armatures in the Hymenopterous genus CoUctes. 45 



smooth space between the basal rugosities and the apex. 

 The apical part of the " petioles " in the seventh ventral 

 plate is very densely ciliated with pale whitish hairs, 

 but not in the style of haltcatus, where the hairs are looser, 

 darker (I think), and commencing much nearer to the 

 base of the segment. 



If a distinct difference in the length of the gense is not 

 to be regarded as a positive specific character in Colletes, 

 it is hard to see on what grounds we are to establish 

 species among them at all : if it is, then certainly cecrops 

 and halteatns differ. Thus we come back to the question, 

 " What is a species ? " to which, for my part, I know no 

 thoroughly satisfactory answer. In the meantime, I hold 

 cecrojjs to be distinct. 



I have only one specimen ($), taken by myself in Attica 

 (April 30, 1901). 



4. Perezi, n. sp. PI. VI, 4, 4«. 



$ Bulteato affinis, sed maior (10-13 mill, long.), abdoinine longe 

 subtilius aiqualiusque punctate (propterea magis opaco), fasciis ab- 

 dominalibus latioribus, flavidis subsquamosis. Segment! ventralis 

 sejitimi petioli nudi, basibus dilatatis. $ (long, circa 15 mill.) a 

 balteato $ notis iisdem, quibus mas, differre videtur. 

 Aegyptus, prope Cairo, iv et v, 1896. 



This is a species which, though it has almost all the 

 other characters of haltcatus, distinguishes itself at a 

 glance by its exceedingly different puncturation. This is 

 intensely fine and close throughout, making the discs of 

 the abdominal segments look remarkably opaque. The 

 fasciaa also are seen at a glance to be very much wider. 

 These in fresh specimens are decidedly fiavescent, and 

 formed of more than one distinct row of very thick 

 (almost scale-like) branched hairs. It is rather larger than 

 haltcatus, and has thirteen alar hooks. I know it only from 

 Egypt, where I have taken both sexes near Cairo. Possibly 

 it may be identical with some of the species described by 

 Spinola (e. g. intricans, not however if Radoszkowski 

 rightly identified that species !), but I cannot fit it satis- 

 factorily to any of them. I have named it therefore 

 after Professor Perez, in acknowledgment of much kind 

 assistance received from him during this inquiry. 



5. Diniidiattis, BruUe. Plate VI, 5, 5a, 5&. 



The largest species of the group, and very distinct by its 



