armatwrs in the Hymenoiiteroits genus CoUetes. 27 



remark in the Revision, and one only, on the couverch 

 genital: viz. that its "form appears to be stable." That, 

 if he means Avhat I suppose him to mean by it, is doubtless 

 true : but it certainly cannot be said to exhaust the subject, 

 nor to indicate that the author had studied it carefully 

 throughout the genus he was revising. 



Perhaps the best thing in Radoszkowski's " Revision " is 

 his recognition of a distinct group of species in which the 

 stipites of the armature are simple, i. c. not divided by a 

 deep sulcation into two apparently distinct portions — an 

 apical and a basal. This is an easy character to see, and 

 I think it is an important one, though I do not believe 

 that all the other species — those with " divided stipites " — 

 should also be regarded as forming a single group. The 

 species with simple stipites are certainly, to some extent, 

 united by other characters also ; and I believe that they 

 are confined to the warmer parts of the palsearctic region. 

 None of them occur in Great Britain, nor, apparently, in 

 Scandinavia. On the other hand, species with divided 

 stipites are found throughout the whole region, and among 

 them are some which seem to have hardly anything else in 

 common (e. g. cunictilarms and fodiens). Still, if it be 

 necessary to classify Colletes-armatures dichotomically, this 

 is perhaps the best " fundamentum divisionis " with which 

 to start. And, as far as I know, Radoszkowski was the 

 first to point it out. 



But to return to the subject of the above-mentioned 

 " seventh ventral plate " ; it was, I think, particularly 

 unfortunate that Radoszkowski, when dealing with GoUetes, 

 should have dismissed it with such slight and inaccurate 

 treatment, instead of figuring it — or at least describing its 

 form — in each of the species with which he dealt. For, in 

 fact, this segment is generally at least as characteristic for 

 specific purposes as the *' armure " itself, and its characters 

 are (for reasons presently to be given) easier both to 

 recognize, to figure, and to describe. One chief object of 

 this paper is to call attention to the very distinctive forms 

 assumed by this segment in different species, and I figure 

 it accordingly, together Avith the armature, in such 

 palsearctic species as I possess — thirty-five in all — and 

 likewise in an American species, which I cannot name, but 

 which differs evidently from any of them, communicated 

 to me from Massachusetts by Prof. C. T. Bruce of New- 

 York. (See PI. IX, 57, 58.) 



