108 R. W. CROSSKEY 



bare. Fore tarsus dilated, fore basitarsus 4-6 times as long as its greatest breadth. $ : cibar- 

 ium armed with some very minute denticles. Tarsal claws with very small basal tooth. Scutum 

 with boldly marked pale grey pattern anteriorly, the grey mark more or less anchor-shaped or 

 as a pair of horse-shoes meeting in mid-line and with open ends directed outwards. Abdomen 

 sparsely fine haired, sixth to eighth tergites shining and preceding tergites matt. Seventh 

 sternite undeveloped. Gonapophyses simple bluntly rounded lobes, occasionally slightly obtuse- 

 angulate at the tips. Paraprocts normal. Spermatheca without reticulate surface pattern or 

 internal hairs, o* : scutum with a pair of bright silvery shoulder patches. Genitalia with very 

 large elongate subparallel-sided styles which are about two or three times as long as the coxites, 

 style with a single apical spinule ; coxite not produced beyond base of style ; ventral plate 

 complex, body of plate narrow and heavily toothed apically, with a coarsely haired ventral or 

 anteroventral process (usually forming a broad tapering projection in profile), basal arms widely 

 divergent ; median sclerite large and subovate, deeply cleft apically and with edges serrate or 

 frayed at posterior end ; parameres broad basally but strongly tapering towards parameral 

 hooks, usually subtriangular, parameral hooks very numerous. Pupa : Gill with 8 simple elong- 

 ate filaments arising near base, nearly always in four regular pairs each with a short common 

 stem ; gill filaments spreading basally but bending so that tips are approximated, directed for 

 wards and extending far beyond cocoon ; gill about as long as pupal body. Abdominal on- 

 chotaxy normal ; segments 7 and 8, and usually also segment 6, dorsally with spine-combs. 

 Cocoon simple, anterior margin usually a little thickened, without anterodorsal median projec- 

 tion, slipper-shaped, loosely woven but not fenestrate. Larva : Head and cephalic fans normal. 

 Hypostomium with usual nine apical teeth, these blunt or only slightly prominent ; 4-10 setae 

 in each hypostomial row, rows subparallel to or slightly divergent posteriorly from lateral margins 

 of hypostomium. Head-spots positive, boldly marked and usually not obscured by infuscation 

 of surrounding areas. Postgenal cleft small, shorter than postgenal bridge or about equal in 

 length to it, subquadrate or evenly rounded. Mandible normal, first three comb-teeth evenly 

 decreasing in size, other comb-teeth very long and fine and sharply differentiated from first three ; 

 two mandibular serrations. Antenna of medium length, with four segments (rarely an apparent 

 fifth because of secondary annulation). Thoracic cuticle bare. Abdomen broadest at sixth or 

 seventh segment, well before posterior circlet. Abdominal cuticle bare. Ventral papillae 

 present but small, bluntly rounded and inconspicuous. Accessory sclerites absent. Rectal 

 scales present. Rectal gills usually without secondary lobules, sometimes from one to four small 

 secondaries on each lobe. Posterior circlet with 60-110 rows of 12-15 hooks. 



Distribution. An entirely Palaearctic subgenus found throughout most of the 

 region, the distribution including Madeira, Canary Islands and North Africa (Map 

 n). Present in Japan, absent from Iceland. 



Discussion. The subgenus Odagmia as here restricted is equivalent only to the 

 ornatum-group of Odagmia as treated by Rubzov (1 959-1 964) in his monograph of 

 Palaearctic Simuliidae, and the variegatum-gvonp (included in Odagmia by Rubzov) 

 is excluded, since it has the pleural membrane bare and has other characters that on 

 balance place it in Simulium s. str. much more satisfactorily than in Odagmia. 

 Davies (1966 : 421) has adduced arguments for not recognizing Odagmia as a valid 

 subgenus, and places the type-species Simulium ornatum Meigen in the subgenus 

 Simulium s. str., thus treating Odagmia as a synonym of Simulium in the restricted 

 sense ; in the present work, as in my paper on the Simuliidae of the Middle East 

 (Crosskey, 19676), it is preferred — while recognizing the undoubted close relationship 

 between the subgenera — to accept Odagmia as a valid subgenus. As here treated it 

 includes ornatum and all its immediate allies that differ consistently from Simulium 

 s. str. by having the pleural membrane haired ; in addition they differ from many 



