4 JAMES WATERSTON. 



stream in a nullah south of the Seres Boad, kilo 58-|. The locality was indicated 

 to us by Captain Carnwath. Larvae were subsequently secured in small numbers 

 near Kocanmah, Mekes and Likovan, and also from pools below the overhanging 

 banks of the Vardar. The imago was only once detected in tents, when a few were 

 taken near Karasouli (Oct.). Lt.-Col. Balfour found it in dug-outs near Langaza, 

 and Captain Valentine detected the species near Hortiack. 



In its early stages bifurcatus is often cannibal. I have seen full-grown larvae 

 gradually devouring those of the first and second instars, although the usual food 

 was present. The victim is seized by the tail and only the empty head-capsule is 

 rejected. One lot of about 25 larvae was kept in conditions under which maculipennis- 

 and palestinensis reached the imaginal state within three weeks. Five or six 

 bifurcatus grew normally, while in the others development was retarded. The smaller- 

 specimens one by one disappeared and when, after two months (in late November), 

 chopped flies were given instead of green weed, the survivors fed up, but only two- 

 imagines resulted — the others dying in the attempt to pupate. 



Anopheles palestinensis (Theo.). 

 Pyretophorus palestinensis, Theobald, Mon. Cul., hi, p. 71 (1903). 

 „ nursei, Theobald, Mon. Cul., iv, p. 66 (1907). 



,, cardamitisi, Newstead & Carter, Ann. Trop. Med., iv, p. 379 (1910). 



Anopheles superpictus var. macedoniensis, Cot & Hovasse, Bull. Soc. Path. Exot.,. 

 x, p. 890, fig. (Dec. 1917). 



This is the Anopheles which has been commonly diagnosed as A. superpictus, 

 Grassi. However superpictus and palestinensis may ultimately be found to compare, 

 the Macedonian specimens examined are referable to the eastern form, in which the 

 tarsi are unhanded. A. nursei and cardamitisi offer apparently no features of 

 specific value to separate them from palestinensis. This synonymy, I find, has 

 already been suggested by Edwards (Journ. As. Soc. Bengal, ix, p. 48, June 1913),, 

 who had before him the types of palestinensis and nursei and a paratype of 

 cardamitisi. For the name macedoniensis there seems to be no justification. 



This is, numerically, the chief autumnal species all over the hilly country west 

 of the Struma — a remark which holds possibly for Macedonia more generally, but 

 I am unable to speak from personal experience. At Karasouli only one example 

 was taken up to and during the second week of August. Later a pair was shown 

 me by Major Bisset from Hortiack (first week in August), where the species was- 

 reported to have appeared in some numbers. During the second half of August 

 and in September, palestinensis was abundant in the Lahana district and the 

 Struma Valley, breeding in clear hill streams, and where these were outside the 

 canalised areas one occasionally found the larvae in shoals. During September 

 a careful count was made of the mosquitos taken in tents, wards, etc., in one- 

 Casualty Clearing Station, and palestinensis there formed from 94 per cent, to 98 

 per cent, of the Anopheline total, the only other species present being macidipennis. 

 Breeding imagines from larvae locally obtained produced practically the same- 

 figures. On returning to Karasouli in October, I found this species common, and 

 during the first fortnight of that month I bred some examples from larvae taken in 



