1890.1 207 



least on its apical portion) apical dorsal valve in the ? of the former, 

 which is distinctly punctured, in Trimmercma, and. has its margins 

 somewhat reflexed. 



First as to the females : not only is the sculpture of the apical 

 dorsal valve of rosee decidedly variable, but even typical Trimmerana 

 is by no means constant in this respect. I have specimens of the 

 latter in which the apical portion of the 6th dorsal segment is hardly, 

 if at all, punctured, others in which the puncturation is very distinct. 

 This common species has been much scarcer than iisual this year, but 

 the few females I have caught are all unusually strongly punctured. 



My specimens of rosce are not highly coloured varieties ; in most 

 cases the red colour is confined to the ajjices of the two basal segments 

 of the abdomen, in some it is absent entirely, and, but for their paler 

 pubescence, the resemblance of these to typical Trimmerana is extreme. 

 One or two specimens are exactly like it, except in one particular, viz., 

 the colour of the scopa beneath. In all my specimens of roscs this is 

 golden, whereas in typical Trimmerana the same part is silvery, but 

 even here there is no constancy, for I have Trimmerana caught in 

 early spring in which the scopa beneath is clothed with the golden 

 hairs, which appear to be normal in the other. As to the sculpture 

 of the apical dorsal valve of roscB, in some specimens it is entirely 

 impunctate, in others it is so on its apical portion ; others again have 

 it distinctly punctured even on this part. 



There seems to be no correlation between the height of coloration 

 and the smoothness of the valve, for in some specimens which have no 

 red coloration at all it is impunctate, while the most highly coloured 

 specimen I possess has it distinctly punctured all over as in Trim- 

 merana ! The reflexion of the edges of the apical dorsal valve in 

 both forms is no more to be trusted than the puncturation, as it is 

 similarly variable. 



With regard to the males, I have no ^ rosce with an entire apex 

 to the 8th ventral segment. In all it is emarginate, more or less, as 

 in (^ Trimmerana. In one specimen, however, the emargination is very 

 shallow, and a comparison of others shows this character to be variable, 

 as in the other species of Andrena, e. g.,A. Givynana, var. bicolor, Fab., 

 and A. angustior, Kirb. In the former of these the 8th ventral seg- 

 ment may be truncate or emarginate at the apex, in the latter rounded, 

 truncate or emarginate ! 



Finally, as to the structure of the cheeks, my specimens of both 

 Trimmerana and rosce are eminently variable. Mr. Saunders has long 

 since sunk A. spinigera, Sm., as a variety of Trimmerana, and I may 



