68 [March, 



described as maxillae by some, and palpi by others. In addition to the 

 difficulties caused hereby, one is also generally faced with great poverty 

 of material to work at. Consequently nothing is known as to the varia- 

 tion of the species, but it is clear to me that size, infuscation of wings, 

 and such-like characters are not always to be relied on. 



Pierce has founded many new species on minute measurements of 

 the 2 cephalothoi'ax, and under Stylops cornii he gives measurements 

 for two individuals, ''the difference" between Avhich " is probably very 

 near to the extreme variation in size." In some British 'species I have 

 found greater variation than this. He assumes that a Stylops does not 

 attack more than one host, and accepts the name S. thwaitei Saunders 

 (undescribed) because the supposed host {A. afzeliella) was named. 

 The description of S. nassonowi Pierce is made from Nassonow's figure 

 of the $ and contains nothing by which the species could be identified. 

 Presumabl}'- we have this Stylops in England, since I have several times 

 taken the host bee (^A. carhonaria=pilipes) stylopized. I do not think 

 that the Stylops in each species of our affected Andrenae is itself a dis- 

 tinct species, but until the S 6 are secured, it is impossible to be certain 

 of this. I cannot detect any good characters to distinguish some females 

 of Stylops that have been taken from quite distinct bees. I have some 

 doubt whether S. thwaitei was really from A. afzeliella at all, and not 

 rather from A. wilkella. Smith, who was the leading authority on the 

 determination of British bees for 40 years of the last century (after 

 Kirby's time, but before 1880), made serious mistakes in his determina- 

 tions. Often at fault in naming typical specimens, his naming of those 

 altered by stylopization is still more open to suspicion. Even when bred 

 from hosts belonging to quite distinct groups of the genus Andrena — 

 I mean natural groups, not the artificial subgenera created by Robertson 

 — the species of Stylops can be at least excessively closely allied and 

 perhaps not even distinct at all. 



The synopsis of the males which is here given is put forward with 

 diffidence. I can only say that great numbers of sketches of the antennae # 

 of the species have been made in different aspects before arriving at any 

 conclusion as to the real proportion of the joints to one another, but it 

 may be repeated that to arrive at cei-tainty in this respect, each joint 

 should be detached and mounted flat in balsam. This I have not been 

 able to do at present. Drawings of the antennae made without these 

 precautions (unless living specimens of the insects are available) must not 

 be considered absolutely correct owing to the flexure and curving of the 

 joints. I regret that I am unable to include S. dalii Curtis in this table, 

 as I have seen no c5" of this species. 



