84 Proceedings of the Royal Physical Society. 



two external and terminal membranous processes whose proximal 

 edges are chitinised. 



The shape of the 9th tergite is seen in Fig. 4. The apex of the 

 " process " is somewhat hooked. The " finger " is roughly tri- 

 angular in shape, with the dorsal edge convex and the sides 

 concave. The posterior (distal) angle has on the inside two 

 conspicuous short, black peg-like spines with below, after a 

 brief interval, a stout curved bristle. One or two other bristles 

 or short hairs are placed as indicated in the larger sketch (Fig. 5). 

 On the 9th sternite ventrally there is a dense tuft of short hairs 

 and spines. 



The inner chitinised portion of the penis shows a characteristic spatulate 

 outline. 



$ — The 8th tergite has about six short hairs above the spiracle. 

 There are two strong bristles below the pygidium with a few 

 weaker ones in front. The lower distal third bears numerous 

 bristles (long and short) externally (about 20). There are fewer 

 bristles on the corresponding inner surface (see Fig. 6). 



The 8th sternite is narrow and inconspicuous. 



The 10th sternite bears, besides the usual long almost terminal 

 bristles, 4 broad short spines which probably lock in copulation 

 with the bunch of spines and bristles on the ventral surface of 

 the 9th st. $ . 



The narrow, region and the bulb of the receptamlum seminis are of 

 almost equal length. 



Legs. — The hind femur bears outside 4-6 bristles in a row and inside 8-10. 



Length. — $ , 2-27 mm. $ , 2'5-27 mm. 



In drawing up this description (as regards the $ ) I have had com- 

 municated to me some critical remarks by Dr K. Jordan, for which my 

 best thanks are due. 



A small series, 1 $ $ and 2 $ $ , from nests 1 of G. urbica, taken from 

 the cliffs at Todhead, near Kinneff, Kincardineshire, by Mr A. M'Innes, 

 in the middle of September 1909. The types {$ and $) have been 



1 Three nests were sent by Mr M'Innes (on 5th Oct.), who had taken them down 

 some six weeks previously. As all three seemed to be swarming with the same flea 

 (farreni), one was burned after a cursory examination. Shortly afterwards, with a view 

 to determining tlie proportions of the sexes, I began a thorough examination of the 

 remaining nests and, after going through some 300 fleas, came on the first 6 of rothschildi. 

 It may be of some interest to give in detail the results of the completed census (see 

 p. 85). 



