n.iR] 187 



only two cells. The hamules of the 2ikI abdominal segment agree 

 with those of C. erytliraea. These specimens seem to belong to the 

 same form as the somewhat smaller ones recorded by Ris from Kashgar, 

 the measm-ements given by him being: d , hind wing, 24 mm.; abdomen, 

 21 mm. ; 2 , hind wing, 25 mm. ; abdomen, 21 mm. ; anq. 8^ ; and in 

 a majority of examples for a short distance only two rows of discoidal 

 cells, in some already only two cells at the triangle. 



I have before me a <S from Quetta (hind wing 27, abdomen 23 mm.), 

 and % 6 6 (hind wing 26-27, abdomen 22-24 mm.), and 1 $ (hind 

 ■p'ing 26i, abdomen 22 mm.) from Kerki, Bokhara, all of which are 

 referable to the same form, all agree in having 8| anq. and all excepting 

 the $ have part of the discoidal field with only two rows of cells. The 

 hamules appear also to be those of C erijtliraea, differing from the less 

 pointed hamules of those examples referred to below as G. servilia, 

 De Selys agreed that this point of distinction between C. erytliraea and 

 C. servilia was right, but it is to be noted that his statement was made 

 in connection with the large Japanese form in whieli I am inclined to 

 think the hamules are sometimes suggestive of those of G. erytliraea, 

 but the difference in this respect is very slight. In the younger phases 

 of this striking Japanese form the shouldei'-stripes, judging from a few 

 specimens before me, are hardly noticeable, and the wings are entirely 

 suffused with a yellow tint. 



13. — Grocotkemis servilia Druiy. 



A long series of both sexes, October 29th-November 12th, 1917 ; 

 April 5th-June 29th; August 17th-September 11th, 1918, all Amara ; 

 1 ?, Jebel Hamrin, N.E. of Baghdad, November 20th, 1918; 1 $ 

 December 20th, 1 J December 31st, 1918, Kizil Robat nearer the 

 Persian frontier {HJvans) ; 2 $ $ , Qurnah, Kiver Tigris, May 17th,. 

 1918 (Buxton) ; Basrah, October 10th-14th, 1918, 2 d d ; October 

 22nd, 1 2 ; Long Island, R. Tigris, November 1st, 1918, 2 $ $ 

 (^Barraud). Received later : 1 $, Basrah, October 10th, 2 $ $, Long 

 Island, November 1st, 1918 {Barraud). 



Captain Evans states that this dragon-fly was fairly common when 

 he arrived at Amara in the end of October 1917, and it increased in 

 abundance till the cold weather in December, and the rains which com- 

 menced in that year on the 12th of that month put a stop to his 

 rambles. It was found chiefly about gardens, but might be seen any- 

 where, a favourite situation being among long grass (two to four feet) 

 beside irrigation ditches just on the outskirts of the palm groves. 

 He writes : •* On one occasion in the evening, I passed through about 



