscope preparations, Bathurst male pinned, remainder in 2 per cent, formaldehyde solution. The arrangement of the outer cross- veins of the fore wing in steps, anterior cross- veins nearer apex, seems reasonably constant, as does the position of separation of Rs and M in hind wing. The type and female paratype are rather paler than the others, probably due to the degree of maturity. Leptoperla reticulata sp.n. FIG. 9 c unknown. $ (in fluid). Head yellowish-brown, marked with dark brown between the ocelli and on the occiput. Antennae brown, about 7 mm. long, a little shorter than the anterior wing. Pronotum dark fuscous, a little broader than long, narrowed anteriorly ; a narrow, impressed, piceous median line linking similar anterior and posterior transverse lines. The latter do not extend more than half the width of the pronotum. Meso- and metanota shining piceous brown. Legs yellowish -brown, marked with dark brown at apices of femora, extending along posterior surface, bases and apices of tibiae and on the tarsi. Abdomen whitish above, except the tenth and sides of seventh to ninth tergites, which are fuscous, and the pale fuscous sternites of the first seven segments. Subgenital plate and centre of ninth sternite brown. Sub-anal plates light fuscous, margined exteriorly with dark brown. Wings rather short, pale greyish, with fuscous venation ; membrane slightly clouded with brownish in costal area, around the pterostigma and bordering some of the cross- veins. In the fore wing, the most noticeable feature is the arrangement of the cross-veins in the apical half, where they tend to form an irregular network between the main veins. Cross-veins in apical half of hind wing normal. 58 A REVISION OF THE AUSTRALIAN AND TASMANIAN Subgenital plate broad, only slightly produced ; margin with a shallow excision, strongly pigmented. Ninth sternite with a small dark brown lobe at its centre, separated from the remainder of the ster- nite by notches in the apical margin. Sub- anal plates broad, apices stout, blunt and divergent. Cerci about 5 mm. long, twenty- six-segmented, yellowish-brown, monili- form at base, becoming filiform. Tenth tergite triangularly produced to a small rounded apex. Length of fore wing, 8-5 mm., of hind wing, 7-25 mm. AUSTRALIA: New South Wales, Mt. Kosciusko, Spencer's Creek, xii.1932, i $ (R. J. Tillyard). Type in form of microscope preparation. This species should be easily separable from other described species by the irregu- FIG. 9. Leptoperlareticulata sp.n. ?. a, wings, lar reticulation of the fore wing, and by b, genitalia, ventral. the lobe of the ninth sternite. Leptoperla nigrifrons sp.n. FIG. 10 cJ (in fluid). Head dark brown, pitchy-black between epicranial suture and labrum ; antennae brown, basal segment blackish above. Pronotum brown, with blackish rugosities, median line, anterior and posterior furrows blackish ; quadrate, parallel- sided, about as long as broad, anterior margin slightly convex, anterior angles rounded. Meso- and metanota brown. Legs brown, femora blackish. Wings hyaline, with considerable pale brownish suffusion as in Fig. 10 ; venation brownish, apical cross- veins rather weak. In posterior wing the stalk of Rs+M is rather long, there are two or three cross-veins between Cui and Cu2, and the angle between the posterior margin and the anal fan is rather abrupt. Abdomen and cerci brownish. Ninth tergite somewhat excised and membranous at the centre of its apical margin. Ninth sternite moderately produced in a parallel-sided subgenital plate, whose apex in ventral view is rounded, with a shallow median excision. Tenth tergite small, dorsally almost divided into two lobes by excisions of the apical and basal margins. Supra- anal lobe broad at base, in dorsal view with a finger-like process arising from an elevated trapezoidal base ; below this the lobe is bent downward and tailward. About midway on the upper surface is a small basally-directed hook, and the lobe terminates in a downturned hook. Cerci long, slender, with more than thirty segments (incom- plete), which become progressively longer and more slender towards the apex. The second segment is obliquely truncate, causing the cerci to be angled in dorsal view. Sub-anal plates upcurved in side view, gradually dilating to a truncate apex, with a pointed upper angle. In ventral view they are slender, with out-turned apices. GRIPOPTERYGIDAE AND NEMOURIDAE (PLECOPTERA) 59 Length of fore wing, 8 mm. TASMANIA (J. W. Evans), 2 $. Type male in form of microscope preparation, paratype in 2 per cent, formaldehyde solution. FIG. 10. Leptoperla nigrifrons sp.n. Q\ a, wings, showing pattern of fore wing ; b, genitalia, lateral (left cercus omitted), c, genitalia, dorsal, d, the same, ventral. The venation of the posterior wing suggests that this species be grouped with L. rugosa and L. reticulata, but it may be separated from them by the pattern of the fore wing and the shape of the anal fan. Leptoperla sp. No. i FIGS. 11-13 Nymph (? about half -grown). General colour yellowish-brown, with obscure, slightly darker markings; of rather slender build (about 7 mm. long), legs rather stout. Body sparsely clothed with short hairs, and in some areas very short, stout spinules. Head rather wider than the pronotum, brownish, smooth, developing ocelli visible as black spots beneath the integument. Antennae incomplete, fairly stout, tapering towards apex : first segment large, second smaller, but still larger than any of the next few succeeding ones ; a few scattered setae, but no dorsal fringe. Labrum about twice as broad as long, anterior angles rounded. Labium with glossae a little shorter than paraglossae, about the same width, with an apical tuft of setae ; excision between glossae very narrow. Paraglossae pointed, inner margins straight, outer rounded. Labial palpus three-segmented, terminal segment large, ovate, as long as first and second together. Maxilla well developed : lacinia with three ENTOM. 2, 2 G 6o or four blunt plate-like teeth at its apex and near them, on its inner margin, a row of strong acute setae. Galea about as long as lacinia, tapering to a setose apex. Palpus five-segmented, first, second, and fourth short, second half as long again as the first, third twice as long as the first, fourth equal to the second, and fifth a little longer than first and second. Mandibles strongly chitinized, with a number of acute teeth near apex. Pronotum about as long as broad, slightly narrowed in front, angles rounded, anterior and posterior margins both some- what convex; median impressed line about half as long as pronotum, anterior and posterior impressed lines present, angled towards each other at their centres. Surface of pronotum smooth. Wing pads small, anterior elongate, posterior triangular ; vena- tion typical of the genus. Mesonotum with an irregular row of small spinules along the anterior margin and over base of wing pad. Posterior margins of meso- and metanota produced backward, plate-like, a narrow incision separating their lateral angles from the wing pads. Legs short and strong, yellowish-brown, femora dorsally with short spinules and setae, rather deep ; tibiae a little longer than femora, with short acute spinules on inner surfaces, and a sparse row of fine setae exteriorly. First tarsal segment short, second very short, third about three times as long as first. Abdomen cylindrical, tergites two to nine elevated along the median line, apical margin of each produced in a short spur. This elevated area and the apical margin bears short spinules as well as hairs, sternites bearing hairs only. Margin of tenth tergite triangularly produced. Anal gills dull whitish. Cerci incomplete. Sub-anal plates broad, spatulate. TASMANIA: Scottsdale, Cuckoo Falls Creek, 3i.vii.i93i (R. J. Tillyard). Five examples, probably all males ; it is possible that these nymphs may belong to the species L. varia Kimmins, which occurs in this locality. The somewhat irregular arrangement of the cross-veins in the nymphal fore wing suggests this, but further confirmation is needed. Leptoperla sp. No. 2 FIGS. 12-13 Full-grown male nymph. General colour fuscous of slender build, with rather short legs: length about 6 mm. Body smooth, sparsely clothed with short setae, whose apices are slender and curled. Head about as wide as pronotum, antennae incomplete, moderately stout, tapering. FIG. ii. Leptoperla sp. No. i. Nymph. GRIPOPTERYGIDAE AND NEMOURIDAE (PLECOPTERA) 61 Labrum about twice as broad as long, anterior angles rounded. Labium with glossae a little shorter and narrower than paraglossae, with an apical tuft of setae. Paraglossae less pointed than in sp. No. i. Labial palpus rather stouter than in sp. No. i ; maxillae with palpus rather stouter, mandibles similar. FIG. 12. Leptoperla spp. Nymphal venation, a, sp. No. i ; b, sp. No. 2. FIG. 13. Leptoperla spp. Nymphal structures, a-g, sp. No. i ; h-k, sp. No. 2. a, mandibles; b, labrum ; c, maxilla ; d, labium ; e, leg ; /, dorsal armature of abdominal tergites, lateral ; g, the same, dorsal ; h, mandibles ; i, labrum ; _;', maxilla ; k, labium. Pronotum broader than long, 'somewhat narrowed anteriorly, angles rounded, margins straight. Strong median and transverse impressed lines, anterior shorter than posterior, both angled towards each other at their centres. Wing pads of medium size, anterior elongate, overlapping basal two-thirds of posterior; fuscous, with longitudinal veins slightly darker. Imaginal venation less distinct than in some species, few cross-veins in apical half of wings. No incision between anal angle of wings and nota. Legs paler than body, similar to sp. No. i, but with rather fewer short setae. Abdomen long, cylindrical, somewhat ridged along dorsum, apical margins not 62 A REVISION OF THE AUSTRALIAN AND TASMANIAN produced. Tenth tergite much elevated in a median dorsal ridge, terminating in an acute finger, projecting beyond the triangularly produced margin. Basal margin with a pair of acute triangular apodemes. Anal gills pale pinkish, cerci incomplete. Sub-anal plates about twice as long as broad, parallel-sided for most of their length, tapering near apex. TASMANIA: Scottsdale, a number of male nymphs. Smaller than sp. No. I, rather more slender; pronotum is definitely broader in proportion to its length, and tenth tergite is more produced ; sub-anal plates narrower. DINOTOPERLA Tillyard Dinotoperla Tillyard, 1921, Canad. Ent. 53: 43, fig. 4&. Type species: Perla opposita Walker, 1852 (fixed by Tillyard, 1921). Tillyard, 1921, Trans. Roy. Soc. S. Aust. 45: 270-273, figs. 2-3. Fore wing with Rs simple, M forked from near the middle of wing, GUI forked ; a thickened cross-vein between IA and 2A; cubito-anal space with a basal cross-vein only. Hind wing with Rs simple; M and Rs separating some distance from base; M forked, posterior branch fused for a space with Cui, separating again before the margin of the wing ; generally only one cross- vein between the branches of cubitus. Anal fan rather narrow. Cerci shorter than abdomen. Distribution'. Australia, Tasmania. Dinotoperla serricauda sp.n. FIG. 14 Head, thorax, and apex of abdomen brown, segments one to eight of abdomen whitish. Eyes blackish, antennae brown, about as long as fore wing, basal segment enlarged. Pronotum almost rectangular, broader than long, slightly narrowed anteriorly. Legs brownish, slightly darker at the knees. Wings pale fuscous, with darker venation ; in anterior, R, Rs, and most cross- veins broadly shaded with darker fuscous. Veins in hind wing not shaded. <$. Apical margin of tenth tergite produced at its centre in a pale process, triangular and acute from the side ; from above broad and pigmented at its base, tapering to a narrow, rounded apex, upper surface closely set with minute setae. Supra-anal lobe projecting beyond apex of tergite, slender and blade-like from above, apex abruptly hooked downwards. Cerci brownish, about i J mm. long, eleven-segmented, tapering slightly, segments near base shorter than broad, slightly serrately produced on inner surfaces. Sub-anal plates scimitar-like, apices somewhat incurved; from beneath, they are seen to arise from broad bases and are contiguous for most of their length. Margin of ninth sternite produced in a parabolic subgenital plate about as long as its sternite, from side appearing as a stout, blunt finger. ?. Wings more elongate. Subgenital plate of eighth sternite very short and broad, scarcely projecting beyond the margin. Sub-anal plates triangular, from beneath apices produced in slender fingers : near the upper basal angle is a small, rounded, whitish pit. Cerci less serrate than in male, twelve-segmented. Margin of tergite produced, rounded and bent down, whitish, with minute setae as in male. Length of fore wing, <$ 9-5 mm., $ n mm. GRIPOPTERYGIDAE AND NEMOURIDAE (PLECOPTERA) 63 Length of hind wing, N.Z.: 119, pi. 10, fig. 14. The type and allotype have suffered severely from the depredations of Anthrenus, little remaining apart from the wings. I am therefore quoting the original description and adding extra notes where possible. '$ Holotype: Length of body 20 mm., of forewing 25 mm., antennae 17 mm., cerci barely 3 mm. Body dark fuscous, marked with dull brown; antennae dull brownish with slight fuscous annulations; legs dull brownish varied with dark fuscous. Wings pale greyish with dark fuscous veins, those of the basal half of the hind wing inclining to brownish ' (the whole of the fore wing and to a lesser degree, the distal part of the hind wing, speckled with dark brown spots, and most of the cross-veins margined with darker brown. Posterior margin of fore wing with a small distinct angular lobe between Cui and Cu2, rather more pronounced than in Till- yard 's figure). 'Forewing with a complete series of costal veinlets; hindwings with hm and three or four distal costal veinlets only. ' (Costal margin of hind wing very ENTOM. 2, 2 I 7 6 A REVISION OF THE AUSTRALIAN AND TASMANIAN slightly scalloped.) 'Thickened cross-vein between IA and 2A in forewing blackish, very prominent.' ' c Allotype. Smaller and slightly paler and less strongly irrorated than female. Length of body 17 mm., forewing 21 mm. Supra-anal plate with a slender copulatory process directed upwards and ending in a small hook directed posteriad; paraprocts upcurved, forming two flatly rounded lobes directed forward and upward. Tenth tergite with a raised flap distally.' AUSTRALIA: New South Wales, Mt. Kos- ciusko, 5,000-5,500 ft., 24.xi.ig2i (R. J. Tilly ard). Holotype female, allotype male in British Museum (Nat. Hist.), from the Till- yard collection. FIG. 25. Trinotoperla irrorata Tillyard. Type. Wings. Trinotoperla australis Tillyard FIGS. 26-28 Trinotoperla australis Tillyard, 1924, Trans. Roy. Soc. S. Aust. 48: 194; 1926, Ins. Aust. &> N.Z.: 119, pi. 10, fig. 15. Female holotype : Head, thorax, and abdomen fuscous ; antennae fuscous, three- quarters as long as fore wing. Pronotum a little broader than long, quadrate, anterior FIG. 26. Trinotoperla australis Tillyard. genitalia. a, lateral, b, apex of supra-anal lobe, lateral, c, dorsal, d, subgenital plate, ventral. FIG. 27. Trinotoperla spp. $ Wings. a, T. australis Tillyard ; b, T. nivata sp.n. GRIPOPTERYGIDAE AND NEMOURIDAE (PLECOPTERA) 77 and posterior margins slightly rounded, a median impressed line, not reaching anterior margin, and posteriorly ending in a transverse groove. Legs fuscous. Wings pale brownish-fuscous, venation darker, cross-veins narrowly bordered. Very few costal cross-veins, no pronounced angular lobe on posterior margin of fore wing. Costal margin of hind wing evenly arched. Thickened cross-vein between IA and 2A in fore wing brownish, less conspicuous than in irrorata. Subgenital plate of the female holotype is creased and probably distorted. The apical margin is slightly produced and rounded, and appears to have a small rounded excision at its centre. I strongly suspect that this excision is accidental, as I have examined a number of examples from South Australia, agreeing in other respects, in which the subgenital plate is evenly rounded and not excised. I am figuring the sub- genital plate of one of these for comparison. In most of these there is a strongly pig- mented bar near each lateral margin. Sub-anal plates broad at their bases, with rather broad rounded apices, which are divergent. Cerci filiform, brownish, incomplete in the type, in other examples about fifteen-segmen- ted. Apical margin of tenth tergite deflexed, rounded, with a slight pro- jection at its centre. Male. Tenth tergite produced in a pale triangular process, rather flattened and acute from the side. Supra-anal lobe long, slender from above; from the side rather deep and somewhat dilated at its apex. Just before the apex on the ventral surface is a small acute excision, giving the apex the appearance of a blunt hook. Basad of this excision are a few minute teeth. Sub-anal plates long, slender, and a FIG. 28. Trinotoperla spp. $. a-b, T. australis Tillyard ; c, T. nivata sp.n. a, genitalia ventral of holotype ; b, subgenital plate of another example ; c, genitalia ventral. little upcurved from the side, apices incurving and acute. From beneath they are broad at their bases, inner lower surfaces membranous. Cerci short, brownish, curv- ing downwards, basal segment rather constricted from the side, about seventeen segments. Subgenital plate broad, apical margin rounded, and moderately produced. The basal margins are more heavily pigmented than the remainder of the tergite. Length of fore wing, $ 18 mm., $ 18-20 mm. Length of hind wing, $ 16 mm., $ 17-18 mm. AUSTRALIA : The locality of the holotype is New South Wales, Towac, near Mount Canoblas, 7.x.i9i6(R. J. Tillyard). I have seen other specimens from Mt. Kosciusko, Spencer's Creek, xii.1932, and Bolaro, 22.xii.i935, and from the Murrumbidgee River, F.C.T., 1.1922, all collected by R. J. Tillyard, and other pinned examples from the McLachlan Collection, labelled Melbourne, S. Australia, or merely Australia. This species bears a superficial resemblance to the new species nivata and minor, but 78 A REVISION OF THE AUSTRALIAN AND TASMANIAN may be distinguished by the male genitalia, and by the rounded sub-anal plates of the female. The wings are rather narrower. Trinotoperla nivata sp.n. FIGS. 27, 28 <$ unknown. $ (in fluid). Head reddish-brown, with a small area on the clypeus, and the post- ocellar sutures pale. Antennal brown, long (incomplete in the type). Thorax fuscous, pronotum a little broader than long, slightly narrowed anteriorly, angles not pro- duced ; a short, median, impressed line, all margins somewhat elevated. Legs medium fuscous, knees darker, a pale spot in centre of anterior surface of each femur. Lower anterior apical angle of each femur produced in a small tooth. Abdomen whitish, except tenth tergite, eighth to tenth sternites, sub-anal plates and cerci, which are brownish. Wings greyish-brown, not irrorated, anterior darker than posterior ; veins darker, faintly shaded. A number of costal cross- veins in fore wing ; posterior margin of this wing scarcely produced in a lobe about the end of Cui. Angle between anal fan and posterior margin of hind wing clearly greater than a right angle. Subgenital plate about twice as broad as long, apical margin not very produced, broadly excised at its centre ; brown, with a quadrate area at the centre of the apical margin, and the lateral margins pale. Sub-anal plates broad at bases, apices produced in acute fingers, curving upward and slightly inward. Cerci short, about 1-7 mm., sixteen or seventeen segments, basal segments much broader than long. Length of fore wing, 16 mm., of hind wing, 15 mm. AUSTRALIA: Victoria, Snowy River, 3.1.1933, i ? (R. J. Tillyard). Type is a microscope preparation. T. irrorata Tillyard, which this species somewhat resembles, may be separated by its larger size, distinctly mottled fore wing, a small but distinct rounded lobe on the posterior margin of fore wing, and by the angle between the anal fan and the posterior margin of the hind wing being only a little greater than a right angle. The female holotype of irrorata suffered greatly from the ravages of Anthrenus whilst in Till- yard 's possession, and little now remains of the body. The sub-anal plate appears to be of the pattern of nivata, but the apex is less acute. Trinotoperla minor sp.n. FIG. 29 (In fluid.) General appearance much as in T. nivata. Legs with the pale area of the femur less pronounced, upper surface darker, lower apical angle toothed. Abdomen whitish, except eighth to tenth segments in male, and tenth tergite, eighth to tenth sternites in female, which are brownish. Wings rather more elongate than in nivata, a little paler, and without costal cross-veins, apart from the humeral. In the hind wing there may be a cross-vein between IA and 2A near their apices. The angle formed by the posterior margin and the anal fan is more obtuse. (. Tenth tergite from above produced in a rather narrow, membranous lobe or flap, rather flattened from the side. Supra-anal lobe slender, somewhat dilated before GRIPOPTERYGIDAE AND NEMOURIDAE (PLECOPTERA) 79 its apex, which curves downward in an acute spine. Cerci pale brownish, curving gently downward, seventeen-segmented, segments at the base very short and broad, gradually increasing in length and becoming narrower towards the apex. Sub-anal plates rather broad, especially from beneath, curving upward on either side of the supra-anal lobe, and with rounded apices. Subgenital plate strongly produced, triangular, with a truncate apex. $. Very like T. nivata. Subgenital plate without a pale quadrate area at the centre of the apical margin. In nivata this pale area seems to be more closely set with fine FIG. 29. Trinotoperla minor sp.n. a, wings <$ ; b-d, genitalia <$, b, dorso-lateral, c, dorsal, d, ventral; e, genitalia $, ventral. hairs than its surroundings, but in minor the clothing is uniform. Apices of the sub- anal plates a little more incurved, and slightly longer in comparison with their bases. Cerci similar, seventeen-segmented, about 1-6 mm. long. Length of fore wing, ^ n mm., $ 12 mm. Length of hind wing, $ 10 mm., $ n mm. AUSTRALIA: New South Wales, Bolaro, xii.1935, 2 $, 3 $ (R. J. Tillyard). Type male, paratype female as microscope preparations, remainder in 2 per cent, formaldehyde solution. Trinotoperla minor nymph FIGS. 30-32 (Mature nymph, about to emerge.) General colour brownish, pronotum and legs paler; medium build, legs longer than in Dinotoperla, length of nymph about 9 mm. A conspicuous fringe or crest of fine setae along the median dorsal line of body, and along legs, cerci, and basal segments of antennae. Body clothed with a sparse mixture of fine hairs and short stout ovate setae. Head dark brown, broader than pronotum, 8o A REVISION OF THE AUSTRALIAN AND TASMANIAN widest at compound eyes, slightly rounded behind ; stem of epicranial suture about one-third as long as pronotum. Antennae about as long as body, filiform, tapering ; first segment large, with a dorsal crest of fine setae, second smaller, but longer than any of the next few succeeding segments, which become progressively longer and narrower from base to apex. Labrum short and broad, anterior angles rounded. Labium with glossae slightly shorter than paraglossae, but of about equal width, apices with a dense tuft of fine setae. A narrow incision separating the glossae to their bases, and a wider one divid- ing them from the paraglossae, whose outer margins are convex. Palpus three-segmented, segments of roughly equal width, increasing in length from base to apex in ratio 1:1-5:2. Maxillae well developed, lacinia with a pair of broad truncate teeth at its apex, beneath which on its inner surface is a row of stout setae. Maxillary palpus five-segmented, aris- ing from a membranous palpifer; first and second segments subequal, third somewhat shorter than the first and second together, fourth a shade longer than the second, and the fifth as long as the third. Pronotum pale yellowish-brown, with a short, impressed, dark brown median line, and on either side of it a curved brownish band, its convex surface inwards, and a smaller streak near the posterior angles. Pronotum quadrate, slightly broader than long, angles rounded; posterior transverse furrow more noticeable than anterior. Wing pads extending backwards, almost horizontal. Fore wing pad elongate, cover- ing about half of posterior pad ; the latter is broader, the anal fan extending a full half of the posterior margin. Enclosed wings folded ready for emergence, venation appearing as pale lines on brownish ground of wing cover. (The venation figure has been drawn from a younger nymph, in which the pads are naturally smaller and relatively broader.) Legs moderately short, prothoracic the shortest, each with a dense crest of fine long setae situated on the dorsum of each coxa, trochanter and tarsus, anterior dorsal margin of femur and anterior surface of tibia. Femur shorter than tibia, with a broad, light brown band before the knee, and a faint stripe of the same colour near the fringe, lower apical angle slightly produced. Tibia with a brownish band at the knee and the outer surface narrowly brown. Tarsus brownish, second segment shorter than the first, third about three times as long as first and second together, claws simple. Abdomen cylindrical, brownish, segments one to nine with whitish apical margins, tenth tergite rather longer than eighth and ninth together, the centre of its apical FIG. 30. Trinotoperla minor sp.n. Nymph. GRIPOPTERYGIDAE AND NEMOURIDAE (PLECOPTERA) 81 margin produced in a rounded lobe, which is itself further produced in a slender finger. A dense tuft of whitish anal gill-filaments. Sub-anal plates brown, roughly triangular, apices divergent, rounded, whitish, outer margins broadly excised. Cerci long, with more than forty segments, slender, with a dense dorsal crest of fine, long setae ; each segment ventrally at its apex with a few short setae. AUSTRALIA: New South Wales, Bolaro, 22.xii.i935 (R. J. Tillyard). The material furnishing this description was collected at the same time and place as the adults of T. minor. It consists chiefly of fully matured nymphs, so mature that the wings within the pads are folded ready for emergence. There are also from the same locality two or three smaller nymphs, probably about half-grown, in which the imaginal venation can be discerned, and this venation closely resembles that of T. minor adults. I think there can be little doubt that these are the nymphs of T. minor. Trinotoperla sp. No. i FIGS. 31-32 Last instar nymph : Related to T. minor but differing in details, notably venation of hind wing, form of mouth-parts and vestiture. General colour light brownish, with darker markings, length about 13 mm. Median dorsal fringe less prominent ; body sparsely clothed with a mixture of short curved setae and short dilated hairs. The FIG. 31. Trinotoperla spp. Nymphal wing pads, a, T. minor sp.n. ; b, T. sp. No. i. latter are somewhat excised at their apices and trough-shaped in cross-section ; those of T. minor are relatively larger and cylindrical in section. Head dark brown, paler at centre of vertex, broader than pronotum, stem of epicranial suture between one- third and one-quarter as long as pronotum. Antennae longer than body (about 16 mm.), filiform, tapering, first segment large, with a crest of fine setae, second smaller, but broader and longer than any of the next few succeeding segments, which also carry a dorsal fringe. Terminal segments longer than broad. Labrum short and broad, narrowing to a broadly and shallowly excised apex. Labium with the glossae about the same length but somewhat narrower than the paraglossae, apices more pointed than in T. minor, and armed with a tuft of setae. Outer margin of paraglossae slightly sinuous, so that the apex is more acute than in T. minor. Palpus three-segmented, first about half as long as second, which is a little 82 A REVISION OF THE AUSTRALIAN AND TASMANIAN shorter than third. Maxillae well developed, lacina terminating in a single truncate tooth, and a number of stout setae ; galea much as in T. minor. Maxillary palpus five-segmented, extending about as far as apex of lacinia ; first and second subequal, third twice as long, fourth a little longer than second, fifth about as long as third. Mandibles with a few rather blunt teeth. Pronotum similar to T. minor, but more heavily marked. Lower anterior apical angle of femur not produced. Tibia with an additional narrow brown band on inner anterior surface, parallel to fringe. Abdomen cylindrical, brownish above, tergites FIG. 32. Trinotoperla spp. Nymphal structure, a-g, T, minor; h-wi, T. sp. No. i. a, labrum; b, labium ; c, maxilla ; d, mandibles ; e, leg ; /, portion of cercus ; g, vestiture ; h, labrum ; i, labium ; _;', maxilla; k, mandibles; m, vestiture. one to nine with a whitish apical margin ; sternites light yellowish-brown, with golden pubescence. Tenth tergite longer than eight and nine together, centre of its apical margin produced, obtuse, not produced in a finger as in T. minor. Anal gills whitish. Sub-anal plates much as in minor. Cerci long (10 mm)., slender, about sixty segments, carrying a dorsal fringe as in minor. TASMANIA: Scottsdale, Cuckoo Falls Creek, 3i.vii.i93i (R. J. Tillyard). Four male nymphs, two apparently last instar, and two with very rudimentary wing pads. Up to the time of writing I have not seen any adult Trinotoperla from Tasmania, but from wing-venation these nymphs undoubtedly belong to that genus. The presence of cross-veins in the anal fan suggests Eunotoperla, but the venation of the fore wing is different. The practice of describing new species of Plecopters from the nymphs alone is undesirable, and in consequence I am giving these nymphs numbers instead of names. I have also seen single examples of two more species of Trinotoperla from the same locality, but as material is so scanty I propose to do nothing further with them at present. GRIPOPTERYGIDAE AND NEMOURIDAE (PLECOPTERA) 83 Trinotoperla sp. No. 2 FIGS. 33, 34 Mature nymph: General colour light yellowish-brown, male about 9 mm., female about 12 mm. long ; medium to rather stout build, legs long and sprawling ; clothed throughout with short curved setae, which arise from globular tubercles: setae widened at their bases, apices very slender. Head flattened, only slightly wider than Vc^-~~^y FIG. 33. Trinotoperla sp. No. 2. a, nymphal wing pads; b, tracheal venation. pronotum ; antennae about as long, or a little longer than body, filiform, tapering, a small dorsal fringe at base. Labrum bent downwards, concealed under the head, apex directed caudad ; short and broad, apical margin slightly emarginate, angles rounded. Labium with the glossae slightly shorter and narrower than the paraglossae. Palpus short and stout, not reaching apex of paraglossae, apical segment about twice as long as basal, second rather shorter than third. Maxilla strong, lacinia with about three acute teeth at its apex, and a number of bristles. Galea tapering, rather hairy. Maxillary palpus rather short and stout, first segment half as long as second, which is two-thirds as long as third. Fourth as long as second, fifth as long as third. Mandibles strong, with numerous acute teeth. Pronotum quadrate, side margins parallel or slightly convergent anteriorly, anterior and posterior margins gently rounded, angles pro- nounced; a slight median line and a strong transverse posterior furrow. Wing pads large, anterior elongate, posterior broader, yellowish- brown, venation indicated by rows of brownish setae, arising from globular bases. In the only large male nymph the tracheal venation is quite evident as well as the imaginal venation, and in the hind wing the fusions of Rs and M, and M3+4 and GUI are clearly shown, confirming Tilly ard's study of ENTOM. 2, 2 K FIG. 34. Trinotoperla sp. No. 2. Nymphal structures, a, labrum ; b, labium; c, maxilla; d, mandibles; e, leg ; /, vestiture. 84 A REVISION OF THE AUSTRALIAN AND TASMANIAN Dinotoperla. Cross- veins in the apical half of the wing dense, including several in the pterostigmatic area. Legs rather long and sprawling, in the male giving a somewhat Ecdyonurid appearance. Trochanters, particularly of anterior legs, with an acute spur at lower apical angle. Femora strongly carinate dorsally, all legs with a dense dorsal fringe. Abdomen cylindrical, in male shorter than hind femur, in female longer. In male, segments one to nine very short, tenth tergite as long as tergites three to nine, apical margin produced in an acute, somewhat raised point. Basal margins of tergites nine and ten each with a pair of stout finger-like apodemes. Cerci long, tapering, with at least thirty-six segments, bearing a dorsal fringe. Sub-anal plates moderately long, broad at base, tapering near apex, which is rounded. Anal gills pale mauve. In female, segments one to nine longer than in male, tenth tergite about as long as tergites seven to nine, apical margin evenly rounded. Apodemes of tergites nine and ten more noticeable. Cerci similar to male, sub-anal plates rather shorter and blunter. AUSTRALIA: New South Wales, Bolaro, 22.xii.i935, 3 ; Rule's Point, 4,450 ft., 30.xii.i934, i & 2 ? (R. J. Tillyard). The rather sprawling appearance of the nymph, the backwardly deflected labrum, and the acute spine on the anterior trochanter separate this species from any other Trinotoperla nymph which I have seen. Fam. NEMOURIDAE Klapalek 1909 This family, which occurs so abundantly in the Northern Hemisphere, is represented in the Southern by about eight genera, comprising at present comparatively few species. They differ to some extent from the typical Northern Nemouridae, particu- larly in the venation, the fore wing resembling more that of the family Leuctridae. They are nevertheless true Nemouridae, as in the hind wing the media is clearly forked and there are five veins on the anal fan. In the Leuctridae the media is apparently simple, Cui being apparently forked, and there are only three veins on the anal fan. It is probable that in the latter family the cross-vein from the media to the cubitus is the branch M3-J-4, which then fuses with Cui and subsequently diverges again to form the apparent fork of Cui. In the hind wing, Rs and M are united at base, whereas in Leuctridae M arises from a cross- vein uniting Rs and Cu. In view of the differences in venation, it seems desirable to place the Nemouridae of the Southern Hemisphere in a separate subfamily, characterized as follows : NOTONEMOURINAE Ricker 1950' Fore wing with Cu2 long, generally extending well into the apical half of the wing ; cross- vein in the pterostigmatic area of both wings either absent or present. Hind wing with the media forking at or before the radio-medial cross- vein. The subfamily will include the following genera : Aphanicerca Tillyard S. Africa. Aphanicercella Tillyard ,, Aphanicercopsis Barnard ,, 1 See Appendix, p. 92. GRIPOPTERYGIDAE AND NEMOURIDAE (PLECOPTERA) 85 Desmonemoura Tilly ard S. Africa Neonemura Navas Chile. Notonemoura Tillyard New Zealand. Spaniocerca Tillyard ,, ,, Australia, Tasmania. Spaniocercoides Kimmins ,, ,, Udamocercia Enderlein S. America, Tasmania. Neonemura Navas is probably a synonym of Udamocercia Enderlein. It is described from a female, which possesses an ovipositor similar to that of Udamocercia. Nemura pirioni Navas, from Chile, should also be transferred to the Notonemourinae on the evidence of the long Cu2 figured in the fore wing ; it is probably a Udamocercia. The venation of the Nemouridae as a whole offers few characters of value for generic separation. It is comparatively simple, though not primitive in type, and tends to be variable in the position and number of the cross- veins. The males of the three species of Spaniocerca from Australia and Tasmania which I have examined, whilst obviously related to each other, differ somewhat in genitalic pattern from the New Zealand species ; had the females shown similar affinities, it might have been desirable to separate generically the Australian and Tasmanian species from those occurring in New Zealand. Unfortunately the females differ in genitalic pattern not only from each other but also from S. zelandica Tillyard. Pending the acquisition of further material from these regions, it seems preferable to retain the Australian and Tasmanian forms in Spaniocerca and to extend its generic diagnosis to accommodate them, rather than to make a new genus based on male genitalic characters only. The two remaining Nemourid species with which I am dealing in this paper are placed for convenience into the genus Udamocercia Enderlein. The male genitalia of one of the species (U. albomacula] are almost identical in pattern with those of two South American species of this genus. The other species shows a striking develop- ment of the supra-anal lobe, and the cerci are modified, but the general pattern is similar. Both species possess a cross-vein in the pterostigma of both wings (contrary to Enderlein's diagnosis) , but this is probably variable as in both the South American species which I have examined it is present, although sometimes obscure. In view of the scanty material available two males of different species, and a female which possibly belongs here it seems unwise to erect a new genus for these Tasmanian species now. SPANIOCERCA Tillyard Spaniocerca Tillyard, 1923, Trans. N.Z. Inst. 54: 216, fig. ; Kimmins, 1938, Ann. Mag. Nat. Hist. (ii) 2: 572. Type species: Spaniocerca zelandica Tillyard, 1923 (Monotypic genus). levised generic diagnosis Sc arching up so as to meet or almost meet costa ; cross- vein M Cu in fore wing ending above on M/3+4 at an angle to it ; the latter vein generally curved as it leaves the transverse cord. Pterostigma in both wings without, or with only one cross- vein. Anal fan moderately narrow, somewhat excised at apex of fourth anal vein. <$. Ninth tergite with apical margin generally produced in a pair of spines or lobes. ENTOM. 2, 2 K 2 86 A REVISION OF THE AUSTRALIAN AND TASMANIAN Ninth sternite more or less produced apically as a subgenital plate, with a median ventral lobe. Titillators of varying length, two-branched (New Zealand) or unbranched (Australia, Tasmania), arising near the apex of the ninth sternite. Cerci simple, sometimes angled. Supra-anal lobe in form of upturned hook. $. Margin of eighth sternite simple, or produced in a subgenital plate. Ninth sternite sometimes produced. Spaniocerca tasmanica Tillyard FIGS. 35-36 Spaniocerca tasmanica Tillyard, 1924, Trans. Roy. Soc. S. Aust. 48: 195. fi g- 3- The holotype female has been reduced to two anterior and one posterior wings by the action of Anthrenus. The following descriptions have been made from a male and a female in fluid. FIG. 35. Spaniocerca tasmanica Till- FIG. 36. Spaniocerca tasmanica Tillyard. Genitalia <$, a, yard. Wings. lateral; b, dorsal; c, ventral; d, genitalia $, ventral. <$. Head dark brown, occiput orange-brown ; antennae piceous, finely annulated with yellowish at sutures. Pronotum dark brown, quadrate, a little broader than long, anterior transverse furrow deep. Meso- and metanota shining piceous. Legs yellowish-brown to dark brown, femora with a conspicuous median ring of yellowish. Wings shining smoky-brownish, pterostigma darker, a few inconspicuous paler areas, especially distad of transverse cord and along cubitus of anterior wing. Venation brown ; generally no cross- vein in pterostigma of either wing, or if present, indistinct ; generally two medio-cubital cross-veins between M5 and the transverse cord. Abdomen whitish except segments nine and ten, which are brown. Ninth tergite produced on either side of centre in a flattened rounded lobe ; between the lobes is a transparent membranous area. Tenth tergite also with a membranous area at centre of apical margin, supra-anal lobe in form of a short broad hook, upcurved from a broad base. Cerci short, stout, lightly pigmented, clothed with short hairs. Ninth sternite with a strong ventral lobe, from the side dilating triangularly to a truncate apex ; from beneath quadrate, slightly constricted at its base, apex faintly excised. GRIPOPTERYGIDAE AND NEMOURIDAE (PLECOPTERA) 87 From the side, the sternite is produced, its apex bent sharply upward and dilated. From beneath it is slightly narrowed about midway, with an elliptical apex. From the upper apical margin arises a pair of small, scythe-like hooks, probably reduced titillators. $. Similar, larger. Wings with more definite hyaline areas. Abdomen whitish, seventh sternite, eighth and ninth segments brownish. Neither seventh nor eighth sternites produced, margins straight, both with hyaline areas at base. Sub-anal plates triangular, apices rounded, upper margins shallowly excised. Cerci short and stout. Length of fore wing, <$ 7 mm., $ 9-4 mm. Length of hind wing, $ 5-3 mm., $ 8 mm. The holotype was taken at Mt. Wellington, TASMANIA, 31.1.1917 (R. J. Tillyard). Other Tasmanian examples in the British Museum from Sheffield, 8.1.1917 (R. J. Tillyard), Hobart, I5.x.i9i6 (C. E. Cole), 4.xi.i9i6 (G. H. Hardy), Penstock Lagoon, 29.1.1933, and Gouldt County, io.ii.i933 (R. J. Tillyard). From the mainland of AUSTRALIA I have seen examples from New South Wales, Mt. Kosciusko, 24.xi.i92i, 3-9. xii. 1921 ; South Australia, Cudlee Creek, 2o.xi.i934 (R. J. Tillyard), and Adelaide. Spaniocerca tillyardi sp.n. FIG. 37 General appearance similar to S. tasmanica ; there appear to be two forms of this species, one of which is smaller and paler than the other. cJ. Head dark brown, occiput a little paler; antennae piceous, with fine pale annulations at sutures. Pronotum dark brown, quadrate, a little broader than long, anterior margin somewhat convex, anterior transverse furrow deep. Meso- and metanota shining dark brown. Legs as in tasmanica. Wings rather narrower, shining smoky-brownish, with a hyaline area just beyond the transverse cord, and a larger one in the neighbourhood of Cui. Venation dark brown, no cross- vein in pterostigma of either wing; one or two medio-cubital cross-veins between M5 and the trans- verse cord. Abdominal segments pale or light brownish, apical margins whitish, segments eight to ten darker. Ninth tergite produced on either side in a flattened lobe, apices approaching one another calliper- wise. Between them the margin is excised and membranous. Tenth tergite also with a membranous area at the centre of its apical margin, giving the appearance of an excision. Supra-anal lobe broad and triangular at its base, a membranous area at its base separating two rounded hairy elevations. From the apex arises a slender hook, in side view with a rounded excision on dorsal surface near base. Cerci longer than in tasmanica, from the side angled upwards in the form of a foot. Ninth sternite long, more slender than in tasmanica, not dilated at its apex, which terminates in a pair of short incurving fingers. From its upper surface arises a pair of flattened titillators, curving upwards and basally, then downwards and diverging apically. These titillators are broad at their bases, apical two-thirds narrow and blade-like. Ventral lobe much narrower in ventral aspect, slightly con- stricted in its apical half. From the side, it is parallel-sided, apex truncate. 88 A REVISION OF THE AUSTRALIAN AND TASMANIAN $. Similar, except that abdominal segments one to seven are whitish. Eighth sternite slightly produced, centre of apical margin still further produced in a small, triangular lobe ; this lobe is more pigmented than the remainder of the margin, and is connected to the main part of the segment by a narrow band of pigmentation. Ninth sternite strongly produced apically in an acute triangle, a narrow triangular membranous area at its centre from base to apex. In side view, this sternite resembles FIG. 37. Spaniocerca tillyardi sp.n. Genitalia <$, a, lateral ; b, dorsal; c, ventral; d, genitalia $, lateral; e, ventral. the valves of a short ovipositor. Cerci short, sub-anal plates subtriangular, apices rounded. Length of fore wing, $ 5-8-6-6 mm., $ 6-3-7 mm - Length of hind wing, $ 4-9-5-8 mm., $ 5-2-6 mm. AUSTRALIA: F.C.T., Lee's Springs, xi.i932 (R. J. Tillyard). Type male, paratype female in form of microscope preparations, remainder in 2 per cent, formaldehyde solution. Spaniocerca bullata sp.n. FIG. 38 Udamocercia albomacula sp.n. i , head, x 10. FIG. 8a. Cerceris flavonasuta $, first two tergites, x 10. FIG. 86. Cerceris flavonasuta $, pygidium, x 10. 8c. Cerceris flavonasuta ?, first two tergites, x 10. of the thorax strongly reticulate-punctate, the punctures on the mesothorax a little larger and deeper than those on the head. Median area of the epinotum with six large transverse costae. Tergites 1-5 not quite dull, with fairly large punctures, the interspaces for the greater part twice as wide as the punctures, the sixth finely punctured and dull, the pygidial area nearly twice as long as wide, the lateral margins parallel over the basal half, convex outwardly on the apical half, the apical margin convex. Sternites 1-6 shining, shallowly and sparsely punctured, the sixth with a yellow fimbria on its hind margin and a triangular tooth on each side. Median area of the clypeus oval, feebly convex, nearly half as long again as wide, the apical margin with three large and rounded teeth. Inner orbits parallel. Vertex wide, the interocular distance there equal to the length of the first five joints of the flagellum. Posterior ocelli twice as far from the eyes as from each other. Second joint of the flagellum a little more than twice as long as wide, two-fifths longer than the third, the apical joint half as long again as wide, transversely truncate at the apex. Shoulders ENTOM. 2. 3 M io 4 SPHECIDAE AND POMPILIDAE (HYMENOPTERA) COLLECTED of the pronotum angular, raised to form a subcarinate tubercle. Mesopleura with a small tooth below. Scutellum depressed lengthwise in the middle. Epinotal dorsum short, about as long as the scutellum. First tergite three-eighths wider behind than long, second tergite seven-ninths wider than long. Hind tibiae with five serrations. Moggio, Ethiopia, April 1948, i $. Not related to any other Ethiopian species. The tuberculate pronotal shoulders and the shape of the outer paramera of the genitalia are distinctive. The outer paramera have on the inside, a little behind the middle of their length, a lamelliform fimbriated lobe, which can be seen only when those parts are widely splayed out. Cerceris insignita sp. n. (FiG. 7 ; PL. 4, FIG. 3) <.ii mm. long. Black. Antennae, occiput, posterior half of the temples, pro- notum, tegulae, scutellum, metanotum, sixth and seventh abdominal segments, and the legs, ferruginous. Mandibles ferruginous, last three joints of the nagelmm blackish above, median area of the clypeus, sides of the face, frontal triangle, and carina, yellowish-white ; the lateral sclerites of the clypeus pale brown and with a fine yellow- ish pubescence, the moustache brown. Tergites 2-5 with a narrow pre-apical pale yellow band which is abruptly dilated laterad, the dilatation almost reaching the base, the sixth with a yellow spot on each side. Sternites fusco-ferruginous. Wings hyaline, the veins brown, the pterostigma ochreous, the apical third of the forewing pale fuscous. Clypeus and yellow areas of the face dull, shallowly, finely and sparsely punctured. The rest of the head, and the thorax, reticulate-punctate and dull, the punctures on the vertex about one-third as wide as the anterior ocelli, the punctures on the mesonotum and scutellum a little larger, and largest on the mesopleura. Triangular area of the epinotum slightly shining, closely and transversely rugose, the middle third of the epinotal declivity shining and almost smooth. Tergites 1-6 dull, sparsely punctured, the punctures as large as those on the vertex. Pygidial area shining, sparsely punctured, nearly parallel-sided but a little wider over the posterior half, the apical margin convex. Sternites smooth and shining, the sixth with a curved tooth on each side below which there is a bundle of ferruginous hairs; seventh sternite arcuately emarginate at the apex, the lateral margins raised or carinate. Median area of the clypeus convex, suboval, half as long again as wide, the apical margin tridentate. Inner orbits parallel. Posterior ocelli twice as far from the eyes as from each other. Interocular distance on the vertex equal to the length of the first five joints of the flagellum. The second joint is twice as long as wide, one-fourth longer than the third, the apical joint truncate. Dorsum of the pronotum short, as long as the first joint of the flagellum, the shoulders subangular. First tergite as long as wide behind, the second two-thirds wider behind than long. Hind tibiae with five serrations. Moggio, Ethiopia, April 1948, i <. The colour pattern of the tergites is unlike that of any other Ethiopian species. Although so different in colour from C. moggionis, it is allied to that species, but the genitalia are different. The outer paramera are elongate triangular over the distal BY MR. K. M. GUICHARD IN W. AFRICA AND ETHIOPIA 105 third and acute at the apex. Another specimen, also from Moggio, represents a variety, opulenta var. n., which differs from the type as follows: c. 12 mm. long. Occiput and temples black, the latter with a small ferruginous macula on top. The yellow areas on the tergites are more extensive, so much so that on the second tergite the black area is reduced to a transverse band at the base and a small triangular pre-apical macula, its base caudad. On the third tergite the black covers only the middle third of the segment, a little more on the fourth and fifth, but on the latter the black merges caudad into ferruginous. The first tergite is a little broader in proportion to its length, and the sculpture on the head and thorax is also a little shallower. Cerceris rufiscutis Cameron, var. decolorata Arnold C. rufiscutis Cameron, 1910, Sjostedt's Kilimandjaro-Meru Exped. 8: 278. Arnold, 1931, Ann. Transv. Mus. 14: 182. Haramaia, April 1948, i $, i$; Lake Bishoftu, Ethiopia, August 1946, I $; Moggio, April 1945, i<; Lekempti, Ethiopia, June 1946, i <. Cerceris flavonasuta sp. n. (Fics. 8, 8a-c ; PL. 4, FIG. 9) $. 12-5 mm. long. Head ferruginous, with a large macula extending from the antennal sockets to a little distance behind the ocellar area, and the anterior half of the temples black. Lamina of the clypeus pale yellow, its apical margin blackish; the median area below it black. Frontal carina pale yellow, mandibles fusco-ferru- ginous, antennae ferruginous, the fifth and following joints of the flagellum fuscous above. Thorax black, the metanotum pale yellow. Tergites 1-4 and the basal half of the fifth black, the first to third with pre-apical yellow bands, narrowly interrupted in the middle and moderately dilated laterad, the fourth with a very thin pre-apical yellow band ; the sternites 1-4 black, the second and third with a large yellow macula on each side, the fifth and the whole of the apical segment ferruginous. Antennae, apices of the femora, tibiae, and tarsi ferruginous. The hind coxae with a pyriform yellow macula on the upper side. Wings hyaline, tinged with dull yellow, the ptero- stigma ochreous, the veins brown, the tegulae ferruginous. Labrum wider than long, with a large tubercle on each side. Median area of the clypeus below the lamina smooth and shining. The lamina is free to its base, as long as wide, convex transversely and strongly so lengthwise, a little wider in front than behind, shining, sparsely punctured and broadly carinate lengthwise in the middle, the apical margin shallowly excised. Lateral sclerites of the clypeus shining and sparsely punctured. Head dull and reticulate-punctate, finely on the face, more strongly on the vertex, with the meshes emphasized longitudinally, and more finely and closely on the occiput and upper part of the temples ; the underside of the head is longitudinally rugose and fairly strongly punctured in between. Thorax dull, pronotum closely and finely punctured, mesonotum and scutellum longitudinally and finely rugose, punctate, the punctures on the middle part of the io6 SPHECIDAE AND POMPILIDAE (HYMENOPTERA) COLLECTED mesonotum largest and not so large as on the upper part of the face. Metanotum finely and sparsely punctured. Mesopleura and epinotum closely rugulose-punctulate, with a shallow larger puncturation superimposed. Triangular area of the epinotum closely and obliquely rugulose, finely punctured between the rugae. First tergite shining, with sparse and large punctures. The second to fifth closely and finely punctured over the basal half, more sparsely with larger punctures apically. Pygidial area widened towards the straight apical margin, and half as long again as wide there. Sternites shining, shallowly and sparsely punctured, the sixth deeply and narrowly excised. Mandibles with a broad rounded tooth at the middle of the inner margin. Interocular distance on the vertex equal to the length of the first six joints of the flagellum, the second joint five-eighths longer than the third. Ocelli half as far again from the eyes as from each other. Temples wide, at their widest clearly wider than the eyes. Pronotal collar short, the shoulders rounded. Mesopleura with a feeble denticle below. First tergite as long as wide, the second three-fourths wider than long. Posterior tibiae with seven spiniferous serrations. c. 9-5 mm. long. Clypeus, frontal triangle and carina and lower two-thirds of the face, excepting a black streak extending from the antennal sockets to the clypeus, pale chrome yellow ; the mandibles ochreous, black at the apex, the rest of the head black. Pronotal collar with a transverse yellow band, narrowly interrupted in the middle. Apical margin of the fifth abdominal segment, and the whole of the sixth and seventh ferruginous. Posterior coxae and trochanters pale yellow on the underside. Eighth to eleventh joints of the flagellum infuscated on the upperside. Otherwise coloured like the ?. The sculpture is everywhere a good deal more coarse than in the ?. Yellow areas of the face and the clypeus dull, shallowly and sparsely punctured, the interspaces about three times wider than the punctures. Thorax dull, the abdomen nearly so. Pronotal collar sparsely and shallowly punctured. Mesonotum and scutellum clearly longitudinally and finely rugose with a shallow and not close puncturation in between the rugae. Mesopleura coarsely reticulate-punctate, the punctures about half as wide as one of the posterior ocelli. Triangular area of the epinotum closely, obliquely, and finely rugose, grooved down the middle ; the rest of the epinotum with a large and deep puncturation. The abdominal segments have a microscopic and close funda- mental puncturation on which is superimposed a larger one, the punctures as large as those on the epinotum, but more widely spaced, the interspaces being for the greater part two and a half times wider than the punctures. Pygidial area slightly widened towards the transverse apex, half as long again as wide, the sides slightly convex. Median area of the clypeus slightly flattened, a little longer than wide, the apex tridentate. Interocular distance on the vertex equal to the length of the first six joints of the flagellum. Apical joint of the flagellum curved, twice as long as wide, the apex transversely truncate. First tergite nearly one-fourth longer than wide. Posterior tibiae with six serrations. Aselle, Ethiopia, July 1947, i ? ; Lekempti, Ethiopia, May 1946, 2 <$<$. This species belongs to the small section B in my key to the species (1931, Ann. BY MR. K. M. GUICHARD IN W. AFRICA AND ETHIOPIA 107 Transv. Mus. 14: 137), which also includes C. andersoni Turner. It differs from that species and the other four species by the shape of the clypeal lamina and pygidial area, and from proteles Brauns, bicolor Smith (PI. 4, fig. 4), and pictiventris Ger- staecker by the colour of the abdomen. Cerceris fulviventris Guerin (PL. 4, FIG. 4) 1844, Icon, regne anivn. 7: 444. C. bicolor Smith, 1856, Cat. Hymen. B.M.: 447. Guiglia, 1948, Ann. Mus. Stor. Nat. Genova, 63: 187. Kaolack, August 1943, i <; Kaffrine, August 1943, i $ and i $. Guiglia's redescription of fulviventris Guerin, based on the type, appears to me sufficient evidence that bicolor Smith should sink in synonymy. The colour of the < in regard to the yellow markings is variable. In the from Kaolack all the tergites have a short yellow streak on each side at the hind margin. It should also be noted that whereas in the ? the epinotal triangle is transversely costate, in the $$ that area is coarsely reticulate-rugose. The $ is easily recognizable by the large semi-elliptical lobe, directed ventrad, on the inner inferior margin of the outer paramera of the genitalia. Cerceris severini Kohl 1913, Rev. zool. afr. 3: 207. Brauns, 1926, Ann. Transv. Mus. 11: 289. Diaforabe, French Sudan, August 1944, i cJ, i ? ; Abbai Gorge, Ethiopia, October 1945, i $ ; Niamey, Niger Colony, August 1944, i $ ; Tillembeya, River Niger, French Sudan, September 1944, 4 cftj. The specimen from Abbai Gorge has the first tergite, excepting the apical margin, black, and third to fifth tergite infuscated in the middle. It should be observed that there is an error in Dr. Brauns's translation of the original description, which runs The fourth (i.e. third) abscissa of the radius is one and a half times longer than the first three (i.e. two) united, which are very short.' Kohl says: 'Der Abschnitt der Radialader, welcher die 3 Cubitalzelle vorne be- grenzt, ist sehr viel kiirzer als das folgende Stuck, welches mehr als i| Mai so lang als wie jenes ist.' Cerceris zavattarii Guiglia (PL. 4, FIG. 5) 1939, R- Accad. Ital. Miss. biol. Borana, 3 (2) : 58. I ascribe to this species, known to me only from the description, a specimen to which I refrain from giving a varietal name, since it differs only in minor details and greater size from one of the paratypes mentioned by Guiglia. cJ. 14 mm. long. The dark areas of the head and thorax are black, without any ferruginous parts. The vertex has two small yellow spots behind the ocelli. The median maculae on the tergites are shorter, not reaching the apical margin, and also more semicircular, that on the first black and on the following fusco-ferruginous. The interocular distance on the vertex is equal to a little more than the length of the io8 SPHECIDAE AND POMPILIDAE (HYMENOPTERA) COLLECTED first four joints of the flagellum. The second joint of the flagellum is fairly slender, or two and three-fifths longer than wide at the apex, the apical joint is short and obliquely truncate. The dense exserted and yellow pilosity on the sternites, including even the eighth, is a distinctive character. Bishoftu, 7,000 ft., Ethiopia, May 1948. This species bears a strong resemblance to C. nugax Arnold and like it has the sternites densely pilose. It can be distinguished, however, by the very different sculpture of the epinotal triangle, by the more slender antennae, the narrower first tergite, and the shape of the outer paramera of the genitalia, which are more acute at the apex and strongly dentate on the inner margin. \ Cerceris rhodesiae Brauns (PL. 4, FIG. 6) 1929, Ann. Transv. Mus. 11: 330. Arnold, 1942, /. Ent. Soc. S. Afr. 5: 20. I reduced this species to the rank of a subspecies of C. monocera Kohl, but on re- examination, and a more careful comparison of the genitalia, find that the latter organs are not identical as I believed. Since the stipites or outer paramera are in most cases curved, it is necessary in comparing them to arrange them in exactly the same plane. As a result, it can be seen that though the genitalia are very similar, in those of rhodesiae the dilatation or lobe on the inner margin of the outer paramera is less abrupt, and the distal portion beyond it is longer and rounder at the apex. Also the differences in the puncturation and in the colour pattern are too great to be considered merely subspecific. Furthermore, another character which I overlooked is to be seen in the radial cell. In rhodesiae it is truncate at the apex, whereas in monocera the apex is bluntly angular and ends on the costal margin. For these reasons rhodesiae must be restored to specific rank. A specimen in this collection represents a subspecies, rhodesiae haramaiae subsp. nov. cJ. ii mm. long. In colour this differs from the species in having pale yellow bands, dilated laterad, on the second to sixth tergites, an elongate macula of the same colour on each side of the epinotal declivity and the scutellum entirely black. The upper sides of the femora are dark brown, almost black on the hind pair. The dorsum of the thorax is moderately shining, the mesonotum more strongly rugose lengthwise, and the punctures between the rugae deeper. The puncturation of the epinotum is larger and deeper and less close, the interspaces smooth and fully as wide as the punctures, whereas in rhodesiae i. sp. the interspaces form raised meshes. The median area of the clypeus is shorter and does not project so far beyond the lateral sclerites, it is one-fourth longer than wide, in rhodesiae very nearly one-half longer than wide. The ocelli are one and three-quarter times farther from the eyes than from each other, in rhodesiae one and a half. The first tergite is wider, or as wide as long, and in rhodesiae one-fourth longer than wide. The genitalia barely differ, the outer paramera being a little less widely dilated anteriorly and slightly rounded at the apex. Haramaia, Ethiopia, May 1948, I $. BY MR. K. M. GUICHARD IN W. AFRICA AND ETHIOPIA 109 Cerceris campsomeroides sp. n. (PL. 4, FIG. 7) <. 14 mm. long. Black. Clypeus, excepting a wide anterior margin, sides of face, frontal triangle and carina, wide apical bands moderately dilated laterad on the second, third and fourth tergites and an apical band covering only the middle third on the fifth tergite, chrome yellow. Fore and middle femora black, their apical fourth, like the tibiae and tarsi dull yellow, the hind legs also of that colour, but the femora more or less stained reddish on the apical half. Apical joint of the flagellum and ochreous below. Wing hyaline, the veins reddish ochreous, the pterostigma pale yellow, the apical margin of both wings faintly fuscous. Head and thorax with a long, erect, yellowish-grey pilosity, the abdomen with a shorter, oblique and sparser one, on the sternites somewhat denser on the apical margins. Yellow area on the head dull, longitudinally rugulose and finely and fairly closely punctured. The upper part of the face, excepting a small smooth area below the anterior ocellus, closely punctured and rugulose, the vertex and upper part of the temples less closely, with larger punctures which are about one-third as wide as one of the posterior ocelli, the interspaces a little wider than the punctures. Lower half of the temples strigose lenghtwise and closely punctured, the punctures smaller than on the vertex. Sides of the pronotum coarsely wrinkled, the mesopleura obliquely costate. The rest of the thorax strongly punctured, the largest punctures, on the scutellum and metanotum, are half as wide as one of the posterior ocelli, the inter- spaces wider than the punctures. Those on the pro- and mesonotum slightly smaller, and a little closer, on the epinotum equally large but closer, or almost reticulate- punctate. The triangular area on epinotal dorsum is dull, obliquely rugose and punctured, its lateral margins not impressed and therefore rather indistinct. First tergite with a puncturation like that of the epinotum, the second to fifth more strongly punctured on the apical than on the basal half, the interspaces a little wider than the punctures ; the sixth and seventh tergites sparsely and coarsely punctured, the pygi- dial area a little narrower at the base than at the apex and fully twice as long as wide at the base, the apical margin straight. Tumid areas of the sternites with a shallow and fairly large puncturation. The whole body fairly dull. Median area of the clypeus fairly flat, one-seventh longer than wide, the apical margin strongly tridentate. Temples a little wider than the eyes. Posterior ocelli very little farther from the eyes than from each other, the interocular distance on the vertex equal to the length of the first five joints of the flagellum. The apical joint of the flagellum curved, excavated below, the two preceding joints also concave below and feebly carinate lengthwise. Dorsum of the pronotum sloping cephalad, as long in the middle as the third joint of the flagellum, the shoulders rounded. First tergite one-fourth wider behind than long, the second a little shorter than the first and fully twice as wide behind as long. Hind tibiae with seven spiniferous serrations. Second abscissa of the radius two- thirds longer than the first. Accra, Gold Coast, 1941, i <$. Not closely related to any other Ethiopian species. In the colour and narrow no SPHECIDAE AND POMPILIDAE (HYMENOPTERA) COLLECTED abdomen this insect has a superficial resemblance to the males of some species of the Scoliid genus Campsomeris, e.g. C. clotho Saussure. C. iniqua Kohl 1894, Ann. naturh. Hofmus. Wien, 9: 288. Lekempti, 6,500 ft., Ethiopia, May 1946, i $. Cerceris nobilitata walegae subsp. n. C. nobilitata Cameron, 1905, Trans. S. Afr. Phil. Soc. 15: 216. $. 10-5 mm. long. Larger than nobilitata, and differing from that as follows: Dorsum of the pronotum raised on each side into a transverse pale yellow torus, lOb I2a lOa FIGS. 9 and ga. Cerceris faceta $, head, x 12. FIG. 96. Cerceris faceta, first two tergites, x 12. FIG. gc. Cerceris faceta, pygidium, x 12. FIG. 10. Cerceris electra $, head, Xi2. FIG. loa. Cerceris electra, first two tergites, x 12. FIG. n. Cerceris maia $, pronotum, x 12. FIG. 12. Cerceris alcyone <$, head, x 12. FIG. I2a. Cerceris alcyone, first two tergites, x 12. similar to that in C. osiris Arnold but larger. Metanotum pale yellow, hind margin of mesopleura, the epinotum excepting the middle, and the first tergite and sternite and the second sternite, castaneous red. Second tergite black, the macula at the base yellow, not interrupted in the middle and larger. The apical band on the third tergite and the apical macula on the fifth pale ochreous, instead of yellowish-white, and larger. The puncturation of the abdomen is closer, the interspaces on the second to fifth tergites being, for the greater part, narrower than the punctures (wider in the type of the species). The first tergite is widest behind the middle, instead of at the BY MR. K. M. GUICHARD IN W. AFRICA AND ETHIOPIA in middle. The fore and middle femora are stained with black above, and the hind femora are castaneous red. Otherwise like the type of the species. Gambeila, 700 ft., in the territory of the Walega and Nuer tribes, Ethiopia, February, 1948, I $. The yellowish parts of the head have been discoloured to reddish-yellow by cyanide fumes. In the presence of the tori on the pronotal dorsum this race resembles C. osiris, which is closely allied to nobilitata. Cerceris faceta sp. n. (Flos. 9, ga-c ; PL. 4, FIG. 8) $. 9 mm. long. Black. The following parts are bright lemon-yellow: mandibles excepting the apex, clypeus excepting the median apical margin and the underside of the median cone, frontal carina, scapes, sides of the face, a small spot behind each eye, pronotal collar, tegulae, a round macula on the mesopleura below the wing-base, a round macula on each side of the scutellum, the metanotum, an ovoid macula on the sides of the epinotum, a band on the first tergite covering nearly the apical half, a transverse apical macula on each side of the second tergite, a wide apical band on the third, slightly dilated laterad, a much narrower one on the fourth, the apical half of the fifth, more or less triangular lateral maculae on the second, third, and fourth sternites, and the legs. The hind femora on the upper side, and the hind tibiae on the underside, streaked with black. The flagellum brownish-yellow above, ochreous below, the first two joints blackish. Wings hyaline, the forewing faintly fuscous at the apex, the pterostigma reddish-brown. Clypeus, face, and temples and mesopleura with fairly dense, whitish decumbent pubescence, elsewhere the pubescence is scanty. Yellow areas of the clypeus and face dull, not very closely punctured. Metanotum finely and sparsely punctured, metapleura obliquely rugose. Triangular area of the epinotum smooth and shining, grooved down the middle. The rest of the body is deeply and closely punctured. Second to fourth tergites are reticulate-punctate with punctures half as wide as one of the posterior ocelli, a little smaller but less close on the thorax and fifth tergite, smallest on the face and vertex which are reticulate- punctate. Pygidial area nearly twice as long as wide, the sides feebly convex, the apical margin straight, dull, coarsely punctured except over the apical third, which is rugulose. Sternites sparsely and coarsely punctured on the tumid areas. Mandibles with a low and wide tooth on the inner margin. Median area of the clypeus inflected in the anterior third, its anterior margin with a triangular tooth on each side ; the basal two-thirds of the median area raised ventrad and ending in a short acute cone. Inner orbits feebly divergent above and below. Posterior ocelli half as far again from the eyes as from each other. Vertex wide, the interocular distance there equal to the length of the first seven joints of the flagellum. Pronotal dorsum convex transversely and lengthwise, in the middle as long as the second joint of the flagellum. First tergite a little wider behind than long, the second as long as the first and twice as wide as long. Fifth sternite with an acute tooth on each side near the apical margin. Posterior tibiae with five spiniferous serrations. ENTOM. 2. 3 N H2 SPHECIDAE AND POMPILIDAE (HYMENOPTERA) COLLECTED ; Moggio, 2 , 1948 ; Aselle, i 7'5~9 mm - l n g- Closely related to preceding species, T. griseola, from which it differs as follows. Third to ninth joints of the flagellum dark brown above, reddish- yellow below. Apices of the femora and the tibiae reddish-yellow, like the tarsi. Pubescence on the clypeus and face pale golden. Pubescence and pilosity on the thorax yellowish-grey. Tergites 1-4 with yellowish-grey fasciae covering the apical third of the segments, the basal portion with only a microscopic black pubescence. Sculpture of the epinotum like that of griseola, but stronger, the declivity moder- ately shining. Interocular distance at the base of the eyes two and a third times greater than on the vertex, where it is equal to the length of the first two joints of the flagellum plus one-fourth of the third. Pygidial area more rounded at the apex, covered with pale golden pubescence. Outer paramera of the genitalia with the apical portion more acute, not bluntly angular on the outer margin but evenly convex. Otherwise like griseola. Tillembeya, River Niger, French Sudan, August 1944, 5 <$<$. BY MR. K. M. GUICHARD IN W. AFRICA AND ETHIOPIA 151 Tachytes discrepans sp. n. (FiG. 42) c. 9-5 mm. long. Black. Face and clypeus with silvery pubescence. Thorax with a scanty yellowish-grey pubescence, the spinotum also with a sparse pilosity of the same colour. Wings hyaline, the veins ochreous. Tergites 1-4 with apical fasciae of grey pubescence, the seventh tergite with silvery. Vertex and dorsum of the thorax very finely recticulate-punctate. Sternites not quite dull, very finely punctured, the FIG. 42. Tachytes discrepans, genitalia, x 27. FIG. 43. Tachytes falcigera, genitalia, x 18. FIG. 43. Tachytes falcigera, apical joints of the flagellum, x 18. FIGS. 44 and 44^. Odontosphex bidens <$, head, x 18. FIG. 446. Odontosphex bidens <$, flagellum, X2y. FIG. 44C. Odontosphex bidens <$, wings, x 12. second, third, and fourth with a thin greyish pubescence, more abundant on the apical margins, forming fasciae. Anterior margin of the clypeus produced into a very short lobe over the median fourth. Galea shorter than the scape. Ninth, tenth, and eleventh joints of the flagellum compressed dorso-ventrally on their inner halves and strongly carinate on the inner margin, the second joint twice as long as wide at the apex. Interocular distance at the base of the eyes five-sevenths wider than on the vertex, where it is equal to the length of the first two joints of the flagellum. Anterior meta- tarsi with four white spines on the outer margin. Apical portion of the outer para- mera of the genitalia very thin, lamelliform, twisted, and slightly dilated close to the apex. $. 10 mm. long. Tarsi, excepting the basal joint, fusco-ferruginous. Pubescence ENTOM. 2. 3 S 152 SPHECIDAE AND POMPILIDAE (HYMENOPTERA) COLLECTED and pilosity on the head and thorax yellowish-silvery, on the underside of the fore and middle femora silvery. Tergites 1-4 with a thin greyish pubescence on the basal part, and with broad apical silvery fasciae. Pygidial area with golden pubescence. Second sternite dull, microscopically punctured, the remaining sternites shining, fairly coarsely and very sparsely punc- tured. Interocular distance on the vertex slightly less than the length of the first two joints of the flagellum. The second joint of the latter two and a half times longer than wide, the ninth to eleventh joints not dilated nor carinate. Anterior margin of the clypeus without a median lobe, otherwise like the <$. Kaolack, Senegal, August 1943, i <$, i ?. The (J differs from all the other African species in the shape of the genitalia and, excepting instabilis and the following species, by the ninth to eleventh joints of the flagellum. The $ resembles T. silverlocki Turner and T. disputabilis Turner, from both of which it can be distinguished by the colour of the pubescence on the sixth tergite, the sculpture of the thorax, particularly of the epinotal dorsum, and by the number of spines on the outer margin of the anterior matatarsi. From instabilis Turner $ it may be distinguished by its smaller size and wider vertex, which in instabilis is equal to only the length of the second joint of the flagellum. Tachytes falcigera sp. n. (Fics. 43, 430) (J. 14 mm. long. Black. Wings hyaline, and faintly tinged with yellow, the veins ochreous. Face and clypeus with yellowish-silvery pubescence. Thorax with a rather sparse yellowish-grey pubescence, and on the epinotum with a sparse pilosity of the same colour. Tergites 1-4 with apical fasciae of silvery pubescence, the seventh with yellowish silvery. Vertex, mesonotum, and scutellum very closely and finely punctured, the following sternites shining, sparsely and coarsely punctured. Galea a little more than half as long as the scapes. Anterior margin of the clypeus flattened. Interocular distance at the base of the eyes twice as great as on the vertex, where it is equal to the length of the second joint of the flagellum. The second joint is two and a third times longer than wide at the apex, and the ninth to eleventh joints are compressed on their inner halves and carinate, as in the preceding species. Dorsum of the epinotum a little shorter than the scutellum and metanotum united. Anterior metatarsi with five white spines on the outer margin. Outer paramera of the genitalia abruptly attenuated apically, the attenuated part sickle-shaped and slightly widened at the apex, the inner rod-shaped paramera emit two long and thin hairs at the apex. Aleg, S. Mauritania, September 1943, i . In my key to the species (1923, Ann. Transv. Mus. 9) this runs down close to T. instabilis Turner, but the interocular distance is shorter than in that species. The genitalia of the specimen from Willowmore, Cape Province, which I identified as instabilis, are quite unlike those of falcigera, resembling those of T. bulawayoensis Bischoff. BY MR. K. M. GUICHARD IN W. AFRICA AND ETHIOPIA 153 Tachytes seminuda sp. n. $. 10-5 mm. long. Black. First two abdominal segments and the third excepting the apical margin red. Last three joints of the anterior tarsi and apical joint of the lower and hind tarsi, ferruginous. Wings hyaline, the veins pale brown. Clypeus and sides of the face and the bottom of the temples with a short and by no means dense silvery pubescence. On the thorax there is a little grey pubescence on the lateral margins of the mesonotum and on the epinotum, but the outstanding pilosity present in nearly all other species of the genus is absent. The first three tergites have inconspicuous apical fasciae of silvery pubescence, and the pygidial area has dark golden pubescence. Vertex microscopically and closely punctured (30 diameters), the mesonotum closely and transversely rugulose (25 diameters), the dorsum of the epinotum reticulate-punctate and obliquely striate at the sides of the base (30 diameters). Second sternite dull, the third and following sternites shining, with a few large punctures near the apical margin. The abdomen is widest in the middle, or lanceolate, instead of being obconical or widest across the basal segment, as in the other species of the genus. Galea less than half as long as the scapes. Eyes widely divergent below, the inter- ocular distance at the base of the eyes being very nearly three and a half times greater than on the vertex, where it is equal to the length of the second joint of the flagellum. Clypeus short, nearly four times wider than long. Flagellum slender, the second joint fully two and a half times longer than wide at the apex, all the following joints at least twice as long as wide. Second to fifth sternites and the fifth tergite with a row of brown setae on the apical margin. Anterior metatarsi with five yellow spines on the outer margin. Pygidial area triangular, longer than wide at the base. Tillembeya, River Niger, French Sudan, August 1944, 2 ??. This species differs from all the others known to me by the paucity of pubescence and pilosity, and the shape of the abdomen. Superficially it looks much more like a Tachysphex, e.g. T. miniatulus Arnold, than a Tachytes. PHYTOSPHEX gen. n. Characters ?. Anterior tarsi without a comb, the metatarsus with a few small spines on the underside and at the apex. Hind tibiae and metatarsi with a row of long spines on the upper side. Pygidium without a marginate dorsal face. <$. Anterior metatarsi without long spines on the outer margin, the hind tibiae and metatarsi as in the $. In both sexes the vertex is wide, the eyes not being strongly convergent above as in Tachysphex, and the sculpture of the head and thorax is strong, consisting of large and widely spaced punctures on the head and thorax, and a coarse reticulation on the epinotum. Anterior margin of the clypeus produced in the middle into an acute, wide, and triangular tooth. Otherwise like Tachysphex. Genotype : Tachysphex turneri Arnold. Not only does this species differ from Tachysphex in the sculpture, but the nesting habits are, according to the late Dr. Brauns's observations, totally different. Instead of burrowing in the ground, it nests in hollow stems of Aloe, Datura, &c., and the partitions between the cells are formed of earth and little pebbles. 154 SPHECIDAE AND POMPILIDAE (HYMENOPTERA) COLLECTED Phytosphex turner! Arnold 1923, Ann. Transv. Mus. 9: 165. A slight variety, differing from the type of the species in having the second to fourth tergites blackish in the middle and on the apical third of the segments. Adis Ababa, Ethiopia, April 1948, i . ODONTOSPHEX gen. n. Characters $. Mandibles not excised on the outer margin. Clypeus short, not longer than the distance between its hind margin and the antennal sockets, the anterior margin with two long and thin teeth near the middle. Eyes strongly convergent above, the vertex narrower than the anterior ocellus. Posterior ocelli oval, flattened, and almost obsolete. Second to seventh joints of the flagellum short, wider than long, the seventh and tenth swollen behind. Dorsum of the pronotum linear, lying well below the level of the mesonotum. Episternal suture absent. Dorsum of the epinotum with a defined and broad median area. Eighth sternite emarginate and bidentate. Anterior tarsi without cilia or spines. Middle tibiae with two calcaria, the inner one very small. Posterior tibiae and tarsi without long spines on the upper side, the tibiae with a row of very short spines hardly longer than the pubescence. Forewings with three cubital cells, the third less than twice as long on the cubitus as on the radius, the second receiving both recurrent veins. Hindwing : the cubitus emitted well before the end of the submedial cell. Genotype: 0. bidens. The genus is remotely related to Tachysphex, from which it differs, inter alia, by the venation, the absence of the episternal suture, and the form of the flagellum. Odontosphex bidens sp. n. (Fics. 44, 44-c) $. 6 mm. long. Black. Middle third of the mandibles rufescent. Flagellum ochreous, the first four joints slightly infuscated. Apical margins of the abdominal segments testaceous. Tibiae, tarsi, and the hind femora pale reddish-ochreous. Wings hyaline, tegulae and pterostigma pale yellow, the veins pale brown. Clypeus and face up to the anterior ocellus, temples, underside of the anterior femora, sides and anterior angles of the mesonotum, mesopleura, and base of the epinotal dorsum, covered with silvery pubescence. A small patch of similar pubescence on each side of the scutellar and metanotal disks. Sides of the epinotum, and the declivity excepting the middle, with exserted and fairly dense silvery pilosity. Abdomen with a very fine, grey, and decumbent pubescence, and on the apical margins of the first five tergites fasciae of silvery pubescence. Face, from the antennal sockets to the anterior ocellus, dull, micro- scopically and closely punctured, slightly convex but without any swellings. Ocellar area and vertex shining, sparsely punctured. On the vertex a carinula on each side margins the eyes. Mesonotum, scutellum, and metanotum shining, finely punctured, the interspaces three or four times wider than the punctures, except on the posterior fourth of the BY MR. K. M. GUICHARD IN W. AFRICA AND ETHIOPIA 155 mesonotum and on the metanotum, where they are smaller. Mesopleura dull, closely punctured (35 diameters), a short transverse crest extends from the hind margin forwards, just above the middle coxae. Dorsum of the epinotum nitidulous, a little more strongly and more closely punctured than the scutellum, the margins of the median area indicated by a narrow depressed line. A median carina extends from the base to near the apex of the area. Declivity dull, with a wide pit below the brow. Metapleura and sides of the epinotum fairly dull, closely punctured (45 diameters). Tergites and sternites almost dull, very closely and finely punctured (30 diameters). Head wider than the thorax. Clypeus five times wider than long, the anterior margin feebly convex and with an acute tooth, longer than wide at the base, on each side of the middle. Eyes strongly divergent below, the interocular distance across the base of the clypeus being eight times greater than across the vertex, where it is equal to the length of the second joint of the flagellum. First joint of the flagellum very small, not globose, the second joint three-fourths wider than long, the first to seventh wider than long, the seventh to tenth swollen below, the apical joint one and a half times longer than wide. Scutellum twice as wide as long, dorsum of the epinotum about three times wider at the base than long, as long as the scutellum, the junction with the sub vertical declivity rounded. First tergite about one and a half times wider behind than long. Basal joint of the anterior tarsi thin and very long, as long as the remaining joints united. Hind femora and tibiae slightly compressed transversely, the tibiae with a row of about twelve very small scalariform teeth on the upper margin (45 diameters), apical margin of seventh tergite transverse. First abscissa of the radius slightly shorter than the third, and a little longer than the second. Aleg, S. Mauritania, September 1943, I <. Tachysphex fluctuates Gerstaecker 1857, Mber. Akad. Wiss. Berl.: 510. Arnold, 1945, Sphecidae of Madagascar: 97. Labadi, Gold Coast, March 1941, ic?; Meisso, Ethiopia, September 1945. The second specimen has the epinotum much more strongly and more widely reticulate, and more pilose than in the typical form. Tachysphex bmneiceps Arnold 1923, Ann. Transv. Mus. 9: 153. River Hawash, Ethiopia, June 1946, i $. Tachysphex egregius Arnold 1924, Ann. Transv. Mus. 11: 273 <$. 1929, Ann. Transv. Mus. 13: 384 . Labadi, Gold Coast, May 1941, i & i ? ; Tillembeya, River Niger, French Sudan, August 1944, i ?. 156 SPHECIDAE AND POMPILIDAE (HYMENOPTERA) COLLECTED Tachysphex theseus sp. n. (FiG. 45) $. 10 mm. long. Black. Apical margin of the fourth and the whole of the fifth and sixth abdominal segments ferruginous. Tarsi ferruginous, the apical joints a little darker. Wings hyaline, the veins pale brown. Face and clypeus with pale golden pubescence, thorax with a very short, sparse, and grey pubescence. Clypeus and face dull, very finely and closely punctured, the vertex nitidulous and not quite so closely punctured as the face, the temples smooth and shining. 48 FIG. 45. Tachysphex theseus $, xc. 10. FIG. 46. Gastrosericus fluviatilis $, x 16. FIG. 46^. Gastrosericus fluviatilis $, clypeus, x 16. FIG. 466. Gastrosericus fluviatilis $, wing, xc. 10. FIG. 47. Gastrosericus senegalensis $, Xi6. FIG. 48. Palarus rothschildi croesus $, pygidium, X 10. Mesonotum shining, punctured, the punctures about twice as large as those on the face, the interspaces from two to three times wider than the punctures, except on the margins which are very closely punctured. Scutellum and metanotum shining, their puncturation a little smaller on the face. Mesopleura finely and very closely punc- tured. Dorsum and declivity of the epinotum dull, fairly closely reticulate, the sides of the epinotum shining, transversely striate. Abdomen dull, impunctate, the pygidial area twice as long as wide at the base, the apex narrowly truncate. Median area of the clypeus nearly twice as wide as long, impressed behind the anterior margin which is feebly concave in the middle. Supra-antennal tubercles prominent. Interocular distance on the vertex very nearly as long as the first two joints of the flagellum. Behind the ocellar area there is a well-defined angular depres- sion. Flagellum slender, the second joint nearly three times longer than wide at the apex, the third joint a little longer than the second. Dorsum of the epinotum two and two-fifths wider at the base than long. Fourth joint of all the tarsi very short, BY MR. K. M. GUICHARD IN W. AFRICA AND ETHIOPIA 157 wider than long, the claws asymmetrical, the outer claw of the anterior tarsi larger than the inner, and smaller than the inner of the middle and hind tarsi. Accra, Gold Coast, September 1941, i ?. Related to T. hippolyta Arnold and T. harpax Arnold and having the claws asymmetrical as in them, but differing from both by the length and sculpture of the epinotum, the slender flagellum, and the colour of the apical abdominal segments. Tachysphex panzeri van der Linden var. aethiopicus Arnold 1923, Ann. Transv. Mus. 9: 167. Labadi, Gold Coast, March 1941, 3 $$. Atelosphex lugubris Arnold 1924, Ann. Transv. Mus. 11: 72. Accra, Gold Coast, September 1941, i ?. This differs to a slight degree from the type of the species, but hardly enough to justify a varietal name. The puncturation of the mesonotum in the middle is sparser, the dorsum of the epinotum is more reticulate, and the pygidial area is narrower. In the coarse puncturation and the absence of spines along the length of the hind metatarsi, the genus shows some affinity with Phytosphex Arnold, but in the latter the epicnemial suture is present. Parapiagetia capensis Brauns var. ferox Arnold Brauns, 1910, Dtsch. ent. Z.\ 666. Arnold, 1922, Ann. Transv. Mus. 9: 135. Ufdem, Ethiopia, August 1945, i $; Aleg, S. Mauritania, September 1943, i ?, Dakar, Senegal, August 1943, i ?. The specimen from Dakar is a slight variety, having the last three abdominal segments blackish. Gastrosericus neavei reversa subsp. n. G. neavei Turner, 1913, Trans. Ent. Soc. Lond.: 754. $. 8 mm. long. The corners of the median lobe of the clypeus are not rectangular as in neavei i. sp. but produced into blunt teeth, and the wings are slightly darker. Otherwise it does not differ except in one very remarkable character in which it also differs from all other Ethiopian Sphecids known to me. In neavei i. sp. the dorsum of the epinotum is glabrous, but in this race on each side of the median transversely rugose area there is a coarse, obliquely exserted and silvery pubescence, directed cephalad. In the $ this pubescence is less abundant, but plainly apparent if the insect is looked at caudad. Otherwise like the type of the species. Tillembeya, River Niger, French Sudan, August 1944, 4 ??, 4 &?. Gastrosericus fluviatilis sp. n. (Fics. 46, 46^ and 466) ?. 8 mm. long. Black. The following parts pale yellow : basal half of the mandibles, the apex of the scapes, tegulae, apical half of the anterior femora, apical third of the 158 SPHECIDAE AND POMPILIDAE (HYMENOPTERA) COLLECTED middle and extreme apex of the hind femora, tibiae and first two joints of tarsi, the last three joints pale ferruginous. Wings hyaline, faintly smoky, the veins brown. Clypeus, lower half of the face, temples, dorsum of the pronotum, mesosterna, and lower half of the mesopleura with silvery pubescence. Epinotum with short, erect and white pubescence. Tergites 1-4 with apical fasciae of grey pubescence, the pygidial area with dark golden pubescence. Sternites 3-6 shining and impunctate, the rest of the body dull, closely and microscopically punctured, the largest punctures being on the scutellum and epinotum (30 diameters). Clypeus short, five times wider than long, carinate lengthwise in the middle, and with a very short lobe on the anterior margin ; between the lobe and the lateral angles there are two small teeth on each side. Interocular distance at the base of the eyes two-thirds greater than across the vertex, where it is equal to the length of the first five joints of the flagellum. Temples with a triangular tooth near the middle. Second joint of the flagellum slightly shorter than the third and not quite twice as long as wide. Lower third of the face excavated, the upper margin of the excavated area continued as a groove margining the eyes as far as the level of the anterior ocellus. Propleura with a transverse torus on each side in front of the hind margin. Anterior coxae triangular, the anterior margin raised and thickened. Pronotal collar barely as long as the first joint of the flagellum and lying much below the level of the meso- notum. Dorsum of the epinotum fully twice as wide at the base as long, joining the oblique declivity in a wide curve. First tergite a little longer than wide at the apex, and like the second tergite a little swollen in front of the hind margin. Pygidial area elongate-triangular, slightly shining and sparsely punctured at the base. Anterior metatarsi with six short spines on the outer margin. Second discoidal cell petiolate, the first abscissa of the radius fully three times longer than the second. $. 6 mm. long. Middle and hind tibiae with a fuscous macula on the inner side, the last four joints of the hind tarsi black. Pubescence on the clypeus and face golden. Clypeus produced in the middle into an acute tooth. Joints of the flagellum very short, the second one and a half times longer than wide. Temples without a tooth, anterior coxae less thickened in front than in the $. Otherwise like the $. Tillembeya, River Niger, French Sudan, August 1944, i <$, 6 $?. Related to turneri Arnold, from which it differs, inter alia, by the shape of the clypeus and anterior coxae, the much finer puncturation of the thorax, and the venation. Gastrosericus senegalensis sp. n. (Fie. 47) $. 6-5 mm. long. Head and thorax black, abdomen pale ferruginous. Mandibles yellow at the base, the apex of the scapes ochreous on the upper side. Femora fusco- ferruginous, the fore and middle pairs yellow at the apex, tibiae and tarsi pale ferru- ginous, the tibiae pale yellow on the upper side. Wings hyaline, the veins ochreous. Head and thorax with a very short adpressed and yellowish-silvery pubescence, without any long pilosity. Tergites with a thin and microscopic greyish pubescence. The whole body dull and, excepting the pygidial area, very closely and microscopically BY MR. K. M. GUICHARD IN W. AFRICA AND ETHIOPIA 159 punctured (50 diameters), the dorsum of the pronotal collar with traces of transverse rugulosity. Pygidial area shining, sparsely punctured, twice as wide at the base as long. Clypeus three times wider than long, subcarinate in the middle, the anterior margin convex and without teeth. Eyes divergent below, the interocular distance at their base half as great again as across the vertex, where it is equal to the length of the first three joints of the flagellum plus half of the fourth. The second joint about twice as long as wide, and as long as the third. Temples unarmed. Dorsum of the pronotum narrowed anteriorly, fairly long, as long as the first joint of the flagellum. Dorsum of the epinotum twice as wide at the base as long, with a very indistinct basal triangular area. Anterior metatarsi with five long spines on the outer margin. First abscissa of the radius twice as long as the second, the recurrent veins contiguous on the second cubital cell. Dakar, Senegal, August 1943, i ?. Smaller than the Egyptian G. waltii Spinola and without the long pubescence on the head and thorax and rugose sculpture of the epinotum present in that species. It differs from G. capensis Brauns in the much finer sculpture and dull surface of the head and thorax and by the absence of a long and woolly pubescence on those parts. Liris (subgen. Motes) nugax Kohl 1894, Ann. naturh. Hofmus. Wien, 9: 302. Dilla, Ethiopia, April 1943, i ? ; Didessa River, Ethiopia, June 1946, I $. Liris (Motes) croesus Smith 1856, Cat. Hymen. B.M. 4: 284. Diafarabe, Senegal, August 1944, i $. Liris (Motes) solstitialis Smith 1856, Cat. Hymen. B.M. 4: 283. Aburi and Accra, Gold Coast, August to September 1941, i <$, i $. Liris (Motes) ragifera Turner 1918, Ann. Mag. Nat. Hist. (9), 5: 362. Accra, Gold Coast, September, i $. Liris (Motes) thysanomera Kohl 1894, Ann. naturh. Hofmus. Wien, 9: 302. Accra and Aburi, Gold Coast, December 1941, 3 ??. Liris (Motes) antaca transvaalensis Cameron Arnold, 1945, Sphecidae of Madagascar: 126. Adis Ababa and River Hawash, Ethiopia, August and June, 2 ?$. ENTOM. 2. 3 T I6o SPHECIDAE AND POMPILIDAE (HYMENOPTERA) COLLECTED Liris (Motes) egregia Arnold 1929, Ann. Transv. Mus. 13: 395. Adis Ababa and Dilla, Ethiopia, June, 2 <$$. Liris (Motes) setigera Arnold 1940, Ann. Transv. Mus. 20: 123. Tillembeya, River Niger, French Sudan, August 1944, i <. Liris (Motes) denticulata Turner 1920, Ann. S. Afr. Mus. 17: 496. Accra and Labadi, Gold Coast, September and March 1941, 2 $$. These represent a slight variety in which the tarsi are ferruginous instead of brownish-black. Liris (Motes) flavitincta Arnold 1940, Ann. Transv. Mus. 20: 141. Accra, Gold Coast, September 1941, I ?. Liris (Motes) abyssinica Arnold iQ33 Ann. Mag. Nat. Hist. (10) 77: 353. Fiche, Ethiopia, October 1945, i <$. Liris (Motes) gracilicornis Arnold 1923, Ann. Transv. Mus. 9: 250. Labadi, Gold Coast, March 1941, i ?. Liris nigricans Walker race reticulata Saussure 1871, List Hymen. Egypt: 21. Arnold, 1945, Sphecidae of Madagascar: 132. Dilla, Ethiopia, April 1948, 2 <. Palarus rothschildi croesus subsp. n. (Fie. 48) P. rothschildi Magretti, 1908, Bull. Mus. Hist. Nat. Paris: 189, . Closely related to dubiosus, from which it differs by the shorter and thicker petiole and the ground colour of the legs, and the position of the yellow bands on the tergites. Crabro (Dasyproctus) stevensonianus occidentalis subsp. n. (Fics. 52 and 520) C. stevensoni Arnold, 1926, Ann. Transv. Mus. 2: 369. C. stevensonianus Arnold (nom. nov.), 1940, ibid. 20: 135. <. 7 mm. long. This race differs from the type of the species as follows : scapes with a brown macula on the inner side below. Anterior tibiae yellowish-red below, dark brown above, all the tarsi brownish-black, all the rest of the legs black. The yellow maculae on the second tergite duller and smaller. The reticulations on the epinotum are much higher and wider apart. As in the type of the species $, the petiole is five times longer than wide across the node. Aburi, Gold Coast, December 1944, i . Subfamily OXYBELINAE Oxybelus coniferus sp. n. (Fics. 53, 53a-c) ? 6-5-7-8 mm. long. Black. The following parts yellowish-white: a short and transverse streak on each side of the pronotal dorsum, small spots, sometimes absent, on the anterior corners of the scutellum, the metanotum, transverse and more or less lanceolate maculae on each side of the first four tergites. Anterior tibiae dirty white on the upper side. Apical joint of the tarsi pale brown. Wings hyaline, slightly smoky, veins dark ochreous to black. Tegulae fusco-ferruginous. Mandibles some- times yellowish-red on the basal half. Mandibles, clypeus, lower half of the face and temples with silvery pubescence. Metanotum and scutellum with erect, brownish-grey pubescence, the pleura, sterna, and femora below with white and thin pubescence. Median area of the clypeus shining, the anterior margin convex and with a blunt tooth on each side, the base with a transversely compressed, and slightly curved conical protuberance, its apex rounded. The space behind the scapes shining and very finely punctured, the rest of the head dull, very closely punctured, the interspaces not wider than the punctures ; on the occiput and temples striated between the punctures. Pronotal dorsum very short, almost linear in the middle, the lateral angles acute. Mesonotum and scutellum shining, a little more strongly punctured than the vertex, the interspaces narrower than the punctures and slightly raised longitudinally, or subrugose. Metapleura nitidulous, coarsely reticulate-rugose. Metapleura strongly, sides of epinotum finely and transversely striate. Metanotum shining, longitudinally and not closely costate. The lateral squama black, rostrate and acute at the apex. Epinotal process canali- culate, narrowed towards the apex which is excised, half as long again as wide at the base. Median triangular area of the epinotal dorsum with eight transverse costae which extend over the lateral areas, the interspaces dull, very finely reticulate. Declivity of the epinotum dull and transversely rugose outside the median triangular space. Abdomen shining, the tergites very sparsely punctured on the yellow areas at the sides, fairly closely over the basal half and at the apical margin. Pygidial area dull, closely punctured, triangular, the apex truncate. Second sternite sparsely and coarsely punctured, excepting the sides which are closely and finely so. Eyes 166 SPHECIDAE AND POMPILIDAE (HYMENOPTERA) COLLECTED moderately divergent above and below, the interocular distance on the vertex equal to the length of the flagellum, less the apical joint. Posterior ocelli twice as far from each other as from the eyes. Between the posterior ocelli and the eyes there is a low and oblique torus. Comb of the anterior tarsi ochreous, the basal joint with six spines on the outer margin. $. 6-2 mm. long. Femora with a dull yellow macula on the upper side at the base. Face and mesonotum with a slight aeneous tint. Pale maculae on the first to fifth tergites, smaller than in the $ and of a chrome-yellow colour. Median area of the clypeus with a longitudinal carina, wide at the base, projecting as a tooth on the concave anterior margin. Pygidial area trapezoidal. Otherwise like the $. Adis Ababa, Wondo, Ethiopia, August and September, 1945, 12 $$, 9 <. Related to 0. pilosus Arnold, from which it differs in both sexes by the greater size, in the $ by the shape of the clypeus and its basal process, black pronotal tubercles, stronger puncturation of the mesonotum and tergites, entirely black flagellum, and colour of the tarsi. In the $, it differs from pilosus by the colour of the legs and of the flagellum. Oxybelus acutissimus Bischoff var. propinquus Arnold 1927, Ann. Transv. Mus. 12: 83. Haramaia, Ethiopia, May, 1948, 3 $$. Oxybelus pictus Arnold 1927, Ann. Transv. Mus. 12: 115. Tillembeya, River Niger, French Sudan, September 1944, i $. This specimen is a slight variety, differing from the type of the species by the less close puncturation of the mesonotum, by the stronger sculpture of the lateral areas of the epinotal dorsum, and by the presence of narrow pre-apical yellow bands on the third and fourth tergites as well as on the first and second. Oxybelus curviscutis Arnold 1927, Ann. Transv. Mus. 12: 109. Accra, Gold Coast, Tillembeya, River Niger, French Sudan, Dakar, Senegal, and Haramaia, Ethiopia, 5 $$, 4 <$$. The $ from Tillembeya is a variety with a reddish pygidium. Oxybelus pallidus Arnold 1927, Ann. Transv. Mus. 12: 101. Ufdem, Ethiopia, August 1945, 2 $$. PART II. POMPILIDAE This interesting collection of Pompilidae made by Mr. K. M. Guichard and now the property of the British Museum (Natural History) comprises 162 specimens of which a considerable proportion is too defective, lacking antennae or tarsi or wings, to BY MR. K. M. GUICHARD IN W. AFRICA AND ETHIOPIA 167 permit of identification ; but it has, nevertheless, been possible to work out the great bulk of it quite satisfactorily. Subfamily PEPSINAE Genus Hemipepsis Dahlbom 1843, Hymen. Eur. 1: 123. Arnold, 1932, Ann. Transv. Mus. 14: 318-319. Arnold, 1948, Occ. Pap. Nat. Mus. S. Rhod. 14: 233. The collection contains twelve specimens of which seven are males. These do not agree with any of the species I have seen, and in view of our present knowledge of the genus and also of the absence of the corresponding females, it would be unwise to describe them as new. I have elsewhere (1932) expressed the opinion that the genus is of recent origin on account of the uniformity of the male genitalia. It is therefore difficult to correlate the sexes unless they are taken in copula. The copulatory act is probably of very short duration, as in the Sphecidae, for in nearly forty years of collecting in S. Rhodesia I have only once taken a Pompilid species copulating, Anoplius fuscus montivagus Arnold. Both sexes may occasionally be seen on trees of which the flowers have short corollas, such as Rhus and Zizyphus, but as the repro- ductive phase in both sexes is then past, they pay no attention to one another, so that their proximity is of little use in correlating the sexes. Hemipepsis aethiops Kohl 1913, Rev. Zool. Afr. 3: 193. Didessa River, western Ethiopia, 25 May 1946, i ?. The type was taken at Mpika, N. Rhodesia, not Belgian Congo as stated in Kohl's work. Hemipepsis fallax Saussure 1892, in Distant: Nat. Transvaal: 221. Didessa River and Lekempti, Ethiopia, May 1946, 2 <<. Hemipepsis iodoptera meridionalis Arnold Arnold, 1932, Ann. Transv. Mus. 14: 361. Kpeve, Gold Coast, June 1942, i $. A variety of this race in which the colour of the head and pronotum is castaneous red. Hemipepsis glabrata anchietae Radoszkovsky Radoszkovsky, 1881, /. Sci. Mat. Phys. Nat. Lisboa, 8: 213. Meisso, Ethiopia, 4 September 1945, i <. Cyphononyx flavicornis Fabricius var. antennatus Smith 1781, Spec. Insect, 1: 450. Smith, 1855, Cat. Hymen. B.M. 3: 186. Lake Bishoftu, 7,000 ft., May 1946, 2 ?$; Asba Tafari, Ethiopia, 7,800 ft., Sep- tember 1945, i c. ENTOM. 2. 3 u 168 SPHECIDAE AND POMPILIDAE (HYMENOPTERA) COLLECTED Cyphononyx optimus Smith 1855, Cat. Hymen. B.M. 3: 141. Ghedo, 7,000 ft., May 1946, I <$; Gore, Ethiopia, 6,000 ft., February 1948, i , wings, x 7. FIG. 56. Pseudagenia accraensis $, head, x 12. FIG. 560. Pseudagenia accraensis ?, first tergite, x 12. at the apex and fully one-third longer than the third joint. Temples nearly as wide as the eyes. Postnotum with a median groove, as long as the metanotum. The junction of the epinotal dorsum with the declivity not clearly delimited, the two forming in profile a low and unbroken arc, the whole segment one-seventh longer than wide. First tergite a little longer than wide at the hind margin, four and a half times wider there than at the base. Second abscissa of the radius four times longer than the first and one-third longer than the third. The first recurrent vein meets the cubitus a little before the middle of the second cubital cell, the second meets the third cubital cell at the end of its proximal third. Nervulus distinctly postfurcal. Cubitus of the hindwing emitted a little beyond the end of the submedial cell. Hind tibiae with a row of nine scales each overlapping a spine, and with a row of spines on each side of the row of scales. Claws with an acute tooth in the middle. Dilla, Ethiopia, April 1948, I $. 170 SPHECIDAE AND POMPILIDAE (HYMENOPTERA) COLLECTED Not closely related to any other Ethiopian species. The colour resembles that of P. inermis Arnold, which is a larger species and with much stronger sculpture. Subfamily CTENOCERATINAE Genus Marimba Pate Pate, 1946, Trans. Amer. Ent. Soc. 72: 93 and 120. Turner (Cryptosalius), 1918, Ann. Mag. Nat. Hist. (9) 1: 354-355. Arnold (Parapompilus) , 1932, Ann. Transv. Mus. 15: 89, 104-106. Marimba guichardi sp. n. (Fics. 55, 550 and 556) $. 1 1 -5 mm. long. Black, head and prothorax very dark reddish-brown. First six joints of the antennae, legs, and extreme apex of sixth tergite, fusco-ferruginous. Wings hyaline, bifasciated, the inner fascia covering the apical two-fifths of the medial, first submedial and anal cells and the base of the second submedial and first discoidal cells, the outer fascia covering the radial cell excepting the apex, the second and third cubital, the apex of the first cubital cell and first discoidal cell, and the second discoidal cell excepting the base. Flagellum, mesosternum, and coxae covered with a decumbent, very fine and silvery pubescence. Lower corners of the epinotum with exserted silvery pubescence, the apical margins of the first three tergites with decumbent silvery pubescence at the sides. Labrum semicircular, the galea extending beyond the closed mandibles, the latter with a tooth behind the apex. Clypeus microscopically and closely punctured, with a greasy lustre, nearly three times wider than long, the anterior margin straight. Face finely and very closely punctured (25 diameters), the interspaces as wide as the punctures. Inner orbits parallel. On each side of the bottom of the face there is a deep circular pit, half on the face and half on the clypeus. Inter-antennal tubercle prominent ; a narrow impressed line extends from it to the anterior ocellus. Posterior ocelli very little farther from each other than from the eyes. Interocular distance on the vertex equal to the length of the second joint of the flagellum plus two-ninths of the third joint ; the second joint is slightly more than three times longer than wide at the apex and the third joint is one-sixth longer than the second. The puncturation of the dull promesonotum and scutellum is like that of the face but a little larger. The dorsum of the pronotum merges gradually into the anterior face, the sides are nitidulous, with a few large punctures, and the transverse groove is wide. Mesonotum three times longer than the pronotum. Scutellum very convex, subcarinate lengthwise in the middle. Postnotum much shorter than the metanotum and depressed. Mesopleura coarsely rugose, pitted between the rugae. Metapleura obliquely costate. Dorsum of the epinotum two-thirds wider than long, shining, very coarsely and transversely rugose, with numerous anastomoses or almost reticulate-rugose, the interspaces wide, about as wide as the ocelli, the sides more closely rugose, the declivity oblique, shining, and with seven transverse rugae. Abdomen lanceolate, two and two-thirds longer than wide across the middle, the first tergite as long as wide behind and four times wider there than at the base. Sixth tergite longer than wide at the base, the apex subacute. Transverse groove of the second sternite between the first and second third of the BY MR. K. M. GUICHARD IN W. AFRICA AND ETHIOPIA 171 segment. Second abscissa of the radius a trifle longer than the third and four times longer than the first. Cubitus of the hindwing emitted a little before the end of the submedial cell. Claws bifid. Dilla, Ethiopia, April 1948, I ?. Apparently closely related to M. indocilis Turner, which I have not seen, but differ- ing therefrom according to Turner's description as follows: the smaller size, the colour of the legs and basal joints of the antennae, the shorter second abscissa of the radius, and the close puncturation of the promesonotum. From M. contristans Turner it differs also by the puncturation of the pro- and mesothorax, the smaller size, and the colour of the wings. These two species and two others from Africa were described by Turner (1918) and placed in his genus Cryptosalim (genotype C. rava Bingham). In my monograph of the Ethiopian Pompilidae I placed Cryptosalius in synonymy with Parapompilus. However, Pate (1946) indicated that Turner's African species were not congeneric with C. rava and proposed for them the new generic name Marimba. These species and the one described above differ from Parapompilus in the absence of a tarsal comb, the fasciated wings, and the long and very coarsely sculptured epinotum. Another distinction mentioned by Pate, 'the humeral angles of the pro- notum not broadly rounded as in Parapompilus ', does not hold good, as in this species, guichardi, they are rounded. Until the males are known, the relationship of Marimba with Parapompilus cannot be decided. Subfamily PSEUDAGENIINAE Pseudagenia pseudocyphononyx Turner 1918, Ann. Mag. Nat. Hist. (9) 1: 352. Adis Ababa, 3 August 1945 ; Zuquala, Ethiopia, October 1945, 2 $?. Somewhat smaller and with shorter wings than in the type and specimens from S. Rhodesia. Pseudagenia accraensis sp. n. (Pics. 56, 56) ?. 16 mm. long. Mesopleura, metanotum, metapleura, epinotum, and the first three abdominal segments black, the rest of the body, the antennae, and legs pale ferruginous. Wings very pale yellowish-brown, slightly darker over the radial, second cubital, and the upper half of the second discoidal cells, but not fasciated. Face and clypeus with a microscopic reddish pubescence. Mesopleura closely and very finely reticulate rugose, metapleura closely striate, epinotum closely, trans- versely, and finely rugose, the rugae strongest on the posterior half of the epinotal dorsum. Clypeus very convex transversely, nearly half as wide again as long in the middle, the anterior margin convex, with a slight projection in the middle. Inter- ocular distance across the base of the clypeus one-third greater than across the vertex, where it is equal to three-fourths of the length of the second joint of the flagellum. Posterior ocelli as far from the eyes as from each other. Flagellum long and slender, all the joints excepting the first at least six times longer than wide, the second joint nearly seven times longer than wide at the apex and one-seventh longer than the third. 172 SPHECIDAE AND POMPILIDAE (HYMENOPTERA) COLLECTED Scutellum fairly flat, two-thirds as long as wide at the base. Postnotum shining, finely and transversely striate, as long as the metanotum. The whole epinotum is as long as wide at the base, the dorsum merges gradually into the declivity, the whole forming a low curve in profile. First tergite one-fifth longer than wide, very narrow basad. Second abscissa of the radius three times longer than the first, and half as long again as the second. The wings extend back to the apex of the abdomen. Accra, 1941, i ?. The species nearest to this one is P. lujae Arnold, from which it differs by the longer clypeus, the narrower vertex, the much more slender antennae, the longer peduncle of the first tergite, and the colour of the wings. Pseudagenia stigmalis R. Lucas 1898, Deutsch Ost-Afr. (3) 4: 61. Arnold, 1934, Ann. Transv. Mus. 15: 342, 346. Adis Ababa, 3 August 1945, 2 ??. It is with some hesitation that I ascribe these two specimens to this species, since the original description is quite inadequate. They appear to differ only in having the hind femora entirely ferruginous, and the face also of that colour excepting a black streak adjacent to the antennal sockets. Pseudagenia laevigata Smith 1855, Cat. Hymen. B.M. 3: 143. Accra, Gold Coast, June 1941, 2 $?. Pseudagenia flavotegulata Bingham 1902, Ann. Mag. Nat. Hist. (7) 9: 208. Kaolack, Senegal, August 1943, i ?. A slight variety in which the base of the hind coxae on the upper side, the first abdominal segment and the sides and pre-apical bands on the second and third tergites, and the middle of the second and third sternites are black. Pseudagenia infantula Kohl 1894, Ann. naturh. Hofmus. Wien, 9: 307. Lake Bishoftu, Ethiopia, July 1945, i ?. This specimen represents a slight variety in which the vertex is a little narrower, being equal to the length of the second joint of the flagellum, and the first tergite as long as wide behind. The distinctive shape of the clypeus is exactly like that of infantula i. sp. Pseudagenia albosignata sp. n. (Fics. 57, 570 and 576) (. 8 mm. long. Black. Clypeus, excepting the basal half of the middle, and the sides of the face to a little beyond the level of the antennal sockets, dirty white. Legs ferruginous, the coxae, trochanters, and tarsi brown. Wings hyaline, with a pale BY MR. K. M. GUICHARD IN W. AFRICA AND ETHIOPIA 173 fuscous cloud over the basal vein and nervulus, and an outer fuscous cloud covering the basal third of the radial cell, the greater part of the second cubital cell and a small area on the upper part of the second discoidal cell ; the apex of the forewing also slightly fuscous. Clypeus and face with a sparse brownish pilosity, the temples, coxae, posterior half of the mesosternum and posterior angles of the epinotum with a decumbent silvery pubescence, the latter and the mesosternum and metapleura 59 FIG. 57. Pseudagenia albosignata $, head, x 15. FIG. 570. Pseudagenia albosignata $, head, x 15. FIG. 576. Pseudagenia albosignata <$, genitalia, xc. 45. FIG. 58. Poecilagenia longicollis $, thorax, X 15. FIG. 59. Poecilagenia rugosa <$, pronotum, x 12. FIG. 593. Poecilagenia rugosa ; Diafarabe, French Sudan, i ?; Dilla, Ethiopia, 2 $$. GUICHARDIA gen. n. Characters $. Palpi short, the last three joints of the maxillary palpi not more than three times longer than wide. Labrum exposed. Mandibles with a tooth behind the apex. Clypeus feebly convex, as wide as the bottom of the face. Cheeks very short. Face without a supra-antennal tubercle. Occiput not concave, its junction with the vertex rounded. Pronotum not much shorter than the mesonotum, convex lengthwise. Lateral margins of the mesonotum feebly reflexed posteriorly. Sides of the scutellum steep, raised above the level of the metanotum. Postnotum linear. Dorsum of the epinotum convex lengthwise and transversely, forming in profile a low and unbroken arc with the very short declivity, its posterior angles dentiform. Apical ventral valve produced into a long spine. Femora of the trachyscelid type. Claws bifid, the pul- villus small and with about ten short cilia. Wings with three cubital cells, the third much wider on the cubitus than on the radius ; nervulus slightly prefurcal. Cubitus of the hindwing emitted a little beyond the end of the submedial cell, the basal lobe small. Genotype: G. macilenta Arnold. The subdentate posterior angles of the epinotal dorsum are very similar to those of Epiclinotus Haupt, but the wide clypeus and the rounded junction of the slightly convex occiput with the vertex show that the genus belongs to the tribe Pompilini. Guichardia macilenta sp. n. (Fics. 61, 6ia-d) 8 'e &0 8? V is i 8 "e too o ^ Ci Host ^ ^> Ors.* 2 i Anthus spinoletta (L.) /B.M. I Rock-Pipit \0rs. fB.M. 6 i Motacilla alba yarrelli ^Ors. i i Gould B.M. 3 Ors. i Pied Wagtail Motacilla cinerea cinerea /B.M. /"B.M. 9 2 Tunst. \Ors. ^Ors. 3 j Grey Wagtail B.M. 2 Certhia familiaris (L.) /B.M. Tree-Creeper \0rs. i Ors. i Sitta europoea L. B.M. i Nuthatch /"B.M. 29 i 4 Parus major L. /B.M. 15 \Ors. 13 I 2 Great Tit lOrs. 10 Parus coeruleus L. /B.M. 13 Ors. I Blue Tit \Ors. 23 Parus ater L. Ors. fB.M. j 6 Coal-Tit \0rs. I 3 Parus atricapillus L. Ors. i B.M. I i Willow-Tit Aegithalos caudatus (L.) /B.M. Ors. I Long-tailed Tit \Ors. i Panurus biarmicus (L.) B.M. /B.M. 2 Bearded Tit \ Ors. I Lanius collurio L. Ors. J B.M. 3 6 Red-backed Shrike \Ors. 2 i 4 Muscicapa striata (Pall.) /B.M. 3 Ors. i Spotted Flycatcher \Ors. i Muscicapa hypoleuca (Pall.) Ors. i B.M. 2 i i Pied Flycatcher Ors. 2 i 7 Regulus regulus (L.) Ors. i "B.M. i Goldcrest Ors. i Phylloscopus collybita B.M. i "B.M. 2 I (Viell.) Ors. I Chiff-chaff "B.M. 29 if 3 Phylloscopus trochilus (L.) /B.M. Ors. jy Willow- Warbler \Ors. Ors. 3 i Phylloscopus sibilatrix J B.M. (Bechst.) \Ors. i /B.M. 2 i 5 Wood-Warbler \Ors. 4 Locustella naevia (Bodd.) Ors. i Ors. i Grasshopper- Warbler Acrocephalus schoenoboenus /B.M. i /B.M. I 2 3 (L.) \Ors. \Ors. I 2 Sedge-Warbler bo Q i i A COLLECTION OF FLEAS FROM THE BODIES OF BRITISH BIRDS 203 TABLE 3 (cont.) Hnst Sylvia nisoria (Bechst.) Barred Warbler Sylvia borin (Bodd.) Garden Warbler Sylvia atricapilla (L.) Blackcap Sylvia communis Lath. Whitethroat Sylvia curruca (L.) Lesser Whitethroat Turdus viscivorus L. Mistle-Thrush Turdus ericetorum Turton Song-Thrush Turdus merula L. Blackbird Oenanthe oenanthe (L.) Wheatear Saxicola torquata (L.) Stonechat Phoenicurus phoenicurus (L.) Redstart Phoenicurus ochrurus (L.) Black Redstart Erithacus rubecula (L.) Robin Prunella modularis (L.) Hedge-Sparrow Troglodytes troglodytes (L.) Wren Cinclus cinclus (L.) Dipper Hirundo rustica L. Swallow Delichon urbica (L.) House-Martin Riparia riparia (L.) Sand-Martin Apus apus (L.) Swift Picus viridis L. Green Woodpecker Dryobates major (L.) Great Spotted Wood- pecker Jynx torquilla L. Wryneck Athene noctua (Scop.) Little Owl Asia otus (L.) Long-eared Owl Ors. /B.M. \0rs. Ors. /B.M. \_Ors. B.M. Ors. B.M. Ors. B.M. Ors. B.M. Ors. B.M. Ors. B.M. Ors. B.M. Ors. B.M. B.M. Ors. B.M. Ors. B.M. I s I 6C O O Q B.M. i 2 Ors. 3 2 B.M. 4 4 Ors. 6 3 5 B.M. 16 2 15 Ors. 6 2 12 B.M. I Ors. I 4 :B.M. i Ors. i B.M. 2 3 Ors. I 2 14 19 IO 7 6 3 i i 2 I B.M. 2 Host 1/3 ) orchard Kincardine on Forth C. gallinae Beneath beech Hylton Bridge, y.xi.os R. S. Bagnall (i?)* bark Sunderland C. gallinae In rotten oak- Sherwood 30.V.27 C. E. Stott (i ^) tree C. gallinae In hollow tree Wroxeter, 23.viii.i2 G. A. K. Marshall (i . ( J ?) glasshouse Morayshire * Although there is only one specimen in the Rothschild Collection, Bagnall (1921) records it 'in numbers under beech tree bark' (Bagnall, R. S., Trans. Nat. Hist. Soc. Northumberland & Durham, 1921, 5: 181-198). A COLLECTION OF FLEAS FROM THE BODIES OF BRITISH BIRDS 211 TABLE 8 (cont.) Species of flea Location Locality Date Collector Collection C. gallinae (in numbers) C. gallinae In barns Hole in an oak- tree Holton, Suffolk 1913 Holton, Suffolk 1913 Strickland and^j f Merriman 1 Strickland and | Merriman J C. gallinae (i?) From Vespa vulgaris Harpenden, iy.iii.2i Herts. R. Stenton (Min. of Agr.) British Museum C. garei C. garei (i?) C. garei (I . 5-. ga//t- nw/ae Swept off Vac- cinium Long Hill, Co. 2I.V.23 Wicklow A. W. Stelfox Nat. Mus., Dublin D. . #*- nulae (i $) D. . a/ft- nulae (i 9) Dry bark of cherry-tree Herbage Craibstone, 5.iv.43 Aberdeen Craibstone, i.v.4i Aberdeen G. D. Morison Morison Coll. D. . gaWi- nulae (i $) Pine log Drum, I5.V-42 Aberdeen " D. ^. galli- nulae (i $) Among herbage Newtonhill, 2I.V.49 Kincardine ' " f These records were published in Parasitology, 1913, 6 (i) : 1-19 (Observations on British rat-fleas, by G. H. F. Nuttall, C. Strickland, and G. Merriman), but the whereabouts of the specimens is not known. D. STRAGGLERS FROM MAMMALIAN HOSTS 1. Nosopsyllus fasciatus (Bosc. D'Antic) 1801. One specimen was collected from Troglodytes t. troglodytes in Arran. The normal host is the Brown Rat (Rattus norve- gicus). There are three females and one male of this species in the British Museum collection from a Blackbird's nest in Holland (Smit, 1949). 2. Orchopeas wickhami (Baker) 1895. One specimen of the Grey Squirrel flea, which was originally imported into this country from the U.S.A. with its host, Sciurus carolinensis, was collected from Parus major newtoni Prazak. This species of flea is a more frequent straggler on to avian hosts. In the Charles Rothschild collection there are specimens collected from 'a Magpie's and Hawk's nest', and an unnamed bird's nest in the New Forest, Hampshire. An exchange of fleas between birds and squirrels is, in fact, a usual accident, and Monopsyllus sciurorum (Schrank) 1803 has been known to breed in large swarms in an Owl's nest (Waterston, 1910), 212 A COLLECTION OF FLEAS FROM THE BODIES OF BRITISH BIRDS while C. gallinae has frequently been recorded breeding in Squirrels' dreys in Britain, although not in large numbers (Freeman, 1941). SOME GENERAL CONSIDERATIONS RELATING TO C. GALLINAE, C. GAREI, AND D. G. GALLINULAE Geographical distribution. With regard to the three common species of bird-fleas C. gallinae, C. garei, and D. g. gallinulae in Britain, the maps in Figs. 1-3 all tell much the same story. These fleas are widely distributed throughout the British Isles and certain obvious concentrations mark the home territories of well-known col- lectors such as C. Rothschild, H. G. Jeffrey, R. Newstead, H. Britten, and J. Water- ston. In Ireland, according to E. O'Mahony, C. garei is relatively scarce, but unless lists of negative records are provided such statements may be somewhat misleading. If a long series of Ducks' nests, or Snipes' nests, yielded no specimens of C. garei this would be significant. In Cornwall, Turk (1946) considered C. gallinae 'not so common as it seems to be in other counties. Its place seems to be taken by D. galli- nulae, which is a very common species. ... I have records from many small Passerine birds and from Blackbirds, the domestic pigeon and the Wood-Pigeon.' However, if we pool all records for the British Isles (see Table 3), D. g. gallinulae is seen to be the dominant species in the nests of the Blackbird (G. B. Thompson, 1937, records 350 specimens from one nest), and the same may apply to the unnamed 'small Passerines ' to which Turk refers earlier. Since going to press Dr. Turk has sent me a list of these species. They include the Robin, Wren, and Tree-Creeper, in which D. g. gallinulae is usually dominant ; the Thrush, Mistle-Thrush, and Goldcrest, in which nests both species have been recorded an approximately equal number of times ; and the Great Tit, Blue-Tit, Jackdaw, Carrion-Crow, Raven, and Rook, in which nests C. gallinae is usually the dominant species of flea. In all, Dr. Turk found D. g. gallinulae ' in nearly a hundred nests '. In Ireland O'Mahony (1939) has come to a somewhat similar conclusion with regard to these two species, namely that D. g. gallinulae is 'the commoner of the two in point of numbers', but again the seven species of birds' nests he has listed in which D. g. gallinulae outnumber C. gallinae include the Wren, the Greenfinch, the Chaffinch, the Mistle-Thrush, the Song-Thrush, and the Blackbird. D. g. gallinulae is usually dominant in five of these, and in the nest of the Song-Thrush the two species are about equally represented (see Tables 3 and 6). He did not record D. g. gallinulae at all from nests of the House-Sparrow or the Blue-Tit records which would have been far more significant. It would be of much interest if further collecting in the west of England and Ireland confirmed the suggestions made by O'Mahony and Turk, since the distribution of D. g. gallinulae among its hosts in Britain and the distribution of the subspecies D. g. perpinnatus in North America suggest that it requires rather a high mean humidity in the nest. Thus it might possibly increase the range of its Passerine hosts where the mean annual rainfall is greater, and also occur in larger numbers in the nests of certain species where it is numerically sparse in drier conditions. In the meantime, how- ever, the distribution maps of C. gallinae, C. garei, and D. g. gallinulae (Figs. I, 2, A COLLECTION OF FLEAS FROM THE BODIES OF BRITISH BIRDS 213 and 3) show the distribution of the collectors rather than that of the fleas, and indicate that all three species occur where they are looked for. A few observations on the distribution of C. vagabunda and the Martin fleas are included in the description of the collection on pages 195 and 196. With all the species concerned further collecting is necessary before any conclusions can be drawn. From the little we know, however, it is obvious that the flea fauna of Scotland is of considerable interest. There are several species, such as C. borealis, Orneacus waterstoni, and Frontopsylla laetus, which display a so-called alpine-boreal type of distribution, occurring in Scotland and the Swiss Alps and in some cases along the northern fringe of Europe, but not apparently in England or the inter- vening areas of the Continent. This type of distribution is even more striking in the case of fleas than in other insects, since the host itself is widely distributed. On the other hand, C. hirundinis occurs on the House-Martin throughout its range. It is, however, clear, as we have seen in the case of D. g. perpinnatus, that other factors limit the distribution of the fleas, presumably requirements of the develop- mental stages. Although many of the Passerine hosts of bird-fleas migrate to South Africa and other warm countries in the winter, and (we may surmise) take their fleas with them, the Ceratophyllid bird-fleas have remained a group which is charac- teristic of temperate climates. In east Africa over a period of fourteen years, G. H. E. Hopkins and a team of his friends examined over 2,000 birds shot for food or scientific purposes. Ecto- parasites were collected by the method described in the introduction. Not a single specimen of Ceratophyllus was found, thus confirming the fact that the apparent absence of this genus on birds, in tropical Africa at least, is genuine. It should be noted that a number of migrants were included in this collection. Thus fifteen specimens of the Sand-Martin (Riparia riparia) and six specimens of the Swallow (Hirundo rustica] were among the negative records. There were also smaller numbers of birds such as the European Cuckoo (Cuculus c. canorus), the Whinchat (Saxicola rubetra], the Green Sandpiper (Tringa ochropus], and the Wood Sandpiper (Tringa glareola) . Host-preference. Among the three species of common bird-fleas which are not host-specific we can discern two different trends of host-preference. On the one hand there is a marked predilection on the part of D. g. gallinulae and C. gallinae for the Order Passeriformes, and on the part of C. garei for the Charadriiformes, Anseri- formes, and Galliformes. Within this framework we can despite a big overlap and many exceptions discern a preference for different types of nesting sites. Broadly speaking, C. garei favours damp nests on the ground, D. g. gallinulae those built near the ground, and C. gallinae nests built in drier situations some distance from the ground. There is also one generalization which may be made regarding the host relationship of all these three common species: those birds which are the most frequently infested and may be termed the favoured hosts are also the most heavily infested. Thus C. gallinae, which can breed in such a wide variety of nests, is found regularly in greatest numbers and obviously 'does best' in the nests of Passer domesticus, Sturnus vulgaris, Parus, and certain Strigidae. This fact is reflec- ted in the number of specimens of C. gallinae collected from individual birds by iy ir 11 10 9 8 7 6* s e 4 3 a* i" o* * 3 WLONO. 9 1E.LONG. FIG. i. Distribution map of C. garei in Britain, based on 108 records. In the case of multiple records from one locality, for example twenty-three separate records of C. gallinae from Blue-Tits' nests in Cambridge, only one circle is used. This accounts for the fact that there are more records than circles on each map. Massed records of this type merely indicate intensive collecting in the area concerned, and have no bearing on the question of distribution. 60' 59 58 55 53' 52 51 LA7 50- y ii it" ip ? a* y* 6 s 4 3* z" i" i" 2* 1OOKM 12 - 1OOKM O 1AT 60 59 58 -57 56' 55' -54" 53" 52' 51 -50 W.LONO. 9" 8 7 6" 5 4 3 2 1 1E.LONG. FIG. 2. Distribution map of C. gallinae in Britain, based on 363 records. IT." tr 10 6o 59 58 5 1 55 54 53 52 51' 50- 10 ?* 8* 7' (f 5* 4 3* 2* 1" 1 2 T 100KM 12 - 10OKM O _,- ,- W. LONG. 9 8" 7 6 5 4 e 3 Z* 1 o" 1'E.LONQ. FIG. 3. Distribution map of D. g. gallinulae in Britain, based on 266 records. A COLLECTION OF FLEAS FROM THE BODIES OF BRITISH BIRDS 217 Meinertzhagen and Pollen, thus six, four, and four specimens from Starlings, four, three, and two specimens from Parus spp., and nine and six specimens respectively from individual Little Owls (Tables i and 9). Bacot (1914) has pointed out that in some cases egg-production of Nosopsyllus fasciatus falls off in drought, and Buxton (1938) has shown that in the case of X. cheopis 'even at one temperature and within tolerable limits of humidity the production of the first filial generation of fleas is closely dependent on humidity, the production being less as the fatal limit is approached'. Edney (1947) has shown that the life of the adult X. brasiliensis is shortened if the pupal stage is kept at low humidities, and these types of phenomena may account for the fact that all three common bird-fleas occur in smaller numbers in some nests than in others, only the favoured hosts providing the optimum conditions. It is of course possible that we are concerned with a combination of many factors, and not only the humidity require- ments of the developmental stages. However, in this fact we can see indications of incipient or vestigial host-specificity either a legacy from the past or a hint of things to come. There are also indications of a third type of host-preference (such as we find in C. vagabunda] which at times seems to cut across the two main trends already mentioned. Undoubtedly certain birds appear to be more attractive or susceptible than others to one or all of the species concerned. Thus the Shag (Phalacrocorax aristotelis) and Cormorant (Phalacrocorax carbo) are placed in the same genus, and build the same type of nest in the same type of habitat, frequently in colonies which adjoin. There are three records of C. garei from the former (including one very heavy infestation where over 500 fleas were counted) and none at all from the latter. There are also more than six records from the Partridge (Perdix perdix} a bird which is not infested with any other flea (Table 5). However, both C. gallinae and D. g. gallinulae as well as C. garei have been collected from the Pheasant (Phasianus colchicus) and Red Grouse (Lagopus scoticus). Here again there would appear to be little if any difference between these three nests, which are no more than sheltered scrapes lined with grass; in fact Pheasants and Partridges not infrequently lay in the same nest. It is also curious that D. g. gallinulae should have a marked predi- lection for the Moorhen (six records), since this bird is not a Passerine and (more important still) favours unusually damp nesting-sites. The records from the Wood-Pigeon (Columba palumbus) suggest that this may also be a case in which the bird itself is attractive to fleas, since the type of nest it builds is exceptionally unsuitable for nidicoles. In Finland Nordberg (1936) has examined the nests of over fifty species of birds. He has shown that nests built in holes and what he terms 'half-holes' (halbholen N ester) support the heaviest populations of nidicoles generally, and also the greatest variety of species. The Stock-Dove (Columba oenas) is one of the most heavily infested of all, with a total of fifty-eight species of nidicoles and 6,573 specimens per i dm. 3 of nesting material. The Wood-Pigeon, with one species and only twenty specimens per i dm. 3 , is unquestionably the least infested out of all those he examined in Finland. In Britain there are only four records of C. gallinae from the Stock-Dove, but from the Wood-Pigeon we have records of the following species : C. gallinae (i but 6 females ENTOM. n. 4 D d 218 A COLLECTION OF FLEAS FROM THE BODIES OF BRITISH BIRDS TABLE 9 C. gallinae, C. garei, and D. g. gallinulae in the British Museum (Natural History) from named Bird Hosts CORVIDAE Corvus corone L. (*Carrion-Crow) C. gallinae, 2 $, N. Wales, vii.sof (N. H. Joy), nest; 7 <$, n $, Ashton, Oundle, 1919 (N. C. Rothschild), nestj; Ashton, Oundle, summer, 1919 (J. Stuart), nest; 3 , Ashton Wold, Oundle, vi.26 (F. J. Cox), nest; < , Netherfield, Sussex, i.vii.oy (W. R. Butter- field) ; 8 , Lochgelly, 25.vi.og (J. Brown); 3 c?> 5 ? Arniston, Midlothian, 28.v.og (A. Bennett), nest; all B.M. ^14.517. FRINGILLIDAE Coccothraustes coccothraustes (L.) (Hawfinch) D. gallinulae, 1^,1$, Eaton, nr. Chester, 5.v.oi (R. Newstead), nest; Eaton, nr. Chester, g.v.oi (R. Newstead), nest; all Rothschild Coll. * Name of host as in Witherby and not necessarily as on slides. Thus Turdus musicus becomes Turdus ericetorum since Waterston's publications make it clear that the Song-Thrush is the host from which he obtained the specimens in question. Certain specimens from Corvus corone are merely marked 'Carrion- Crow'. f With regard to material in the British Museum, the registered number is appended in all cases unless the material referred to pertains to the Rothschild Collection (1923. 615). It must be explained that after Charles Rothschild's death in ig23 material continued to flow in to Karl Jordan, which he worked out and then incorporated in the Rothschild Collection. Although technically British Museum material, the author has made no attempt to separate it from the rest, since she feels that the date is sufficient indication of the fact that it was incorporated after the Rothschild Collection passed into the absolute control of the Trustees. J When it is known for certain if specimens were collected from the body or nest of the host this in- formation is added whether or not it is present on the slides or tubes. A COLLECTION OF FLEAS FROM THE BODIES OF BRITISH BIRDS 219 TABLE 9 (cont.) Chloris Moris (L.) (Greenfinch) C. gallinae, i , Colinton, Midlothian, ii.v.od (J. Waterston), nest; D. gallinulae, i $, Lothianburn, 26.v.o6 (?) ; i <$, i$, Arniston, Midlothian, 28.v.og (A. Bennett), nest; Arnis- ton, Midlothian, S.vi.og (A. Bennett), nest; all B.M. 1914.517 ; 2 9, Abbey St. Bathans, Berwick- shire, I4.v.oy (J. Waterston), nest; 3 $, Tresco, Isles of Scilly, I2.vi.o7 (F. J. Cox), nest; all Rothschild Coll. Carduelis carduelis (L.) (Goldfinch) C. gallinae, < $, < $, Redland, Bristol, 1917 (P. Stonelake) ; C. garei, 2 <$, 2 $, Fancott, nr. Luton, 1.44 (R. B. Laurence), nest; all Roths- child Coll. (B.M.). Carduelis cannabina (L.) (Linnet) C. garei, < cJ, < $, Newport, I.O.W., vi.io (H. G. Jeffrey), nest; Rothschild Coll. (B.M.). Fringilla coelebs L. (Chaffinch) C. gallinae, i (J, Ross-shire, i6.iii.34 (W. H. Pollen), body; Rothschild Coll.; 2 $, Arniston, Midlothian, vi.og (A. Bennett), nest; 3 $, Loudoun, Ayrshire, summer, 1909 (? collector) ; D. gallinulae, 8 , Tring, Herts., vii.o3 (N.C.R.), nest; Tring, Herts., vi.o6 (F. J. Cox), nest; D. gallinulae, i $, Windsor Forest, Surrey, 23.vi.28 (F. W. Frohawk) ; all Rothschild Coll. PLOCEIDAE Passer domesticus (L.) (House-Sparrow) C. gallinae, 3 <$, 5 $, Carie, Rannoch, 24.vii.o8 (F. J. Cox), nest; Rothschild Coll. (B.M.) ; < cj, < $, Mull, Argyllshire, 28.viii.i2 (L. G. Esson) ; Rothschild Coll. (B.M.) ; 8 $, 9 , Wilber- lea, St. Andrews, Fife, i8.viii.o8 (J. Waterston), nest; 3 $, Auchinblae, Kincardineshire, 27^.09 (J. Campbell) ; 4 $, 10 $, Kennetpans, Clack- mannan, 8-io.viii.o8 (J. Waterston), nest; I <$, Gorebridge, nr. Dalkeith, I7.vii.o6 (J. Water- ston), nest; 8 , < $, Colinton, Midlothian, 7.v.o6 (J. Waterston), nest; Colinton, Mid- lothian, n.v.o6 (J. Waterston), nest; 2 , Tring, Herts., 1901 (N.C.R.), nest; Tring, Herts., vi.o6 (K. Jordan), nest; Tring, Herts., ix.io (K. Jordan), nest; < , nr. Haylor, Ronas Voe, Cumberland, I3.vii.i2, nest; all B.M. 1914.517. Motacilla alba yarrelli Gould (Pied Wagtail) C. gallinae, 5 <$, 6 $. Kennetpans, Clack- mannan, 6.viii.o8 (J. Waterston), nest; 3 $, 8 $, Arniston, Midlothian, S.vi.og (A. Bennett), nest; all B.M. 1914.517; 2 $, Westfield, Sussex, 2.viii.o7 (W. R. Butterfield) ; C. garei, 3 , Bere Regis, Dorset, I5.vii.ig (H. Russell); < <$, < $, Dorset, vi.i4 (H. Russell) ; all Rothschild Coll. ; D. gallinulae, 2 (J, 2 $, Cowdenbeath, io.vi.o7 (J. Waterston), nest; i (J, i $, Aberdour, Fife, v.07 (J. Water- ston) ; i $, Arniston, Midlothian, 4.vi.og (A. Bennett), nest; all B.M. 1914.517. Parus coeruleus L. (Blue-Tit) C. gallinae, 8 Reigate, Surrey, I2.vi.oi (C. Reeves) ; all Rothschild Coll. Phylloscopus trochilus (L.) (Willow-Warbler) D. gallinulae, 2 <$, 3 $, Glencorse Pond, 3.vii.o7 (R. Cochrane), nest; i <$, 3 $, Arniston, Mid- lothian, 3~4.vi.og (A. Bennett), nest; all B.M. 1914.517; < <$, < 9, Tring, Herts., vi.28 (K. Jordan), nest; < (J, < $, Tring, Herts., vii.28 (K. Jordan), nest; all Rothschild Coll. Phylloscopus sibilatrix (Bechst.) (Wood-Warbler) D. gallinulae, < $, < $, Arniston, Midlothian, vi.og (A. Bennett), nest; B.M. 1914.517. Acrocephalus schoenobaenus (L.) (Sedge- Warbler) D. gallinulae, 2 $, Tring, Herts., v.34 (L. Goodson), nest; Rothschild Coll. Sylvia borin (Bodd.) (Garden Warbler) D. gallinulae, 3 (J, 4 ?, Arniston, Midlothian, 25.vi.og (A. Bennett), nest; B.M. 1914.517. Sylvia communis Lath. (Whitethroat) D. gallinulae, i $, Skokholm Bird Observa- tory (R. M. Lockley), body; Rothschild Coll. Sylvia curruca (L.) (Lesser Whitethroat) D. gallinulae, 9 $, 10 $, Tansor Wold, Oundle, v.o6 (F. J. Cox), nest; Rothschild Coll. TURDIDAE Turdus viscivorus L. (Mistle Thrush) C. gallinae, 1^,1$, Colinton, Midlothian, 24.v.o6 (J. Waterston) ; D. gallinulae, i (J, 6 $, Hillend, Fife, 25.v.o8 (R. Cochrane), nest; 2 9, Loganlea Waterfall, 25^.07 (R. Cochrane), nest; all B.M. 1914.517. Turdus ericetorum Turton (Song-Thrush) C. gallinae, i <$, Colinton, Midlothian, 7.v.o6 (J. Waterston), nest; B.M. 1914.517; i $, Nethercraigs, Paisley, 4.v.og (J. Waterston), joint nest with T. merula; B.M. 1914.517; < <$, < $, Loch Ranza, Arran Isle, N.B., vi.27 (F. J. Cox), nests; 2 $, Tring, Herts., io.xi.i4 (E. Hartert) ; all Rothschild Coll. ; D. gallinulae, 2 (J, Kinneff, Kincardineshire, I4.vi.og (J. Water- ston), body; i , Carie, Rannoch, 23.vii.o8 (F. J. Cox), nest; Rothschild Coll.; i $, Woodburn House, nr. Dalkeith, I2.v.o8 nest; i <$, 3 $, Colinton, Midlothian, 15.^.07 (J. Waterston), nests; Colinton, Midlothian, 26.vii.og (J. Waterston), body; i $, Nether- craigs, Paisley, 4.v.og (J. Waterston), joint nest with T. ericetorum; all B.M. ^14.517; i $, Armitage, Staffordshire, 1.33 (L. C. E. Stott), nest; 5 $, Tring, Herts., vii.igoo (N.C.R.), nest; Tring, Herts., I5.vii.ii (H. Mugford) ; 4 , Colinton, nr. Edin- burgh, 24.v.o6 (J. Waterston); Colinton, nr. Edinburgh, 25.v.o6 (J. Waterston); all Roths- child Coll. ; 4 cj, 2 $, Arniston, Midlothian, 3.vi.og (A. Bennett), nest; Nethercraigs, Paisley, 4. v.og, joint nest with T. musicus (= ericetorum) ; all B.M. igi4.5i7; i ^, i $, Tring, Herts., 222 A COLLECTION OF FLEAS FROM THE BODIES OF BRITISH BIRDS TABLE 9 (cont.) i5.vii.n (H. Mugford) ; i <$, Newport, I.O.W., iv.io (H. G. Jeffrey), nest; all Rothschild Coll. Oenanthe oenanthe (L.) (Wheatear) D. gallinulae, i $, Gluss Voe, west side, Olla- berry, 28.iv.i3 (J. Waterston), body; B.M. 1914.517. Saxicola torquata (L.) (Stonechat) D. gallinulae, < $, < $, Loudoun Estate, Ayr- shire, I2.vi.og (J. Gloag), nest; B.M. 1914.517. Phoenicurus phoenicurus (L.) (Redstart) C. gallinae, i $, Gorebridge, nr. Dalkeith (? collector), nest; <$,<$, Arniston, Mid- lothian, 25.vi.o9 (? collector), nest; D. gallinulae, 9 $, < $, Arniston, Midlothian, i.vii.o6 (A. Bennett), nest; Arniston, Midlothian, 28.v.og (A. Bennett), nest; all B.M. 1914.517. Erithacus rubecula (L.) (Robin) C. gallinae, i <$, Auchinblae, Kincardineshire, I7.iv.og (J. Duff us), nest; D. gallinulae, i $, 1 $, Auchinblae, Kincardineshire, i7.iv.og (J. Duffus),nest; all B.M. 1914.517; 3 $, 3 $, Hillend, Fife, 2g.v.o6 (? collector), nest; 3 $, Flotterstone, Fifeshire, 2g.v.o6 ; 5 ^, 2 , Colinton, Midlothian, vi.o6 (J. Waterston), nest; all B.M. 1914.517; 2 (J, 8 $, Carie, Rannoch, N.B., vii.oS (F. J. Cox), nest ; i $, Banbury, 4.xii.34 (W. H. Pollen), body ; 2 $, < $, Loch Ranza, Arran Isle, vii.27 (F. J. Cox), nests; < , Colinton, Mid- lothian, I3.v.o6 (J. Waterston) ; Colinton, Mid- lothian, 24-v.o6 (J. Waterston); Colinton, Midlothian, ig.v.og (J. Waterston) ; Colinton, Midlothian, 3i.vi.og (J. Waterston), body; all B.M. 1914.517; i $, Dumbarton, Scotland, 2i.ix.34 (W. H. Pollen), body; Rothschild Coll. ; 3 cJ, 4 $, Arniston, Midlothian, vi.og (A. Bennett), nest; B.M. 1914.517; < <$, < $, Talacre, nr. Prestatyn, i6.v.o3 (R. Newstead) ; i $, Newport, I.O.W., v.io (H. G. Jeffrey), nest; all Rothschild Coll. TROGLODYTIDAE Troglodytes troglodytes (L.) (Wren) C. gallinae, i $, Larnba, Shetland, 25.ix.i2 (J. Waterston), nest; B.M. ^14.517; 6 (J, 7 $, Ashton, Oundle, vi.o6 (F. J. Cox), nest; 2 <$, < $, Tring, Herts., 5-x.oi (N.C.R.), nest; D. gallinulae, 2 $, i $, St. Kilda, igoo (W. Eagle Clarke) ; all Rothschild Coll. ; i $, Gluss Voe, Ollaberry, 2o.ix.io (J. Waterston), body; 4 <$, < $, between Hillend and Flotterstone, Fife- shire, 2g.v.o6 (J. Waterston), nest; 2 $, 3 $, nr. Loganlea Waterfall, 25^.07 ( J. Waterston), nest ; all B.M. 1914.517; i $, Lesmahagow, Lanark- shire, 5.viii.o6 (J. Clark, per Watkins and Don- caster), nest; Rothschild Coll.; i <$, Pettadale Water, nr. R. Roe, 24.iv.i2 (J. Waterston), body; B.M. 1914.517; i <$, i , Cardiff, i4.vi.3i (H. M. Hallett, A. H. Salmon); 2 ?, Tring, Herts., vi.oi (N.C.R.), nest; all Rothschild Coll. CINCLIDAE Cinclus cinclus (L.) (Dipper) C. gallinae, i $, Auchinblae, Kincardineshire, I4.iv.og (L. Duffus), nest; i $, Abbey St. Ba- thans, Berwickshire, 26.V.O7; all B.M. ^14.517; i (J, i $, Rothbury, Sunderland, vii.o6 (per Watkins and Doncaster), Rothschild Coll. ; D. gallinulae, g <$, < $, Torduff Reservoir, Mid- lothian, i2.v.og (J. Waterston), nest; B.M. 1914. 517:1$, nr. Granthouse, 14^.07 (J. Waterston), nest; B.M. 1914.517; i 9. Nantyffrith, nr. Min- era, Flintshire, 5-v.oi (R. Newstead) ; Roths- child Coll. HIRUNDINIDAE Hirundo rustica L. (Swallow) C. gallinae, 2 <$, 7 $, Dirleton, East Lothian, 2.ix.og (J. Gardiner), nest; 3 $, 7 $, Howletburn, Galston, Ayrshire, 4-ix.io (J. Gloag), nest; all B.M. igi4.5i7; 8 ^, < $, Newport, I.O.W., igi2 (H. G. Jeffrey), nest; Newport, I.O.W., igi4 (H. G. Jeffrey), nest; all Rothschild Coll. Delichon urbica (L.) (House-Martin) C. gallinae, 7 , 5 $, Mull, Argyllshire, ix.i2 (L. G. Esson) ; Mull, Argyllshire, 7.1.13 (L. G. Esson) ; all Rothschild Coll. ; 2 Ashton Wold, Oundle, vi.26 (F. J. Cox), nest; < Berkshire, n.iv.34 (W. H. Pollen), body; Rothschild Coll. Columba oenas L. (Stock-Dove) C. gallinae, i 9 Colinton, Midlothian, ig.v.og (J. Waterston), nest; Rothschild Coll. Columba livia Gm. (Rock-Dove) C. gallinae, < cJ, < 9. Little Roe, Yell Sound, Shetland, I7.vii.n (J. Waterston), nest; i 9 Caves in cliffs, Todhead, Kinneff, i5.vi.og (J. Waterston), nest; all B.M. 1914.517; i 9, Belfast, Ireland, 24^.35 (W. H. Pollen), body; C. garei, 4 $, < 9 Little Roe, Shetland, I7.vii.n (J. Waterston), nests; D. gallinulae, 4 , 5 9 in a cave NE. end of Little Roe, Shetland, i7.vii.n (J. Waterston), nest; all Rothschild Coll. SCOLOPACIDAE Scolopax rusticola L. (Woodcock) D. gallinulae, 8 <$, < 9. Aberdour, Fifeshire, 25^.07 (J. Waterston), nest; B.M. 1914.517. Capella gallinago (L.) (Snipe) C. garei, i 9> Northmavine, Shetland (R. H. McNair) ; 5 9> Tankerness, Orkney, 8.vi.o6 (J. Waterston), nest; < Aberlady, nr. Edin- burgh, 22.V.O9 (J. F. Cormack), nest; all B.M. 1914.517. Vanellus vanellus (L.) (Lapwing) C. garei, 2 $, Tents Moor, Fifeshire, I2.vi.og (J. F. Cormack), nest ; 4 $, 5 9, Lochhead, Loch- gelly, Fifeshire, 25.iii.og (J. Brown), nest; 5 Orkney Islands, vi.n (N. H. Joy), nest; Rothschild Coll. Larus canus L. (Common Gull) C. garei, < (J, < 9> Little Roe, Shetland, v.n (J. Waterston), nest; B.M. 1914.517. Larus fuscus L. (Lesser Black-backed Gull) C. garei, 3 Walls, Shetland, viii.oG (per Watkins and Doncaster), nest; Rothschild Coll. STERCORARIIDAE Stercorarius parasiticus (L.) (Arctic Skua) C. garei, < Arniston, Midlothian, 26.vii.o6 (A. Bennett), nest; i < ^ < ^ Tring> Herts ( summe r f Phasianus colchicus L. (Pheasant) 1911 (F. J. Cox), nest; Rothschild Coll. C. gallinae, i Rutland County Report on Wild Birds for 1944: 1021. JORDAN, K. 1925. New Siphonaptera. Novit. zool. 32: 96-112, figs. 1-46. 1926. New Siphonaptera. Novit. zool. 33: 385-394; figs. 1-22. 1928. Siphonaptera collected in the Dolomites. Novit. zool. 34: 173-177; figs. 1-6. 19290. On some problems of distribution, variability and variation in North American Siphonaptera. Trans. 4th Int. Congr. Ent. (1928), 2: 489-499; figs. i-n. 19296. Further records of North American bird fleas, with a list of the Nearctic birds from which fleas are known. Novit. zool. 35: 89-92. I929C. Notes on North American fleas. Novit. zool. 35: 28-39; 2 pis. 19290*. On a small collection of Siphonaptera from the Adirondacks, with a list of the species known from the state of New York. Novit. zool. 35: 168-177; n s - 5~8. 1932. Die aus der arktischen Zone bekannten Flohe. Fauna Arctica, Jena, 6 (2) : 115-118. 1937- Three new bird fleas from Kashmir. Novit. zool. 40: 299-306; figs. 87-92. and ROTHSCHILD, N. C. 1911. List of Siphonaptera collected in Portugal. Novit. zool. 18: 551-554: figs. 1-3. 19200. On American Bird-Ceratophylli. Ectoparasites, 1: 65-76; figs. 65-72. 19206. A preliminary catalogue of the Siphonaptera of Switzerland. Ectoparasites, 1: 78-122; figs. 75-101. MALLOCH, J. R. 1907. *Diptera in Dumbartonshire in 1906. Ent. mon. Mag. 43: 86. (Fleas determined by N. C. Rothschild.) NORDBERG, S. 1934. Aphanipterologisches aus Finnland. I. Verzeichnis der bisher in Finnland gefundenen Aphanipteren nebst Diagnosen iiber 8 fur die Wissenschaft neue Arten. Memor. Soc. Fauna Flor. Fenn. 10: 354-369; 15 figs. 1936. Biologisch-okologische Untersuchungen iiber die Vogelnidicolen. Ada Zool. Fenn. 21: 1-168; 4 figs., 20 diagrams. NUTTALL, G. H. F., STRICKLAND C., and MERRIMAN, G. 1913. ""Observations on British rat- fleas. Parasitology , 6: 1-19. O'MAHONY, E. 19390. *A preliminary list of Irish fleas. Ent. mon. Mag. 75: 124-126. 19396 *Notes on Irish Siphonaptera, I. Ent. mon. Mag. 75: 253-254. 19410. *Notes on Irish Siphonaptera, III. Ent. mon. Mag. 77: 208. 19416. *Notes on Irish Siphonaptera, IV. Ent. mon. Mag. 77: 231. 1948. *Some miscellaneous Siphonaptera. Ent. mon. Mag. 84: 89. 19490. *The fleas of North Bull Island, Dublin Bay. Ent. mon. Mag. 85: 139-140. 19496. * Ectoparasites from Fair Island. Ent. mon. Mag. 85: 140. PHILIP, C. B. 1938. A parasitological reconnaissance in Alaska with particular reference to varying Hares. II. Parasitological data. /. Parasit. 24: 483-488. (C. vagabunda Boh. determined by K. Jordan.) ROTHSCHILD, M. 1947. *A note on Ceratophyllus rusticus Wagner, 1903 (Siphonaptera) at Ashton Wold. Entomologist, 80: 253-255. 1948. *Bird fleas collected by Miss Theresa Clay, Colonel Richard Meinertzhagen, and Captain W. H. Pollen in the Island of Ushant, Brittany, France, with a note on the distribution of Ceratophyllus borealis Rothschild, 1907. Entomologist, 81: 84-95; figs. 1-3. and CLAY, T. (in press). *Fleas, flukes and cuckoos. (New Naturalist Series.) London. ROTHSCHILD, N. C. 1900. *Notes on Pulex avium Tasch. Novit. zool. 7: 539-543- 1901. *A new British flea (Ceratophyllus newsteadi). Ent. Rec. 13: 284; i pi. 1902. *New British fleas. Ent. mon. Mag. 38: 225; figs. 1-7. (Ceratophyllus garei.) * Papers containing records of British bird-fleas. A COLLECTION OF FLEAS FROM THE BODIES OF BRITISH BIRDS 231 ROTHSCHILD, N. C. 1903. Types of Siphonaptera in the Daleian collection. Ent. man. Mag. 39: 144146. 1906. *A new British flea: Ceratophyllus insularis, spec. nov. Ent. mon. Mag. 42: 5960, i pi. 1912. A note on Ceratophyllus vagabundus, Boheman. Ent. mon. Mag. 48:67. - 1915. *A synopsis of the British Siphonaptera. Ent. mon. Mag. 51: 49-112; pis. vii-xiv. SALMON, H. M. 1931. *The Manx Shearwater's flea. Brit. Birds, London, 25: 171. SHIPLEY, A. E. 1909. *The ectoparasites of the Red Grouse (Lagopus scoticus). Proc. zool. Soc. Land. 1909: 39~334. P ls - 35~47- *Skokholm Bird Observatory Reports for 1938, p. 5, and 1939, p. 10. (Siphonaptera determined by C. Oldroyd and G. B. Thompson.) SLATER, H., 1925. *Proc. Somersetsh. Archaeol. Nat. Hist. Soc. 70: liv. 1927. *Proc. Somersetsh. Archaeol. Nat. Hist. Soc. 72: Ixviii. SMIT, F. G. A. M. 1949. Monstrosities in Siphonaptera. Tijdschr. Ent. (1946), 90: 35-42; figs. 1-2. THOMPSON, G. B. 19370. *The parasites of British birds and mammals. XII. On some parasites from the burrows of Puffins. Ent. mon. Mag. 73: 87-88. 19376. *The parasites of British birds and mammals. XIII. Records of Siphonaptera bred from birds' nests. Ent. mon. Mag. 73: 105-107. !937C. "The parasites of British birds and mammals. XV. Bird fleas and their hosts. Ent. mon. Mag. 73: 137-142. TURK, F. A. 1946. *The Siphonaptera of Cornwall and the Isles of Scilly with the description of a new subspecies. Ent. mon. Mag. 82: 97-100; i fig. VERDCOURT, B. 1945. *Records of Bedfordshire Siphonaptera. Ent. mon. Mag. 81: 74. *Victoria History of the County of Hertford, 1920: 170. "Victoria History of the County of Oxford, 1939: 178. WAGNER, J. 1929. tJber neue palaearktische Floh-Arten ( Aphaniptera) . I. Annu. Mus. zool. Akad. Leningrad, 30: 21-33; T 3 n s - 1930. Katalog der palaearktischen Aphanipteren. 55 pp. Wien. WAHLGREN, E. 19030. Aphanipterologische Notizen, nebst Beschreibung neuer Arten. Ark. Zool. 1: 181-196; pis. 7-9. 19036. t)ber Pulex vagabunda Bohem. Ent. Tidskr. 24: 219. 1907. Svenska Siphonaptera. Ent. Tidskr. 28: 85-91 ; figs. 1-2. WALSH, G. B. 1938. *A further contribution to a list of Yorkshire Siphonaptera. Naturalist, Land. 977: 185-187. WATERSTON, J. 1906. *On some Scottish Siphonaptera. Ann. Scot. Nat. Hist.: 211-214. 1909. *On some Scottish Siphonaptera. Ann. Scot. Nat. Hist. 18: 226-228. 1910. *On some habits and hosts of bird Ceratophylli taken in Scotland in 1909, with description of a new species (C. rothschildi) and records of various Siphonaptera. Proc. R. phys. Soc. Edinb. 18: 73-91. 1914. *Some records of Scottish Siphonaptera. Ent. mon. Mag. 50: 88-91 and 159-166. 1916. Fleas as a menace to man and domestic animals, their life-history, habits and control. Brit. Mus. (Nat. Hist.) Econ. Ser. 8:21 pp. ; 6 figs. 1923. *On the occurrence, near London, of the flea Ceratophyllus vagabundus Boh., under unusual circumstances. Trans. Ent. Soc. Lond. (1922), parts hi, iv: 454-460; figs. 1-2, i pi. WILLIAMSON, K. 1940. *Numbers of Black Redstarts on passage in Man. Brit. Birds, 33 (9) : 252-254. * Papers containing records of British bird-fleas. 232 A COLLECTION OF FLEAS FROM THE BODIES OF BRITISH BIRDS LIST OF TABLES TABLE i : Fleas collected by Colonel Richard Meinertzhagen and Miss Theresa Clay from the bodies of British birds. TABLE 2: Records of C. vagabunda Boheman (1866). TABLE 3 : British bird hosts of C. gallinae, C. garei, and D. g. gallinulae. TABLE 4: Comparative records of C. gallinae, C. garei, and D. g. galli- nulae (compiled from all known British records). TABLE 5: Selected hosts on which C. garei is the dominant species of flea. TABLE 6: Selected hosts on which C. gallinae is the dominant species of flea. TABLE 7: Selected hosts on which D. g. gallinulae is the dominant species of flea. TABLE 8: British records of C. gallinae, C. garei, and D. g. gallinulae collected away from any host or nest. TABLE 9: C. gallinae, C. garei, and D. g. gallinulae in the British Museum Collection from named bird hosts. TABLE 10: Monthly records of C. gallinae, C. garei, and D. g. gallinulae from the bodies of bird hosts in Britain. TABLE 1 1 : Proportion of birds infested with fleas. (Collected by Colonel Meinertzhagen and Miss T. Clay.) TABLE 12 : Proportion of Great Tits (Parus major newtoni) infested with fleas. (Collected by Colonel Meinertzhagen and Miss T. Clay.) PRESENTED 7 FEB 1952 PRINTED IN GREAT BRITAIN AT THE UNIVERSITY PRESS OXFORD BY CHARLES BATBY PRINTER TO THB UNIVERSITY .1 7 APR 1952 THE SUB-GENUS STEGOMYIA (DIPTERA : CULICID AE) IN THE ETHIOPIAN REGION (PART I) P. F. MATTINGLY BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ENTOMOLOGY Vol. 2 No. 5 LONDON : 1952 THE SUB-GENUS STEGOMYIA (DIPTERArCULICIDAE) IN THE ETHIOPIAN REGION I. A PRELIMINARY STUDY OF THE DISTRIBUTION OF SPECIES OCCURRING IN THE WEST AFRICAN SUB-REGION WITH NOTES ON TAXONOMY AND BIONOMICS BY P. F. MATTINGLY Pp. 233-304; 16 Text-figures BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ENTOMOLOGY Vol. 2 No. 5 LONDON : 1952 THE BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY), instituted in 1949, is issued in five series, corresponding to the Departments of the Museum. Parts appear at irregular intervals as they become ready. Volumes will contain about three or four hundred pages, and will not necessarily be completed within one calendar year. This paper is Vol. 2, No. 5 of the Entomology series. PRINTED BY ORDER OF THE TRUSTEES OF THE BRITISH MUSEUM Issued April 1952 Price Fifteen Shillings. THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) IN THE ETHIOPIAN REGION I. A PRELIMINARY STUDY OF THE DISTRIBUTION OF SPECIES OCCURRING IN THE WEST AFRICAN SUB-REGION WITH NOTES ON TAXONOMY AND BIONOMICS By P. F. MATTINGLY CONTENTS INTRODUCTION ........ 235 NOTES ON TAXONOMY . . . . . . 238 Key to adults ....... 244 Key to larvae ....... 247 DISTRIBUTION RECORDS ...... 249 Doubtful records . . . . . . -257 Records based on misidentifications . . .258 Unidentified and misquoted localities . . . 258 Distribution outside the Ethiopian Region . . 259 LIST OF LOCALITIES WITH TOPOGRAPHICAL DETAILS . 259 BIONOMICS IN RELATION TO DISTRIBUTION . . . 269 Breeding-places . . . . . . .269 Seasonal distribution . . . . . .271 Biting-habits ....... 275 ZOOGEOGRAPHY ....... 277 SUMMARY ........ 293 ACKNOWLEDGEMENTS ...... 294 REFERENCES ........ 295 APPENDIX ........ 303 SYNOPSIS This paper deals with the distribution of the species of Stegomyia occurring in the West African sub- region. All the species of wide distribution in the Ethiopian Region are, however, included and the topo- graphical data therefore cover most of the region. Aides aegypti is treated very briefly since it is regarded as a special case meriting a separate paper. Zoogeography is discussed mainly in relation to rainfall and altitude. Temperature and vegetational relationships will be discussed in later papers. Such notes on taxonomy and bionomics are included as appear to be necessary for an understanding of distribution. These also will be discussed at greater length in later papers in the series. INTRODUCTION THE objects of the present study are firstly to render possible a more rational approach to the taxonomy of the sub-genus Stegomyia, and incidentally of other African insects, and secondly to discover what light may be thrown by this comparatively well-known group on general problems concerning the distribution of the Ethiopian fauna. It is perhaps not too much to hope that some small quota may be added to 236 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) the growing accumulation of evidence regarding past climates in Africa which is of fundamental importance for the understanding of phylogeny and of the human, plant, and animal geography of the continent. 1 Ernst Mayr (1942) has clearly shown that it is as a population that the species is encountered in nature, and it is upon population studies that our knowledge must ultimately be based. A taxonomy based solely on the arbitrary selection of morphological types, while indisputably essential, can only be of ancillary importance. In the author's opinion the greatest need for mosquito studies at the present time is the application of genetical methods (on this point see Bates, 1949). The upper limit to our knowledge of distribution is imposed by the supply of collectors. In the present instance records are available from more than 600 localities (Fig. i), but not all of these have been investigated with equal thoroughness. In general the British territories have rendered the best service, but it is unhappily true that the most serious gap in our knowledge at the present time is provided by Tanganyika, of which only a narrow fringe along the northern border and along the coast is known. Vast areas of unknown territory exist in Angola, the Cape Province, and French Equatorial Africa, but these are so situated that discoveries there would be likely to add to rather than to integrate our existing knowledge. From the phylogenetic point of view most knowledge could probably be gained from the study of high-altitude faunas. At present only those of Ruwenzori, the Kigezi district, and a few parts of Kenya are known. The present paper deals only with those species occurring in the West African sub-region. Those which are known only from the East and South African sub-regions will be dealt with in a later communication. This does not mean that only the West African districts are considered. All the Stegomyia species of wide distribution in Africa enter this sub-region, and these species must necessarily be considered in relation to their range as a whole if their distribution is to be understood. The dis- tribution of one species, A edes aegypti, is so artificial and so many records are available that to deal with these in full, as has been done for the other species, would serve little useful purpose, besides taking up a wholly disproportionate amount of space. Details of the distribution of this species may be found in Kumm (19316) supplemented by Edwards (1941) and by the notes on recent records given below. Some small attention is paid, in the section on breeding-places, to the occurrence of wild popula- tions. It is possible that some idea of the natural distribution of the species, which is almost certainly Ethiopian in origin, might be gained if more records of this kind were available. Apart from one or two important collections which the author has not as yet had the opportunity to examine, all known sources of distribution records have been tapped. It is not believed that the collections in question will add to any great extent to our knowledge, as much of the material which they contain has been published. Great difficulty has been experienced in tracing many localities and in ascertaining the known facts regarding their climate and topography. In order to save others this labour the best available data have been included in tabular form. The major factors affecting distribution may be considered to be rainfall, temperature, humidity, 1 For a summary of this work see Zeuner (1944). IN THE ETHIOPIAN REGION 237 and vegetation. The last named is itself dependent to so large an extent on the other factors that it is preferred to discuss it separately in a later communication. It is here referred to only where, as in the case of the Guinean forests, it appears to con- stitute a gross physical barrier to distribution. Attention has been paid to the possible influence of human transportation, but this is considered to be only a minor 5.0 20 . t FIG. T. Distribution of collectors of Stegomyia spp. other than Aedes aegypti. factor. Its importance for the mosquitoes is very liable to exaggeration and only the merest handful of species can be shown ever to have become established outside their natural boundaries, while very striking cases of failure to become established even under apparently favourable conditions can be quoted. Among the species here discussed only the domestic Aedes aegypti, and Aedes simpsoni which commonly breeds in the leaf axils of food plants, show any sign of having been affected in this way. Man as a destroyer of forests is perhaps more important, but this factor has relatively little significance for the present study, which is concerned rather with distribution at the present time and the factors by which this is controlled than with distribution in the past. The consideration of temperature and humidity has had to 238 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) be postponed for a subsequent communication owing to the relative inaccessibility of the necessary data, but altitude is considered for the sake of the light which it throws on temperature limits. Rainfall, for which quite abundant figures have recently become available, is considered in full, and rainfall boundaries are found to provide a surprisingly, in some cases almost dramatically, accurate fit with known Stegomyia distributions, especially when the seasonal distribution of rain is taken into account. It must, however, be emphasized that the observations here recorded are purely empirical. It is not claimed that rainfall ^>r se governs distribution. On the contrary, such evidence as is available suggests that for the most part it operates indirectly through its influence on vegetation. Detailed taxonomic study of the sub-genus is in progress and is not yet complete. The treatment here adopted is therefore largely conventional except as regards dendrophilus , which has been the subject of much confusion. Work already carried out has, however, revealed a considerable number of errors in published descriptions, and these are corrected in the notes on taxonomy. The keys to adults and larvae cover all Ethiopian members of the sub-genus and have been made as up-to-date as possible. No descriptions of new species are at present known to be in the press, but the position of a new form from the East African Highlands is under consideration and will be discussed in the second paper. The section on bionomics is confined to those aspects of the subject having an obvious and immediate bearing on distribution. In this connexion it may be said that the most serious gap in our knowledge concerns seasonal distribution, the proper understanding or even the mere recording of which would illuminate every aspect of mosquito studies. NOTES ON TAXONOMY Edwards (1932) divided the sub-genus Stegomyia into four groups. Of these, group A contains all the purely Ethiopian mainland species and the cosmotropical Aedes aegypti. Group B contains certain Oriental and Palearctic species -and Aedes mas- car ensis of Mauritius. Group C contains Oriental and Australasian species and Aedes albopictus, granti, and unilineatus, part of whose range is Ethiopian. Group D contains only Aedes vittatus, which is Palearctic, Oriental, and Ethiopian, and shows certain characters anectant to the sub-genus Aedimorphus. In so far as this grouping was based on the male terminalia it appears to have been a good one, but colour charac- ters in Aedes are somewhat unsatisfactory and mascarensis should perhaps have been placed in group A. In any case a new system of grouping has been rendered necessary by the discovery of Aedes amaltheus (De Meillon & Lavoipierre, 1944), which is anectant between groups A and B, having the colour characters of one and the male terminalia of the other, the discovery of certain aberrant species of Stegomyia in the Mariana Islands (Bohart & Ingram, 1948), and the proposed transfer to this sub- genus of certain species hitherto assigned to Armigeres (Brug & Bonne- Wepster, 1947). Such a regrouping is, however, beyond the scope of the present paper, and the following notes refer only to matters of detail. Aedes apicoargenteus. Specimens from Kampala have a number of pale scales on the lateral lobes of the scutellum. Edwards (1941) states that in the 'typical form' IN THE ETHIOPIAN REGION 239 these lobes are entirely black-scaled but gives no further explanation. This has led to some confusion. Other specimens with the lateral lobes white-scaled are recorded from Kenya (Van Someren, 1946), and Wolfs (1949) notes that they are white-scaled in ssp. denderensis from Costermansville. This is a rather unstable species, perhaps of comparatively recent origin, and the occurrence of this type of peripheral variation does not seem to have much taxonomic significance. It is, however, noteworthy that pale scaling has only been recorded from the higher parts of the range, above 3,000 ft. As will be shown below, apicoargenteus seems to be more strictly confined to low altitudes than, for example, africanus, whose range is otherwise similar. Its highland counterpart, schwetzi, has white lateral lobes. The occurrence of pale scales on the lateral lobes in apicoargenteus unfortunately renders separation from fraseri rather more difficult since this species frequently exhibits a number of dark scales in this position (see below) . The extent of the pale area on the posterior surface of the hind femur is used by Edwards (1941) as a key character, but this is so variable in different parts of the range as to be extremely unreliable. The best diagnostic character from fraseri, as pointed out by Van Someren (1946), is the metallic sheen on the pale markings of the femora; schwetzi is easily separated by the very extensive pale scaling in the pre-scutellar region, but the differences in the male terminalia given by Edwards (1941) do not appear to be significant. The twisting of the style observed in schwetzi is due merely to the angle from which this was drawn and apicoargenteus also shows it when drawn from the same angle. The number and length of the bristles on the basal lobe is very variable in apicoargenteus and overlaps with that found in schwetzi. An average figure would perhaps be 8, but some specimens may have as many as 12. West coast specimens sometimes have fewer bristles than those from farther East, but the difference is not constant. The only genitalic character which seems to be of much diagnostic value concerns the IXth tergite, which is rather more deeply excavated in schwetzi, but even here there is some variation in both species. It is rather doubtful whether schwetzi should be treated as more than a sub-species, but further material is required from intermediate localities before the point can be settled. The larvae of apicoargenteus, fraseri, and schwetzi are all very similar and closely resemble those of the East African calceatus, soleatus, and langata. As far as can be ascertained from available material, apicoargenteus and fraseri can generally be separated from the other species on the position of the antennal bristle. In fraseri this arises at 0-82-0-85 x the distance from base to apex of the antenna. In apico- argenteus the corresponding figure is 0-67-0-82. Specimens of this species from Uganda show a rather greater average displacement from the base than do those from Nigeria, Cameroons, and Belgian Congo, and this may be an altitudinal effect comparable to that described in connexion with the scutellar lobes. Average positions in material so far available are 0-80 for Uganda, 0-74 for Nigeria, and 0-75 for Belgian Congo and Cameroons, but the samples are too small for proper statistical analysis. In all specimens of the calceatus group examined the figure is less than 0-70. schwetzi overlaps the two groups. As noted by Van Someren (1946), those larvae of fraseri so far examined can be separated from apicoargenteus by the greater reduction of the median denticle of the comb spines (with proportional emphasis of the lateral fringe) , 240 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) but all specimens so far available come from Kenya and it is by no means certain that this character would be diagnostic for West Coast larvae. From all other members of the group, except ssp. denderensis , calceatus can be separated by the possession of a wholly dark siphon. This character is not mentioned in the type description of denderensis (Wolfs, 1949), but it is shown by both larvae available for examination and is very probably constant. It is a striking character, and although the differences from the type form with respect to male terminalia given in the type description are not constant, it is considered that this form should be named as a sub-species. Further material from the Kivu Highlands and Ruanda-Urundi would, however, be very welcome. Differences in colour characters between the sub-species and West Coast specimens of the type form include smaller anterolateral scutal patches, more frequent occurrence of pale scaling on the lateral lobes of the scutellum, narrower pale bands on third hind tarsi, and reduced pale scaling of fifth hind tarsi. None of these are, however, absolutely diagnostic from Kenya and Uganda specimens of the type form. Aedes fraseri. This species has already been discussed, but a further word is perhaps necessary regarding the scaling of the lateral lobes of the scutellum. Edwards (1941) states that the scales in this position are 'always white', a statement belied by his figure (Fig. 39). In fact some dark scales are usually present. It is probable that West Coast apicoargenteus can be reliably distinguished from fraseri by having the lateral lobes wholly or very largely black, but the distinction does not hold good in the eastern part of their range. As noted by Edwards (1941) and Van Someren (1946), the shape of the anterolateral pale scutal patches and the extent of the pale area on the posterior surface of the hind femur are subject to much variation. Aedes dendrophilus. This species has been misidentified by Van Someren (1946) and others by reason of the misleading description given by Edwards (1941). This description should be ignored and that given by Van Someren (loc. cit.) substituted for it. Edwards appears to have been misled by the very striking sexual dimorphism occurring in dendrophilus. This is an unusual phenomenon in Stegomyia, although a small degree of dimorphism with respect to, for example, the abdominal tergites is common. An even greater degree of dimorphism characterizes some oriental F inlay a, however (see photographs in Barraud, 1934). The characteristic markings of the male as described by Van Someren for the Kenya and Uganda forms are also shown by the Gold Coast form, and Edwards's type $, although damaged, shows them well. By reason of the same error of observation Edwards failed to recognize the identity of dendrophilus with Aedes trinidad. Through the kindess of Dr. Ceballos and Dr. Peris of the Madrid Museum it has been possible to examine the two cotype males of this species and it is clearly conspecific with dendrophilus. It should be noted that the white spot on the clypeus referred to in the type description of trinidad (Gil Collado, 1935) is due to frosting of the integument and not to the presence of scales. Aedes aegypti thus remains distinguishable from all other Stegomyia by the presence of scales in this position. A careful comparative study reveals that Edwards's type series of Aedes demeilloni from Eshowe, Zululand, contains a specimen of dendrophilus which can be recognized by-the character given in the key. The correctness of this diagnosis was suggested by the fact that the specimen in question was bred out from IN THE ETHIOPIAN REGION 241 a tree-hole, whereas all those of demeilloni came from axils of Dracaena or 'Fern Tree '. Both the identity of the Eshowe specimen with dendrophilus and the distinctiveness of the breeding-places of the two species have been confirmed by the receipt of further specimens of both sexes kindly sent by Dr. De Meillon, and Mr. Muspratt has made extensive collections of both in Natal which add further confirmation. This curious but by no means unique distribution is discussed more fully in the appropriate section. The Kenya-Uganda, Sierra Leone-Gold Coast-Nigeria, Fernando Po, and South African forms all show minor differences in colour markings, and the latter at least should perhaps be named as a sub-species. The point will be discussed in a later paper to be devoted to taxonomy. The problem is complicated by the fact that Edwards's type series of dendrophilus consists of a number of dwarfed specimens, some with slightly atypical markings. These have the appearance of specimens bred out in the laboratory on an inadequate larval diet. The presence of thickened, spine- like setae on the inner angle of the basal lobe of the male terminalia of dendrophilus is not sufficiently indicated either by Edwards's drawing or by his figure. Such spines are characteristic of a number of species in this group (dendrophilus, deboeri, bambusae, heischi, demeilloni, keniensis, angustus) which cannot safely be separated either by the degree of development of the spines or the shape of the lobe, both of which are very variable. The larva of Aedes deboeri is not constantly separable from that of dendrophilus. Kenya and Uganda specimens of the latter commonly have the sub- ventral seta of the siphon single (usually double in deboeri), but it is frequently double in the Gold Coast form. The best partial character appears to be the possession of more than one pecten tooth beyond the subventral seta. This is of freqent occurrence in deboeri but has never been observed in dendrophilus. In bambusae the subventral seta is commonly bifid but may be single, and there may be either i or 2 pecten teeth beyond it. Aedes africanus. Edwards (1941) gives a number of striking colour differences between this species and luteocephalus. Most of these differences are, however, very inconstant and this has led Haddow and Mahaffy (1949) and others to suggest that the two forms are interbreeding sub-species. With this opinion the present author is in strong disagreement. An examination of a large series of both species from many parts of their joint range shows that the extent of the pale basal marking on the hind femur affords an absolutely constant diagnostic character. Even in the most extreme forms the difference is striking, and in almost the whole of the material it is so gross as to be easily detectable with the naked eye (Fig. 2). Of the other characters pro- posed by Edwards the most reliable is perhaps the extent of the antero-median silver stripe on the mesonotum. Unfortunately, however, too many of the available specimens are rubbed in this position to be certain on the point. A fairly reliable character is the extent of the pale scaling at base of the 4th hind tarsal. This nearly always forms a complete ring in africanus, but in a few specimens from peripheral parts of the range the ring is interrupted ventrally. In luteocephalus the whole segment is commonly dark, but a small basal pale spot may be present on the dorsal surface, and in some specimens from the West Nile Province of Uganda this is fairly extensive and resembles the rather narrow pale interrupted band sometimes found in africanus. The remaining characters are of local value only, although their totality, ENTOM. ii. 5 G g 242 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) in the great majority of cases, renders identification both certain and easy. The larval characters proposed by Hopkins (1936) are very largely unreliable and should be ignored except where experience has shown them to be locally applicable. Aedes pseudoafricanus. The small difference from africanus with respect to scutal markings noted by Chwatt (1949) appears to be constant in all available specimens. The whole of these, however, come from the Lagos area, and it is by no means certain that the character would prove constant should the species be discovered elsewhere. All male specimens attributed to africanus in the British Museum collection have been dissected and these have yielded only a few pseudoafricanus from Yaba. The resem- blance of the male terminalia to those oifraseri is striking but it is not absolute. The FIG. 2. Left hind-femur, a-b. Aedes africanus; c-d. Aedes hiteocephalus. a from Ndola, b and c from Godia, d from Likoma Island, d is very exceptional. posterior edge of the IXth tergite, for example, is quite different in shape in the two species. The combination of colour characters resembling those of one species with male terminalia resembling those of another is not unique among the Ethiopian Stegomyia. It is also shown strikingly by amaltheus (De Meillon & Lavoipierre, 1944) . In view of the very curious results of cross-breeding experiments between certain species of Stego- myia, such cases must necessarily carry a suggestion of naturally occurring hybridiza- tion, and this is lent some colour in the case of pseudoafricanus by the distribution of the species concerned. In the experiments concerned Toumanoff (1937, &c.) showed that hybrid offspring of Aedes aegypti and Aedes albopictus exhibit all the morpho- logical characters of the female parent in whichever direction the cross is made. This was confirmed for the aegypti $ X albopictus $ cross by Hoang Tich Try (1937) and for the reverse cross by Downs and Baker (1949). The genetical and cytological background is obscure, although some light is thrown on the problem by the discovery that in the mosquitoes, or at least in those species so far examined, there is no differentiation of sex chromosomes (Sutton, 1942 ; Callan & Montalenti, 1947 ; White, 1949). Gilchrist & Haldane (1947) consider that sex in these insects is determined by a single dominant gene for male. Downs & Baker (1949) have suggested that the phenomenon is one of parthenogenesis following mechanical fertilization of the ovum. This would not, however, explain the very interesting observation made by Bonnet (1950) that hybrid offspring of aegypti $ X albopictus $, although themselves identical IN THE ETHIOPIAN REGION 243 with aegypti, produce offspring with albopictus characters when back-crossed to pure line albopictus <$s. While these observations are of considerable interest in relation to the phylogeny of the sub-genus it is by no means certain that hybridization takes place in nature. Rozeboom and his fellow workers are of the opinion that it does not since females of aegypti and albopictus in cage colonies show an exclusive preference for males of their own species when given the choice, even though they may mate with males of the other species when no others are available (Mattingly et al., in press). In the present author's opinion the most probable explanation of the occurrence of these intermediate forms is that they are genuinely anectant in the phylogenetic sense. The Lagos populations of africanus and pseudoafricanus are effectively sympatric at the present time, since, although the larval breeding-places are segregated, adults of both sexes of both species are taken together near the fringes of the mangrove swamps (Mattingly, 19496). It does not, however, follow that this is a case of sym- patric speciation. The area of low rainfall occurring between Lagos and Cape Three Points and serving to separate the Upper from the Lower Guinean forests is commonly supposed to result from a comparatively slight change in the direction of the coastline in relation to the direction of rain-bearing winds. If this is true, then small changes in the level either of the sea or of the land may well have served in the past to isolate extensive areas of mangrove ' forest ' from the true forest with consequent speciation of mangrove and true forest Stegomyia, an interesting example of the possible effect of small changes of elevation on the climate, flora, and fauna. Aedes simpsoni. Haddow, Van Someren, et al. (in the press), note that this is a very variable species and express doubts as to the usefulness of applying the name ' var. lilii ' to one among many varieties. It would seem, however, that the simpsoni of the Transvaal, Natal, and Cape Province may be largely isolated from those occurring in the rest of Africa (Fig. 12), and it is felt that if good representative collections could be assembled these would repay careful study. It is proposed to consider the matter from this angle in the course of the taxonomic work on the group. For a note on variation in this species see Edwards (1941). Aedes luteocephalus. This species has already been discussed under africanus. The type locality is wrongly quoted in the type description (see under ' Doubtful Records ', P- 257). Aedes unilineatus. Difficulties experienced in identifying the larvae of this species (De Meillon & Lavoipierre, 1944 ; Lewis, 1945) were due in part to an error in the legend to Fig. 61 of the first volume of Mosquitoes of the Ethiopian Region (Hopkins, 1936). An examination of the original drawings for this figure and the specimens on which they were based shows that the comb and pecten spines attributed to unilineatus were in fact drawn from a specimen attributed by Ingram & De Meillon (1929) to pseudonigeria and reassigned by Edwards (1936) to demeilloni. The identity of this specimen is still uncertain, but it is certainly not unilineatus. The larva illustrated, in the same figure, as demeilloni is in fact angustus, and neither demeilloni nor uni- lineatus is figured. Typical comb spines of unilineatus have very small secondary denticles which are almost or quite invisible under low powers of the microscope. An occasional spine may have one or two large denticles, but such spines are clearly aberrant and should cause no difficulty. An extreme example is shown in Fig. 3 244 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) together with a normal comb spine. The pecten spines are very variable. The larva of unilineatus is very distinctive and should not be confused with any other occurring on the mainland. It can be recognized by the characters given in the key. The degree of development of the barred area of the ventral brush varies even in specimens from the same locality, but it is never as well developed as in the majority of species. Barraud (1934) notes that in Ethiopian unilineatus the saddle hair is usually 5- branched while in Indian larvae it is usually bifid. Such South African larvae as have been examined, however, show the bifid condition. Sudan and West Nile larvae have 4-5 branches, but Gold Coast larvae appear to produce either type of FIG. 3. Aedes unilineatus. a. Comb spines, b. Pecten spines, c. IXth tergite of male. The lower comb spine is aberrant. This condition is only seen in an occasional spine. saddle hair at random. The descriptions of the male terminalia given by Edwards (1941) and Barraud (1934) are misleading since both appear to have overlooked the presence of two small, very strongly sclerotized setiferous lobes on the outer edges of the IXth tergite (Fig. 3). These are not shown in the only dissection left by Edwards in the British Museum owing to the way in which this is mounted, but they show clearly in fresh mounts from the same locality (Accra) and from Gadau, Erkowit, and Lahore. KEY TO ADULTS 1. Scutum with 3 pairs of small white spots; tibiae white ringed near middle; proboscis usually pale in middle vittatus Bigot 1 Scutum and tibiae otherwise ; proboscis dark 2 2. Proboscis and tibiae with white lines granti Theobald Proboscis all dark ; tibiae dark except, in some cases, at base . ,.' 3 3. Front half of scutum white scaled . . . mascarensis MacGregor Scutum otherwise marked . . . . . . . . .4 4. Front half of scutum with a median white stripe only . . . '.5 Scutum otherwise marked . ......... 6 1 Species dealt with in the present paper are shown in heavy type. IN THE ETHIOPIAN REGION 245 5. Middle femur with a white spot in middle in front . unilineatus Theobald Middle femur without such a spot . . . . . albopictus Skuse 6. Middle femur with a white stripe in front reaching nearly to tip ; clypeus of female scaly aegypti Linnaeus Middle femur without such a stripe ; clypeus bare .... 7 7. Third hind tarsal much more broadly pale than the others ... 8 Third hind tarsal only narrowly pale . . . . . . .11 8. Hind femur with a well-marked pale patch at base broadly expanded below and behind (Fig. 2) luteocephalus Newstead Hind femur at most with a small pale spot at base .... 9 9. Posterolateral lines of scutum broad, deep yellow, conspicuous, reaching forward almost or quite to the posterior ends of the anterolateral patches ; thorax laterally compressed ..... ruwenzori Haddow & E. C. C. Van Someren These lines shorter, less conspicuous ; thorax of normal width . . 10 10. Anterior median stripe of scutum very short ; antero-lateral patches on scutum slightly crescentic, concave behind . . africanus Theobald This stripe longer ; scutal patches wedge-shaped, straight-edged behind . pseudoaMcanus Chwatt 11. Fourth hind tarsal segment all black . . . . . . .12 Fourth hind tarsal segment pale at least at base ..... 14 12. Lateral lobes of scutellum white scaled ; anterior median spot on scutum composed of broad or narrow scales ....... 13 Lateral lobes of scutellum dark scaled ; this spot composed of narrow scales only .......... woodi Edwards 13. Hind tibia all dark below ...... simpsoni Theobald Hind tibia extensively pale below at base . . . strelitziae Muspratt 14. Prescutellar area of scutum covered with broad metallic scales metallicus Edwards This area with narrow, non-metallic scales ...... 15 15. Posterior half of scutum with pale scales forming three or four narrow longitudinal lines . . . . . . . . . 16 Posterior half of scutum with pale scales not aggregated into lines chaussieri Edwards 16. Median area of scutum with paired yellow longitudinal lines inside the usual posterolateral lines ........ 17 This area with a single median line . . . . . . .18 17. Fourth hind tarsal segment white at base only . . subargenteus Edwards This segment all white . . . subargenteus ssp. kivuensis Edwards 18. Each tibia with a narrow pale ring slightly removed from base pseudonigeria Theobald White marks on tibiae confined to extreme base (except sometimes on the hind leg) ........... 19 19. Hind tibia all dark (occasionally with a very small number of pale scales at base below in keniensis] ......... 20 246 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) Hind tibia with a well-marked pale patch at base, at least on the under surface ........... 22 20. Third hind tarsal pale all round on at least the basal fth keniensis E. C. C. Van Someren Third hind tarsal all dark or, at most, with one or two pale scales at base below ............ 21 21. Median longitudinal line of scutum white ; prescutellar bare patch with a conspicuous border of pale scales . amaltheus De Meillon & Lavoipierre Median longitudinal line yellow ; pale scales round prescutellar bare patch reduced or absent ....... masseyi Edwards 22. Pale basal patch on 6th abdominal tergite prolonged backwards nearly to the apex of the segment ........ 23 This patch shallower, at most half the depth of the segment ... 26 23. First mid-tarsal pale behind nearly to the tip, pale in front on only about the basal gth ........ soleatus Edwards This segment pale above on at most the basal fths, usually much less, more narrowly pale below, without any noticeable extension of pale scaling behind .......... 24 24. Pale markings of femora snowy white ..... fraseri Edwards These markings metallic silver ........ 25 25. Prescutellar bare patch with a broad border of yellow scales ; 5th hind tarsal entirely dark ....... schwetzi Edwards Prescutellar bare patch with at most a very narrow border of pale scales ; 5th hind tarsal at least narrowly pale at base above (except sometimes in ssp. denderensis) ...... apicoargenteus Theobald 26. First mid-tarsal segment pale behind for most of its length ... 27 This segment more or less uniformly ringed at base, without any noticeable extension of pale scaling on the posterior surface .... 28 27. 5th hind tarsal entirely dark ..... calceatus Edwards 5th hind tarsal pale at base . . heischi E. C. C. Van Someren (in part) 28. Hind femoral knee-spot suppressed, represented by at most two or three pale scales ; mid-femoral knee-spot, where present, at most very small, not reaching to the tip of the femur ...... 29 Hind femoral knee-spot well developed ; mid-femoral knee-spot reaching to the tip of the femur above ........ 30 29. Thorax laterally compressed ; mid-femur nearly always with a white spot on the anterior surface at about half-way between base and apex angustus Edwards Thorax of normal width ; mid-femur without such a spot bambusae Edwards 30. Mid-femur with a conspicuous white spot on the anterior surface at about two-thirds the distance from the base ; 5th hind tarsal dark . . 31 Either with the anterior surface of the mid-femur dark except for the knee- spot or with the 5th hind tarsal pale at base or both .... 33 31. Middle tibia pale below on at least the basal ^th . . . poweri Theobald Middle tibia usually all dark, at most with a small pale spot at base . 32 IN THE ETHIOPIAN REGION 247 32. Hind tibia pale below on at least the basal ^ths, this pale stripe expanded distally ; median longitudinal pale line of scutum expanded in front of prescutellar bare patch and fused with the pale border round this patch ; posterolateral lines of scutum yellow ; second mid-tarsal pale above on at least a half, usually more . . . langata E. C. C. Van Someren Hind tibia pale on at most J, the pale stripe at most very slightly expanded distally ; median pale line of scutum of uniform width throughout, not fused with the pale scales at the sides of the prescutellar bare patch; posterolateral lines usually white or whitish; second mid-tarsal pale above on at most |, usually less .... contiguus Edwards 33. Fifth hind tarsal largely dark, at most narrowly pale at base; anterior surface of mid-femur dark except for knee-spot . . deboeri Edwards Fifth hind tarsal pale on at least the basal ^rd, often more, or, if not, then anterior surface of mid-femur with a silvery spot at about f rds . . 34 34. Pale streak at base of hind tibia at most th the tibial length, usually less ; 2nd mid-tarsal broadly pale above on at most the basal half, apical half dark or at most with a very narrow dorsal line of pale scales . . 35 Either with the pale streak at the base of the hind tibia much more than gth the tibial length or with the 2nd mid-tarsal broadly pale above nearly to tip or both .......... 36 35. Anterior surface of mid-femur usually with a spot of silvery scales heischi E. C. C. Van Someren (in part) Anterior surface of mid-femur usually dark except at tip demeilloni Edwards 36. Anterolateral scutal patches pointed in front, varying in colour from bright yellow to creamy white; basal bands on abdominal tergites yellowish . ... bambusae ssp. kenyae E. C. C. Van Someren Anterolateral scutal patches rounded in front, snowy white ; tergal bands white . dendrophilus Edwards KEY TO FOURTH-STAGE LARVAE 1. Ventral brush with at least the great majority of setae single . . 2 Ventral brush with at least the great majority of setae branched . . 5 2. Lateral seta of saddle and sub ventral seta of siphon single . . angustus These setae each with at least 2 branches ...... 3 3. Comb spines with a small number of well-developed lateral denticles, at least the submedian ones easily visible under low powers of the micro- scope ; antennal seta with 2-4 branches . ... amaltheus (in part) Comb spines with the lateral denticles less strongly developed, invisible or visible only with difficulty under low powers of the microscope; antennal seta single ......... 4 4. Barred area of ventral brush absent or poorly developed ; saddle hair often with more than 2 branches ....... unilineatus Barred area of ventral brush well developed ; saddle hair bifid . albopictm 5. Ventral brush with 2 or more isolated tufts arising proximal to the barred area 6 248 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) Ventral brush without unpaired tufts proximal to the barred area . . 7 6. Antenna lightly spiculate ; setae of ventral brush with numerous branches ; comb spines with lateral denticles only moderately developed . vittatus Antenna smooth ; setae of ventral brush single or bifid ; comb spines with submedian teeth strongly developed, often appearing trifid amaltheus (in part) 7. Comb consisting of elongated scales fringed around the apex . africanus, pseudoafricanus , luteocephalus Comb consisting of spines with the median denticle more strongly developed than the lateral denticles . . . . . ... .8 8. Comb spines with a small number of rather coarse lateral denticles of which the submedian are usually visible under low powers .... 9 Comb spines with a fringe of more delicate denticles on either side of the main median denticle, these lateral denticles usually invisible under low powers and often almost invisible under high ..... 19 9. Pecten spines very deeply cleft, typical spines with two or more large median denticles subequal in size; saddle hair and subventral seta of siphon single ........ subargenteus Pecten spines otherwise ; saddle hair usually and subventral seta always with at least two branches ........ 10 10. Pecten teeth with poorly developed secondary denticles invisible under low powers of the microscope . . . . . . .11 Pecten teeth with well-developed secondary denticles visible under low powers of the microscope ........ 14 11. Lower caudal seta with 2 or more branches . . . . . .12 Lower caudal seta single ......... 13 12. Typical pecten teeth short and broad, not more than 4Xas long as their breadth at base ....... simpsoni, demeilloni Typical pecten teeth longer and narrower, at least 6xas long as their breadth at base ......... strelitziae 13. Spines at bases of pleural setae large and curved .... heischi These spines smaller ......... keniensis 14. Distal edge of saddle with a group of coarse spicules near the base of the caudal setae, easily visible under low powers; antennal seta with 2-4 branches ........ amaltheus (in part) Distal edge of saddle smooth or at most with very minute spicules not or scarcely visible under low powers ; antennal seta single . . 15 15. Saddle hair with at least 3 branches; pleural spines very large and con- spicuous .......... metallicus Saddle hair with at most 2 branches ; pleural spines smaller 16 16. Submedian denticles of comb spines strongly developed, giving the majority of these spines a trifid appearance under low powers . . 17 Submedian denticles not so strongly developed, only an occasional spine appearing trifid under low powers . . . . . . . 18 17. Pecten with 8-12 spines ....... mascarensis Pecten with 14-20 spines ........ aegypti IN THE ETHIOPIAN REGION 249 18. Upper caudal seta with about 6 branches ; saddle hair with 3-4 . granti Upper caudal seta with 2-4 branches ; saddle hair bifid . . . contiguus 19. Pecten spines arranged in discontinuous groups . . bambusae kenyae Pecten spines more or less equally spaced, at most with one or two terminal spines detached ........ 20 20. Pecten spines with at least one secondary denticle near the base easily visible under low powers . . . . . . . .21 Pecten spines often almost simple, secondary denticles, when present, visible only under high power . . . bambusae, dendrophilus, deboeri 21. Siphon more or less uniformly dark ....... 22 Siphon dark only on about the basal fths, apical gth pale ... 23 22. Subventral seta of siphon with 3-4 branches ; antennal seta situated at not more than i 7 5 ths the distance from base to apex; saddle hair bifid or trifid ........... calceatus Subventral seta of siphon bifid; antennal seta at more than ^ths the distance from base ; saddle hair single or bifid apicoargenteus ssp. denderensis 23. Median denticle of comb spines much reduced, lateral denticles well developed forming a fringe which is visible under low power . fraseri Median denticle strongly developed in relation to the fringe which is invisible under low power . . apicoargenteus, schwetzi, langata, soleatus DISTRIBUTION RECORDS The list which follows is believed to contain all published records except those which are merely repetitive and do not indicate that a species has been found more than once in a particular locality but merely that the same record has been published on more than one occasion. Doubtful records are marked with an asterisk and are discussed in a separate section. Records clearly based on misidentifications are omitted and are discussed in another section. Some of the records are based on unpublished specimens in various collections. For these collections the following abbreviations are employed. B.M. British Museum (Natural History) ; Berl. Berlin Museum; H.D. Hope Department, Oxford; H.I. Tropical Institute, Hamburg; L.S.H. London School of Hygiene and Tropical Medicine ; Liv. Liverpool School of Tropical Medicine ; M.L. Entomological Laboratory, Morogoro, Tanganyika. Records from the Washington Museum were kindly supplied by Dr. Stone. These are marked W.M. All references are listed in full in the bibliography. Addresses of correspondents who have sent records are given in the acknowledgements. Aedes apicoargenieus SIERRA LEONE. Daru (Simpson, 1913 ; Evans, 1925), Batkanu, Koinadugu (Butler et al., 1915), Mabang (Gordon, 1929), Moyamba, Pujehun (Davey, MS.). LIBERIA. Du River (Bequaert, 1930). GOLD COAST. Kumasi, Obuasi (Theobald, 1909), Lorha (Carter, 1916), Nsawam (Macfie & Ingram, 19230), Accra (Macfie & Ingram, 19236), Aburi, Sunyani (Edwards, 1941), Asuboi (Mattingly, 1947). TOGOLAND. Klein Popo (= Anecho, as Amelia togoensis, Enderlein, 1923). NIGERIA. Lagos (Wesche, 1910; ENTOM. n. 5 H h 250 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) Dunn, 19276; Philip, 1931), Yaba (Graham, 1911; Dunn, 1926; Haltingly, 1947), Bende (Simpson, 19126), Kaduna*, Katagum*, Zungeru* (Johnson, 1919), Aba, Ibadan, Ikeja (Mattingly, 1947), Itowolo (Haltingly, 19496), Agbor (L.S.H.). BR. CAMEROONS. Hisselele (Zumpt, 1937). FR. CAMEROONS. Kribi or Wuri (as Kingia maculoabdominalis , Enderlein, 1923), Yaounde (B.H.). BELGIAN CONGO. Unnamed locality (as Kingia albertii, Theobald, 1912), Kayembe Hukulu (Rodhain & Bequaret, 1913), Kabinda (Edwards, 1926), Leopoldville (Duren, 1929), Stanleyville (Schwetz, 19300; Liegeois, 1944), Lomami River between Obenge-Benge and Isangi (Schwetz, 19306), Eala, Kisantu, Ponthierville (Edwards, 1941), Hongbwalu (Schwetz, 1944), Aba, Aketi, Ango, Basoko, Bayenga, Bondo, Buta, Irumu, Kimvula, Libenge, Nian- gara, Titule, Yangambi (Liegeois, 1944), Coquilhatville (Wolfs, 1946), Haluku, Nta, Pongwema (Fain & Henrard, 1948), Kimbao (Edwards, HS.), Luluabourg (B.H.), Katompe (Liv.). RUANDA-URUNDI. Kisenyi* (Seydel, 19290). UGANDA. Unnamed locality (Neave, 1912), Entebbe (Bequaert, 1930), Kilembe (Edwards & Gibbins, 1939), Chagwe, Kampala (Edwards, 1941), Bundibugiyo (Haddow, 19450), Bubuk- wanga (Haddow, 19456), Hongiro (Smithburn & Haddow, 1946), Hamirimiri (Had- dow, 1947), Zika (Entebbe area, Haddow & Dick, 1948), Budonga and Siba Forest, Bugala I., Bugoma Forest, Bujuko Forest, Busi I., East and West Busoga Woodlands (Jinja and Busia areas), Buvuma I., Bwamba Lowlands and North Spur Ruwenzori to 5,000 ft., Fort Portal, Itwara Forest, Kalinzu Forest, Kasyoha and Kitomi Forests, Kiansonzi, Kibale-Hima Forest, Kitgum, Kome I., Kyenjojo, Hasaka, Hasindi Port, Hubende, Hukono, Sango Bay Forests (Kaiso, Halambigambo, Namalala and Toro) (Haddow, in litt.), Hpumu, Namanve Payida (B.H.). SUDAN. Khor Kobkwa ( = Khor Kobwa, as Kingia maculoabdominalis, Theobald, 1913). KENYA. Kaimosi Forest, Kisii (Garnham et al., 1946), Kakamega (E. C. C. Van Someren, 1946), Kabwoch, Sondu River (E. C. C. Van Someren, in litt.). TANGANYIKA. Bukoba (HcHardy, 1932; Harris, 1942). Aedes fraseri FRENCH GUINEA. Dubreka (B.H.). SIERRA LEONE. Freetown (as, fraseri, Bacot, 1916; as fraseri and blacklocki, Evans, 1926), Dam (as blacklocki, Evans, 1925), Njala (Edwards, 1941), Habang, Hoyamba Town (Liv.). GOLD COAST. Takoradi (Edwards, 1941; Haltingly, 1947). NIGERIA. Unnamed locality in the south (Edwards, 1917), Oshogbo (L.S.H.). UGANDA. Unnamed locality (Neave, 1912), Hpumu (Edwards, 1912), Kampala (Hancock, 1930), Hamirimiri, Hongiro (Haddow, 1947), Bubuk- wanga (Smithburn et al., 1949), Bukakata, Bwamba lowlands, and North Spur Ruwenzori to 5,000 ft., Bwigi I., Hasindi Port (Haddow in litt.). KENYA. Kaimosi Forest, Kakamega (E. C. C. Van Someren, 1946), Sondu River (E. C. C. Van Someren in litt.). Aedes dendrophilus SIERRA LEONE. Freetown (as fraseri, Edwards, 1917 ; as dendrophilus, Edwards, 19210), Kissy (Liv.). GOLD COAST. Aburi, Nsawam, Oblogo (Edwards, 19210), Dodowah (Hacfie & Ingram, 19236). NIGERIA. Lagos (Philip, 1931), Ibadan (Edwards, 1941). FERNANDO Po. Santa Isabel (as trinidad, Gil Collado, 1936). IN THE ETHIOPIAN REGION 251 UGANDA. Bwamba County (as deboeri demeilloni, E. C. C. Van Someren, 1946), Mongiro (as deboeri demeilloni, Smithburn & Haddow, 1946), Mamirimiri (as deboeri demeilloni, Haddow, 1947), Bubukwanga (Smithburn et al., 1949), Bugoma Forest, Bujuko Forest, Busi L, Bwamba lowlands to 5,000 ft., Kitgum (fide Harper) (Haddow in litt.). KENYA. Kaimosi Forest (as deboeri demeilloni, E. C. C. Van Someren, 1946). ZULULAND. Eshowe (as deboeri demeilloni, Edwards, 1936). NATAL. Isipingo Beach, Margate (B.M.), Amanzimtoti (Muspratt in litt.). CAPE PROVINCE. Embotyi (Muspratt in litt.). Aedes africanus FRENCH GUINEA. Kindia (Wilbert & Delorme, 1927). SIERRA LEONE. Freetown (Theobald, 1901; Evans, 1926; Davey, 1933), Koinadugu (Butler et al., 1915), Daru (Evans, 1925), Mabang (Gordon, 1929), unnamed localities (Davey, 1939), Moyamba, Port Loko, Pujehun (Davey, MS.), Segbwema (Liv.), Bo (B.M.). LIBERIA. Kaka Town (Bequaert, 1930), Roberts Field, Kakata (Briscoe, 1947). GOLD COAST. Ashanti (Macfie & Ingram, 19166), Tamale, Yape (Simpson, 1918), Broomassie* (Liv.). DAHOMEY. Bismarckburg (as Aniella africana, Enderlein, 1923). NIGERIA. Old Calabar (Theobald, 1901), Bonny, Duke Town (Theobald in Annett et al., 1901), Lagos (Theobald, 1901 ; Wesche, 1910 ; Connal, 1916 ; Dalziel, 1920 ; Dunn, 19276 ; Philip, 1931; Kerr, 1933), Oshogbo (Mayer, 1911), Yaba (Graham, 1911; Dalziel, 1920 ; Dunn, 1926 ; Philip, 1929 ; Mattingly, 1947), Bende, Forcados (Simpson, 19126), Opobo (Connal, 1915), Afaka, Kaduna, Ubassa (Johnston, 1916), Brass, Ikom (Connal, 1917), Ikoyi (Dunn, 1926; Kerr, 1933), Ibadan (Kumm, 19310; Kerr, 1933), Iwonran, Mushin (near Yaba, Kerr, 1933), Kateri (Edwards, 1941), Aba, Apapa, Ikeja (Mattingly, 1947), Itowolo (Mattingly, 19490), Ogbomosho (Bugher et al., in press), Sapele (L.S.H.). FR. CAMEROONS. Duala and unnamed localities (Grunberg, 1905), Yaounde (B.M.). FR. EQUATORIAL AFRICA. Brazzaville (Sice & Vaucel, 1928). BELGIAN CONGO. Ruwe (Theobald, 1907), Kabinda (Schwetz & Carter, 1915), Elisabethville, R. Vakila (Schwetz, 19270), Stanleyville (Schwetz, 19276; Liegeois, 1944), Tshela, Thysville (Schwetz, 1938), Coquilhatville, Lualaba River, Matadi (Edwards, 1941), Ilambi (Schwetz, 1942), Aba, Aketi, Bondo, Buta, Inongo, Irumu, Kimvula, Niangara, Titule, Yangambi, Yatolema (Liegeois, 1944), Balumi, Ban- ningville, Black River, Dumi-Mato, Edudu, Pongwema, Tsunikitoko (Fain & Henrard, 1948), Costermansville, Lembwa River, Luluabourg (B.M.). RUANDA-URUNDI. Dendezi (B.M.). UGANDA. Kampala (as Kingia africana, Theobald, 19110; as Aedes africanus, Hancock, 1930), unnamed locality (= Chagwe, Mabira Forest, Neave, 1912), Bugomolo (Hancock, 1930), Kilembe (Edwards & Gibbins, 1939), Entebbe (Edwards, 1941; Haddow & Dick, 1948), Bundibugiyo (Haddow, 19450), Bubuk- wanga (Haddow, 19456), Mongiro (Smithburn & Haddow, 1946), Mamirimiri (Haddow, 1947), Kitinda, Zika (both near Entebbe, Smithburn et al., 1948), Amag- burugburu, Manjuguja North, Mbango (Smithburn et al., 1949), Kyarumba (Haddow & Van Someren, 1950), Dwaji I. (Gillett in litt.), Budonga & Siba Forests, Bugala I., Bugoma Forest, Bujuko Forest, Bukasa I., Busi I., East and West Busoga Wood- lands (Jinja and Busia areas), Buvuma I., Bwamba Lowlands, and North Spur Ruwenzori to 5,000 ft., Bwigi I., Fort Portal, Itwara Forest, Jana I., Kalinzu Forest, 252 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) Kasyoha and Kitomi Forests, Kiansonzi, Kibale-Hima Forest, Kiryandongo, Kitgum, Rome I., Kyenjojo, Masaka, Masindi Port, Mityana, Mubende, Muhangi Forest, Mukono, Nyamugasani Valley, Sango Bay Forests (Kaiso, Malambigambo, Namalala and Toro) (Haddow in litt.), Godia, Kabarole, Kitasa, Midia, Payida (B.M.), Jinja, Mboga Stream (Jinja-Mumias Road) (Liv.). SUDAN. Li Rangu, Sopo (Lewis, 1945). ABYSSINIA. Sidamo Prov. (Bevan, 1937). KENYA. Chepahmgu, Fort Ternan, Kaimosi Forest, Kisii (Garnham et al., 1946), Kabwoch, Kakamega, Kibigori, Kisumu, Kitale, Koru, Lela, Maseno, Mugusi River, Muhuroni, Sondu River (E. C. C. Van Someren in litt.). TANGANYIKA. Bukoba (McHardy, 1932 ; Harris, 1942), Ukara I.* (B.M.). MOZAMBIQUE. Unnamed localities* (Howard, 1912). NYASALAND. Zomba* (Theobald, 1901). ANGOLA. Bihe (as Stegomyia africana and Stegomyia dubia), Katemar (= Katema's, Bihe area, as S. dubia} (Theobald, 1910), Kibokolo do Zombo (Gamble, 1912). N. RHODESIA. Ndola (Robinson, 1950). Aedes pseudoafricanus NIGERIA. Yaba (as africanus, Graham, 1911), Lagos area (as africanus, Connal, 1915, &c. ; Kerr, 1933; Edwards, 1941; as pseudoafricanus , Chwatt, 1949), Mushin (near Yaba, as africanus, Kerr, 1933), Itowolo (as africanus, Mattingly, 19490), Ilado, Middle Point (both in Lagos area, Chwatt in litt.), Apapa, Gbohun-Gbohun, Ikoyi, Onigbongbo (Lagos area) (B.M.). Aedes simpsoni GAMBIA. Bathurst (Edwards, 1941). SIERRA LEONE. Freetown (Kennan, 1915; Bacot, 1916; Evans, 1926), Daru (Blacklock, 1925), Kaballa (L.S.H.), Makeni, Port Loko, Pujehun (Davey, MS.). GOLD COAST. Dodowa (Simpson, 1914), Accra (Macfie, 1917), Takoradi (Pomeroy, 1932), Aburi (Edwards, 1941), Asuboi, Kumasi (Mat- tingly, 1947). NIGERIA. Afaka, Kudemsa (Johnston, 1916; Johnson, 1919), Yaba (Dunn, 1926; Philip, 1929; Mattingly, 1947), Ibadan (Philip, 1931; Kumm, 19310; Kerr, 1933), Oshogbo, Shaki (Philip, 1931), Lagos (Kerr, 1933), Gadau (Taylor, 1934), Ikeja, Kaduna, Port Harcourt (Mattingly, 1947), Ogbomosho (Bugher et al., in press). BR. CAMEROONS. Kumba (fide Macnamara), Babango (L.S.H.). FR. CAMEROONS. Yaounde (B.M.). GABOON. Mayumba Forest (Galliard, 1931). BELGIAN CONGO. Kasongo, Tshumbiri, Wathen (Newstead et al., 1907), Leopoldville (Newstead et al., 1907; Duren, 1929), Matadi (Newstead et al., 1907; Duren, 1938), Kivanda Kapepula (Rodhain & Bequaert, 1913), Kabinda (Schwetz & Carter, 1915), Stanley- ville (Schwetz, 19276; Liegeois, 1944), Katanga Prov. (Seydel, 19296), Popokabaka (Schwetz, 1938), Karimi (as Karumi, Uganda, Edwards, 1941), shores of Lake Kivu (Schwetz, 1944), Costermansville (Schwetz, 1944; Wolfs in litt.), Aba, Aketi, Basoko, Bondo, Buta, Irumu, Kimvula, Niangara, Titule, Yangambi, Yatolema (Liegeois, 1944), Coquilhatville (Wolfs, 1946), Kingangati, Kwamouth (Fain & Henrard, 1948), Kimbao (Edwards in litt.). RUANDA-URUNDI. Usumbura (Edwards, 1941). UGANDA. Kampala (as bromeliae, Theobald, 19110; as simpsoni, Hancock, 1930), unnamed locality (Neave, 1912), Kilembe (Edwards & Gibbins, 1939), Bundibugiyo (Haddow, 19450), Bubukwanga (Haddow, 19456), Bundinyama (Smithburn & Haddow, 1946), Soroti (Hopkins, MS.), Budongo and Siba Forests, Bugala I., Bugoma Forest, IN THE ETHIOPIAN REGION 253 Bujuko Forest, Bukakara, Bukasi I., Bus! I., East and West Busoga Woodlands (Jinja and Busia areas), Bwamba Lowlands and North Spur Ruwenzori to 5,000 ft., Entebbe, Fort Portal, Itwara Forest, Jubiya Forest, Kalinzu Forest, Kalungu, Kasyoha and Kitomi Forests, Kiansonzi, Kibale-Hima Forest (= Mpanga Forest auctt.), Kiryandongo, Kitgum, Rome I., Kyenjojo, Masaka, Masindi Port, Mityana, Mubende, Muhangi Forest, Mukono, Nyamugasani Valley to 5,000 ft., Sango Bay Forests (Kaiso, Malambigambo, Namalala, and Toro) (Haddowin litt.), Godia, Metu, Midia, Rumogi (B.M.). SUDAN. Bor (zslilii, Theobald, 1910), Khartoum* (Whitfield, 1939), Juba (Edwards, 1941), Kau (Findlay, Kirk, & MacCallum, 1941), Heiban, Kauda, Malakal, Meridi, Roseires, Sennar (Lewis, 1945), Torit (Knight in litt.). ABYSSINIA. L. Hardin (Edwards, 1941). ERITREA. Nefasit (Lewis, 19436). KENYA Digo, Eldoret*, Kiambu*, Meru*, Nairobi* (Symes, 1935), Kisumu (Symes, 1935; Haddow, 1942), Kakamega, Kilifi (Symes, 1935; E. C. C. Van Someren in litt.). Mombasa (Symes & Roberts, 1937; Wiseman el al., 1931; Teesdale, 1941), Fort Ternan, Kaimosi, Kisii (Garnham et al., 1946), Gede (Bailey, 1947), Taveta (Heisch, 1948), Butere, Kerio Valley, Luanda, Makindu, Malindi, Msambweni, Tiwi, Yala (E. C. C. Van Someren, in litt.). TANGANYIKA. Unnamed locality (Neave, 1912), Dar-es-Salaam (Pomeroy, 1920; Edwards, 1923; McHardy, 1927; Harris, 1942), Lindi, Tanga (Haworth, 1924), Moshi (McHardy, 1932), Mtibwa Forest (Harris, 1942), Morogoro (Harris, 1942; Hocking, 1947), Amani (Berl.), Mwanza (M.L., E. C. C. Van Someren in litt.). PEMBA I. Unnamed locality (Edwards, 1941). ZANZIBAR. Various localities including Zanzibar Town (Aders, 19176). MOZAMBIQUE. Mozambique, Nampula (De Meillon & Rebelo, 1941), Lourengo Marques, Umbeluzi (Rebelo & Pereira, 1943), Busi R., Guara Guara, Mambone, Nhemancone, Quelimane (W.M., fide Stone), Chindio (Liv.). NYASALAND. Unnamed locality (Neave, 1912), unnamed locality in uplands (Lamborn, 1930), Fort Johnston (Edwards, 1941), Fort Maguire, Likoma I. (L.S.H.). N. RHODESIA. Livingstone (Muspratt, 1945), Balovale (Robinson, 1948), Ndola (Robinson, 1950). ANGOLA. Bihe (Theobald, 1910), San Salvador (Gamble, 1914). S. RHODESIA. Shamva, Umtali (Leeson, 1931), Salisbury (Edwards, 1941), Bulawayo, Ndanga, Sinoia, Umtali, Victoria Falls (Meeser in litt.). TRANSVAAL. Unnamed locality (Theobald, 1905), Abundant in low veldt (Bostock & Simpson, 1905), Onderstepoort (Theobald, 19116), Roberts Heights (Pratt- Johnson, 1921), Leysdorp Road, Nelspruit (Ingram & De Meillon, 1929). ZULULAND. Empan- geni (Ingram & De Meillon, 1927), Mhlatuze, Ntabanana (Bedford, 1928). NATAL. Durban (Pratt- Johnson, 1921 ; Ingram & De Meillon, 1927; Bedford, 1928), Margate (Muspratt 1950), Amanzimtoti (Muspratt in litt.). BECHUANALAND. Kasane (De Meillon, 19470). CAPE PROVINCE. Mkanduli (De Meillon & Lavoipierre, 1944). Aedes luteocephalus GAMBIA. Bathurst (Edwards, 1941). FRENCH GUINEA. Dubreka (B.M.). SIERRA LEONE. Freetown (Bacot, 1916; Evans, 1926; Davey, 1933), unnamed localities (Davey, 1939), Bonthe, Makeni, Moyamba, Port Loko, Pujehun (Davey, MS.), Kissy (Liv.), Aberdeen (L.S.H.). LIBERIA. Roberts Field (Briscoe, 1947). GOLD COAST. Accra (Macfie & Ingram, 19160:), Ashanti (Macfie & Ingram, 19166), Christians- borg (Accra area, Ingram & Macfie, 1917), Tamale, Yape (Simpson, 1918), Nsawam 254 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) (Macfie & Ingram, 1923*2), Saltpond (Edwards, 1941), Aburi (B.M.). NIGERIA. Bende (Simpson, 19126), Yaba (Simpson, 19126; Dunn, 1926; Kerr, 1933; Haltingly, 1947), Lagos (Connal, 1915, &c. ; Dalziel, 1920; Dunn, 19274:; Philip, 1931; Kerr, 1933), Adun (Connal, 1917), Ibi (Connal, 1924), Ikoyi (Lagos area, Dunn, 1926; Kerr, 1933), Oshogbo (Kumm, 19316), Ibadan (Kumm, 19310; Kerr, 1933; Mattingly, 1947), Iwonran, Kano, Mushin (near Yaba) (Kerr, 1933), Gadau (Taylor, 1934), Lokoja (Edwards, 1941), Ikeja (Mattingly, 1947), Ogbomosho (Bugher et al., in press), Badagry (Liv.). BELGIAN CONGO. Kumba* (Newstead et al., 1907), Jadotville (De Meillon, 1943), Kasongo (Liv.). UGANDA. Mpumu (as Kingia luteocephala, Theobald, 1910), unnamed locality (Neave, 1912), Mamirimiri (Haddow, Smithburn, et al., 1948), Bwamba Lowlands and North Spur Ruwenzori to 5,000 ft. (Haddow in litt.), Godia, Laufori, Metu, Rumogi (B.M.). SUDAN. Bor (Theobald, 1910), Juba (Hewer, 1934; Edwards, 1941), Khartoum* (Whitfield, 1939), Heiban (Lewis, 19430), Danaglia, Malakal, Sennar, Wad el Magdub (Lewis, 1945), Jebel Marra (Abbott, 1948), Singa (B.M.). ERITREA. Nefasit (Lewis, 19436). KENYA. Fort Hall, Kerio Valley, Kisumu (E. C. C. Van Someren in litt.). N. RHODESIA. Balovale, Livingstone (Robin- son, 1948), Ndola (B.M.). NYASALAND. Likoma I. (L.S.H.). S. RHODESIA. Victoria Falls (Meeser in litt.). BECHUANALAND. Kasane, Maun (De Meillon, 19470). Aedes unilineatus GOLD COAST. Accra (Macfie, 1916), Tamale* (Ingram, 1919), Aburi (Edwards, 1941). NIGERIA. Kaduna, Katagum, Zungeru (Johnson, 1919), Ibadan (Philip, 1931), Gadau (Taylor, 1934), Baro, Lokoja (Edwards, 1941), Ogbomosho (Bugher et al., in press). UGANDA. Metu, Rumogi (B.M.). SUDAN. Bahr-el-Ghazal (Theobald, 1906), Jebelein (Theobald, 1910; Lewis, 1948), Erkowit, Juba (Edwards, 1941), Heiban (Lewis, 19430), El Fasher, Jebel Moya, Sennar, Tembura, Wad el Magdub, Wad Medani (Lewis, 1945). KENYA. Unnamed locality (Neave, 1912), Dolo, Juba River (Edwards, 1941). NYASALAND. Port Herald (Edwards, 1941). N. RHODESIA. Livingstone (De Meillon & Lavoipierre, 1944). S. RHODESIA. Fort Victoria (De Meillon, 1943). TRANSVAAL. Letaba, Tzaneen (Ingram & De Meillon, 1929). Aedes metallicus SENEGAL. Dakar (Kartman et al., 1947). GOLD COAST. Accra (Macfie, 1916), Navarro (Ingram, 1919), Asuboi (Mattingly, 1947). NIGERIA. Kano (Kumm, 19310), Gadau (Taylor, 1934), Maiduguri (Mattingly, 1947). BELGIAN CONGO. Jadotville (De Meillon, 1943). UGANDA. Mbarara (Edwards, 1941), Soroti (Haddow in litt.). SUDAN. Bor (as dubia, Theobald, 1910), Darfur Prov., Juba (Edwards, 1941), Nuba Mts., Wad el Magdub (Lewis et al., 1942), Heiban (Lewis, 19430), Abu Usher, Dagu, El Fasher, Kadugli, Kalora, Kassala, Khartoum, Kheiwok, Kudring, Malakal, Mug, Nahud, Raga, Roseires, Sennar, Shaloat, Sheikh Gamal, Sheikh Karim, Wad Banda, Wad Ganatir, Wad Medani (Lewis, 1945), Dueim, Kosti (Lewis, 1948), Yambio (Abbott in litt.), Umm Sunt (B.M.). KENYA. Mombasa (Wiseman et al., 1939; Teesdale, 1941), Kisumu (Haddow, 1942), Fort Ternan (Garnham et al., 1946), Isiolo (Heisch, 1947), Gede (Bailey, 1947), Garissa, Garsen, Gotani, Kakamega, Kerio Valley, Kibos, Kilifi, Kipini, Koru, Langata Forest, Makindu, Malindi, Maseno, IN THE ETHIOPIAN REGION 255 Msambweni, Taveta, Witu (E. C. C. Van Someren in litt.). TANGANYIKA. Dar-es- Salaam (Edwards, 1923 ; Scott, 1926 ; McHardy, 1927), Tanga (Nixon, 1923 ; Ha worth, 1924), Lindi (Haworth, 1924), Morogoro (Harris, 1942), 'Widely distributed from coast to Moshi and Mwanza' (Harris, 1942), Mwanza (M.L., E. C. C. Van Someren in litt.). ZANZIBAR. 'Various localities' (Aders, 19170), Zanzibar Town (Aders, 19176), Mnazi Moja (= European Club, Zanzibar Town, Edwards, 1941). MOZAMBIQUE. Nova Choupanga (Seguy, 1931). N. RHODESIA. Livingstone (De Meillon & Lavoi- pierre, 1944). ANGOLA. Bine (as Quasistegomyia dubia, Theobald, 1910). S. RHODESIA. Bindura, Gatooma, Shamva (Leeson, 1931), Fort Victoria, Odzi, Salisbury (Meeser in litt.). TRANSVAAL. Komatipoort, Letaba (Ingram & De Meillon, 1929). ZULULAND. Mhlatuze, Ntabanana (Bedford, 1928). BECHUANALAND. Francistown, Kasane, Maun, Nata (De Meillon, 19470;). S. W. AFRICA. Fransfontein, Grootfontein, Karabib, Okahandja, Okimbahe, Omatjette, Otjimbingwe, Outjo, Usakos (Muspratt in litt.). Aides vittatus GAMBIA. Bathurst (Theobald, 1903). SIERRA LEONE. Freetown (Theobald, 1901; Butler et al., 1915; Bacot, 1916; Evans, 1926; Davey, 1939), Batkanu, Benikoro, Firiwa, Kaballa, Serakolia, Sonkonia, Tirikoro, Waliki (Simpson, 1913), Koinadugu (Butler et al,, 1915), Daru (Edwards, 1941), Kamabai, Yiraia (B.M.), Boala Karafaia, Kamasiki (L.S.H.), Bo, Kailahun, Makeni, Moyamba, Port Loko (Davey, MS.), Dunkiawallia, Pujehun (Liv.). LIBERIA. Monrovia (Kumm, 19310), Firestone Plantation (Du Group) (Barber et al., 1932), Roberts Field (Briscoe, 1947). IVORY COAST. Bobo-Dioulasso (B.M.). GOLD COAST. Bole (Ingram, 1912), Cape Coast, Tamale (Simpson, 1914), Wa (Carter, 1916), Batiasan, Dogankade, Gambaga, Jefisi, Kulmasa, Lilixia, Mayoro, Nacon, Nandaw, Nandawli, Nangudi, Pinna, Sambolugu, Savelugu, Tanina, Tishi, Tumu, Ulu, Winduri, Yeji, Zouragu (Ingram, 1919), Sunyani (Edwards, 1941), Accra, Kintampo (Liv.). NIGERIA. Oshogbo (Mayer, 1911; Mattingly, 1947), Baro (Simpson, 19120), Zungeru (Simpson, 19120; Johnson, 1919), Lokoja (Connal, 1915), Kakuri (Johnston, 1916), Katagum (Johnson, 1919), Kaduna (Johnson, 1919, 1920), Gadau (Taylor, 1930), Abeokuta, Funtua, Ibi, Kano, Katsina, Shaki, Yola (Kumm, 19310), Ibadan (Kumm, 19310; Mattingly, 1947), Ikoyi (Kerr, 1933), Ogbomosho (Bugher et al., in press). FR. CAMEROONS. Yaounde (B.M.). BELGIAN CONGO. Kisui (as sugens, Newstead et al., 1907), Leopoldville, Matadi (Newstead et al., 1907; Duren, 1929), Elisabethville (Schwetz, 19270), Stanleyville (Schwetz, 19276, 19300; Liegeois, 1944), Boma (Duren, 1929), Kinshasa (Bequaert, 1930), Lukula (Schwetz, 1938), Kaparata (Schwetz, 1944), Aba, Bondo, Buta, Libenge, Niangara, Titiile (Liegeois, 1944), Coquilhatville* (as sugens, Wolfs, 1946). UGANDA. Entebbe (Theobald, 1907), unnamed locality (Neave, 1912), Sipi (Hancock & Soundy, 1929; Hargreaves, 1932), Kyere, Mukongoro (Hancock, 1930), Busi I., Fort Portal, Kitgum, Masindi Port, Soroti (Haddow in litt.), Ntotoro Valley (Lumsden in litt.), Buvuma I. (Gillett in litt.), Chagwe, Jinja, Metu, Rumogi (B.M.). SUDAN. Erkowit (Theobald 1910), Darfur Prov., Juba, Mongalla Prov. (Hewer, 1934), Nuba Mts. (Hewer, 1934; Lewis, 19430), Malakal (Hewer, 1934; Lewis, 1945), Khartoum* (Bedford, 1938), Dar Shol, Delami, Kajo Kaji, Talodi (Edwards, 1941), JebelMeidob (Edwards, 1941 ; Lewis, 1945), Jebelein (Edwards, 1941; Lewis, 1948), 256 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) Wau (Edwards, 1941 ; Abbott in litt.), Damazin Cataract, Doka, El Obeid, Gedaref, Geneina, Jebel Gurein, Jebel Handera, Kassala, Kosti*, Roseires, Sennar, Umm Ruaba (Lewis, 1945), El Fasher, Jebel Marra (Abbott, 1948), Rumbek, Yambio (Abbott in litt.), Loka (Liv.). ARABIA. Aden, Nobat (= Nobat Dakim), Sheikh Othaman, Ulub (Patton, 1905), Wadiyain (as Wadi Yain, Edwards, 1941), Jebel Jihaf (B.M.). SOKOTRA I. Hadibo, Kallansiya, Kathub (Leeson & Theodor, 1948). ABYS- SINIA. Harrar (as brumpti, Neveu Lemaire, 1905), L. Metahara area* (as Stegomyia sp., Doreau, 1909), Meta Gafersa (La Face, 1939). ERITREA. Arisateb Valley, Tes- senei, Ugaro (Lewis, 19436), Ras Malalu, east slopes of central massif, western plains (Jannone et al., 1946), Mt. Ramlu at 3,000-4,000 ft. (B.M.). BRITISH SOMALILAND. Buran (Edwards, 1941), Gedeis, Upper Sheikh (G. R. C. Van Someren, 1943), Las Anod (B.M.). KENYA. Shimoni (as sugens, Edwards, 1914), Digo, Kisumu, Meru (Symes, 1935), Mombasa (Symes, 1935 ; Wiseman et al., 1939), Nairobi (Symes, 1935 ; H. I., fide Weyer, E. C. C. Van Someren in litt.), Fort Hall, Kakamega, Kilifi (Symes, 1935; E. C. C. Van Someren in litt.), Central Kavirondo (Symes & Roberts, 1938), 'Third Camp' (Edwards, 1941), Kaimosi Forest (Garnham et al., 1946), Isiolo (Heisch, 1947), Kisumu (Grainger, 1947), Eldoret, Gotani, Kericho, Kiambu, Kisii, Kitale, Kitui, Langata, Makindu, Malindi, Mara R., Maragua, Maseno, Narok, Taveta, Thika, Turbo, Wesu (E. C. C. Van Someren in litt.), Malakisi (fide Leeson). TANGANYIKA. Lindi (Haworth, 1924), Morogoro (Scott, 1926; Harris, 1942), Mwanza (McHardy, 1932; Harris, 1942), 'Widely distributed from coast to Bukoba' (Harris, 1942), 'All along shores of Mwanza Gulf and on Ukara Island' (Smith in litt.), Dar-es- Salaam (B.M.). PEMBA I. Unnamed locality (Edwards, 1941). ZANZIBAR. Zanzibar Town (Aders, 19170) . MOZAMBIQUE. Unnamed localities (Howard, 1912), Mocuba, Quelimane (De Meillon & Rebelo, 1941), Namaacha (Rebelo & Pereira, 1943), Ressano Garcia (W. M., fide Stone). NYASALAND. Uplands (Lamborn, 1930), Fort Johnston (B.M.), 'Abundant above 3,000 ft.' (Lamborn, MS.). N. RHODESIA. Living- stone (Robinson, 1948) . ANGOLA. Bihe (Theobald, 1910). S. RHODESIA. Mashonaland (Theobald, 1901), Gatooma, Salisbury, Shamva (Leeson, 1931), Bindura, Chibi, Mt. i)arwin, Fort Victoria, Goromonzi, Gwelo, Maranka Reserve, Matopo Hills, Msoneddi, Ndanga, Triangle Estate, Umtali, Victoria Falls, Zaka (Meeser in litt.), Epworth (L.S.H.). TRANSVAAL. Onderstepoort (Theobald, 19116; Bedford, 1928), Brits (Ingram & De Meillon, 1927), Yokeskei River (De Meillon, 1928), Komatipoort, Letaba, Louis Trichardt, Nelspruit, Skukuza, Tomango, Waterpoort, Waterval Boven (Ingram & De Meillon, 1929), Pietersburg (Liv.). ZULULAND, Begamuzi Drift (Ingram & De Meillon, 1927), Ntabanana (Bedford, 1928). NATAL. Weenen (Bedford, 1928), Margate (Muspratt in litt.). BECHUANALAND. Kasane (De Meillon, 19470). S.W. AFRICA. Karabib, Okimbahe, Otjimbingwe, Outjo (Muspratt in litt.). Aedes aegypti Apparently more or less throughout the region except at high altitudes and, perhaps, in the extreme south and south-west. The following records will be found to fill most of the gaps in the table given by Edwards (1941). CAPE PROVINCE. Grahamstown, Port St. Johns (Edwards, 1941), Kimberley (De Meillon, 1942), Upington (De Meillon, 1943). S.W. AFRICA. Tsumeb, Windhoek (De Meillon, 1943), IN THE ETHIOPIAN REGION 257 Fransfontein, Karabib, Okimbahe (Muspratt in litt.). BECHUANALAND. Kasane, Nata (De Meillon, 19470). N. RHODESIA. Balovale, Livingstone (Robinson, 1948), Ndola (Robinson, 1950). TANGANYIKA. 'Throughout' (Harris, 1942). SUDAN. Widespread in Darfur Province (Abbott, 1948). DOUBTFUL RECORDS Aedes apicoargenteus NIGERIA. Kaduna, Katagum, Zungem. Johnson does not state from which of these localities his specimens came. Katagum seems much too dry. RUANDA-URUNDI. Kisenyi. No specimens are at present available for examination. In the distribution maps (Figs. 4 and 8) the record is ascribed to ssp. denderensis. Aedes africanus GOLD COAST. Broomassie. The specimen in question is not marked with the locality but was contained in a box with other specimens similarly labelled and known to have come from Broomassie. There is nothing intrinsically unlikely about the record. TANGANYIKA. Ukara Island. This record is based on a single larva and requires confirmation. MOZAMBIQUE and NYASALAND. No specimens can now be found by means of which these interesting records could be confirmed or disproved. They are discussed below under zoogeography. Aedes simpsoni SUDAN. Khartoum. This record is based on a single specimen found in an aeroplane and may be presumed to be the result of casual introduction. KENYA. Records given by Symes (1935) from localities above about 5,000 feet have not been confirmed and are thought to have been due to misidentification, although, in the absence of speci- mens, this cannot be proved. Aedes luteocephalus BELGIAN CONGO. Kumba. In his type description Newstead states that a single specimen was taken at Kumba. This is clearly an error since the Liverpool School collection contains two cotypes both marked by Newstead on the actual mounts as taken at Kasongo. In his type description of Aedes (Dunnius] albomarginatus, in the same paper, Newstead states that two specimens were taken at Kasongo, but the Liverpool collection contains only a single holotype marked as taken at Kumba. It is quite evident that Newstead confused the data relating to the two species in his published descriptions and that the type locality of luteocephalus is Kasongo. The point is of some importance since a record from Kumba would be entirely at variance with what is known of the distribution of this species, while Kasongo is perfectly acceptable. Edwards (1941) sought to equate Kumba with Tumba, but this appears to have been quite gratuitous. Kumba is, in fact, a small village whose position is clearly shown on Newstead's map (see also under misquoted localities infra). The reference by Theobald (1910) to a 'type <' of luteocephalus in the British Museum concerns a specimen of King's from the Sudan apparently chosen by Theobald as a lectallotype. SUDAN. Khartoum. This record is based on a single specimen found in a plane and may be taken to represent a casual introduction. ENTOM. ii. 5 i i 258 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) Aedes unilineatus GOLD COAST. Tamale. This record was based on larvae only and is queried by Ingram. He was, however, familiar with the larva of unilineatus which he helped to describe (Ingram and Macfie, 1917) and the record seems perfectly acceptable. Aedes vittatus BELGIAN CONGO. Coquilhatville. Wolfs quotes this record from a report of the hygiene service at Coquilhatville for 1941. It is not supported by specimens and vittatus has not been found again in Coquilhatville. SUDAN. Khartoum and Kosti. Lewis (1945) considers that these were chance introductions. ABYSSINIA. Lake Metahara area. The species in question is stated to have been found breeding in large numbers in rock-holes and it appears very probable that it was vittatus although its identity cannot now be established. RECORDS BASED ON MISIDENTIFICATIONS The following records have been rejected. Aedes fraseri SIERRA LEONE. Freetown (Edwards, 1917). = dendrophilus (see Edwards, 19210). S. RHODESIA. Unnamed localities (Garnham et al., 1946). = langata. Aedes africanus N. RHODESIA. Unnamed locality (Edwards, 1912). = chaussieri from Lake Young area. S. RHODESIA. Mashonaland (Theobald, 1901). = contiguus. UNIDENTIFIED AND DOUBTFUL LOCALITIES SIERRA LEONE. Dunkiawallia, Kamasiki, Waliki. Records of Aedes vittatus from these localities are not thought to be of sufficient interest to merit an intensive search. It is evident that this species is widely distributed throughout the territory. Kumm (19316) takes Waliki to be the name of a place, but from Simpson's account it seems equally possible that it is the name of a river. FR. CAMEROONS. Wuri (A. apicoargenteus) . This is presumed to refer to some locality on the Wuri River, probably at or near Duala. BELGIAN CONGO. Kimba (A. luteocephalus, Theobald, 19100;) and Tumba (A. luteocephalus , Schwetz, 19276). These are clearly inaccurate renderings of the type locality, Kumba, as given by Newstead et al. (1907) (for a further discussion see above under doubtful records). Lualaba River (A africanus, Edwards, 1941). In its present form this record is too vague to be of any value. No specimens are at present available. KENYA. 'Third Camp' (A. vittatus}. The significance of this record could only be ascertained by reference to relevant correspondence or records. These are not available to the author. The specimen was collected by Dr. Chell in about 1913. ? NYASALAND. Zoruta (A. africanus, Theobald, 1910). This is presumed to be a misspelling of Zomba from which Theobald (1901) has a previous record. IN THE ETHIOPIAN REGION 259 ARABIA. Wadi Yain (A. vittatus, Edwards, 1941). The correct rendering should be Wadiyain. Mr. Philby, who collected the specimens, kindly supplied this information together with the approximate map reference given below. DISTRIBUTION OUTSIDE THE ETHIOPIAN REGION Aides unilineatus. Punjab, Delhi Prov., Bombay area (Barraud, 1934). Aedes vittatus. Corsica (Bigot, 1861, not since found there) ; Spain (Gil Collado, 1935 ; Clavero, 1946) ; India, Ceylon (Barraud, 1934) ; Cochin China, Annam (Borel, 1930) ; Algeria (Senevet, 1936). Aedes aegypti. Throughout tropics and sub-tropics (Barraud, 1928; Kumm, 19316; Farner et al., 1946; Brug & Bonne-Wepster, 1947). LIST OF LOCALITIES WITH TOPOGRAPHICAL DETAILS The list is as accurate as it has been possible to make it from readily available sources. It could undoubtedly be improved, especially by persons with local know- ledge. Figures enclosed in brackets have been read from maps or, in the case of rainfall figures, refer to an adjacent station. In the case of regions or provinces an average figure is given and such figures are also enclosed in brackets. Where more than one rainfall station is available for a particular locality an average figure is given. Published figures and figures sent in litt. are unenclosed. Altitudes are given in feet, where possible to the nearest 100 ft., otherwise to the nearest 500 ft. Rainfall is given as the mean annual total to the nearest 5 inches and all references in the text should be read in this sense. For many of the data included in the table the author is indebted to correspondents. Aside from this the main sources have been Gossweiler & Mendonga (1939), Grandidier (1934), Hurst & Black (1943), Nash (1948), Vanden- plas (1943), The Times Gazetteer, the Atlas General du Congo Beige, the Atlas do Portugal Ultramarino, the Gold Coast Atlas, the Atlas of the Tanganyika Territory, the East and South African Handbooks, the Boletim Mensal das Observances Meteorologicas (Lourengo Marques), the annual rainfall summaries of the British East African Meteorological Department, the climatological summaries of the Rhodesia and Nyasaland Meteorological Service, the summary of rainfall normals issued by the South African Department of Irrigation, and the handbooks and reports of various government departments. Finally, where no published figures were available, reference was made to the excellent large-scale maps of nearly every part of Africa prepared during the last war for use by the forces and now available to the public. Among these the Rainfall Map of Eastern Africa prepared by the East African Survey Group (E. A. F. No. 1518 of 1943) deserves special mention. LIST OF LOCALITIES Locality A Ititude Latitude Longitude Rainfall Aba, Belgian Congo .... Aba, Nigeria ..... Abeokuta, Nigeria ..... Aberdeen, Sierra Leone .... (3-500) < 500 < 500 < 5 3-50 N. 5-0? N. 7.09 N. 8.30 N. 30.13 E. 7.21 E. 3-21 E. 13.18 W. (55) 85 40 < IOO 2 6o THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) LIST OF LOCALITIES (cont.) Locality A Ititude Latitude Longitude Rainfall Aburi, Gold Coast ..... 1,400 5-51 N. o.io W. 40 Abu Usher, Sudan ..... (1,200) (14.54 N.) (33-14 E.) (10) Accra, Gold Coast ..... < 5o 5-33 N. 0.13 W. 25 Aden, Aden Prot. ..... < 5 12.45 N. 45.04 E. (5) Adun, Nigeria ..... < 5 5-59 N. 8.09 E. (80) Afaka, Nigeria ..... (2,000) (10.41 N.) (7-23 E.) (50) Agbor, Nigeria ..... < 500 6.15 N. 6.14 E. (80) Aketi, Belgian Congo .... (1,500) 2.42 N. 23-52 E. (65) Amagburugburu, Uganda (3,000) (0.45 N.) (30.03 E.) (55) Amani, Tanganyika .... 3,000 5-05 s. 38.40 E. 75 Amanzimtoti, Natal .... < 5 30.03 S. 30.53 E. (40) Anecho, Togoland ..... < 500 6.15 N. 1.39 E. (30) Ango, Belgian Congo .... (2,500) 4.03 N. 25-53 E. (75) Apapa, Nigeria ..... < 500 (6.20 N.) (3-19 E.) (70) Arisateb Valley, Eritrea .... (2,000) (15.13 N.) (36.45 E.) (15) Asuboi, Gold Coast ..... < 500 (5.46 N.) (0.22 W.) (50) Babango, Br. Cameroons .... (5,000) (6.20 N.) (10.30 E.) (70) Badagry, Nigeria ..... < 500 6.23 N. 2.58 E. (60) Bahr el Ghazal, Sudan .... (1,200) (9.30 N.) (30.00 E.) (30) Balovale, N. Rhodesia .... 3,400 (I3-34S.) (23.07 E.) 40 Balumi, Belgian Congo .... (2,000) 4.10 S. 15.50 E. (55) Lake Bangweulu, N. Rhodesia . 3,800 11.20 S. 30.00 E. (50) Banningville, Belgian Congo 1,100 3.198. 17.21 E. 50 Baro, Nigeria ..... < 500 8.35 N. 6.18 E. (55) Basoko, Belgian Congo .... (i,5oo) i. ii N. 23.40 E. (75) Bathurst, Gambia ..... < 500 I3.28N. 1 6.40 W. 45 Batiasan, Gold Coast .... (1,000) (10.43 N.) (1-44 W.) (40) Batkanu, Sierra Leone .... (500) (9-05 N.) (12.27 W.) no Bayenga, Belgian Congo .... (1,500) 2.19 N. 21.46 E. (70) Begamuzi Drift, Zululand (3,000) (28.108.) (31.20 E.) (25) Bende, Nigeria ..... < 500 5-32 N. 7.36 E. (90) Benikoro, Sierra Leone .... (1,500) (9-43 N.) (11.27 W.) (80) Bih6, Angola ..... (5,500) 12.188. 17.00 E. (70) Bindura, S. Rhodesia .... 3,7oo ( I7 .2I S.) (31-21 E.) (35) Bismarckburg, Dahomey 2,300 8.12 N. 0.51 E. 55 Mt. Bizen, Eritrea ..... (5-8,000) (15-18 N.) (39.05 E.) (15) Black River, Belgian Congo (1,500) 3-53S. 15-55 E. (55) Bo, Sierra Leone ..... < 5 7-55 N. H.44W. 1 20 Boala Karafaia, Sierra Leone . (1,500) (9-39 N.) (11.04 W.) (80) Bobo-Dioulasso, Ivory Coast . 1,400 II. 12 N. 4.17 W. 40 Bole, Gold Coast ..... 1,000 9.02 N. 2.29 W. (40) Boma, Belgian Congo .... < 500 5-50 s. 13.09 E. (35) Bondo, Belgian Congo .... (1,500) 3-55 N. 23-55 E. (65) Bonny, Nigeria ..... < 5 4-25 N. 7.13 E. > IOO Bonthe, Sierra Leone .... < 5 (7-33 N.) (12.37 W.) 150 Bor, Sudan . . . . 1,700 6.12 N. 31-33 E. 35 Brass, Nigeria ..... < 500 4.19 N. 6.15 E. > IOO Brazzaville, Fr. Eq. Africa (1,000) 4-15 S. I5.20E. 60 Brits, Transvaal ..... 3,600 25.38 s. 27.47 E. 25 Broomassie, Gold Coast .... (500) 8.21 N. 0.58 W. (45) Bubukwanga, Uganda .... (3,000) (0.45 N.) . (30.03 E.) (55) Budongo Forest, Uganda (3,500) (1-45 N.) (31-34 E.) (60) Bugala I., Uganda ..... (3,800) (0.24 S.) (32.I6E.) (80) Bugoma Forest, Uganda .... (4,000) (1.20 N.) (31.00 E.) (55) Bugomolo, Uganda .... (3,500) 1.31 N. 31.27 E (55) Bujuko Forest, Uganda .... (4,000) 0.25 N. 32-I5E (45) IN THE ETHIOPIAN REGION LIST OF LOCALITIES (cont.) 261 Locality Altitude Latitude Longitude Rainfall Bukakata, Uganda .... (4,000) 0.158. 32.02 E. 55 Bukasa I., Uganda .... (3,800) (0.26 S.) (32.30 E.) (75) Bukoba, Tanganyika .... 3,800 1.20 S. 31.48 E. 80 Bulawayo, S. Rhodesia .... 4,400 20.09 S. 28.37 E. 25 Bundibugiyo, Uganda .... 3,600 0.42 N. 30.04 E. 55 Bundinyama, Uganda .... (3,000) (0-47 N.) (30.05 E.) (55) Buran, Br. Somaliland .... 3,000 10.13 N. 48.47 E. 10 Busia, Uganda ..... (4,000) (0.28 N.) (34-05 E.) (60) Busi I., Uganda ..... (3,8oo) (0.03 N.) (32.20 E.) (60) Busi R., Mozambique .... < 500 (19-55 S.) (34-20 E.) (40) Buta, Belgian Congo .... (1,200) 2.48 N. 24.46 E. 60 Butere, Kenya ..... (4.5oo) 0.13 N. 34-30 E. (70) Buvuma I., Uganda .... (3,8oo) o.n N. 33-18 E. 65 Bwigi I., Uganda ..... (3,8oo) (0.28 S.) (32.24 E.) (75) Calabar, Nigeria ..... < 500 4-57 N. 8.20 E. 125 Cape Coast, Gold Coast .... < 500 5.06 N. I.I6W. 25 Chagwe, Uganda ..... (4,000) (0.25 N.) (33-00 E.) (55) Chepalungu, Kenya .... 5,5oo (i.oo S.) (35-15 E.) (35) Chibi, S. Rhodesia ..... (4,000) 20.19 S. 30.25 E. (30) Chindio, Mozambique .... < 5 I7-39S. 35-14 E. (35) Christiansborg, Gold Coast. See Accra. Coquilhatville, Belgian Congo . 1,200 0.04 N. 18.20 E. (70) Costermansville, Belgian Congo 5,300 2.308. 28.51 E. (55) Dagu, Sudan . . . . (2,000) (13-39 N.) (30.08 E.) (15) Dakar, Senegal ..... < 5 14.40 N. 17.26 W. 20 Damazin Cataract, Sudan (1,000) (11.44 N.) (34.25 E.) (30) Danagla, Sudan . . . . (1,300) (14.25 N.) (33-31 E.) (15) Dar-es-Salaam, Tanganyika < 500 6.508. 39-I7E. 45 Darfur Prov., Sudan .... (2,500) (13.00 N.) (25.00 E.) (15) Dar Shol, Sudan ..... (2,500) 11-54 N. 30.28 E. (30) Daru, Sierra Leone . . . 500 (7-59 N.) (10.50 W.) 90 Mt. Darwin, S. Rhodesia. 4,900 16.45 S. 31-38 E. (35) Delami, Sudan . . . . (2,300) (".52 N.) (30.20 E.) (30) Dendezi, Ruanda-Urundi (6,000) (2.288.) (29.03 E.) (55) Digio, Kenya ..... < 500 i. 228. 40.56 E. (35) Diredawa, Abyssinia .... (4,000) 9.40 N. 41.59 E. (30) Dodowa, Gold Coast .... (500) 5-53 N. 0.05 W. (40) Dogankade, Gold Coast .... (500) (8.42 N.) (0.29 W.) (45) Doka, Sudan ...... (1,500) 13.30 N. 34-45 E. (25) Dolo, Kenya ...... (1,000) (4.08 N.) (41.48 E.) (10) Du River, Liberia ..... (0-1,500) (6.15 N.) (10.25 W.) > 100 Duala, Fr. Cameroons .... < 500 4.03 N. 9.41 E. 1 60 Dubreka, Ivory Coast .... < 500 9.48 N. 13.32 W. (170) Dueim, Sudan ..... 1,500 I3.59 N. 32.20 E. IO Duke Town, Nigeria .... < 500 (4-55 N.) (8.19 E.) > IOO Dumi-Mato, Belgian Congo (2,000) 4-3oS. 15-45 E. (50) Dunkiawallia, Sierra Leone. Not identified. Durban, Natal ..... < 5 29.52 S. 31.03 E. 45 Dwaji I., Uganda ..... (3,8oo) (o.oo) (32-56 E.) (80) Eala, Belgian Congo .... I.OOO o.oi N. 18.30 E. 70 Edudu, Belgian Congo .... (2,000) (3.10 S.) (16.30 E.) (60) Eldoret, Kenya ..... 6,900 0.32 N. 35.16 E. 40 El Fasher, Sudan . 2,800 13.38 N. 25.21 E. 10 Elisabethville, Belgian Congo . 4,000 11.40 S. 27.34 E. 50 El Obeid, Sudan 2,200 13.11 N. 30.14 E. 15 Embotyi, Cape Province < 500 (31-308.) (29.50 E.) (45) 262 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) LIST OF LOCALITIES (cont.) Locality Altitude Latitude Longitude Rainfall Empangeni, Zululand .... < 5 28.46 S. 31-55 E. 40 Entebbe, Uganda ..... 3.900 0.05 N. 32.29 E. 60 Enugu, Nigeria ..... 700 6.27 N. 7.29 E. 70 Epworth, S. Rhodesia .... (5,000) (17.56 S.) (31.36 E.) (35) Erkowit, Sudan ..... 4.300 18.46 N. 36.06 E. (5) Eshowe, Zululand ..... 1,700 28.53 S. 31.28 E. 60 Firestone Plantation, Liberia . < 500 (6.18 N.) (10.25 W.) > IOO Firiwa, Sierra Leone .... (1,500) (9.20 N.) (11.18 wo (80) Forcados, Nigeria ..... < 500 5.26 N. 5.26 E. 130 Fort Hall, Kenya 4,100 0.42 S. 37.10 E. 55 Fort Johnston, Nyasaland 1,700 14.25 S. 35-16 E. 25 Fort Maguire, Nyasaland i, 600 13.40 S. 34-57 E. (30) Fort Portal, Uganda .... 5,100 0.40 N. 30.17 E. 55 Fort Ternan, Kenya .... 7,000 0.12 S. 35-20 E. 50 Fort Victoria, S. Rhodesia 3.70 20.04 S. 30.46 E. 25 Francistown, Bechuanaland 3.30 2I.I 3 S. 27.30 E. 20 Fransfontein, S.W. Africa (3,500) (20.13 S.) (15.01 E.) (10) Freetown, Sierra Leone .... < 500 8.30 N. 13.10 W. 1 60 Funtua, Nigeria ..... (2,500) (11-32 N.) (7-15 E.)- (55) Gadau, Nigeria ..... (1,000) (11.50 N.) (IO.I2E.) 30 Gambaga, Gold Coast .... 1,300 10.33 N. 0.27 W. (45) Garissa, Kenya ..... (500) 0.27 S. 39.38 E. 10 Garsen, Kenya ..... < 500 2.16 S. 40.06 E. (3o) Gatooma, S. Rhodesia .... 3,800 18.208. 29.51 E. 30 Gbohun-Gbohun, Nigeria. See Lagos. Gedaref, Sudan ..... 2,400 14.02 N. 35-24 E. 25 Gede, Kenya ...... < 500 3.198. 40.01 E. 45 Gedeis, Br. Somaliland .... (3,500) (10.03 S.) (45.30 E.) (15) Geneina, Sudan ..... 3,100 13.29 N. 22.27 E. 25 Godia, Uganda ..... (4,000) 3-25 N. 30.58 E. (60) Goromonzi, S. Rhodesia .... (4,800) (17.51 S.) (31.22 E.) (35) Gotani, Kenya ..... (1,000) 3-47S. 39-32 E. (45) Grahamstown, Cape Province . i, 800 33-iS S. 26.32 E. 30 Grootfontein, S.W. Africa (4,000) 19.31 s. 18.06 E. 20 Guara Guara, Mozambique < 500 (25.00 S.) (32.30E.) (25) Gwelo, S. Rhodesia .... 4,70 19.28 S. 29.48 E. 25 Hadibo, Sokotra I. < 500 (12.37 N.) (54-05 E.) (15) Lake Hardin, Abyssinia .... (3,000) (8.52 N.) (39.50 E.) (40) Harrar, Abyssinia ..... 5,5oo 9.22 N. 42.02 E. 35 Heiban, Sudan ..... 3,400 11.05 N. 30.05 E. 35 Ibadan, Nigeria ..... 700 7.23 N. 3-50 E. 50 Ibi, Nigeria ...... < 5 8.ii N. 9-43 E. (45) Ikeja, Nigeria ..... < 5 (6.29 N.) (3.14 E.) (60) Ikom, Nigeria ..... < 500 6.00 N. 8.40 E. (90) Ikoyi, Nigeria. See Lagos Ilado, Nigeria. See Lagos. I Iambi, Belgian Congo .... (i,5oo) 0.34 N. 24.15 E. 80 Inongo, Belgian Congo .... I.IOO 1-55 S. 18.30 E. 65 Irumu, Belgian Congo .... (3.000) 1.25 N. 30.00 E. 50 Isangi, Belgian Congo . . (1,000) 0.42 N. 24-15 E. (80) Isiolo, Kenya . . . . (4,000) 0.22 N. 37-35 E. 20 Isipingo, Natal ..... < 500 30.00 S. 30.55 E. (40) Itowolo, Nigeria ..... < 500 (6.28 N.) (3-17 E.) (60) Itwara Forest, Uganda .... (5,000) (0-45 N.) (30.25 E.) (55) Iwonran, Nigeria. See Lagos. Jadotville, Belgian Congo (4,300) 11.00 S. 26.44 E. (50) IN THE ETHIOPIAN REGION LIST OF LOCALITIES (cont.) 263 Locality A Ititude Latitude Longitude Rainfall Jano I., Uganda . . . (3,8oo) (0.15 S.) (32-36 E) (75) Jebelein, Sudan ..... 1,500 12.37 N. 32.50 E. (15) Jebel Gurein, Sudan .... (1,400) (13-39 N.) (34.41 E.) (25) Jebel Jihaf, Aden Prot. .... 7,100 (13-45 N.) (44.40 E.) (10) Jebel Mandera, Sudan .... (1,300) (14-25 N.) (33-31 E.) (15) Jebel Marra, Sudan .... (5-12,000) 13.00 N. 24.20 E. 25 Jebel Meidob, Sudan .... (2,500) (15-18 N.) (26.25 E.) 10 Jebel Moya, Sudan .... (i,5oo) (13.28 N.) (33.22 E.) (15) Jefisi, Gold Coast . . . . (1,000) 10.45 N. 2.i6W. (40) Jinja, Ugana ...... 3,900 (0.26 N.) (33." E.) 45 Juba, Sudan ...... i, 800 4-51 N. 31-37 E. 35 Jubiya Forest, Uganda .... (4,000) (0.15 S.) (3L59 E.) (55) Kaballa, Sierra Leone .... (1,500) 9-39 N. 11.41 W. 80 Kabarole, Uganda. See Fort Portal. Kabinda, Belgian Congo .... 3,100 6.08 S. 24.30 E. 65 Kabwoch, Kenya ..... (4,000) 0.45 S. 34.27 E. (65) Kadugli, Sudan ..... 2,000 11.00 N. 29.43 E. 30 Kaduna, Nigeria ..... 1,900 10.30 N. 7.28 E. 45 Kailahun, Sierra Leone .... (1,000) (8.16 N.) (10.36 W.) 1 20 Kaimosi, Kenya ..... 5-500 0.05 N. 34-50 E. 70 Kajiado, Kenya ..... (6,000) 1.51 S. 36.47 E. (20) Kajo Kaji, Sudan ..... 4,100 3-53 N. 31.40 E. 50 Kaka Town, Liberia .... (1,000) 6.34 N. 10.20 W. 100 Kakamega, Kenya ..... 5,100 (0.14 N.) (34.44 E.) 70 Kakata, Liberia ..... (1,000) (6.30 N.) (10.25 W.) IOO Kakuri, Nigeria ..... (i,5oo) (10.26 N.) (7.26 E.) (45) Kalinzu Forest, Uganda .... (4,000) (0.25 S.) (30.05 E.) (55) Kallansiya, Sokotra I. . . . < 500 (12.39 N.) (53.38 E.) (15) Kalora, Sudan ..... (4,000) (11-55 N.) (31-18 E.) (25) Kalungu, Uganda ..... (4,000) (0.20 S.) (31.46 E.) (45) Kamabai, Sierra Leone .... (i,5oo) (9.20 N.) (11.58 W.) (10) Kamasiki, Sierra Leone. Unidentified. Kampala, Uganda ..... 3,900 0.19 N. 32-35 E. 45 Kano, Nigeria ..... 1,500 I2.O2 N. 8.32 E. 35 Kaparata, Belgian Congo 3,800 (0.25 N.) (29-33 E.) (60) Karabib, S.W. Africa .... 3,800 (21.55 S.) (15.40 E.) 5 Karimi, Belgian Congo .... (4,5oo) (0.07 N.) (29.47 E.) (55) Kasane, Bechuanaland .... (3,000) 17.49 S. 25.09 E. 20 Kasongo, Belgian Congo .... (2,000) 4.278. 26.39 E. (50) Kasyoha Forest, Uganda. (4,000) (0.20 S.) (30.10 E.) (55) Kassala, Sudan ..... 2,000 15.28 N. 36.24 E. 15 Katagum, Nigeria ..... (1,000) I2.I8N. 10.20 E. 20 Kateri, Nigeria ..... 1,700 (9-37 N.) (7-25 E.) (60) Kathub, Sokotra I. < 500 (12.36 N.) (54.00 E.) (15) Katompe, Belgian Congo (2,500) 6.06 S. 26.24 E. (40) Katsina, Nigeria ..... i, 800 13.01 N. 7.30 E. 35 Kau, Sudan ...... (1,000) (10.40 N.) (31-30 E.) (25) Kauda, Sudan ..... (2,500) (11.01 N.) (30-33 E.) (35) Kayembe-Mukulu, Belgian Congo (2,500) (9.30 S.) (25-36 E.) (50) Kericho, Kenya ..... 6,400 0.23 S. 35-17 E. 65 Kerio, Kenya ..... (4,000) 0.38 N. 35-37 E. (45) Khartoum, Sudan ..... 1,500 15-35 N. 32-35 E. 5 Kheiwok, Sudan ..... (2,000) 12.42 N. 29.09 E. (15) Khor Kobwa, Sudan .... (2,500) (4.02 N.) (30.35 E.) (55) Kiambu, Kenya ..... 6,300 1.138 36.52 E. 4 Kiansonzi, Uganda .... (4,000) (0.14 N.) (31-45 E.) (35) 264 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) LIST OF LOCALITIES (cont.) Locality A Ititude Latitude Longitude Rainfall Kibale Forest, Uganda .... (4,500) (0.30 N.) (30.26 E.) (50) Kibigori, Kenya ..... (4,000) 0.05 S. 35-03 E. (65) Kibokolo do Zombo, Angola 3,800 6.22 S. I5-I7E. 50 Kibos, Kenya ..... 3,800 0.04 S. 34-51 E. 45 Kilembe, Uganda ..... 4-500 (0.25 N.) (29-59 E.) (60) Kilifi, Kenya ..... < 500 3-37 S. 39-52 E. 40 Kimbao, Belgian Congo .... (3.500) (5-30 S.) (17.30 E.) 65 Kimberley, Cape Province 4,000 28.42 S. 24.59 E. 15 Kimvula, Belgian Congo .... (2,500) (5-44 S.) (16.04 E.) (50) Kindia, Fr. Guinea ..... 1,400 10.03 N. 12.50 W. 80 Kingangati, Belgian Congo (2,500) (4.22 S.) (15-50 E.) (50) Kinshasa, Belgian Congo. See Leopoldville. Kintampo, Gold Coast .... 1,200 8.04 N. 1.42 W. (55) Kipini, Kenya ..... < 500 2.32 S. 40.32 E. (45) Kiryandongo, Uganda .... 4,000 1-53 N. 32.03 E. 50 Kisantu, Belgian Congo .... 1,900 5.07 S. 15.07 E. (50) Kisenyi, Ruanda-Urundi 4,800 1.308. 29.10 E. 50 Kisii, Kenya ...... 5,70 0.41 S. 34-47 E. 65 Kissy, Sierra Leone .... < 500 (8.29 N.) (13-12 W.) > IOO Kisui, Belgian Congo .... (1,500) (0.22 S.) (25-38 E.) (70) Kisumu, Kenya ..... 3,800 0.06 S. 34-45 E. 45 Kitale, Kenya ..... 6,200 i. 02 N. 35.00 E. 45 Kitasa, Uganda. See Entebbe. Kitgum, Uganda ..... 3,600 3.20 N. 32-53 E. 50 Kitinda, Uganda. See Entebbe. Kitui, Kenya ..... 3,900 1. 21 S. 38.01 E. 30 Kivanda-Kapepulu, Belgian Congo . (4,000) (10.00 S.) (24.00 E.) (50) Koinadugu, Sierra Leone (1,500) (9.30 N.) (11.25 W.) (80) Komatipoort, Transvaal .... 500 25.26 S. 31-57 E. 25 Kome I., Uganda ..... (3,8oo) 0.05 S. 32.45 E. (80) Koru, Kenya ..... 5,900 o.n S. 35.16 E. 60 Kosti, Sudan ..... 1,500 13.10 N. 32.40 E. 15 Kribi, Fr. Cameroons .... < 500 2-59 N. 9.56 E. no Kudemsa, Nigeria ..... (2,000) (10.23 N.) (7-33 E.) (45) Kudring, Sudan ..... (2,700) (11.39 N. (30.32 E.) (35) Kulmasa, Gold Coast .... (1,000) (9-42 N.) (2.25 W.) (40) Kumasi, Gold Coast .... (800) 6.43 N. 1.37 w. 70 Kumba, Belgian Congo . (i,5oo) (I-50S:) (25-50 E.) (70) Kumba, Br. Cameroons .... 800 (4.40 N.) (9-35 E.) > IOO Kwamouth, Belgian Congo i, 600 3.128. 16.15 E. 60 Kyarumba, Uganda .... (5,5oo) 0.09 N. 29.58 E. (50) Kyenjojo, Uganda .... 4,800 0.36 N. 30-39 E. 45 Kyere, Uganda ..... 3,500 1.29 N. 33-37 E. 50 Lado, Sudan ...... (i,5oo) (5-01 N.) (31-43 E.) (35) Lagos, Nigeria ..... < 500 6.20 N. 3.20 E. 70 Langata Forest, Kenya .... (5,500) 1.20 S. 36.45 E. (35) Las Anod, Br. Somaliland . . ' (2,500) (8.31 N.) (47-26 E.) (5) Laufori, Uganda ..... (2,500) 3.36 N. 31-35 E. (50) Lela, Kenya ...... (4,500) 0.02 S. 34.36 E. (65) Lembwa River, Belgian Congo (4,000) 6.01 S. 17.06 E. (60) Leopoldville, Belgian Congo 1,100 4.20 S. I5.I8E. 55 Letaba, Transvaal < i, 800 23-52 S. 30.18 E. (25) Leysdorp Road, Transvaal (2,000) (24.00 S.) (30.22 E.) 30 Libenge, Belgian Congo .... 1,300 3.40 N. 18.39 E. 55 Likoma I., Nyasaland .... 1, 600 12.04 S. 34-44 E. (75) Lilixia, Gold Coast ..... (1,000) 10.46 N. 2.06 W. (40) IN THE ETHIOPIAN REGION LIST OF LOCALITIES (cont.} 265 Locality Altitude Latitude Longitude Rainfall Lindi, Tanganyika ..... < 500 10.00 S. 39.42 E. 35 Li Rangu, Sudan ..... (2,000) (4.40 N.) (28.21 E.) (55) Livingstone, N. Rhodesia 3,200 I7-50S. 25.49 E. 25 Loka, Sudan ..... (1,000) 4.14 N. 31.01 E. (50) Lokoja, Nigeria ..... < 500 7.48 N. 6.44 E. 50 Londiani, Kenya ..... 7.500 o.io S. 35.36 E. 45 Lorha, Gold Coast ..... < 500 10.39 N. 2.52 W. (45) Louis Trichardt, Transvaal 3,200 23.03 S. 29-54 E. 30 Louren9o Marques, Mozambique < 500 25.58 S. 32-36E. 25 Luanda, Kenya ..... (4.500) 0.02 N. 34-34 E. 65 Lukula, Belgian Congo .... < 500 5-22 S. 12.59 E. 45 Luluabourg, Belgian Congo 2,100 5-56 S. 22.I8E. 65 Lumbwa, Kenya ..... 6,300 0.12 S. 35-29 E. 50 Mabang, Sierra Leone .... < 500 (8.22 N.) (12.52 W.) (130) Mabira Forest, Uganda .... (4,000) (0.28 N.) (32.58 E.) (55) Maiduguri, Nigeria ..... 1,200 11.49 N. 13.09 E. 20 Makeni, Sierra Leone .... < 500 (8-55 N.) (12.05 W.) I2O Makindu, Kenya ..... 3.300 2.178. 37-49 E. 25 Malakal, Sudan ..... 1,500 9.32 N. 31-39 E. 30 Malakisi, Kenya ..... (4,000) (0.25 N.) (34.20 E.) (70) Malindi, Kenya ..... < 500 3.178. 40.07 E. 40 Maluku, Belgian Congo .... (1,000) 4.04 S. 15-33 E. (55) Mambone, Mozambique .... < 500 20.58 S. 35.00 E. (35) Mambrui, Kenya ..... < 500 3-07 S. 40.09 E. (40) Mamirimiri, Uganda .... (3,000) (0.48 N.) (30.09 E.) (55) Manjuguja North, Uganda (3,000) (0-45 N.) (30.03 E.) (55) Mara River, Kenya .... (6,500) 1.078. 35-08 E. (45) Maragua, Kenya ..... (4,000) 0.47 S. 37.07 E. (35) Maranka Reserve, S. Rhodesia (4,000) (19.15 S.) (32-15 E.) (35) Margate, Natal ..... < 5 (30.51 s.) (30.22 E.) (45) Masaka, Uganda ..... 4.300 0.20 S. 31-44 E. 45 Maseno, Kenya ..... 6,000 o.oi S. 34.36 E. 50 Masindi Port, Uganda .... 3,400 1.42 N. 32.05 E. 40 Matadi, Belgian Congo .... < 500 5-508. 13-31 E. 45 Matopo Hills, S. Rhodesia (4,000-5,000) 20.40 S. 28.30 E. (25) Maun, Bechuanaland .... (3,000) 19-59 S. 23.25 E. (15) Mayoro, Gold Coast (5oo) (10.56 N.) (1.04 W.) (40) Mayumba Forest, Gaboon < 500 3-24 S. 10.37 E. (60) Mbango, Uganda ..... (3,000) (0-45 N.) (30.03 E.) (55) Mbarara, Uganda ..... 4,700 0.37 S. 30.39 E. 35 Meridi, Sudan ..... 2,900 4.52 N. 29.28 E. 55 Meru, Kenya ...... 6,200 0.03 N. 37-39 E. 60 Meta Gafersa, Abyssinia .... (5,000) (4-50 N.) (38.45 E.) (30) Lake Metahara, Abyssinia (3,000) 8.52 N. 39-51 E. (40) Metu, Uganda ..... (3,000) 3-39 N. 31-47 E. (50) Mhlatuze Settlement, Zululand (1,000) (28.40 S.) (31.45 E.) 55 Middle Point, Nigeria. See Lagos. Midia, Uganda ..... (4,000) 3.27 N. 31.01 E. (50) Misselele, Br. Cameroons < 500 (4.06 N.) (9.25 E.) 85 Mityana, Uganda ..... (4,000) (0.24 N.) (32-05 E.) (45) Mnazi Moja, Zanzibar. See Zanzibar Town. Mocuba, Mozambique .... (500) (16.49 S.) (37.01 E.) 40 Mombasa, Kenya ..... < 500 4-03 s. 39.40 E. 50 Mongalla Prov., Sudan .... (1,500) (5-00 N.) (32.00 E.) (35) Mongbwalu, Belgian Congo 4,000 (1-54 N.) (30.04 E.) (55) Mongiro, Uganda ..... (3,000) (0.49 N.) (30.10 E.) (55) ENTOM. II. 5 Kk 266 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE LIST OF LOCALITIES (cont.) Locality A Ititude Latitude Longitude Rainfall Monrovia, Liberia ..... < 500 6.18 N. 10.45 W. > 100 Morogoro, Tanganyika .... 1,700 6.50 S. 37-47 E. 35 Moshi, Tanganyika ..... 2,700 3.208. 37-32 E. 40 Moyale, Abyssinia ..... 3-900 3.30 N. 39.02 E. 25 Moyamba, Sierra Leone .... < 500 8.09 N. 12.22 W. IOO Mozambique, Mozambique < 500 15.00 S. 40.42 E. 40 Mpumu Forest, Uganda .... (4,000) O.I2 N. 32.51 E. (55) Msambweni, Kenya .... < 500 4.278. 39.27 E. 60 Msoneddi, S. Rhodesia .... (4,800) (17.08 S.) (30-53 E.) (35) Mtibwa Forest, Tanganyika (2,000) (6.00 S.) (37-45 E.) (35) Mubende, Uganda ..... 5,100 0.35 N. 31.22 E. 45 Mug, Sudan ...... (2,000) (12.42 N.) (29.09 E.) (15) Mugusi River, Kenya .... (4.500) (0.30 S.) (34-42 E.) (50) Muhangi Forest, Uganda. (4,000) 0.50 N. 30.45 E. (50) Muhoroni, Kenya ..... 4.300 0.09 S. 35-12 E. 60 Mukongoro, Uganda .... 3.300 i. 20 N. 33-53 E. 55 Mukono, Uganda ..... 3.900 O.2I N. 32.45 E. 55 Mumias, Kenya ..... (4.5oo) 0.21 N. 34.28 E. 70 Mushin, Nigeria. See Yaba. Mwanza, Tanganyika .... 3.7oo 2.32 S. 32.53 E. 40 Nacon, Gold Coast ..... (1,000) (10.50 N.) (1.25 W.) (40) Nahud, Sudan ..... 2,IOO 12.42 N. 28.26 E. 15 Nairobi, Kenya ..... 5.500 1.178. 36.50 E. 35 Naivasha, Kenya ..... 6.2OO 0.43 S. 36.21 E. 30 Namaacha, Mozambique .... (800) 25.58 s. 32.08 E. 40 Namanve, Uganda ..... 3.700 O.2I N. 32.41 E. 50 Nampula, Mozambique .... (2,000) 15.068. 39-17 E. 45 Nandaw, Gold Coast .... (l.OOO) (10.44 N.) (2.18 W.) (40) Nandawli, Gold Coast .... (1,000) (10.19 N.) (2-35 W.) (45) Nangudi, Gold Coast .... (1,000) (10.51 N.) (0.40 W.) (40) Nanyuki, Kenya ..... 6,400 o.oi N. 37-05 E. 30 Narok, Kenya ..... (6,500) 1.05 S. 35-51 E. 25 Nata, Bechuanaland .... (4,000) 25-158. 25-I5E. (15) Navarro, Gold Coast .... (1,000) (10.41 N.) (1.50 W.) (40) Ndanga, S. Rhodesia .... (4.000) 20. 1 1 S. 31.20 E. (40) Ndola, N. Rhodesia .... 4,100 12.598. 28.38 E. 50 Nefasit, Eritrea ..... (5,000) (15.17 N.) (39.03 E.) (30) Nelspruit, Transvaal .... 2,300 25-278. 30.58 E. 40 Nhemancone, Mozambique (500) (17.008.) (37.00 E.) (45) Niangara, Belgian Congo (2,500) 3-41 N. 27-55 E. (65) Njala, Sierra Leone .... < 500 (8.05 N.) (I2.IOW.) (no) Nobat Dakim, Aden Prot. 1,100 (I3-I5 N.) (44.40 E.) (5) Nova Lisboa, Angola .... 5.700 12.45 S. 15-48 E. 60 Nova Choupanga, Mozambique (500) (17.098.) (34.40 E.) (30) Nsawam, Gold Coast .... < 500 5-48 N. O.2O W. (45) Nta, Belgian Congo .... (2,500) 4.10 S. 15-45 E. (55) Ntabanana, Zululand .... (1,000) (28.35 S.) (31.40 E.) (30) Ntotoro Valley, Uganda .... (2,000-4,000) (0.50 N.) (30.00 E.) (55) Nuba Mts., Sudan . . . . (2,000-4,000) 11.00 N. 30.30 E. (35) Nyamugasani Valley, Uganda . (3,000-12,000) (o.oi S.) (29.54 E.) (50) Obenge-benge, Belgian Congo . (1.500) 1-33 S. 25.05 E. (75) Oblogo, Gold Coast .... < 500 (5-36 N.) (0.17 W.) (40) Obuasi, Gold Coast .... (700) 6.12 N. I.40W. (60) Odzi, S. Rhodesia 3,100 18.55 S. 32.25 E. (35) Ogbomosho, Nigeria .... (1,100) 8.06 N. 4.12 E. 40 Okahandja, S.W. Africa . . . 4,400 21.598. 16.56 E. 1 (15) IN THE ETHIOPIAN REGION LIST OF LOCALITIES (cont.) 267 Locality A Ititude Latitude Longitude Rainfall Okimbahe, S.W. Africa .... (2,500) (21.21 S.) (15-23 E.) (10) Omatjette, S.W. Africa .... (3.500) 21.038. 15-32 E. (10) Onderstepoort, Transvaal 4,200 (24.47 S.) (3I-I5E.) 30 Opobo, Nigeria ..... < 500 4-34 N. 7-35 E. IOO Oshogbo, Nigeria ... . . . 1,000 7-47 N. 4.29 E. 45 Otjimbingwe, S.W. Africa (2,500) (22.21 S.) (16.08 E.) (10) Outjo, S.W. Africa ..... 4,000 (20.07 S.) (16.09 E.) 15 Payida, Uganda ..... 5,000 2.25 N. 30.59 E. 55 Pemba I. ...... < 500 5.00 S. 39-50 E. 80 Pietersburg, Transvaal .... 4,200 23-54 S. 29.25 E. 20 Pinna, Gold Coast ..... (1,000) 10.53 N. 1.45 W. (40) Pongwema, Belgian Congo (2,000) 4.15 S. 16.00 E. (55) Ponthierville, Belgian Congo . i, 600 0.22 S. 25.28E. (75) Popokabaka, Belgian Congo (1.500) 5-42 s. 16.40 E. (55) Port Harcourt, Nigeria .... < 500 4.46 N. 7.11 E. 90 Port Herald, Nyasaland < 500 17.02 S. 35-15 E. 35 Port Loko, Sierre Leone < 500 8.46 N. 12.44 W. 105 Port St. Johns, Cape Province < 5 3I-38S. 29-33 E. 50 Pretoria, Transvaal .... 4,400 25-45 S. 28.12 E. 30 Pujehun, Sierra Leone .... < 500 7-35 N. n.o6W. > IOO Quelimane, Mozambique .... 500 (17-51 s.) (36.54 E.) 55 Raga, Sudan ...... i, 800 8.28 N. 25-41 E. 45 Mt. Ramlu, Eritrea .... (T 0006,000) (13.15 N.) (41 .4S E.) (20) Ras Malalu, Eritrea .... < 500 (15.12 N.) \T T^ ~f 1 (39-51 E.) \~ / (10) Ressano Garcia, Mozambique . < 500 25.268. 32.00 E. (25) Roberts Field, Liberia .... < 5 (6.50 N.) (11.22 W.) IOO Roberts Heights, Transvaal (5,000) 25-47 S. 28.08 E. 25 Roseires, Sudan ..... i, 800 11.51 N. 34-23 E. 30 Rumbak, Sudan ..... 1,900 6.48 N. 29.42 E. 40 Rumogi, Uganda ..... (3,000) 3-34 N. 31.21 E. (50) Salisbury, S. Rhodesia .... 4,800 17-508. 31.01 E. 35 Saltpond, Gold Coast .... < 500 5-12 N. 1.05 W. (30) Sambolugu, Gold Coast (1,000) (10.48 N.) (0-53 W.) (40) Sango Bay Forests, Uganda (4,000) (0.50 S.) (31.40 E.) (45) San Salvador, Angola .... 1,900 6.20 S. 14.47 E. 50 Santa Isabel, Fernando Po < 500 3-40 N. 8.46 E. > IOO Sapele, Nigeria ..... < 500 5-53 N. 5.46 E. (100) Savelugu, Gold Coast .... (5oo) 9.38 N. 0.50 W. (40) Segbwema, Sierra Leone .... < 500 (7-59 N.) (10.58 W.) (100) Sennar, Sudan ..... 1,700 13-34 N. 33-34 E. 20 Serakolia, Sierra Leone .... (i,5oo) (9-37 N.) (11.08 W.) (80) Shaki, Nigeria ..... (1,500) (8-34 N.) (3-19 E.) (40) Shaloat, Sudan ..... (2,000) (12.42 N.) (29.09 E.) (15) Shamva, S. Rhodesia .... 3,200 17.20 S. 31-37 E. (35) Sheikh Gamal, Sudan (2,000) (12.42 N.) (29.09 E.) (15) Sheikh Karim, Sudan .... (2,000) (u.oi N.) (30-33 E.) (35) Sheik Othaman, Aden Prot. < 500 (12-53 N.) (45-03 E.) (5) Shimoni, Kenya ..... < 500 4-38S. 39.23 E. (55) Sidamo Prov., Abyssinia 5,900 6.10 N. 38.45 E. (40) Singa, Sudan ..... (1,400) 13.10 N. 33-55 E. 25 Sinoia, S. Rhodesia. .... 3,800 17.22 S. 30.11 E. 30 Sipi, Uganda ...... 6,700 i. 20 N. 34-22 E. (70) Skukuza, Transvaal .... (1,000) 24.59 S. 31.36 E. 20 Sondu River, Kenya .... (6,000) 0.31 S. 35.04 E. (60) Sonkonia, Sierra Leone .... (1,500) (9-45 N.) (11.26 W.) (80) Sopo, Sudan ...... (2,000) (8.06 N.) (26.08 E.) (45) 268 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) LIST OF LOCALITIES (cont.) Locality Altitude Latitude Longitude Rainfall Soroti, Uganda ..... 3,700 1-43 N. 33-37 E. 50 Stanleyville, Belgian Congo 1,400 0.26 N. 25.14 E. 65 Sunyani, Gold Coast .... 1,200 7.22 N. 2.21 W. 60 Tagoi, Sudan . . (3.500) (11-53 N.) (30.55 E.) (25) Takoradi, Gold Coast .... < 500 4-53 N. 1.46 W. 40 Talodi. Sudan ..... 2.OOO 10.37 N. 30.21 E. 30 Tamale, Gold Coast .... 6OO 9.28 N. 0.51 W. 45 Tanga, Tanganyika .... < 500 5-05 s. 39.08 E. 55 Tanina, Gold Coast. .... (1,000) 9-55 N. 2.28 W. (45) Taveta, Kenya ..... 3.500 3-248. 37.40 E. 30 Tembura, Sudan ..... (2,000) (5.36 N.) (27.30 E.) (55) Tessenei, Eritrea ..... (2,OOO) (15.08 N.) (36.39 E.) (15) Thika, Kenya 4,9OO 1.03 S. 37-05 E. 35 'Third Camp', Kenya. Unidentified. Thysville, Belgian Congo 2.OOO 5.16 S. 14.54 E. (50) Tirikoro, Sierra Leone .... (1.500) (9-05 N.) (10.50 W.) (90) Tishi, Gold Coast (1,000) (11.02 N.) (0.26 W.) (35) Titule, Belgian Congo .... (1.500) 3-15 N. 25-35 E. (75) Tiwi, Kenya ...... < 500 4-158. 39-35 E. 65 Tomango, Transvaal .... 2,400 (25.278.) (30-58 E.) (30) Torit, Sudan 2,500 (4-25 N.) (32-33 E.) 40 Triangle Estate, S. Rhodesia . (2,500) (21.008.) (31.30 E.) (25) Tshela, Belgian Congo .... 800 4-58 S. 12.56 E. 50 Tshumbiri, Belgian Congo (1,500) (2.51 S.) (16.15 E.) (60) Tsumeb, S.W. Africa .... 4,200 19.14 S. 17.42 E. 20 Tsunikitoko, Belgian Congo (2,000) 4.20 S. 16.20 E. (55) Tumu, Gold Coast ..... 1,000 10.52 N. i-59 W. (40) Turbo, Kenya ..... 5,900 0.38 N. 35-03 E. 50 Tzaneen, Transvaal. .... 2,500 23-50 S. 30.10 E. 30 Ubassa, Nigeria ..... (1,500) (10-35 N.) (7.20 E.) (45) Ugaro, Eritrea . . (2,000) (14.46 N.) (37.20 E.) (20) Ukara I., Tanganyika .... 3.900 1.51 S. 33-03 E. 60 Ulu, Gold Coast (1,000) (10.42 N.) (2-47 W.) (45) Ulub, Aden Prot (3,000) (13.30 N.) (44.42 E.) (5) Umbeluzi, Mozambique .... (500) 26.02 S. 32.20 E. 25 Umm Ruaba, Sudan .... 1, 800 12.53 N. 31.13 E. 15 Umm Sunt, Sudan ..... (1,500) (14.20 N.) (33-35 E.) (15) Umtali, S. Rhodesia .... 3,700 18.588. 32.40 E. 30 Upington, Cape Province 2,600 28.27 S. 21.15 E. 10 Upper Sheikh, Br. Somaliland . 4,700 (9.56 N.) (45.12 E.) 20 Usakos, S.W. Africa .... 2,900 22.OI S. 15-35 E. 5 Usumbura, Ruanda-Urundi 2,600 3.26 S. 29.15 E. 35 River Vakila, Belgian Congo . 5,100 (8.30 S.) (27.15 E.) (50) Victoria Falls, S. Rhodesia 3,000 17.528. 25-51 E. 20 Wa, Gold Coast 1,100 10.05 N. 2.27 W. (45) Wad Banda, Sudan .... (2,000) 13.06 N. 27-55 E. (15) Wad el Magdub, Sudan .... (1.300) (14.23 N.) (33.24 E.) (15) Wad Ganatir, Sudan .... (2,000) (12-35 N.) (28.37 E.) (15) Wadiyain, Arabia ..... (1,000) (15-30 N.) (47.00 E.) (5) Wad Medani, Sudan .... 1, 600 14.24 N. 33-31 E. 15 Waliki, Sierra Leone. Unidentified. Waterpoort, Transvaal .... 2,600 (22.53 S.) (29.37 E.) (15) Waterval Boven, Transvaal 4,800 25.38 s. 30.20 E. 30 Wathen, Belgian Congo .... (2,000) (5-o8 S.) (14.42 E.) (50) Wau, Sudan ...... 1,700 7.42 N. 28.01 E. 45 Weenen, Natal ..... 2,800 28.51 S. 30.05 E. 25 IN THE ETHIOPIAN REGION LIST OF LOCALITIES (cont.) 269 Locality A Ititude Latitude Longitude Rainfall Wesu, Kenya . . . . (5,000) 3.228. 38.20 E. 60 Windhoek, S.W. Africa . . . . 5.5oo 22.34 S. 17-05 E. 15 Winduri, Gold Coast .... (1,000) (10.43 N.) (0.50 W.) (40) Witu, Kenya ..... < 500 2.22 S. 40.27 E. 45 Wuri, Fr. Cameroons. See Duala. Yaba, Nigeria ..... < 5 (6.24 N.) (3-20 E.) (70) Yala, Kenya ..... (4.5oo) 0.06 N. 34-32 E. 65 Yambio, Sudan ..... 2,900 4-34 N. 28.24E. 55 Yangambi, Belgian Congo (1,000) 0.50 N. 24.152. 70 Yaounde^ Fr. Cameroons 2,300 3-52 N. 11.31 E. 65 Yape, Gold Coast ..... (500) 9.08 N. i.ioW. (40) Yatolema, Belgian Congo (1.500) o.iSN. 24.32 E. (80) Yeji, Gold Coast < 500 8.13 N. 0.39 W. (50) Yiraia, Sierra Leone . . (1.500) (9.27 N.) (II.20W.) (80) Yokeskei R., Transvaal .... (5,000) (26.05 S.) (28.00 E.) (30) Yola, Nigeria ..... 900 9.13 N. 12.29 E. 40 Lake Young, N. Rhodesia (4.500) (11.14 S.) (31.44 E.) (40) Zaka, S. Rhodesia ..... (3,000) (20.21 S.) (31.24 E.) (40) Zanzibar Town, Zanzibar < 5 6.10 S. 39.14 E. 65 Zika, Uganda. See Entebbe. Zomba, Nyasaland .... 3,100 15.21 S. 35-25 E. 55 Zouragu, Gold Coast .... (1,000) 10.47 N. 0.47 W. (40) Zungeru, Nigeria ..... (600) 9-45 N. 6.05 E. 45 BIONOMICS IN RELATION TO DISTRIBUTION The following notes are intended only to illustrate certain aspects of bionomics of particular importance to the study of distribution. The subjects discussed include breeding-places, seasonal distribution, and biting-habits (including vertical distribu- tion). It will be seen that our knowledge of all these matters, with the possible exception of breeding-places, is still very rudimentary. It could have been much less so had collectors realized the importance of publishing full data wherever possible. Observations which may by themselves appear insignificant often acquire consider- able significance when related to our knowledge as a whole. BREEDING-PLACES The fact that the majority of Stegomyia breed for the most part in tree-holes or plant axils has had an important bearing on the nature of the distribution records available. Much collecting has been carried out in Africa in the course of malaria surveys, but in work of this kind attention is commonly confined to such breeding- places as ground pools and swamp or stream edges. None of the Ethiopian Anopheles is known to breed in tree-holes. As against this, however, it is interesting to recall that two of the Ethiopian Stegomyia, Aedes africanus and Aedes trinidad, were first obtained from ground pools. Records of this kind are also available for some other species, e.g. luteocephalus (Briscoe, 1950) and apicoargenteus (Evans, 1925), but, except where they refer to rock-pools, they may be assumed to be due almost cer- tainly to flushing of tree-holes by heavy rain. All the records quoted are from places with very heavy rainfall. The converse process, contamination of receptacles by 270 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) ground-pool or stream breeding larvae, also occurs quite commonly. One calabash found by the author in a Ju-ju shrine in Southern Nigeria contained a thriving population of Anopheles gambiae, Megarhinus brevipalpis, and Aedes aegypti. Such cases serve to emphasize the point that, save for certain highly specialized forms, mosquito larvae are capable of a higher degree of tolerance than might be expected from the comparatively restricted range of breeding-places in which they are normally found. The restriction results, in fact, not from the preferences of the larva but from those of the ovipositing female, though no doubt the latter are adapted to providing the larva with optimum conditions for development (see Russell & Rao, 1942, and summary in Bates, 1949) . It is probable that no Ethiopian Stegomyia breeds exclusively in one type of habitat, and cases in which only a single type of breeding-place is known are probably attributable to insufficient collecting. Nevertheless with the exception of certain highly plastic species (e.g. aegypti, albopictus) it is generally possible to distinguish a breeding-place of choice from others which are merely occasional or accidental. The whole question of the selection of breeding-places is a complex one and has never been systematically studied, at least in the present group. It would appear that the adoption of an unusual larval habitat may result from diametrically opposite conditions, namely the absence of suitable breeding- places or the existence of such ideal conditions that a teeming population is produced which 'overflows' into every usable niche. The list of breeding-places which follows makes no pretence to completeness, but an effort has been made to include the preferred breeding-place of each species, where this is known, together with such records of occasional breeding-places as seem necessary to give an idea of the adaptability of the species in question. The ecology of the various classes of breeding-place is discussed briefly under seasonal distribution. Aedes apicoargenteus. Preferred. Tree-holes (Hopkins, 1936 ; Garnham et aL, 1946). Occasional. Rock-holes, utensils, bamboo stumps (Hopkins, 1936; Harris, 1942). Aedes fraseri. Preferred. Tree-holes (Hopkins, 1936 ; Garnham et al., 1946). Occasional. Rock-holes (Garnham et al., 1946), rot-hole in mangrove (Holstein in litt.). Aedes dendrophilus. Preferred. Tree-holes (Hopkins, 1936 ; Van Someren, 1946). Occasional. Banana axils, banana and bamboo stumps (Hopkins, 1936), rock-holes (Garnham et al., 1946). Aedes africanus. Preferred. Tree-holes (Hopkins, 1936, also mentions it as common in bamboo stumps). Occasional. Utensils (Hopkins, 1936), rock-holes (Garnham et al., 1946), bamboo pots (Robinson, 1950). Aedes pseudoafricanus. Preferred. Rot-holes in Avicennia mangrove (Chwatt, 1949). Occasional. No other breeding-places have been recorded. Aedes simpsoni. Preferred. Plant axils, especially Colocasia, Gonja banana, and pineapple (Haddow, 1948). May be locally abundant in tree-holes (Hopkins, 1936) or in utensils (Lewis, 19430;). Occasional. Coco-nut shells, tubs (Hopkins, 1936), rock-pools (Lewis, 19430:), a concrete basin (Abbott, 1948), bamboo pots (Robinson, 1950), axils of Dracaena and Strelitzia (Muspratt, 1950 and in litt.). Aedes luteocephalus. Preferred. Tree-holes (Hopkins, 1936; Lewis, 19430;). Occasional. Cut bamboos, utensils, crab-holes ( ? accidental) (Hopkins, 1936), rock- holes (Hopkins, 1936; Abbott, 1948), temporary pools (almost certainly accidental) (Briscoe, 1950). Aedes unilineatus. Preferred. Tree-holes (Hopkins, 1936; Lewis, 19430;; Barraud, 1934). Occasional. Screw-palm axils (Ingram & De Meillon, 1929), IN THE ETHIOPIAN REGION 271 rot-hole in paw-paw (Hopkins, 1936), bamboo pots (Robinson, 1950), bamboo stumps (author, unpublished). Aedes metallicus. Preferred. Tree-holes (Hopkins, 1936). Occasional. Utensils, coco-nut shells (Hopkins, 1936), rock-pools (Lewis, 19430), banana axils (Teesdale, 1941), a concrete basin (Abbott, 1948), bamboo pots (Robin- son, 1950). Aedes vittatus. Preferred. Rock-pools (including holes in coral, Wiseman et al., 1939) (Hopkins, 1936). Occasional. Utensils, hoof-prints, boats, wells (Hopkins, 1936), tree-holes (Kerr, 1933; Rageau in litt.), bamboo pots (Harris, 1942). Aedes aegypti. Common in all kinds of utensils, gutters, boats, water-tanks, &c. May be locally abundant in wells (Hopkins, 1936). Dalziel (1920) records it, surprisingly, as abundant in crab-holes and this has been confirmed by Dunn (1927, &c.) and by Riqueau (1929). Apart from this the dictum enunciated by Carter (1924), that the species is never found breeding right through from oviposition to adult in pools all the sides of which, at the waterline, are of mud, appears to hold good. Berner (1947) gives records from residual pools in earth drains and the author has had similar experiences, but it is extremely difficult in these cases to rule out contamination. Some interesting wild populations have been described, notably by Haddow (19450, 1948) and Garnham et al. (1946). Haddow found larvae most commonly in tree-holes, less commonly in plant axils. Garnham and his fellow workers found them most commonly in rock-holes and in holes in recently felled trees, less commonly in true tree-holes (for a discussion of this point see below under seasonal distribution). Robinson (1950) found a high proportion of the tree-hole population to consist of this species. He also gives records from rock-pools which seem to constitute a dry season refuge. Adults of Aedes aegypti recently received from Dr. Fain were bred out from holes in old weathered lava on the lava plain north of Lake Kivu near Sake. These holes were said to be about 10 km. from the nearest human settlement. SEASONAL DISTRIBUTION A number of scattered records are available, but their sum total is surprisingly small and little can be gained from attempting to correlate them. Most of our knowledge therefore rests on a few studies pursued continuously over a period in particular localities. There can be no question but that this is a most profitable field for study, and good figures, were they available, would throw much light both on bionomics and on distribution. As it is, so little is known regarding differences between individual species that the subject is best discussed in relation to breeding- places. For this purpose the latter are classified in accordance with common practice as tree-holes, bamboos, plant axils, rock-holes, and artificial containers. It must, however, be emphasized that this is not a natural classification and that even in the present state of our knowledge further subdivision is required. Tree-holes. Two distinct types were recognized by Garnham et al. (1946), who found that the large horizontal tree-hole not uncommonly encountered in fallen tree-trunks was unattractive to most Stegomyia although apparently preferred to the smaller type of tree-hole by Aedes aegypti. This is in accordance with the author's experience in Southern Nigeria except that such holes were not found to contain even Aedes aegypti. The 'dominant species' appeared to be Culex macfiei. These observations appear to indicate that the attraction for the ovipositing female is not purely a 272 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) chemical one depending on the nature of the walls of the breeding-place but is to some extent conditioned by the degree of exposure of the water surface to insolation. If this is so, then it might be expected that tree-holes with a more or less vertical opening and deeply shaded interior would be more attractive to certain species than those in the forks of trees with more or less horizontal opening. It is clear also that the situation of the breeding-place relative to shade trees will be of importance. Dunn (1928) made a quantitative study of the fauna of a tree-hole with a small opening vis-a-vis one with a large opening. His observations appear to indicate a general preference on the part of Stegomyia for the latter, but being based on only a single tree-hole of each type they must be regarded as inconclusive. A number of investigators (Dunn, 19270; Harris, 1942; Garnham et al., 1946; Hocking, 1947; Bailey, 1947) have employed 'artificial tree-holes', made by sawing off joints of bamboo, as traps, and Dunn showed these to be more attractive than containers made of metal. It would be interesting to compare the attractiveness of a container with deeply shaded interior, such as a narrow-necked calabash, with that of a bamboo pot, but this does not appear to have been done. Larvae of Aedes unilineatus, metallicus, aegypti, simpsoni, and/rasm' have all been obtained by immersing dried material from tree-holes, and the larva of Aedes subargenteus was first obtained in this way. It is presumed that these species survive through the dry season by means of drought-resistant eggs (Taylor, 1934 ; Hopkins, 1936 ; Garnham et al., 1946 ; Gillett et al., 1950), but it appears highly probable from the distribution studies discussed below that there is some variation in this respect between individual species. Robinson (1950) appears to regard survival in the adult state as more probable in Northern Rhodesia, and Taylor (1934) has an interesting negative record for luteocephalus. The possibility of obtaining larvae by immersion of tree-hole scrapings during dry-season surveys should be borne in mind by the collector particularly as such scrapings are much more easily transported than living adults or early stages. Hatching does not usually take place immediately the first rain falls. Nor is it to be expected that all the eggs in a tree-hole will hatch on the first immersion. Ingram & Macfie (19166) found larvae at Accra in March, i.e. right at the beginning of the rainy season, but Taylor, working at Gadau, noted that 'although a few showers fell during May and June no rainwater remained more than a few hours in any of the tree-holes until the beginning of July', and Muspratt (1945) notes that at Livingstone, where the rainy season lasts from October to May, larvae are usually found from the middle of November onwards. Robinson (1950) states that at Livingstone he found larvae in tree-holes from December to May inclusive and that 'one to one and a half inches of rain falling over a period of one to two weeks is sufficient to initiate breeding in tree-holes, although most holes do not get sufficient water before about three inches have fallen'. It is to be presumed that the last eggs laid in a large tree-hole during the previous year would be near the bottom and would therefore be flooded well before the tree-hole was full, but this point does not appear to have been investigated. In view of Robinson's findings in Northern Rhodesia it seems almost certain that a number of species of Stegomyia can survive the dry season as adults in wetter parts of the region. There can be no doubt, however, that IN THE ETHIOPIAN REGION 273 the adult population is very drastically reduced even under quite favourable condi- tions. Thus in Bwamba County during a prolonged dry spell lasting from the end of December to the end of March 20 catches made between 2 February and n March yielded only 7 adults of apicoargenteus as compared with 198 from 20 catches made during the preceding wet season (Haddow el al., 1947). At Itowolo 7 catches made between 21 November and 16 May yielded 24 africanus as compared with 501 from 15 catches between 12 June and 25 October. In this case the number of adults taken declined very sharply even during the short dry spell between the large and small rains and 2 catches made between 24 August and 7 September yielded only 7 africanus (Mattingly, 1 949*2). Adults do not, of course, appear in greatest numbers until some time after hatching takes place. At Itowolo greatest numbers of africanus were taken shortly before the period of maximum rainfall in July and a second peak occurred shortly after the heaviest rainfall during the small rains. In this case the seasonal distribution of africanus was closely paralleled by that of the two species of Diceromyia which were the only other tree-hole breeders taken, and it was considered that the type of breeding-place must be the governing factor rather than any intrinsic difference between the species. As against this the figures given by Kerr (1933) for africanus and luteocephalus in the Yaba area, although small, do suggest the possiblity of a specific difference. In this case luteocephalus showed a much smaller peak in relation to the small rains and a correspondingly larger concentration during the period May to July. In addition the average catch obtained during July was more than twice that obtained during June (slightly smaller during July than during June for africanus at both Yaba and Itowolo) . In Bwamba the wet and dry season catches referred to in connexion with apicoargenteus yielded 429 and 34 africanus respectively. Bamboos. Among the Ethiopian Stegomyia the only specialized bamboo breeders are confined to the East African Highlands and cannot therefore be discussed here. Most of the Guinean species appear to breed readily in bamboo pots and several of them are known from bamboo stumps, so that it is clear that this type of habitat closely resembles the small tree-hole. Bored bamboos clearly present a distinct type of habitat, but no information appears to be available regarding their utilization by Guinean species. In Southern Nigeria the most characteristic bamboo breeders appear to be Dunnius spp. Plant axils. These are the subject of a most informative paper by Haddow (1948) devoted to the breeding-places and seasonal distribution of Aedes simpsoni in Bwamba County. This author found that at Bundibugiyo with a rainfall of about 55 inches, local rainfall had little direct effect on the numbers of larvae obtained since a light shower was sufficient to fill the plant axils to capacity and heavy rain could not fill them further. Three or four showers a month appeared to be sufficient under normal circumstances to maintain the larval population. Increased rainfall favoured breed- ing indirectly, however, through its effect on the growth of the plants. A monthly survey of the axils of Xanthosoma sagittifolium showed that these only dried out completely for a short period at the beginning of February and the larval population was renewed within 10 days after the first light rain. During the dry season larvae appeared to be able to survive in the merest film of water. During the short period of drought they could be obtained by immersing scrapings from the dry axils, ENTOM. n. 5 L 1 274 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) presumably from drought-resistant eggs. It seems that in localities with a higher or better distributed rainfall or in a wetter year larvae would occur all the year round and the necessity for a drought-resistant egg would not exist. At Mombasa, with a mean annual rainfall of only 45 inches, Wiseman et al. (1939) found that some banana axils held water throughout the dry season. They found adult simpsoni to be most numerous during the drier months and suggested that heavy rainfall might wash the larvae out of pineapple crowns and leaf axils. Rock-holes. Garnham et al. (1946) showed that these are of two different kinds, each supporting an entirely different fauna. Well-shaded holes in boulders in the Kaimosi Forest were found to harbour typical tree-hole breeding species such as fraseri, africanus, and dendrophilus, but exposed pools outside the forest contained only aegypti and vittatus. Residual pools in stream beds must also be distinguished from those which are directly dependent for their water-supply on rain. Robinson (1950) notes that at Livingstone the occurrence of such pools in combination with tree-holes may serve to provide Aedes aegypti and vittatus with natural breeding- places for nine months of the year, whereas tree-holes alone were available only for six months. In general it would seem that the latter species is likely to be most abundant during the dry season when the water level in stream beds is low (see also Gil Collado, 1935). A rain-filled rock-hole exposed to sunlight is in general more liable to rapid desiccation than a tree-hole, and species utilizing such breeding-places must be capable of very rapid larval development. Such a capacity is certainly possessed by vittatus, pupae of which have been found in rock holes only 3 days after the onset of rain (Lamborn, 1930). Aedes aegypti is also capable of very rapid development, at least when adequate food is available. On a particularly rich diet this species was found to be capable of completing its development from the moment of hatching to the moment of emergence in 85 hours at normal laboratory temperature (author's unpublished experiments at Yaba). Artificial containers. These do not form a natural group. Their attractiveness for particular species depends in part on their intrinsic nature and in part on their situation. Shade and proximity to dwellings are important extrinsic factors. Most species, other than aegypti, very seldom enter dwellings although some, e.g. africanus, simpsoni, luteocephalus , are occasionally recorded. Even aegypti has been shown to prefer natural breeding-places when these are available in sufficient numbers, e.g. in the case of villages situated near forests or shambas (Haddow, 19450 ; Garnham et al., 1946; Dunn, 19270 and c). Receptacles made of natural organic matter, e.g. the bamboo pots used as traps or calabashes, to some extent mimic tree-holes and form a class by themselves. The only species regularly found in small containers made of metal, pottery, or other inorganic substances or in containers in or immediately about houses is Aedes aegypti, and it is evident that its ability to utilize such breeding- places accounts in large measure for its very wide distribution. Nevertheless it cannot account entirely for this phenomenon. It may explain transportation, especially by sea, and occurrence in urban centres with piped or other permanent water-supplies, but it cannot explain occurrence in rural areas where the presence of water in small containers is as much a seasonal phenomenon as its presence in tree- holes. There can be no question that this is a highly adaptable species both with IN THE ETHIOPIAN REGION 275 respect to temperature and to rainfall requirements, probably the most adaptable of all the Ethiopian Stegomyia with the possible exception of vittatus. In a special class are those very large containers used for conservation, e.g. tanks, cisterns, concrete wells, hollow baobab trees, and these may perhaps be essential to the survival of certain species occurring in very arid parts of the region. Thus Patton (1905) records Aedes vittatus only from such places in the neighbourhood of Aden. Even here, however, fig-trees and residual and other rock pools might be expected to provide natural breeding-places at least in some years. BITING-HABITS Under this heading are included three distinct aspects of mosquito behaviour, namely host preference, the biting-cycle, which is an expression of the concentrations of any particular species found biting at various times of day and night, and vertical distribution, which is an expression of the concentrations found at various heights above ground. All these, no doubt, are related reciprocally to distribution, but in the present state of our knowledge the nature and extent of this relationship can only be dimly apprehended. They are accordingly discussed very briefly. Very little is known concerning host selection in this group. It seems reasonable to suppose that forest species may be, in general, more specific in their preferences than those inhabiting the savannah, and certain observations communicated privately by Dr. Chwatt suggest that africanus may be more exclusively addicted to primate hosts than luteocephalus . A detailed investigation of this point would be of consider- able interest. On the whole the vast and confused literature relating to zoophily and anthropophily in mosquitoes suggests that the majority of species are plastic in the sense that host specificity occurs, if at all, at the level of the strain or biotype rather than that of the species (see, in particular, Galliard, 1936, and Wanson & Nicolay, 1937). In this connexion it may be noted that ' Culex molestus' , usually regarded as a highly anthropophilic form, besides showing a preference for birds over man when given the choice will, like Culex pipiens and Culex fatigans , produce many more eggs on a meal of bird blood than on one of mammal blood (Mattingly et at., in press, and author's unpublished observations on Lagos fatigans). Woke (1937) studied the same question using Aedes aegypti and found that of the various hosts employed man and monkey were the least satisfactory. The figures he obtained, expressed as eggs per mg. of ingested blood, were Monkey 28-2, Man 29-2, Canary 42-4, Turtle 46-3, Rabbit 48-9, Guinea-pig 52-4, and Frog 52-6. As against this Toumanoff (1949) obtained more numerous eggs and greater longevity with human blood than with that of two species of lizards (Gecko and Calotes spp.). There seems to be no reason to suppose that most mosquito species are any less plastic in this respect than in their host preferences. If this is so, then the species population may be expected to adapt itself to whatever selection of hosts is available in a given locality and we ought not to be surprised if a species which readily bites man in one part of its range refuses to do so in another. A number of inconsistencies of this kind are in fact to be found in the literature relating to the Ethiopian Stegomyia. Thus De Meillon (19470) states that in South Africa Aedes vittatus does not bite man, but Robinson notes (in litt.) that it bites in Northern Rhodesia. Garnham et al. (1946) noted that 'Adults have 276 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) rarely been found, biting, in catches in the vicinity of breeding places' in the Kaimosi Forest, while, on the other hand, Leeson has an interesting unpublished record from Malakisi, not far from Kaimosi, where vittatus bit in abundance in a wooded ravine but were entirely absent from native huts at a similar distance from the breeding- places. Jannone et al. (1946) record it as biting in Eritrea, and Lewis (1943) considered that it was probably the most important vector of yellow fever in the Nuba Mountains because 'it existed in great numbers and bit people by day and in the evening near their homes'. Gil Collado (1935) records the Spanish form as biting man avidly. Similarly Haddow (19450) and Garnham et al. (1946) note that adults of forest populations of aegypti are very rarely taken biting man. Aedes simpsoni has proved equally elusive in Nigeria and the Gold Coast. Full length 24-hour catches at various heights above the ground failed to yield a single adult although larvae were taken abundantly in nearby banana groves (Bugher et al., in press). Short-period catches at ground level in banana groves at Asuboi also failed to yield any adults although larvae were abundant (author's unpublished observation). On the other hand, it is understood from Dr. F. N. Macnamara that this species was quite often taken biting in the British Cameroons, while Haddow (1945^ was able to take 596 adults in the course of ten 24-hour catches in Bwamba County. Bostock & Simpson (1905) state that this is one of the commonest species in the Low Veld region of the Transvaal, suggesting by implication that here also it bites readily. Aedes unilineatus appears very rarely to bite man. The only records of its biting or attempting to bite are two from the Sudan (Lewis, 1947). This author failed to observe it biting in the Nuba Mountains although larvae were common. A edes metallicus is recorded by Lewis (1943) as biting only very near its breeding-places in the Sudan. Bailey (1947) rarely took adults in Kenya; Wiseman et al. (1939) found it to bite freely at Mombasa. Robinson found it to the extent of 2 per cent, among just over 2,000 mosquitoes taken biting man at Livingstone. A edes apicoargenteus was taken by Haddow et al. (1947) in Bwamba and Garnham et al. (1946) found it to be 'the dominant species in many adult catches' in the Kaimosi Forest. In Nigeria only 7 specimens were taken during a year's catching (Mattingly, 19496). Garnham et al. (1946) took adults oifraseri in abundance in deep forest, but Haddow et al. (1947) record only one specimen from a considerable number of catches in Bwamba. Haddow, Van Someren, et al., however, record it as biting not uncommonly in this part of its range (in press). Garnham et al. (1946) took no adults of dendrophilus in Kaimosi, but Haddow, Van Someren, et al., record it as 'quite common' in Bwamba. Aedes africanus seems to bite man readily in all parts of its range, and the same is true of pseudoafricanus and luteocephalus. Regarding the biting-cycle it will suffice for present purposes to point out that some species are markedly crepuscular with a strongly marked biting-peak shortly after sunset. These include africanus (Haddow et al., 1947; Mattingly, 19490; Kerr, I 933)> Pseudoafricanus (Mattingly, 19496; Chwatt, 1949), luteocephalus (Kerr, 1933; Lewis, 19430 ; Bugher et al., in press), and metallicus (Lewis, 1947). Others feed quite readily during the daytime, e.g. simpsoni (Haddow, I945c), vittatus (Lewis, 19430), apicoargenteus (Garnham et al., 1946; Haddow et al., 1947), dendrophilus and/rasm' (Haddow, Van Someren, et al., in press), and possibly unilineatus (Lewis, 1947), Concerning hosts other than man little is known, but Haddow & Dick (1948) give IN THE ETHIOPIAN REGION 277 an interesting account of africanus and apicoargenteus feeding on anaesthetized monkeys. De Meillon (19470) took luteocephalus on a freshly killed baboon but failed to take any in a Magoon trap baited with a monkey either at 50 ft. or at ground level. Bedford (1928) gives records of simpsoni and metallicus from horses and cattle. With respect to vertical distribution evidence is available only for a few species. Acrodendrophilic species, i.e. those biting in greatest numbers in the canopy, include africanus (Haddow et al., 1947 ; Mattingly, 19490 ; Garnham et al., 1946), luteocephalus (Bugher^ al., in press; Haddow, Van Someren, et al., in press; De Meillon, 1947), apicoargenteus (Garnham et al., 1946; Haddow et al., 1947), and/rasm' (Garnham et al., 1946). It is very dangerous to base any estimate of the size of the population of such species as africanus on ground-level catches. It must be emphasized that such catches can give little or no idea of the relative abundance of a canopy species. Thus in a series of catches in Bwamba Haddow et al. (1947) took only 22 out of 502 africanus at ground level, while the author (19490) took only 39 at ground level out of a total of 525 at Itowolo. It is true that Kerr (1933) obtained considerable numbers of 'africanus' on the ground, but this was almost certainly due to the presence of pseudoafricanus as a contaminant. The presence of small numbers of the latter species, which is less addicted to the canopy than africanus, may also have been responsible for the rather higher percentage taken on the ground at Itowolo than in Bwamba (7-4 per cent, as compared with 4-4 per cent.). It is also essential that catches should be made during the peak biting-time. Species biting mainly on the ground include dendrophilus (Haddow, Van Someren, et al., in press), simpsoni (Haddow, 19450), aegypti, and metallicus (Bailey, 1947). Still less is known regarding the vertical distribution of breeding-places. Bailey (1947) obtained simpsoni in bamboo pots at ground, level only, and Garnham et al. (1946) give a similar observa- tion on forest-dwelling aegypti. Teesdale (1945), however, obtained the latter in bamboo pots placed on the tops of coco-nut palms. Garnham et al. (1946) obtained fraseri and dendrophilus at all heights up to 60 ft. (the greatest height investigated), but note that africanus, while breeding at heights up to 50 f t. , was five times commoner at ground level than higher up. ZOOGEOGRAPHY The existence of two major elements in the Ethiopian fauna, a West African on the one hand and an East and South African on the other, has long been recognized. Chapin (1932), basing his conclusions mainly on the birds, has mapped out an approximate boundary for the West African sub-region and Edwards (1941) attempted to fit this to the distribution of the mosquitoes. The latter author gives a list of species supposed to be confined to the forested part of this area or to spread only into the adjacent savannah. Among the Stegomyia he includes Aedes apicoargenteus, fraseri, luteocephalus, and dendrophilus. Of these, however, only apicoargenteus occurs within the area of more or less continuous closed canopy forest where it is accompanied by Aedes africanus, a species which Edwards does not include and which is certainly found well outside the sub-region. Aedes luteocephalus is not a forest species at all. It is a savannah species which does not enter closed forest. It extends far beyond the boundaries of the sub-region. Of the four species listed by 278 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) Edwards onlyfraseri and apicoargenteus are even approximately confined to the West African sub-region and even these are found in certain isolated forest areas around the Kavirondo Gulf which lie in Chapin's East African Highland District. Aedes dendro- philus resembles fraseri in possessing a restricted forest-fringe type of distribution but, like some birds and Lepidoptera, this species also occurs in the forested parts of Natal, an interesting distribution suggesting that it is a more ancient species than the others, probably a relict from some past pluvial period when the southern forests were more nearly continuous with those of central Africa (see Chapin, 1932 : 376, and Moreau, 1933). For general notes on mosquito distribution in the region see Haddow (19450), De Meillon (19476), Lewis (1947), and Abbott (1948). e Fro-ser'v. . A ScVvwetzl. I West African su.b-vegi.orx. Ha, Su.dLa.wece Sa.vo.vmcL, District. II b Ea.y\dPi.&tri.ct.. FIG. 4. Distribution of the Aedes apicoargenteus group. Of all the West African species only apicoargenteus can be said to be typical of the sub-region in the sense that it occupies the whole or a large part of it and extends for only short distances beyond its boundaries (Fig. 4) . This species is therefore discussed first. Aedes apicoargenteus. In so far as our knowledge goes, the distribution of this species fits Chapin's boundaries very well. The only certain records outside them are from Kitgum and a few localities round the Kavirondo Gulf, all in the East African Highland District, and Lorha in the Sudanese Savanna District. There is also a IN THE ETHIOPIAN REGION 279 probable record from Kaduna in this district. All these localities have mean annual rainfalls of at least 45 in., but if the 45-in. isohyet is taken as the limit of distribution, further exceptions must be made, since Accra, Anecho, and Masindi Port certainly, and Katompe and Kiansonzi probably, have a lower rainfall than this. The explana- tion appears to be that, within limits, the monthly distribution of rain is as important in determining distribution as its total amount. Accra and Anecho lie in the curious dry belt extending approximately from Cotonou to Cape Three Points, where the coast runs parallel to the path of the south-west monsoon instead of cutting sharply across it as it does farther east and west. Although the rainfall in this area is low the apicoo.rgen.teus. o Nega.ti. ou.bt?uk.l Record. FIG. 5. Distribution of Aedes apicoargenteus in West Africa. The shaded area receives less than i in. of rain during each of six or more months in the year. Rainfall contours after Nash (1948). proximity of the sea maintains a high atmospheric humidity and the seasonal dis- tribution of rain is very equable, there being, on an average, only 4 dry months (i.e. months with less than i in. of rain) at Anecho and 3 at Accra. Even so, however, it is unlikely that apicoargenteus occurs permanently within the area since in some years the rainfall is almost certainly too low (e.g. 10-84 m - at Accra in 1926). It is note- worthy that the record from Accra (Macfie & Ingram, 19236) was for December 1918, i.e. the end of a 3-year period of exceptionally heavy rain (41-05 in. in 1916, 44-20 in. in 1917, and 32-37 in. in 1918). Deducing empirical limits from the data available it may be said that apicoargenteus is not known to occur anywhere with less than 25 in. and 3 dry months, 30 in. and 4 dry months, or 45 in. and 5 dry months. On this basis we might expect to find the species everywhere in the western part of its range with 5 dry months or less, and such a distribution is in good accordance with the data available. Reliable negative records are few and, in general, no use is made of these in the present paper. Those shown in Fig. 5 (Gadau and Kano) seem to be the only ones of much value in the present instance. In the eastern and southern parts of its range the distributional limits of apico- argenteus are probably mainly altitudinal. Over almost the whole of Uganda rainfall 280 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) is high and extremely equable in its annual distribution (Masindi Port and the Kiansonzi area have only one dry month), but there is some reason to suppose that the species may be absent from the extreme northern part of the territory. Thus Lumsden in a recent survey failed to find it beyond Payida in the West Nile Province although other localities visited (Godia, Laufori, Metu, Midia, Rumogi) would seem, to judge from nearby rainfall stations, to have plenty of rain (3 dry months in each case). It cannot, however, be said that reliable negative records have been established 36 ^-Forest- * " (*.t attire FIG. 6. Distribution of Aedes apicoargenteus in north-east Uganda and the Sudan. Only those rainfall stations are shown which are in the critical area. for this area, since the survey was a very short one. When the empirical limits deduced for the western part of the range are applied here an exceedingly interesting picture is obtained (Fig. 6). The boundary turns east in the neighbourhood of Juba and again farther north in the neighbourhood of Bor. How far east it runs is unknown since no figures are available from this comparatively little-known part of the Sudan, but it is tempting to suppose that it may join with a similar boundary delimiting the wetter parts of Abyssinia. The similarity of the Abyssinian and West African faunas has often been a matter for comment (see, e.g., Carpenter, 1935), and whether a bridge still exists in this region or not, it seems reasonable to suppose that a relatively slight change of climate would suffice to recreate one. Aedes apicoargenteus is not known from Abyssinia, but so little collecting has been done there that negative records are quite valueless. Aedes africanus is known from the Sidamo Province so that the existence of a fairly recent link between the Abyssinian and Guinean Stegomyia may be safely inferred. The more northerly part of the Bor-Pibor region IN THE ETHIOPIAN REGION 281 is largely swampy and is stated by Lewis (1947) to possess very few trees, but the map (NB 36 of the International Series) shows small isolated patches of forest having the appearance of relicts of a continuous belt running north-east from Mongalla to the Abyssinian border. Collections from these patches of forest as well as from others in the Bor, Mongalla, and Juba areas would be of great interest. From Mt. Elgon southwards the distribution of apicoargenteus is limited well to the atxcoargenteu.s. O SSp- dlsndevens is. | 1 Lotnctover 5000'. rmj tetndl over 3000' Ar*with wall-distributed. ramfo.lt but less than XfO" FIG. 7. Distribution of Aedes apicoargenteus in the southern and eastern parts of its range. west of the 45-in. isohyet by altitudinal factors (Fig. 7). Haddow, Van Someren, et al. (in press), report that it does not occur above about 5,500 f^. on the western slopes of Ruwenzori, and Garnham et al. (1946) record it up to approximately the same altitude in Kavirondo. This is a critical level for many plants and animals at this latitude (see Chapin, 1932). Considerable areas of northern Tanganyika, while having less than 45 in. of rain in the year, have a sufficiently well-distributed rainfall to support the species, at least on the basis calculated for West Africa. No collecting has, however, been done except at and near Mwanza, from which there is a negative record. 1 Once again surveys of isolated patches of forest would be of considerable interest. In the Kivu Highlands apicoargenteus appears to extend up to at most 5,000 ft. Above this it is replaced by schwetzi (very rare as far north as this) or more 1 See Appendix. ENTOM. II. 5 Mm 282 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) commonly by ssp. denderensis. Farther south still, as noted by Chapin for the birds, the altitudinal limit is much lower and it is not known to occur above 2,500 ft. or possibly 3,000 ft. (Kayembe-Mukulu). Jn the Katanga and the Rhodesian Highlands it certainly does not reach 4,000 ft. (Elisabeth ville, Ndola), and it is replaced at these altitudes and, farther south still, at 3,500 ft. (Balovale) by schwetzi. It seems im- 1771 Closed Forest;. Area of insu.PPi'cent rainfall. JJ23 Land, over Sooo*. 3.0 30 30 too FIG. 8. Distribution of Aedes fraseri in the western and eastern parts of its range and putative range as a whole. probable that this species would be found anywhere in the southern Congo above about 3,000 ft. Summarizing the distribution of apicoargenteus it may be said that it shows a much less clear-cut relationship to rainfall and altitude than do a number of other species. The explanation may be that it requires for permanent survival in an area a more purely equatorial type of forest than the dry or mixed forests which suffice for, for example, africanus. This matter is, however, beyond the scope of the present paper and must be left for discussion in a later one. Aedes fraseri. As already noted this species has a peripheral or fringing distribution IN THE ETHIOPIAN REGION 283 relative to the main Guinean forests (Fig. 8). Haddow states (in litt.) that, unlike apicoargentem and africanus, it is never taken more than a short distance inside the Semliki Crown Forest. Garnham et al. (1946) found it in the depths of the Kaimosi Forest, but it would appear that this is a comparatively poor type of relict forest. Chapin (1932: 232) discusses the distribution of birds having a similar habit and states that they follow the northern border of the Congo forests in a narrow band and sometimes extend southwards along the eastern edge of the forests or occur in places along its southern margin. Except at Dubreka fraseri is not known to occur any- where with a rainfall of less than 40 in. or more than 3 dry months. Dubreka has 4 dry months with a rainfall of over 150 in., and here the species is at present known only from mangrove. Applying the limits of 40 in. with 3 dry months or 100 in. with 4 dry months a putative distribution is obtained which is in very good agreement with the known distribution of fraseri and the type of distribution described by Chapin (Fig. 8) . An extensive area with comparable rainfall also exists in the southern Congo, and it is possible that fraseri may be found there although at present there are no records. Similar altitudinal limits appear to apply to this species in Uganda and Kenya to those given for apicoargenteus (Garnham et al., 1946; Haddow, Van Someren, et al., in press). It is to be noted that the dry area in the Gold Coast and Togoland and, farther east, the Adamawa Highlands both constitute serious inter- ruptions to the distribution and it would seem that the Upper and Lower Guinean populations may at times be discontinuous. Aedes dendrophilus. In the northern part of its range this species appears to have a fringing distribution very similar to that of fraseri (Fig. 9), but its putative rainfall limits are even more restricted since it is not known from anywhere with more than 2 dry months in the year. 1 In addition it is also known from an area in Natal with the same type of rainfall and from Fernando Po which suggests that it is a rather ancient species with, at one time, a much wider distribution. This view is also supported by its morphological characters. Its altitudinal limits appear to be much the same as those cited for fraseri (Garnham et al., and Haddow, Van Someren, et al., loc. cit.). Various small, isolated areas in the southern part of the continent appear to possess an adequate rainfall for the species and not all of them are at impossible altitudes. Among the more promising are the Haenertsberg-Magoebaskloof area of the Trans- vaal, the Mbabane region of Swaziland, parts of Mozambique (Vila Paiva d'Andrada and Chinde-Pebane areas), the Inyanga, Morgenster, Bikita, Mt. Silinda, and Umtali regions of Southern Rhodesia, and certain localities in southern Tanganyika at the northern end of Lake Nyasa (Makete, Musekera, Mwitika, Kyela). Aedes africanus. Like apicoargenteus but unlike fraseri and dendrophilus this species is found in the depths of closed canopy forest which does not therefore con- stitute a barrier to its distribution. Among mosquitoes, as noted by Chapin for the birds, this capacity to penetrate thick closed canopy forest is the exception rather than the rule. Unlike apicoargenteus, africanus has not been found in the dry part of the Gold Coast. It is, however, able to surmount the Rhodesian Plateau and in consequence its distribution is very closely and completely defined by the 40-in. isohyet except to the east of Lake Victoria where there is a complete altitudinal 1 See Appendix. 284 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) barrier and south of Lake Nyasa where the rainfall is sufficient but its distribution is unsuitable (Fig. 10) . The only records from outside the distributional area so denned are those from Zomba (Theobald, 1901) and Mozambique (Howard, 1912), of which the former certainly and the latter very probably fall within an isolated area of adequate rainfall which may well support a relict population (Fig. n). It is AeAes Ae wdLr-oV>Kilus. Areas witH lest tKo-n no ra.it* ov more thaw 21 FIG. 9. Distribution of Aedes dendrophilus. unfortunate that these records can no longer be checked. The putative distributional limits of 40 in. with 4 dry months, 45 in. with 5 dry months, or 50 in. with 6 dry months, deduced empirically from the known distribution, fit this distribution so well that it can scarcely be doubted that they are significant. As to whether their effect on the species is direct or indirect, through the effect on vegetation, is not at present known. 1 The existence of a relict population of this species in Abyssinia has already been referred to. Altitudinal limits appear to be similar to those of apicoargenteus , but with a decidedly greater tolerance of high altitude especially in the southern part of the range. Known limits include 5,500 ft. on Ruwenzori (Haddow & Van Someren, 1 See Appendix. IN THE ETHIOPIAN REGION 285 1950), about 6,000 ft. in the Kavirondo region (Garnham et at., 1946), about 6,000 ft. in the Kivu Highlands (Costermansville) and in Ruanda (Dendezi), about 5,500 ft. on the Kibara Massif (R. Vakila) and on the Bihe Plateau (Bihe), and about 4,000 ft. on the Rhodesian Plateau (Elisabethville, Ndola). The record from the Sidamo Province of Abyssinia (Bevan, 1937) is associated with an altitude of 5,900 ft. V / /| Area, with VesstKa- t*O*-a.iv. Land o.v sooo'm E.AFrica.nHigMa\(U Zfl' FIG. 10. Distribution of Aedes africanus. Aedes simpsoni. This is the only species, in addition to those already discussed, which occurs in the area of continuous closed canopy forest. Its presence here is almost certainly the result of introduction by man, since Haddow (1945^, 1950, &c.) has clearly shown that it is a mosquito of the forest edges and especially of plantations and shambas, and its occurrence in the forest proper is confined to stray individuals and is so rare as to be without significance. Haddow has pointed out, in conversation, that very much traffic takes place in young pineapple and banana plants which frequently contain larvae. As a plant-axil breeder it flourishes in areas of very much lower rainfall than can be tolerated by the species so far discussed. The minimum average rainfall tolerated is, as far as present records go, 20 in. with 8 dry months (Sennar), and the 2o-in. isohyet fits the known distribution very well (Fig. 12). These rainfall requirements appear to exclude it entirely from Chapin's Sudanese and Somali Arid Districts and from his South- West Arid District (except in the Bulawayo area). Otherwise it occurs in all the main faunal districts. Its altitudinal limits appear to be much the same as those of africanus and it is known from up to about 286 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) 5,500 ft. on Ruwenzori, 5,000 ft. in the Cameroons and in Eritrea, 6,000 ft. in the Kivu Highlands, 5,500 ft. on the Bihe Plateau, 4,000 ft. on the Rhodesian Plateau, and 4,500-5,000 ft. in the Transvaal (Roberts Heights) and Southern Rhodesia (Salisbury, Bulawayo). It is rare on the Rhodesian Plateau, but Robinson (1950) considers this to be due to the absence of banana plantations. Garnham et al. (1946) appear to have found it up to about 6,000 ft. in Kavirondo but, unfortunately, o Rainfo.lt Stations* |\\N Aria. wiiUu.nsuiio.ble rainfall Rcstri.ct. FIG. 12. Distribution of Aedes simpsoni. lowest rainfall recorded anywhere in the northern part of its range being 15 in. with 8 dry months (Danagla and Wad el Magdub in the Wad Medani area) . In the southern part of its range it is also known from Maun, which has only 15 in. The record from Nefasit shows that in this part of its range it can occur up to about 5,000 ft., but elsewhere it is rare above about 3,500 ft. In western Kenya it has been recorded from two localities at about 4,000 ft. (Kerio, Fort Hall), but on the Uganda Plateau it is very rare and the only records which can be confirmed are from about 3,000 ft. in the Semliki Valley and from between 3,000 ft. and 4,000 ft. in the West Nile 288 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) Province. The record from Mpumu cannot now be confirmed and may have referred to africanus. As against this it seems possible that luteocephalus occurring in this region may occasionally have been misidentified as africanus. On the Rhodesian Plateau it has been recorded from Balovale (3,400 ft.) and Ndola (4,100 ft.). At Ndola it varies in abundance from year to year and in one year only a single specimen was af- tuteocet>Halus. Land above 35OO ( . Arfta. oFcontinu.o*S closed. forttt. (VrtaUtK\e$s tKan 15" 10* 10* FIG. 13. Distribution of Aedes luteocephalus. taken (Robinson in litt.). The eastern limit of its distribution cannot be inferred since no collecting has been done in western Tanganyika. There is a single specimen from Likoma Island off the eastern shore of Lake Nyasa in the London School of Hygiene collection. Nor is it known whether it occurs in Mozambique, but it may be taken as virtually certain that it does not extend nearly as far south as simpsoni. Aedes unilineatus. This species is of interest as being one of the few Ethiopian Stegomyia which also occurs in the Oriental Region. Like luteocephalus it is purely a savannah mosquito, but its resistance to drought appears to be even higher and it is on record from several places with rainfalls of about 10 in. (Dolo, El Fasher, Erkowit) . Unlike luteocephalus it does not occur in deforested areas with high rainfall such as Freetown and it is not, in fact, known from anywhere in Africa with more IN THE ETHIOPIAN REGION 289 than 55 in. per annum (Tembura). It is not known whether it occurs in those parts of Arabia with adequate rainfall and there are no acceptable negative records even from the Aden region since it is not clear that anybody has searched in tree-holes. It is stated by persons with local knowledge that the latter are by no means rare in the wetter parts of Arabia, especially in fig-trees growing in the wadis. It is not known from anywhere above 4,000 ft. except Erkowit, which is in the hottest part of its Aedes L Ara viitH less tkan lO"r Area. With. more. tK an 60 Land, over 3Soo'. FIG. 14. Distribution of Aedes unilineatus. Inset: Putative distribution of unilineatus with that of two related species. range. Elsewhere the highest altitude recorded is at Fort Victoria (3,700 ft.), and the 3,5oo-ft. contour probably provides a good approximate boundary. It does not appear to enter forest and this probably serves to exclude it even from those parts of the Congo Basin with less than the empirical limit of 55 in. of rain per annum. The putative distribution is shown inset in Fig. 14. As to whether the southern populations of the Zambesi and Limpopo valleys are linked with the northern population of the Sudanese and Guinean savannahs is not known, but, in view of the fact that it occurs as far east as Dolo, it seems possible that it may extend southward along the edges of the central plateau. The Indian records seem to indicate similar climatic limits with the exception of that from the Bombay area. The rainfall here is about 70 in. per annum. It appears that unilineatus is a relict of a fauna which was at one time widespread over much of Asia Minor. Related species still occurring in this ENTOM. ii. 5 N n ago THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) area include granti, which is known only from Sokotra, and cretinus, which has been recorded from Crete (Edwards, ig2ib), Macedonia (as delta, Seguy, 1924), and Transcaucasia (as albopictus, Rhoudkhadze, 1926; as lindtropi, Schingarew, 1927, and see Stackelberg, 1937). Aedes metallicus. This species has been recorded from only four localities in the 20 30 acfl or- tor. uf. Aedes metallicus. West African -10 2JOP 10 10 20 30 50 FIG. 15. Distribution of Aedes metallicus. West African sub-region. Two of these (Mbarara and Yambio) are peripheral. The other two (Accra, Asuboi) are in the dry area of the Gold Coast and Togoland (Fig. 15). Its resistance to drought appears to be greater even than that of unilineatus and it is the only species so far discussed which occurs widely in Chapin's South- West Arid District. It is not, however, known from Arabia. Despite its almost complete absence from the West African sub-region it is known from several localities in East Africa with quite high rainfall, e.g. Kakamega with 70 in. and Tanga with 60 in. It seems probable that temperature plays a greater part in controlling the distribution of this species than in the case of those previously discussed. Its altitudinal limits IN THE ETHIOPIAN REGION 291 vary in different parts of the range. In Kenya it has been recorded from an indefinite altitude in the Fort Ternan area (Garnham et al., 1946), apparently about 6,000 ft., and from 5,500 ft. in the Nairobi area. There are no records from the Kivu High- lands and only one from the Rhodesian Plateau (Jadotville, 4,300 ft.). On the Bihe Plateau it apparently occurs at about 5,500 ft. and in South- West Africa it occurs up to about 4,500 ft. (Okahandja). It appears to be absent from the higher parts of the Transvaal where it has not been recorded above about 2,000 ft. (Letaba), but in Southern Rhodesia it occurs up to 4,800 ft. (Salisbury). Morphologically it is unique among the Ethiopian Stegomyia in lacking a ventral process from the paraprocts, a feature which recalls the oriental members of the sub-genus. Its mesonotal ornamentation is also unique and suggests a comparison with the Aedes longipalpis group vis-a-vis the other Ethiopian F inlay a. Aedes vittatus. This is the most widely distributed of the species here discussed (with the exception of aegypti). It combines drought resistance, due at least in part to its ability to breed in rock-pools, with a marked ability to withstand low tempera- tures. It is, however, by no means equally common everywhere. In the Lagos area it is extremely rare, having only been recorded once. In Bwamba only a single speci- men has been taken (Haddow, Van Someren, et al., in press). It appears to be absent from or extremely rare in a large area of the Belgian Congo, a fact which may perhaps be related to the nature of the surface rocks (Lubilash sandstones, see Chapin, 1932 : 30) . It is not clear to what extent it enters forest, but Garnham et al. (1946) specifically mention its absence from shaded forest rock-holes and there are many indirect references in the literature to suggest that it is largely or entirely a mosquito of the open country. That this does not preclude it, as it apparently precludes luteocephalus, from the forested eastern part of the Congo is readily understood from the fact that its breeding-places of preference are commonly found in the exposed beds or edges of streams. The occurrence of this species in Spain is of particular interest since it implies the ability to contend with two sets of adverse conditions, the long, cold winters and the torrential nature of the breeding-places during spring and summer. Gil Collado (1935) notes that it is most abundant in the autumn. It is not known in what stage the winter is passed in this part of the range, although it seems almost certain that it must be in the egg. On the basis of the Ethiopian records alone, this would seem to be more resistant to low temperature than any of the other species discussed, except perhaps aegypti, if only because it occurs in a number of places in the Kenya Highlands well above 6,000 ft. (Eldoret, Kericho, Kiambu, Kitale, Mara R., Maseno, Mem, Narok). There is also an interesting record from Sipi (6,700 ft., Hancock & Soundy, 1929), on the Uganda side of Mt. Elgon. As against this there are some rather striking negative records. Patton (1905) notes that it has not been found on the Jebel Jihaf above Ulub. Mara (1945) failed to find it on Mt. Bizen. (This may have been a coincidence, but it seems to be a very common mosquito at lower levels.) It appears to be very rare on the Rhodesian Plateau, where the only records are from Elisabethville and Livingstone, and there is no record from the Kivu Highlands. The likeliest explanation seems to be that the egg is highly resistant to low temperatures as well as to desiccation but the adults and larvae are not. In this case survival would depend on the coincidence between suitable 292 THE SUB-GENUS STEGOMYIA (DIPTERA, CULICIDAE) temperature conditions and a low water level in the streams. Records from over 4,000 ft. in the southern part of the range include Bihe, Salisbury, Goromonzi, Gwelo, Msonneddi, Onderstepoort, Yokeskei River, Waterval Boven, and Pietersburg. It is interesting to note that it has not been found as high up as metallicus in South- West Africa or as simpsoni in Eritrea. 90 3, 23^1.1929, Glenmore Castle, I ?, i6.vi.i934, Powerscourt Demesne, 2 <$$, 25^.1932, on oak, Powerscourt Deerpark, 2 (?, 27^.1932, on oak, A. W. Stelfox: ENGLAND, unlocalized, i ?, 2 $$; Essex, Colchester, i , s.vi.igog, Harwood ; Staffs., Lichfield district, 9 ??, 6 $$, L. A. Carr ; Yorks., Grassington, i $, 19^.1945, Beedale, i ?, 9^1.1945, W. D. Hincks: SWEDEN, Skane, Hoor district, 2 ??, 2-17.^1.1938, D. M. S. & J. F. Perkins: CZECHOSLOVAKIA, Aussig on Elbe, i , 4.ix.i937, Th. Kupka; unlocalized (probably Germany), i , O. Schmiedeknecht : CZECHOSLOVAKIA, Moravia (unlocalized), i ?; Moravia, Suchdol, i (J, 27.viii. 1930, Fr. Gregor; Bohumin, Haatsch, i <$, 28.viii. 1936, Th. Kupka: SWEDEN, Skane, Hoor district, i <, n.vi, Ringsjo, i $, 19^.1938, D. M. S. & J. F. Perkins: FINLAND, Tavastia australis, Saaksmaki, i , mid-vi.i92i, W. Hellen (Hellen, 1941, under A . parvulus Thorns.) ; Hattula, i $, L. v. Essen. Additional material seen : SWEDEN, Stockholm, type $ ; Skane, i <, C. Boheman ; Skane, Farhult, i $ in Thomson coll. (Thomson, 1888: 1255) redetermined by G. J. K.: FRANCE, Mesnil, nr. Paris, i $, ig.vii.iSgi, J. de Gaulle: GERMANY, Hann- over, i <$, 2i.viii. 1946, R. Hinz: AUSTRIA, Carinthia, i specimen, C. Boheman. I examined the Holmgren type in 1939, and formed the opinion that the species should be transferred to Anisoctenion. Six days later, the Moravian pair was received in Manchester from Professor Gregor : they agreed with the diagnosis made and with my memory of the Holmgren type. The synonymy was confirmed by direct com- parison in 1947. Professor Gregor wrote to me that he had taken the species commonly in warm localities. He found there were great individual differences between different speci- mens and that, in smaller specimens, there was very little difference in length between the mandibular teeth. He regarded the species as a connecting link between Acro- tomus and Cteniscus, but had not published a valid description lest one should already exist in an unsuspected place as, in fact, has proved to be the case. In Schmiedeknecht 's key to Acrotomus (5: 2284-2286) this species runs down with alacer Grav. ; indeed, the Schmiedeknecht specimen in the British Museum collection was so determined by that author. Pumilio differs from alacer principally in having 334 A REVIEW AND A REVISION OF the head short, strongly narrowed behind the eyes ; malar space much shorter than distance between mandibular apices ; mesopleura finely punctate ; sternaulices and mesolcus strongly impressed; gaster spindle-shaped in female; ovipositor sheaths strongly projecting (Fig. 20) ; tarsal claws scarcely pectinate ; in having none of the features mentioned (p. 350) as special for alacer ; and, of course, in the coloration and much smaller size and number of antennal segments. It has a much greater 19 FIGS. 18-19. Head in dorsal view of Fig. 18. Anisoctenion pumilio Hlgr. Fig. 19. Anisoctenion alacer Grav. FIGS. 20-21. Apex of gaster, in sinistro-lateral view, of Fig. 20. Anisoctenion pumilio Hlgr. Fig. 21. Anisoctenion laticeps Grav. resemblance to a small specimen of rubiginosus Grav. which, however, has the head slightly broadened behind the eyes and the mesopleura rather coarsely punctate ; and from which it can also be separated by the key characters. Anisoctenion tenuis sp. n. Head not so transverse as in pumilio Hlgr., but strongly narrowed and rounded behind eyes ; moderately emarginate behind : on vertical region and temples finely and rather sparsely punctate; on genae finely or very finely and very sparsely punctate above, rather finely and much more closely so below ; on upper and lower face moderately and closely punctate, more finely and sparsely so towards orbits. Clypeus as described for pumilio Hlgr., but less sparsely punctate and not sharply or at all deeply delimited. Cheeks and malar space as described for pumilio Hlgr. : hypostomal and genal costae also similar, but hypostomal weakly raised. Antennae with scape stout, rather weakly but sharply excised externally (Fig. 29) ; THE CTENISCINI OF THE OLD WORLD 335 with pedicellus of normal proportions; with flagellum filiform and slender, 27- to 3O-segmented. Pronotum with dorsal transverse impression broad; with epomiae of moderate strength ; laterally sparsely and very finely punctate above, finely reticulate to rugose below. Propleura shallowly punctate ; finely and rather closely so by outer margin, elsewhere more coarsely and sparsely so. Mesoscutum not rising sharply from pro- notum, but rather strongly and evenly convex, though flattened discally behind middle ; with notaulices moderately and rather sharply impressed anteriorly, quite distinct to flattened part ; finely and more or less sparsely punctate. Scutellum weakly convex and not falling sharply behind ; sharply margined to beyond middle ; finely and sparsely punctate discally, very finely and more closely so at sides and especially behind. Mesopleura rather finely to moderately, rather shallowly and sparsely punctate below, obscurely punctate anteriorly above : sternaulices broad and rather deeply impressed : mesosternum finely and not closely punctate : mesolcus of moderate width and moderately impressed, closed behind: prepectus finely rugose: epicnemial carina sharply though not strongly raised. Propodeum long and not at all sharply declived ; with costae of moderate strength in front, stronger behind ; for the most part very finely to finely punctate and elongately griseo-pilose : area postero- media strongly transverse, extending barely a third the way forward ; area supero- media narrow, very much longer than its greatest breadth, basally delimited but not sharply so, expanded to where it emits costulae before middle. Petiolar segment very elongate and slender, 3-5 to 3-8 times length of its greatest breadth; laterally distinctly but not strongly margined, with spiracular tubercles prominent ; with dorsal keels not strong, not reaching base and evanescent well before apex ; finely and sparsely punctate beside keels behind spiracles ; with sparse, elon- gate pilosity. Tergite 2 with thyridia distinctly broader than long ; finely punctate in basal half : gaster more and more finely and sparsely punctate towards apex. Fore wings with basal and disco-cubital veins nearly parallel; with areolet sub- sessile or sessile ; with nervus parallelus emitted above middle of brachial cell. Hind wings with nervellus almost opposite, intercepted about in middle. Tarsal claws very weakly pectinate. Length 6 to 7 mm. Head black, with mouth-parts except mandibular apices, and clypeus except marginally, pale flavous: clypeus deeper flavous in Burmese type. Antennae mostly flavous to base of post-annellus ; beyond that dull fulvous to pale ferrugineous, distinctly darker towards apex. Thorax and propodeum rufo-ferrugineous to black ; with ventral margin of pronotum, propleura, fore and hind margins of mesopleura, and hind margins of scutellum and metascutellum distinctly paler; with scutellum above and propodeum behind dull rufous in Burmese type. Legs mainly fulvous ; with fore and mid coxae, trochanters and trochanterelli whitish, the coxae fulvous around base ; with hind legs slightly darkened in part in Formosan specimen. Wings with veins and stigma ferrugineous; with stigma slightly and veins conspicuously paler at base. Gaster with tergite i rather ferrugineous except at mid apex, and with thyridia pale : remainder rufous in Burmese type but in Formosan specimen mainly infuscate, though with tergites dark rufous apically. 336 A REVIEW AND A REVISION OF Described from the following material: NE. BURMA, Kambaiti, 7,000 ft., i <$, 24^.1934, R. Malaise (holotype): FORMOSA, Hassenzan, i <$, 2^.vi. 1934, L. Gressitt (paratype). Holotype in Royal Natural History Museum, Stockholm: paratype in collection of Dr. H. K. Townes. This species is clearly most closely related to the European pumilio Hlgr. The advance in knowledge of the taxonomy and distribution of the genus gained by describing the species is held to outweigh the disadvantage of having to do this from the male sex alone. W. R. M. Mason has sent me a single female specimen of a species related to pumilio Hlgr. and tenuis sp. n., from a locality at a height of 3,500 ft. in the Nilgiri Hills in southern India. This is a further wide extension of the known distribution of the genus. Anisoctenion ridibundus Gravenhorst 1829 Tryphon ridibundus Gravenhorst, Ichneumonologia Europaea, 2: 188-189. 1906 Acrotomus ridibundus Pfankuch, Z. Hym. Dipt. 6: 219. 1911 Acrotomus sexcinctus Morley, The Ichneumons of Great Britain, 4: 197 (non Gravenhorst, 1829). Head moderately narrowed behind eyes in female, more strongly so in male, and strongly rounded; rather shallowly emarginate behind: for the most part finely punctate, rather closely so above, sparsely so on most of gena ; with lower and upper face moderately punctate, and generally with lower margin of gena densely and a little more coarsely so. Clypeus sparsely beset with punctures of mixed sizes, very fine to rather coarse; elongately pilose near apical margin; sharply delimited basally; strongly projecting in side view, but not strongly convex; not broadly rounded apically in female, though often rather so in male ; and not or scarcely impressed before margin. Cheeks not at all buccate, but in frontal view slightly concave. Malar space of female about three-quarters distance between mandibular apices ; that of male short, about equal to greatest width of fore metatarsus, seen from above near apex. Hypostomal costa rather weakly raised: genal costa scarcely raised, inflected. Antennae relatively slender; with scape strongly convex interno-dorsally and weakly excised externo-ventrally ; with pedicellus almost barrel-shaped, but more elongate than in laticeps Grav. (Fig. 30) ; with flagellum 26- to 29-, most usually 28-segmented ; having the ultimate segment usually elongate, but sometimes imper- fectly and sometimes fully divided (thus sometimes having a segment more in one antenna than in the other) ; and having some middle segments incurved externally as in laticeps Grav. Pronotum with dorsal transverse furrow shining; with epomiae sharp but not strong ; behind epomiae finely punctate above, rather strongly longitudinally rugose below. Propleura irregularly, rather finely and sparsely punctate. Mesoscutum strongly convex, rising rather sharply from pronotum; with notaulices weakly or moderately, seldom at all sharply impressed ; finely punctate discally, less finely so peripherally. Scutellum strongly convex, and falling rather sharply behind; not margined to middle ; finely to very finely punctate. Mesopleura moderately and not THE CTENISCINI OF THE OLD WORLD 337 closely punctate below, finely, sparsely and irregularly so above : sternaulices broad, rather shallowly impressed: mesosternum finely to moderately and closely punctate, prepectus rather coarsely so : mesolcus sharply impressed, widening a little posteriorly, almost always well closed behind : epicnemial carina very slightly elevated, extending about half way up mesopleuron. Propodeum finely punctate above and on area pleuralis, elsewhere weakly rugose ; with costae of moderate strength at sides and behind, usually weak and often incomplete above : area basalis excavate : area supero- media longer than broad. Petiolar segment 1-6 to 2-0 times length of its greatest breadth; finely punctate above, but almost impunctate by hind margin ; sharply but not strongly margined laterally; with dorsal keels not strong, but distinct to well beyond spiracles; not canaliculate between dorsal keels where these are distinct, but with a median depres- sion behind. Remainder of gaster finely to very finely punctate. Tergite 2 strongly transverse ; with a pair of broad, not sharply defined, furrows extending diagonally outward from base ; with thyridia short, sharply impressed and conspicuous, about one-half breadth of intervening space. Female hypopygium sharply folded mid- ventrally. Female third valvulae rather broad, closely punctate ; with ventral margin slightly sinuate but, on the whole, slightly upcurved ; projecting well beyond apex of tergite 8. Fore wings with basal and disco-cubital veins parallel or nearly so ; with radius slightly upcurved at apex and slightly sinuate ; with areolet sub-sessile ; with second recurrent almost regularly arcuate; with nervus parallelus emitted a little below middle of brachial cell. Hind wings with nervellus slightly antefurcal, intercepted below middle. Legs rather stout. Hind tibiae distinctly swollen before and very distinctly narrowed to base. Claws of all tarsi strongly or rather strongly pectinate. Length 6| to 9 mm. Female. Head black, with the following parts flavous, with some fulvous margins: palpi, mandibles except at apex, clypeus, genae except broadly near occiput, and whole of lower face, except for a short median mark just below antennae. Antennae with scape and pedicellus dull flavous to rufous below, rufous to ferrugineous above ; with annellus more or less rufous ; with remainder of flagellum fulvous to testaceous, more or less ferrugineous near base. Thorax and propodeum black, with the following parts flavous, with some fulvous margins: scutellum at apex and metascutellum, numeral angles of pronotum and ventral margin rather broadly at sides, propleura mainly, marks below fore wings, tegulae, and metategulae in part ; usually with the following parts more or less ferrugineous : fore and hind margins of mesopleura in part, lateral extensions of scutellum and metascutellum, and some margins of propleura ; sometimes with paler marks beside mesolcus. Legs mainly fulvous, with hind tibiae decidedly paler at base ; with fore and mid coxae, and all trochanters and trochanter- elli, mainly stramineous to pale flavous, the mid coxae often flushed with fulvous ; with hind coxae except at apex, and hind femora except narrowly at base and apex, dark ferrugineous. Wings with veins and stigma ferrugineous, paler at base. Petiolar segment black, sometimes dark ferrugineous towards apex. Remainder of gaster mainly dark ferrugineous above ; with apex and sides and with thyridia fulvous ; with ENTOM. II. 6 T t 338 A REVIEW AND A REVISION OF slightly broad apical margins of tergites 2 and 3, usually with narrow margins of remaining tergites, and with most of venter paler. The male differs as follows : genae with flavous coloration much more restricted, often only narrowly flavous by the mandibular articulation; clypeus with a pair of dark marks below tentorial pits ; lower face with a broad, percurrent, median black band and with eye margins black for some distance below antennal sockets, the black margin sometimes curved inward towards tentorial pits. Propleura not pale-coloured except by hind margins. Hind tarsi and tibiae dark above, the latter conspicuously pale at base : fore and mid femora and tibiae often paler than in female. Gaster of darker appearance, scarcely fulvous at apex and only narrowly so at sides. Redescribed from the following material : ENGLAND, Desvignes coll., i $ (previously twice misdetermined) ; Surrey, Abinger, i $, i8.viii.i947, G. J. K. : BELGIUM, unlocal- ized, Groenendael, Rouge Cloitre, 9 $$, 7 $$, Wesmael and Tosquinet coll. : SWITZER- LAND, Grisons, Tarasp. i $, Ch. Ferriere : GERMANY, unlocalized (probably Germany) i cJ, O. Schmiedeknecht ; nr. Berlin, i $, Ruthe coll. ; Munich and other localities, i (J, 20. vi, i $, 2 $$, i6.vii. Buchecker coll. ; Bavaria, Oberstdorf, Birgsau, 3 ??, viii.1936, G. E. J. Nixon; Silesia, Ransern, i , 26.viii. 1931, i ?, 26.vii. 1933, G. J. K. ; Norfolk, Catfield, 2 ??, i2-i6.vii.i920, J. J. F. X. King: BELGIUM, unlocalized, 6 $3, 4 ??, Diest, 2 , 8.vii.i937, i $, I9.vii.i938, K. Hedwig; Ruthe coll., i , Buchecker coll., i #, I3.vii; unlocalized (probably Germany), 2 <$<$, O. Schmiedeknecht : CZECHOSLOVAKIA, Bohu- min, Haatsch, i , 24.vii.i938, Th. Kupka: SWEDEN, Skane, Palsjo, i (lectotype), Ortofta, i ^ (allotype), C. G. Thomson. Additional material seen: FRANCE, Arras, Chaville (near Paris), 6 specimens, 23.vi-ix, J. de Gaulle: GERMANY, Schladen, i , R. C. L. Perkins and J. F. Perkins; Lanes., Man- chester, Mersey banks, i , L. Nathan, 6 <$, 2 ??, G. J. K.: BELGIUM, Sysseele, Groenendael, Stockal, Cortenaeken, Charleroy, 12 <(, 2 ??, Wesmael and Tosquinet coll.: GERMANY, ? Bavaria, 3 <<, 2 $, Buchecker coll.; Silesia, Muhlbach, i , ii.viii.ig5o, G. Heinrich: FINLAND, Kangasala, i unlocalized, 7 33, 3 $? ', Riesengebirge, i $ (undated), Morris, Cameron, & Jepson: GERMANY, unlocalized (probably Germany), 2 33, O. Schmiedeknecht (mus. Geneva); Heilbrunn, i , 8.vi.i888, J. Kriechbaumer ; Bay- rische Wald, Waldmunchen, i 3, 9^.1948, i 3> i-8.ix.ig48; Holstein, Trittau, Hahnheide, i $, 29.^.1946, G. Heinrich: SWEDEN, Lapland, Tornetrask, R. Malaise (mus. Paris): FINLAND, Kuusamo, Liikasenvaara, 3 33, i $, 9.vii.i935; Mutkatun- turi, i $, 3i.vii.i935, G. J. K. All the above, except the type, have been designated paratypes. Additional material seen: AUSTRIA, Styria, Admont, Kaiserau, 2 33, 2 $$, i-8.vii. 1950, G. Heinrich. Recorded (under claripennis Thorns.) as established in Canada (Mason, 1951 April: 229). Biology. The Czech material was reared by Morris, Cameron, & Jepson from Gilpinia polytoma Hart, with the exception of the specimen from the Riesengebirge which was from Neodiprion sertifer Geoffr. : these authors give considerable detail of their work (1937). The Finnish specimens were beaten from Picea on cool days. KEY TO EUROPEAN SPECIES OF EXENTERUS HARTIG 1. Petiolar segment trisinuate (Fig. 40): hind tibia bearing a very minute spur: genal costa double, conformed much as in Exyston genalis Thorns. (Fig. 51) : fore wings conspicuously clouded along fore margin . . apiarius Grav. Petiolar segment not thus trisinuate (e.g. Fig. 39) : hind tibia without a minute spur : genal costa not double : wings hyaline, or moderately and evenly inf u- mate ............ 2 2. Clypeus broadly rounded at apex and neither transversely impressed nor re- flexed : antennae in female barely half as long as body, in male much shorter than body: gaster broad, with tergite 2 much broader at base than long (in female nearly twice as broad as long) : claws in female usually pectinate . . (abruptorius group) 3 Clypeus more or less distinctly transversely impressed before the strongly rounded and reflexed apex : antennae in female much more than half as long as body, in male nearly as long as body : gaster generally narrow, with tergite 2 nearly or quite as long as basally broad : claws in female not or weakly pectinate (amictorius group) 5 3. Margin of cavities of fore coxae not raised, except weakly at sides: in side view, greatest length of head behind eye a little less than eye length : malar space almost equal to distance between mandibular apices: gaster in female more than half yellow: mesosternum conspicuously yellow-marked: yellow of first tergite extending far towards base laterally . . . oriolus Hart. Margin of cavities of fore coxae raised: in side view, greatest length of head THE CTENISCINI OF THE OLD WORLD 365 behind eye a little more than eye length : malar space only about half distance between mandibular apices: gaster in female more than half, and mesosternum black : yellow of first tergite not or little wider laterally than medially . . 4 34 35 amict. cU r >p. 36 37 38 Willie. conf. FIGS. 34-38. Antennal scape, &c., in sinistro-lateral view, of Fig. 34. Exenterus amictorius Panz. Fig. 35. Exenterus claripennis Thorns. Fig. 36. Exenterus adspersus Hart. Fig. 37. Exenterus vellicatus Cushm. Fig. 38. Exenterus confusus sp. n. FIGS. 39-40. Petiolar segment of Fig. 39. Exenterus amictorius Panz. $. Fig. 40. Exenterus apiarius Grav. $. 4. Margin of cavities of fore coxae not bidentate : claws in female pectinate : larger species, 8-n mm. ...... abruptorius Thunb. Margin of cavities of fore coxae strongly raised and strongly bidentate: claws simple: smaller species, 7 mm. ..... simplex Thorns. 5. Gaster broad, as in abruptorius group, mainly chestnut-brown in colour, and with segments 1-2 strongly punctate but not rugose : costulae of propodeum absent, or weak and incomplete ....... ictericus Grav. Gaster relatively narrow, as described before, black or black and yellow, and with 366 A REVIEW AND A REVISION OF segments 1-2 strongly rugose-punctate (except in smallest specimens of some species) : costulae of propodeum distinct ...... 6 6. Margin of cavities of fore coxae raised. ...... 7 Margin of cavities of fore coxae not raised, or only with traces of keels laterally 8 7. Margin of cavities of fore coxae bidentate, emarginate between the teeth and appearing discontinuous in middle: scape very weakly excised externo-ven- trally (Fig. 34) ... amictorius Panz. (= marginatorius Fabr.) Margin of cavities of fore coxae not bidentate, appearing continuous and almost straight : scape rather weakly excised externo-ventrally (Fig. 35) claripennis Thorns. 8. Males: see discussion immediately following. Females ............ 9 9. Gaster laterally beyond tergite 3, genitalia and hypopygium rufo-castaneous : hypopygium very acute (Cushman, 1940, fig. B), and without a distinct, broad stramineous or white hind margin: head moderately to strongly narrowed behind eyes, not as in female adspersus Hart. . . . tricolor Roman Gaster not laterally rufo-castaneous : hypopygium not as above . . 10 10. Temple strongly convex, nearly as long as short diameter of eye: hypopygium sharply folded and moderately acute (Cushman, 1940, Fig. A) : third valvulae very distinctly projecting beyond apex of tergites 7 and 8 (they project slightly beyond in tricolor Roman but not in other described European species) . adspersus Hart. Not as above ........... n 11. Hypopygium pinched ventrally near apex, its lower margin in profile very strongly curved, its apex thin, wavy, broadly white (Cushman, 1940, fig. D) : mesosternum and especially propodeum largely yellow-marked . . vellicatus Cush. Hypopygium not at all as above, not dissimilar from that of amictorius (mar- ginatorius) (Cushman, 1940, fig. C), though ovipositor sheaths broader than as there represented: mesosternum and propodeum not largely yellow-marked confusus sp. n. SEPARATION OF MALES OF FOUR SPECIES Whereas the species adspersus Hart., tricolor Roman, vellicatus Cush., and confusu nov. are clearly distinguishable in the female sex upon the form of the hypopygium and third valvulae, and certain other features, such good external characters are not available for the separation of the males. Thus I find that in male adspersus the head may be shorter and distinctly narrowed behind the eyes; and some males that I believe to be vellicatus Cush. have the gaster laterally rufo-castaneous as in tricolor Roman. This condition should be distinguished from that obtaining in some small specimens of other species, in which the whole gaster, except the yellow-coloured parts, is dull castaneous rather than black. Taking specimens and the description of confusus nov. as a starting-point, the THE CTENISCINI OF THE OLD WORLD 367 other species have been compared with these in both sexes. It is hoped that the resulting observations will enable the majority of specimens to be placed with confid- ence : the greatest difficulty appears to be the separation of males of tricolor Roman and vellicatus Cush. E. adspersus Hart, differs from confusus as follows: lower face reticulate to some extent below antennae, but in greater part discretely punctate: clypeus finely to rather coarsely and rather sparsely punctate, on the average less strongly punctate than in confusus: scape somewhat strongly convex dorsally (Fig. 36): scutellum rather weakly to moderately convex, generally much more evenly so than in confusus, i.e. seldom falling at all abruptly behind: posterior face of propodeum more abruptly declived: propodeum more strongly, often quite strongly reticulate-rugose: yellow coloration of gena reaching hypostomal costa, and reaching or approaching genal costa : male mesoscutum more often, like that of female, with a pair of discal yellow marks between tegulae : propleura sometimes pale-marked : mid coxae and femora of male extensively pale-marked, as much so as on fore leg of confusus: hind coxae sometimes pale-marked beneath. E. vellicatus Cush. is best compared with adspersus, with the colour differences of which from confusus it very much agrees, though the extent of genal coloration is often intermediate, as is the form of the scape (Fig. 37). Epistoma alutaceo-punctate, very little raised: notaulices a little more distinct than in adspersus or tricolor: sternaulices more distinctly impressed : mesosternum decidedly flattened, with mesol- cus more sharply impressed, scarcely broadening behind: area pleuralis of propodeum weakly rugose behind and below (quite strongly so in adspersus). The head is often distinctly raised behind the ocelli, but unfortunately this is not invariably so. In female a yellow mark runs along the sternaulices, and in male there is usually a trace of this. E. tricolor Roman differs from vellicatus as follows: scape weakly convex dorsally, as in confusus : epistoma not thus flattened, usually smooth between punctures : genal colour as in adspersus or intermediate, as in vellicatus: mesosternum and pleura as described for adspersus, but speculum extends farther forward and pleuron is more finely punctate in front of it. Genus EXYSTON Schi0dte 1839 Exyston Schi0dte, Guerin's Mag. Zool. (2) 1: 12 (separate pagination). 1877 Tricamptus (Forster) Woldstedt, Bull. Acad. Sci. St. Petersbourg, 23: 454. Our knowledge of this genus in Europe has been in a most unsatisfactory condition. Unfortunately I have not examined the Gravenhorst types. The following specific names require consideration: albicinctus Grav., albocinctus Grav., tricolor Grav., sub- nitidus Grav., cinctulus Grav., sponsorius (Fabr. non Jurine) Grav., carinatus Thorns., genalis Thorns., calcaratus Thorns., phaeorrhaeus Hal., pratorum Woldst. = brevipetio- latus Thorns. = triangulatorius Hlgr. non Grav. (Anisoctenion) . Of the above I am, in the present work, making calcaratus Thorns, the type of a new genus, Parexyston. One of the two commonest European species has probably been sufficiently charac- 3 68 A REVIEW AND A REVISION OF terized for its recognition, and has been best known as brevipetiolatus Thorns. Roman (1939) thought it to be Cteniscus phaeorrhaeus Haliday but, as will be shown below, that conclusion is now disproved. Hellen (1937) placed it in synonymy with pratorum Woldst. ; and there is not, to my knowledge, any valid prior name. The second of the two most common species, which is very variable in colour, has the genal costa continuous or almost so; a fact first recognized by Thomson (1883), who took it to be the typical species and placed it under Gravenhorst's name cinctulus. The type of cinctulus Grav. 1820 is not to be found in the Turin museum (see Kerrich, 1950), so a lectotype should be selected from the Gravenhorst collection at Wroclaw; but I accept the descriptions of 1820 and of 1829 as referring to this species, and I find no clear indication that any of the colour varieties described should refer to genalis Thorns, (see below). An intensive study of cinctulus Grav. confirms what had already been suspected, namely that the form described as carinatus Thorns, is within its limits of variation and is not specifically distinct. Consideration must now be given to the name sponsorius. Pfankuch and Schmiede- knecht place sponsorius Grav. as a variety of cinctulus, but Gravenhorst attributed the name to Fabricius who would, of course, have priority. Dr. Kai Henriksen wrote to me that the type of sponsorius is not in the museum at Copenhagen. I am indebted to Kustos Dr. Olaw Schroder for information from Kiel. It is not known whether Pfankuch ever examined the Fabrician collection there. No specimen labelled Ich- neumon sponsorius is present. The type of Cryptus sponsor is present, labelled in Fabricius 's own writing: it is a female with a distinctly exserted ovipositor, and may well have been correctly identified by later authors. There is also a specimen labelled by Fabricius ' sponsatorius ' : on this the mid and hind tibiae each bear two spurs, so it can be no Cteniscine. It is a large species, about 15 mm. in length excluding appen- dages, and does not appear to agree with Fabricius's descriptions of sponsorius. It seems best to drop the name sponsorius unless or until another type can be discovered. Exyston sponsorius Grav. as identified by most later authors is a form of cinctulus Grav. ; and I find no clear indication in Gravenhorst's descriptions that it could refer to any other species. Thomson's diagnosis of genalis (1883 : 883) was quite inadequate, and has failed to lead subsequent authors to its recognition. In his collection are three specimens, all female and all belonging to the species figured and redescribed as genalis in the present work. Unfortunately, no one of these can with assurance be selected as lectotype: one, however, is labelled ' coll. Ros.' ; and since, as Dr. Kjell Ander informs me, Rosen- schiold collected in Vastergotland, this specimen could be the type. Thomson's words, 'Long. 3-3! lin. E. cinctulo colore statura et magnitudine simillimus', cor- roborate this interpretation of the species. I was able to tell A. Roman that this was the species placed as genalis in the Lund collection ; and his acceptance of this is evidenced by the series he arranged in the Swedish collection at Stockholm. He had known the species as a species, and he told me he had recognized it among Forster's material in the Vienna museum under the manuscript names apicalis (?) and aberrans (, 20.v.-io.vi, 1876-1889, i , 20.viii.i889, Tournier; Champel, i $, 25. iv, i $, iS.vi, Buess; Valais, les Hauderes, i (J, 6-27.vi.i935, R. B. Benson; Engadine National Park, 2 $$, 5 ??, 6.vii-6.viii, 1919-1923, Ch. Ferriere: GERMANY, Berlin dist., i , 8.vi, i $, I5.vi, i ?, 25.viii, i ^ without data, Buchecker coll. ; Erzge- birge, Gorkau, 3 ??, vii.1932, Th. Kupka; Schwerin, i $, ii.vi.i88o, i <$, 30^.1884, S. Brauns; Celle, i $, 2i.vi.i942, R. Hinz: CZECHOSLOVAKIA, Bohumin, i $, viii.1937, Th. Kupka: RUMANIA, 2 , K. Valle ; Taipalsaari, i ^, W. Hellen: SWEDEN, good series from many parts in Stockholm museum ; material under cinctulus and carinatus in Lund museum : GERMANY, Upper Bavaria, Berchtesgaden, i <, 30.v, i <$, 30.vi, i $, 2O.vi.i947, G. Heinrich: FRANCE, Maisons Laffitte, Fontenay, la Bourboule, &c., J. de Gaulle; Brout-Vernet, H. de Buysson ; Haut-Rhin, Alpes Maritimes, A. Seyrig (altogether 24 specimens in Paris museum): AUSTRIA, Styria, Admont, Kaiserau, 1,200 m., i $, vi.ig5o, G. Heinrich: 60 further specimens in British Museum (Natural History). An inordinately variable species. To decide whether it was a single species, the method adopted was to write a description first from the Irish material, which could be assumed to be conspecific; and then to modify this by reference to more than seventy other specimens, which was as diverse a collection as could be assembled. 374 A REVIEW AND A REVISION OF VARIATIONAL TENDENCIES A dark form is found in Ireland, the west of England and Scotland, and the high Alps. This tends to have the clypeus rather broadly rounded at apex. An inter- mediate has been studied from southern Sweden, Degaberga, having the same data as the normal form ; and also from La Bourboule, French Central Massif. A brighter, stouter form is found in Holland, Belgium, Switzerland, and Bavaria. This has the head rather less narrowed behind the eyes, and the petiolar segment relatively less elongate. Small specimens of the species tend to have the clypeus more transverse and the malar space a little shorter. Exyston cinctulus Grav. ab. approximatus, ab. n. Differs from the normal form of cinctulus in having genal costa approaching rather close to mandibular base but thereafter inflected (Fig. 49). RUSSIA: Samara (on R. Volga), i $, viii.1924, Lubischew (collection of A. Seyrig). Holotype in Museum d'Histoire Naturelle, Paris. FIGS. 41-42. Left hind femur, in outer side view, of Exyston cinctulus Grav. These illustrate the great intraspecific variation, and it is instructive to compare them with Figs. 16-17 (P- 3 2 7) f two species of Anisoctenion, where a good inter-specific difference is indicated. Fig. 41 is drawn from a female specimen from Czechoslovakia, and Fig. 42 from the type of ab. approximatus n. from Russia. A second specimen has since been definitely attributed to this aberration : FINLAND, Tammerfors, C. Lundahl: this is a fine female specimen of n mm. length. At first this was thought to be a good species, closely related to cinctulus Grav. but having secondary differences correlated with the primary character. When, however, cinctulus was studied intensively from a series of over eighty specimens, as diverse as could be obtained, it was found that the supposed differences of the Russian specimen were within or but little outside the great range of variation in cinctulus to the west of Russia. Other specimens were found, with the genal costa inflected but the intervening space narrower. Yet other specimens have a mere trace of an inflected genal costa ; and one particularly instructive specimen (in British Museum) has the genal costa THE CTENISCINI OF THE OLD WORLD 375 split into an outer branch running almost normally as in Fig. 50 and an inner branch inflected in the manner of the aberration. It appears that in the normal form the genal costa, though having the condition described as continuous, continues towards the hypostomal costa as a lamina contiguous with the genal margin backing the mandibular articulation. Exyston cinctulus Grav. subsp. clavatus Cresson 1864 Cteniscus clavatus Cresson, Proc. Ent. Soc. Philadelphia, 3: 284. 1897 Exyston clavatus Davis, Trans. Amer. ent. Soc. 24: 238. 1944 Exyston clavatus Townes, Mem. Amer. ent. Soc. 11 (i): 167168. Altogether less strongly hairy than the European form : genal costa not so strongly elevated or hairy: yellow coloration more extensive behind eyes: antennal scapes largely dull fulvous, with a yellowish spot beneath : flagellum also dull fulvous above in apical half : thorax more richly yellow-marked : propodeum decidedly tending to dull rufous (among European material I have only seen this in a few old specimens which may have suffered fading) : coxae darkened to rufous only, not at all black: petiolar segment scarcely infuscate at base, but there is some infuscation on other tergites, particularly 3 and 4. Diagnosis based on I ?, losco Co., Michigan, 16-17.^.1939, R. R. Dreisbach; det. H. K. Townes. W. R. M. Mason kindly informs me that clavatus is very much less variable than the European form : also that it is evidently not an introduced species, for it is known from remote parts of northern Ontario ; yet it is geographically isolated, being un- known west of the Rocky Mountains. These facts strengthen my belief that it is a true geographical sub-species. He has since written that he agrees with my treatment of clavatus, and has sent me a further two females and a male for study. These substantially confirm the above diagnosis; but the male has the hind coxae and petiolar segment pale castaneous except at apex. He has recorded this new synonymy in the North American catalogue (1951 April: 229). Exyston genalis Thomson [883 Exyston genalis Thomson, Opuscula Entomologica, fasc. 9: 883 (non Roman 1931, nee Teunissen 1948). Head rather strongly to strongly narrowed behind eyes ; moderately emarginate behind: on vertex, temples and inner orbits rather finely to moderately punctate; on genae more coarsely, irregularly and in part confluently so: epistoma coarsely reticulate-punctate: upper face with a distinct median keel, moderately punctate, more or less rugose on triangle between median ocellus and antennal sockets. Clypeus rather strongly transverse, distinctly more so than in cinctulus Grav. ; rather sparsely beset with punctures of different sizes, fine to coarse ; rather deeply delimited basally ; more or less narrowly rounded at apex. Malar space about equal to distance between mandibular apices in female, distinctly less in male. Hypostomal costa strongly raised, the elevation falling gradually to the foramen magnum. Genal costa strongly 376 A REVIEW AND A REVISION OF raised, double (Fig. 51) : the outer costa almost continuous, the inner strongly in- flected, meeting the hypostomal costa at the point where the latter is most strongly raised ; the intervening space traversed by a series of ridges : outer and inner costae crenulate ; usually both are upstanding in the female ; but sometimes in that sex, and usually to a greater extent in the male, the inner costa is bent outward, so that the intervening space is in varying degree narrowed. Antennae with scape weakly excised externo-ventrally (Fig. 54) ; with postannellus about 1-8 times length of the following segment ; with flagellum 31- to 37-segmented. Pronotum with epomiae of moderate strength; laterally moderately punctate above, finely punctate to rugose below. Propleura moderately, or rather finely and closely punctate. Mesoscutum strongly and almost evenly convex ; with notaulices very superficial and not clear-cut ; more or less finely punctate medially, very finely punctate and shining laterally. Scutellum moderately convex; rather finely to moderately and often rather closely punctate (in the specimen from the Tatra Mts. largely rugose) ; not margined to middle. Mesopleura moderately punctate, in part rugose-punctate, in front ; more finely punctate behind : sternaulices quite distinctly, though seldom strongly, impressed : mesosternum densely and very finely punctate : epicnemial carina strongly raised, indistinct above sternaulices, medially emarginate but not nearly interrupted: mesolcus very sharply impressed, partly closed behind. The male differs in having the epicnemial carina only moderately raised, and the mesolcus more or less distinctly transcostate, not sharply impressed. Propodeum with costae rather weak but normally complete, though area basalis sometimes in- completely delimited apically ; with area pleuralis finely and closely punctate ; else- where more or less finely rugose. Gaster distinctly clavate. Petiolar segment elongate and slender, 2-5 to 3-1 times length of its greatest breadth near apex ; bearing lateral hairs distinctly longer than half its least breadth; weakly or moderately auriculate basally; with dorsal keels rather sharp, distinct for quite three-quarters length of segment; almost smooth between the keels, finely punctate and obsoletely rugose beside and behind them. Tergite 2 distinctly raised on mid line at base, and impressed to either side of this elevation; rather finely to moderately punctate and often rugulose on basal half. Remainder of gaster more and more finely punctate. Female third valvulae small, vomeriform. Fore wings with basal and disco-cubital veins strongly convergent ; with areolet usually shortly petiolate but sometimes sub-sessile or sessile ; with nervus parallelus emitted in middle of brachial cell. Hind wings with nervellus almost opposite, inter- cepted a little above middle. Hind femur 4-1 to 5-2 times length of its greatest breadth. Tarsal claws not pectinate. Length 6 to 10 mm. Female. Head black above a line running roughly transversely around, just below antennal sockets and foramen magnum ; below this line stramineous to flavous, the pale coloration also extending some way up orbits of upper face: mandibles con- colorous except at apex. Antennae with scape, pedicellus and postannellus mainly black, the scape almost always conspicuously pale-marked below : with annellus cas- THE CTENISCINI OF THE OLD WORLD 377 taneous; beyond postannellus changing to ferrugineous above, dull fulvous below. Thorax black, with the following parts pale-coloured as on head : propleura mainly, ventral margins of pronotum at sides (the marks sometimes interrupted), region of epicnemial carina with backward extensions on sternaulices and beside mesolcus, small marks below wing-base, tegulae, mark at apex of scutellum and usually another on metascutellum. Legs rufous ; with trochanters and trochanterelli, fore and mid coxae mainly, and hind coxae nearly always below, bright flavous ; with knees rather indefinitely pale flavous ; with the following parts infuscate : hind tarsi, hind tibiae at apex and before base, and hind femora before apex and at base : hind coxae gener- ally in greater part black and usually rufous-marked, but sometimes in greater part rufous. Wings with veins and stigma pale ferrugineous, more or less yellow at base. Petiolar segment back, with a small pale mark at mid apex. Remainder of gaster normally mainly rufous ; with tergite 2 black at base and sides, and with lateral black marks at base of tergites 3 and 4 ; with tergites having ill-defined but successively broader pale yellow apical margins, which occupy most of tergites 6 and 7: the yellow coloration may be more extensive, or the pale parts of the tergites may be to a greater or lesser extent overspread with infuscation. Male differs as follows. Black coloration more extensive, reaching behind eye to a line between eye and lower articulation of mandible ; often developed at mid base of mandible, along clypeo-frontal suture, and extending farther on to epistoma: orbits of upper face not pale. Scape with pale mark smaller or almost undeveloped. Thorax black, with tegulae usually wholly or partly pale, but without other pale marks. Fore and mid coxae to a large extent black or blackish at base : hind coxae almost entirely black. Yellow margins of tergites narrower and often more sharply defined: in one specimen from Bavaria tergites 2 to 6 mainly black above, with just a trace of rufous colouring on 3 and 4. The male specimen from the French Ardennes has much of the female structure and coloration. This species is illustrated on PI. 5. Redescribed from the following material: NORWAY, Sognsvatnet, Aker, i $, 3i.vii.i935, T. Soot-Ryen (mus. Stockholm, recorded as E. sp. ign. by Roman, 1942: 14): SWEDEN, Dalarne, i 3, i6.vii, A. Roman; Dalarne, Orsa W., i ?, 5.^.1938, G. J. K. : FINLAND, Keuru, i <$, J. Sahlberg ; Karislojo, 2 33, n.vi.i934, R. Krogerus ; Ruovesi, i <$, 26.vi.i93i, V. Saarinen; Tampere, Aitolahti, i <, 26.vi, i $ 28.vi.i933, A. Saarinen ; Kuusamo, i <$, 6.vii. 1935, R. Krogerus ; Kuusamo, Ruskeakallio, i 3, I7.vii.i935, R. Krogerus, i $, 26.vii.i935, G. J. K. : SWITZERLAND, Nyon, i $, Cheyr: FRANCE, Ardennes, Verdun, i $, io.viii.i925, Benoist: BELGIUM, Severin, i 3, 22.vi. 1899, Bastogne, i 3, 23.viii.i887, J. Tosquinet: GERMANY, Schwerin, i $, viii.i887, S. Brauns (Nat. Mus. Ireland) ; Koln, i $, vii.1945, Aerts; Goslar, i $, 2i.viii.i945, Celle, i $, 26.vii.i942, R. Hinz; Berlin dist., i 3, Ruthe coll. ; Oderberg-Liepe, i $, I9.viii.i938, G. J. K. ; Bavaria, Erzgebirge, Gorkau, i $, n.vi.i932, Th. Kupka: MIDDLE EUROPE (unlocalized), i $, 4.vii.i900, R. von Stein coll.: CZECHOSLOVAKIA, Bohemia, Chodau, i <$, io.vi.i878, R. von Stein coll. ; Erzgebirge, Aussig, i $, n.vi. 1932, Th. Kupka: POLAND, Tatra Mts., Zakopane, i <$, 28.vi.ig32, D. Aubertin & E. Trewavas. Additional material seen: ITALY, Liguria, i $ in Spinola coll. in Turin museum: ENTOM. n. 6 3 A 378 A REVIEW AND A REVISION OF GERMANY, Bodmann am Bodensee, i $, 25.v., i , 1886, i ?, S.vi.iSSg, Tournier; Champel, i $, coll. Buess; CZECHOSLOVAKIA, Moravia, Brno Hady, i $, I3.vi.i936, A. Hoffer. Further material seen: FRANCE, Paris district, St. Michel, 0. Sichel; Haut-Rhin, 3O.V.I920, A. Seyrig (4 specimens in mus. Paris) : GERMANY, Schleswig, Kiel, i <$, 22^.1949, G. Heinrich: AUSTRIA, Styria, Admont district, 1,200 m., on Alnus incana, i $, 25.vi.i95o, G. Heinrich: TuRKEY-in-Europe, Istanbul, i $, coll. de Gaulle; this is a peculiar specimen, with the spiracular tubercles of the petiolar segment sharp and very prominent. \ Exyston subnitidus Gravenhorst 1829 Tryphcm subnitidus Gravenhorst, 1 chneumonologia Europaea, 2i 144-145 (non Morley 1911, nee Morley 1936). 1906 Exyston subnitidus Pfankuch, Z. Hym. Dipt. 6: 90. This species is closely related to, but much less strongly hairy than, phaeorrhaeus Hal. Head rounded and rather strongly narrowed behind eyes (decidedly less so than in phaeorrhaeus Hal.); moderately and rather angularly emarginate behind: before occiput finely punctate above ; on temples and cheeks for the most part moderately punctate, sparsely so between eyes and ocelli ; on inner orbits rather finely punctate and on malar space densely so : epistoma of female rather finely punctate above to rather coarsely so below, on an alutaceous background, the punctures for the most part well separated; of male reticulate-punctate to rugose, the punctures not well separated except in mid line : upper face with median keel absent or very short ; rather coarsely punctate, the punctures for the most part well separated. Clypeus as de- scribed for phaeorrhaeus Hal. Malar space about three-quarters distance between mandibular apices, the gena not angularly raised before mandibular articulation. Genal region of the type described and illustrated (Fig. 52) for phaeorrhaeus Hal., differing as follows : outer costa seen in frontal view as in Fig. 44, not forming a very distinct angle with the normal plane of the gena, not bulging so strongly outward, so that the genal cavity is less wide : inner costa running forward to a sharper point, but incomplete internally, so that the cavity is partly open behind : hypostomal costa more strongly and laminately raised, distinctly bent outward: the keel that arises from point of hypostomal costa does not run very close behind mandibular articula- tion nor appear to close the cavity in front : the keel that crosses the cavity from point of genal costa itself raised to a point. Antennae with scape weakly excised externo-ventrally (Fig. 57) ; with postannellus about 1-7 times length of the following segment ; with flagellum 25- to 27-segmented. Pronotum as described for phaeorrhaeus Hal., but laterally moderately punctate above. Propleura rather finely punctate. Mesoscutum rather strongly convex ; with notaulices scarcely traceable ; medially rather finely punctate, at sides more finely and anteriorly densely so. Scutellum strongly convex ; rather finely to moderately THE CTENISCINI OF THE OLD WORLD 381 punctate ; not margined to middle. Mesopleura not much rugose anteriorly ; moder- ately punctate above, rather coarsely so below, and closely so almost throughout ; sternaulices broad and rather shallow and, like the mesosternum, rather finely and closely punctate : mesolcus rather deeply and sharply impressed, closed by an incised ridge behind : epicnemial carina much as described for phaeorrhaeus Hal. Propodeum with costae moderately strong: finely to very finely punctate above, moderately so on area pleuralis : areabasalis not falling sharply from apex : area posteromedia rugose, not elongate as in phaeorrhaeus Hal. and thus not extending far beyond middle. Caster not strongly clavate as it is in phaeorrhaeus Hal. Petiolar segment 2-0 to 2-1 times length of its greatest breadth ; moderately auriculate basally ; with dorsal keels FIGS. 43-44. Head in facial view of Fig. 43. Exyston phaeorrhaeus Hal. Fig. 44. Exyston subni- tidus Grav. not distinct to near apex, nor marking such an abrupt change of slope as in the previous species ; conspicuously punctate, or in part rugose-punctate, in apical two- thirds. Tergite 2, like those following, normally finely punctate and shining; almost evenly convex between the thyridia, which are not very conspicuous. Female third valvulae weakly vomeriform. Fore wings with basal and disco-cubital veins extraordinarily strongly convergent, so that a continuation of the lower abscissa of the disco-cubital, from point of ramel- lus, would meet the stigma before the point of emission of the radius ; with areolet very distinctly petiolate; with nervus parallelus emitted slightly below middle of brachial cell. Hind wings and tarsal claws as described for phaeorrhaeus Hal. Length 6 to 7 mm. Female. Head coloured as in female phaeorrhaeus Hal., but the pale coloration more distinctly yellowed and rather less extensive; region of tentorial pits black. Antennae mainly black at base ; to some extent paler towards apex above ; in greater part dull fulvous below. Thorax black, with scutellum near apex, tegulae, and lower angle of pronotum yellowish-ivory: propleura black with inner margins flavous, or flavous with outer margins black. Gaster mainly castaneous, or with petiolar segment and bases of following tergites blackish: tergites somewhat paler at sides. Wings with costa yellow, with other veins and stigma more or less ferrugineous, paler at base. Male. Head coloured as in male phaeorrhaeus Hal., but neither specimen before me has more than a sub-triangular apical mark pale on clypeus. Thorax black with 382 A REVIEW AND A REVISION OF tegulae more or less castaneous. Tergites 2 and following with distinct pale margins, these becoming broader and paler towards apex, which is mainly bright flavous. Legs coloured much as in phaeorrhaeus Hal., sex for sex; but the mid and hind coxae and the trochanters are blackish, a little paler beneath. Redescribed from the following material: SWITZERLAND, Engadine national park, Zernetz dist., i $, 25.vii.i92i, i <$, 23.vii.i922, Grupshum, i <, i8.vii.i925: SPAIN, Valle de Ordesa, i $, viii.i93i, J. M. Dusmet. Exyston albicinctus Gravenhorst 1820 Ichneumon albicinctus Gravenhorst, Mem. Accad. Sci. Torino, 24: 375 (non Ichneumon albicinctus Gravenhorst 1829 = Barichneumon locutor Thunb.). 1829 Tryphon albocinctus Gravenhorst, Ichneumonologia Europaea, 2 1 204-205, syn. n. 1829 Tryphon tricolor Gravenhorst, ibid. 2: 207-208, syn.. n. 1878 Exyston cinctulus <$ var. Brischke, Schrift. naturf. Ges. Danzig, 4: 106. 1906 Exyston albocinctus Pfankuch, Z. Hym. Dipt. 6: 222. 1906 Exyston tricolor Pfankuch, ibid. 6: 223. Head slightly narrowed or very slightly broadened behind eyes, and then strongly rounded ; shallowly or rather shallowly emarginate behind : from level of lateral ocelli shining, with temples rather finely punctate: genae moderately to rather coarsely punctate, the punctures in many cases separated by less than their own diameters : malar space and inner orbits rather finely punctate : lower face coarsely, and in large specimens very closely punctate, the punctures somewhat squashed out of shape by their mutual proximity : upper face very closely beset with moderate punctures, often almost trans-rugose between antenna! bases and median ocellus, and with a median keel running to the median ocellus. Clypeus irregularly punctate, rather coarsely so above, rather finely and sparsely so below ; scarcely convex ; very superficially, and very seldom at all sharply, delimited basally, and not set distinctly at an angle with lower face ; apically distinctly bi-tuberculate. Malar space about equal to or, more usually, less than distance between mandibular apices in male, about one-fifth greater than such distance in female. Genal costa moderately raised ; inflected much as in pratorum Woldst. (Fig. 48), but more angularly so in female: the gena not angled between costa and mandibular base, in position where in other species there is an outer costa. Hypostomal costa, where met by genal, bent outwards and raised rather to a point. Antennae with scape (Fig. 55) moderately excised externally ; with postannellus about twice length of the following segment, both measured along inner margin: flagellum 31- to 35-segmented. Pronotum with epomiae of moderate strength or rather weak, not sharp ; laterally moderately punctate above, closely punctate to rugose-punctate below. Propleura moderately and closely punctate, but rather sparsely so in the small specimen from Finland. Mesoscutum moderately and almost evenly convex ; with notaulices not clearly indicated ; finely to moderately punctate, sparsely so across disk, more closely so in front and behind. Scutellum rather strongly and evenly convex, though some- times slightly buckled ; rather coarsely and closely punctate ; not distinctly margined beyond basal furrow. Mesopleura in front coarsely and closely punctate, often in part THE CTENISCINI OF THE OLD WORLD 383 rugose-punctate ; behind more moderately and less closely punctate, but without a distinct impunctate speculum : sternaulices broad but generally superficial : meso- sternum densely and rather finely punctate: epicnemial carina moderately raised between and along fore margin of sternaulices, then with margin curving inward; medially emarginate: mesolcus sharply impressed, often weakly transcostate, with either a strong costa closing it behind or else a series of weaker transverse costae in that position. Propodeum with areolation weak and incomplete in middle ; in particu- lar, the area superomedia is not or very indistinctly bordered basally and laterally, and the costulae are usually incomplete though always indicated : area posteromedia well bordered and transversely rugose: area angularis longitudinally rugose: area pleuralis moderately and rather closely punctate : area externa more finely and other lateral areas more coarsely punctate, all less closely so : area superomedia represented by a more or less distinctly shining patch, impunctate or sparsely punctate, often flanked by weak transverse rugosities. Petiolar segment of female 2-2 to 2-3 times, of male 1-9 to 2-1 times length of its greatest breadth ; very strongly auriculate at base ; with dorsal keels distinct in about basal third of segment, thereafter not or scarcely indicated in male, though with clearer indications in female ; for the most part moderately punctate, but in middle third between positions of keels shining, and not or sparsely punctate. Remainder of gaster with tergite 2 discretely and for the most part moderately punctate; with remaining tergites more and more finely punctate towards apex. Female third valvulae vomeriform. Fore wings with basal and disco-cubital veins strongly convergent ; with areolet distinctly petiolate; with nervus parallelus emitted below middle of brachial cell. Hind wing with nervellus almost opposite, intercepted about in middle. Tarsal claws not pectinate. Length 6| to n|, normally 8 to 10 mm. Female. Head black above a line running roughly transversely around, just below antennal sockets and foramen magnum ; below this line ivory-white, much modified by flavous or fulvous, the pale coloration also extending some way up orbits of upper face: mandibles concolorous except at apex. Antennae with scape, pedicellus and postannellus mainly black, the scape conspicuously pale-marked in apical half below ; with annellus mainly castaneous ; with remainder of flagellum fulvous, ferrugineous for a short distance at base beneath and a much greater distance above. Thorax black, with the following parts pale-coloured as on head : pronotal collar, widely near hind angles, propleura totally, tegulae, small mark below fore wings and hind margin of mesopleura narrowly above, and large apical marks of scutellum and metascu- tellum ; also on mesothorax a pair of marks that go broadly down lower halves of fore margins of pleura, then narrow on upper halves of sternaulices, then greatly broaden along fore margin of mesosternum, then continue more narrowly backward beside mid line to closure of mesolcus. Legs with the following parts pale flavous : coxae, trochanters and trochanterelli, fore and mid femora anteriorly, and knees ; with fore and mid femora behind, and with hind femora fulvous to castaneous ; with tibiae and tarsi flavous to fulvous, the tarsi somewhat darkened, and the hind tibiae darkened above near apex and with a broad darker ring near base ; with dark marks on femora, 384 A REVIEW AND A REVISION OF hind trochanters and mid coxae at base, and with hind coxae mainly almost black above. Wing veins pale ferrugineous, except for the yellow costa: stigma ochreous. Petiolar segment black, with a median apical spot coloured as on following : remainder of gaster bright, pale fulvo-castaneous, the tergites more and more broadly margined apically and laterally with yellowish-white, tergites 6 and 7 being mainly so coloured, but the colour margins are not sharply denned : tergite 2 blackish at base, with yellow thyridia: sterna mainly yellowish. Male differs as follows. Head for the most part black, including a very broad band inside basal margin of clypeus ; with pale colour on remainder of clypeus, on middle and along basal margins of mandibles, and on maxillary and labial palps. Antennal scape without a pale mark beneath, or with an inconspicuous mark that does not reach apex of segment. Thorax black ; with a small pale apical mark on scutellum, and sometimes an obscurer one on metascutellum : tegulae partly ferrugineous. Legs coloured as in female, but with fulvous coloration darkened ; with coxae wholly or for the most part, trochanters usually in part and hind femora behind, black or almost so. Gaster coloured as in female, but usually tergites 3 to 5 are blackish at base and tergite 2 more than half black : the palest coloration is always less extensive and often more yellow. Redescribed from the following material: BELGIUM, Steenbrugge, i <$, 24^.1875, Ghlin, 5 $$, ij.vi.i8yj, Mont 1'Enclus, i , 26^.1893, J. Tosquinet: HOLLAND, Exaeten, i $, vi.i884, i <$, 5.vi.i887, A. Wasmann: GERMANY, Schwerin, i <$, ii.vi.i88g, S. Brauns: SWITZERLAND, Geneva, Peney, i $, 29^.1876, i , 5.vi.i88g, i <$, 2O.vi.i889, Tournier; Champel, i <$, coll. Buess; without data, i ? : HUNGARY, Jaszbereny, i $, Moczar (mus. Stockholm): FINLAND, Abo district, Uskela, i $, E. J. Bonsdorff, coll. Maklin: EUROPE, unlocalized, i <$ bought from Staudinger (Brit. Mus.). Additional material seen : FRANCE, Paris district, Maisons-Lamtte, Rueil, Nantesse ; Varennes, St. Baume, J. de Gaulle (4 specimens in mus. Paris) : GERMANY, Crefeld, i ?, vi.1928, A. Ulbricht. Originally described from Piedmont, Italy, and the female synonym from the Rhineland. Brischke's insect was from the Danzig district : from his diagnosis I have no doubt that it belonged to this species. Evidently a very scarce species : Morley has already, quite rightly, rejected Stephens's British records. Exyston sp. 1931 Exyston genalis Roman, Ark. Zool. 23A (6): 14 (non Thomson 1883). It is thought desirable to correct an existing record, but inopportune to validate this species with a name based on a single imperfect male. In order, however, to stimulate the search for further specimens, and to increase our understanding of the genus as a whole, the species has been separated in my key to palaearctic species, and is here further diagnosed. A larger, more elongate species than pratorum Woldst.: length of unique male 10 mm. Upper face reticulate-punctate, not rugose. Clypeus sharply delimited basally. Malar space a little less than distance between mandibular apices. Post- THE CTENISCINI OF THE OLD WORLD 385 annellus 1-7 times length of following segment. Pronotum laterally moderately punc- tate above. Notaulices short and extremely superficial. Mesopleura coarsely punctate in front. Propodeum with costae normally complete and rather strong. Petiolar segment about 2-5 times length of its greatest breadth ; weakly auriculate basally ; with dorsal keels strong and almost percurrent. Tergite 2 reticulate-rugose in basal half, thereafter rugose-punctate. Fore wings with basal and disco-cubital strongly convergent. Tarsal claws not pectinate. So closely related to the nearctic E, boreotis Davis (of which I have seen two pairs determined by H. K. Townes) that I thought it might be a form of that species. In the American specimens, however, the temples are finely or rather finely punctate ; whereas in the Siberian they bear coarse punctures separated by less than their own diameters. I have been allowed to submit the specimen to W. R. M. Mason, who is of the opinion that it is not boreotis Davis but that, like that species, it belongs to the group of variatus Prov. Exyston pratorum Woldstedt 1858 Exenterus triangulatorius Holmgren, Svensk Vet.-Akad. Handl. 1: 235 (non Gravenhorst 1829). 1874 Exenterus pratorum Woldstedt, Bidr. Kdnn. Finl. Natur Folk, 21: 52-53. 1877 Cteniscus (Tricamptus) pratorum Woldstedt, Bull. Acad. Sci. St. Peter sbourg, 23: 454. 1883 Exyston brevipetiolatus Thomson, Opuscula E-ntomologica, fasc. 9: 883 (synonym by Hell6n 1937)- 1911 Exyston brevipetiolatum Morley, The Ichneumons of Great Britain, 4: 192-193. 1911 Exenterus phaeorrhoeus Morley, ibid. 4: 207 (non Haliday 1838). 1911 Exyston triangulatorius Schmiedeknecht, Opuscula Ichneumonologica, 5: 2300-2301. 1937 Exyston pratorum Helle'n, Notul. Ent. 17: 124. 1939 Exyston phaeorrhaeus Roman, Ent. Tidskr. 60: 195 (non Haliday 1838). Head slightly to moderately narrowed behind eyes and strongly rounded ; rather shallowly to moderately emarginate behind : vertical region and temples very finely punctate, genae finely to moderately so ; upper face finely to moderately and closely punctate, sometimes transrugulose just above antennae ; epistoma rather coarsely punctate, sometimes reticulate in part. Clypeus for the most part rather coarse- ly punctate ; sharply and rather deeply delimited basally ; more or less weakly convex ; often quite broadly rounded at apex. Malar space of female a little less than distance between mandibular apices ; of male much shorter, about equal to width of annellus. Genal costa moderately raised, strongly but seldom angularly inflected (Fig. 48), often bearing short carinae in front : hypostomal costa strongly and roundedly raised, bent slightly outward. The gena is roundedly angulate between costa and mandibular base, in the position in which in other species there is an outer costa (see again Fig. 48) : in male this angulation is carried farther back (dorsal), is much sharper near mandi- bular base, and bounds an intervening area that is more distinctly excavate behind articulation of mandible and is often distinctly though imperfectly transcarinate. Antennae strongly hairy ; with scape weakly excised externo-ventrally (Fig. 58) ; with postannellus generally less than 1-5 times length of following segment; with flagellum of female 23- to 26-, of male 22- to 25-segmented. Pronotum with epomiae rather strong and sharp ; laterally very finely punctate ENTOM. II. 6 3 B 386 A REVIEW AND A REVISION OF above, rugose in part below. Propleura finely punctate. Mesoscutum very strongly convex anteriorly, flattened discally; with notaulices shallowly impressed, but extending to beyond middle ; more or less finely punctate, often with some faint rugulosity. Scutellum weakly to moderately convex ; not margined to middle ; gener- ally finely punctate above, closely punctate or rugose-punctate behind, longitudinally rugose at sides. Mesopleura each with a large, almost impunctate area above ; finely to moderately punctate below, moderately so above : sternaulices broad and moder- ately impressed: mesosternum very finely punctate: epicnemial carina moderately FIGS. 45-47. The very variable petiolar segment of Exyston pratorum Woldst. Fig. 45 drawn from a male from Orsa, middle Sweden, resembling the type and illustrating the applicability of Forster's diagnosis of Tricamptus to this species. Fig. 46 drawn from a normal female from Paanajarvi, NE. Finland, and Fig. 47 from a male from the New Forest in southern England, illustrating the other extreme condition. raised between and along fore margin of sternaulices, then bent upward on to meso- pleura, medially almost completely interrupted: mesolcus moderately to strongly impressed, broadening posteriorly, closed behind. Propodeum with costae of moder- ate strength ; mainly more or less finely rugose, but with some upper areae shining and very finely punctate, the area pleuralis generally discretely punctate in front, and the area posteromedia finely reticulate-rugose : area superomedia not distinctly delimited basally. Petiolar segment (Figs. 45-47) short for an Exyston, 1-3 to 1-7 times length of its greatest breadth ; dorsally reticulate-rugose, sometimes finely so ; not strongly auricu- late basally ; with dorsal keels broad and flattened at base, thereafter not very strong though sometimes almost percurrent: spiracular tubercles undeveloped or more or less prominent. Remainder of gaster moderately punctate, finely so towards apex ; with tergites 2, generally to beyond middle, and 3 at base longitudinally finely rugose. Female third valvulae small, vomeriform. Fore wings with basal and disco-cubital veins almost parallel ; with areolet sessile or sub-sessile ; with nervus parallelus emitted below middle of brachial cell. Hind wings with nervellus strongly antefurcal, intercepted below middle. Tarsal claws finely pectinate. Length 5 to 7^ mm. Female. Head black, with the following parts ivory-white, usually in part suffused THE CTENISCINI OF THE OLD WORLD 387 ith flavous and margined with fulvous: palpi, mandibles except at apex, clypeus, genae broadly, and lower face largely. The pale colour on the gena usually extends beyond the angulation and on to the hypostomal costa ; it extends broadly more than half-way up the sides of the lower face, which also bears a rather heart-shaped mark almost covering the epistoma: the pale marks on the lower face may be so largely confluent as to confine the lower black coloration to the tentorial pits. Antennae with basal segments mainly black, often castaneous at apex and sometimes more exten- sively below : beyond postannellus ferrugineous above, often paler towards base, dull fulvous below, often darker at apex. Thorax black, with the following parts pale- coloured as on head : propleura largely, tegulae, and scutellum and metascutellum at apex ; usually lower angles of pronotum and often an extension along ventral margin, and generally a pair of marks on anterior margin of mesopleura ; occasionally also marks on epicnemia, below bases of fore wings and down posterior margins of meso- pleura. Fore and mid legs with coxae, trochanters and trochanterelli similarly pale- coloured, the trochanters dark-marked above, the coxae darkened usually only at base but sometimes more extensively ; with femora fulvo-castaneous and with knees, tibiae, and tarsi fulvous to flavous: hind legs with coxae, trochanters, and trochan- terelli black or almost so, pale at apex and on trochanterelli beneath ; with femora, tibiae, and tarsi ferrugineous, the femora and tibiae more or less dull castaneous in middle or almost entirely, the tibiae sometimes much paler than the femora, and the knees conspicuously stramineous-white. Wing veins and stigma ferrugineous to fus- cous, basally paler. Petiolar segment black. Remainder of gaster normally having tergite 2 black at base and sides, and 3 with isolated black marks at mid base and at anterior angles ; with tergite 2 mainly ruf ocastaneous, the following similar but succes- sively duller; with tergites having pale margins, yellowish to ivory, broader on successive segments to apex, covering most of tergite 6. In some specimens the black coloration is much more extensive, covering all of tergite 2 but the apical margin and thyridia ; and in some the pale apical margins are much less pronounced. Male differs as follows. Head without the pale marks on genae and sides of lower face, with heart-shaped mark on epistoma smaller, and with clypeus narrowly black along sutural margin. Thorax black ; with tegulae generally pale ; sometimes (as in type) with a transverse pale mark near apex of scutellum. Mid, and generally fore, coxae dark except at extreme apex. Gaster sometimes mainly black dorsally ; pale margins of segments, on the average, narrower and sharper. Redescribed from the following material: IRELAND, S. Tipperary, Ballinacourty, i ?, W. Cork, Glengarriff, i ?, Waterford, Dungarvan, i $, Queens Co., Portarlington, i (J, Woodbrook, i ?, N. of Emo Park, 2 ?$, Wicklow, Clara, 3 $<$ (one seen by A. Roman, see Roman, 1939), i ?, Arklowdist., i <, Ballinclea, i <, Deputy's Pass, i $, Athdown, 2 $$, 4.vi-26.vi and 26. vii-i6.viii. 1925-1945, A. W. Stelfox: SCOTLAND, Argyll, Puck's Glen, i $, 9^1.1925, J. J. F. X. King; Inverness-shire, Aviemore, i $, 26.vi.i934, R. B. & J. E. Benson: ENGLAND, Devon, Cornworthy, i $, T. A. Marshall coll. ; Kingsteignton, 2 <$$, i ?, Hilber, i $, Lustleigh, i , I3.viii.i890, J. Tosquinet: SWEDEN, Dalarne, Orsa, i #, 4.viii.i938, G. J. K. : FINLAND, Ladoga-Karelia, Sortavala, i $, 27.vi, Woldstedt (type) ; Kuusamo, Liikasenvaara, i $, 3.viii.i935, Paanajarvi, Ruskeakallio, i $, I7.vii.i935, G. J. K. : FRANCE, Paris district, Claman, i , 5.vi.i892, Mesni