170 [August, 



Veronica, BelJis, and other small flowers, as well as TJmhelliferas, and 

 many German examples were found on dandelions. This species can- 

 not be an intermediate brood of parvula-minutnla, for I received 

 specimens taken in Baden on March 31st, when parimla ^ was still 

 in fresh condition, and this year in Devonshire it appeared at a time 

 when the $ $ of parvida were still in perfect condition. I suspect 

 that only in exceptional seasons or in special localities, a second brood 

 is produced, but I took a perfectly fresh ^^ in August, 1886, at 

 Sidmouth, which differs in no way from the Spring foi-m. About the 

 burrows of the isolated colony above mentioned, Nomada Jk(vo(jvtiata 

 was taken, and this bee also affects A. nana, but I doubt whether it 

 attacks A. parvvia, since it never appears xmtil the latter has been 

 long on the wing. All the English examples I have seen have the 

 anal fringe of the "i dark-, but in very early (Mar(;h) specimens from 

 Germany, it is more yellow. It is possible that these are the descen- 

 dants of a second generation, whereas in England, in most localities, 

 only one brood (May — July) occurs. 



A. 7iana> is always readily distinguished from either of the pre- 

 ceding by its superficial appeai-ance in the ^, owing to the rather 

 different pubescence, and especially the pure white hairs of the clypeus. 

 The abdominal puncturation varies in amount and intensity, but is 

 always very evident. The ? not only differs from them in the 

 pubescence, but, in addition, the smooth, shining apical impressions 

 of the abdominal segments and the yellow middle portion of the 

 stigma are evident charactei-s. It can always be told at a glance from 

 the other species. It appears at the end of April or in the first half 

 of May, according to the weather, and is partial to daisies and Veronica 

 chamxdrys, and later the $ $ abound on white TJmhelliferie.. I have 

 seen no examples of a second brood in England. It is much attacked 

 by Stylops and Nomada flavognttafa. 



A. spreta, Per., by a lapsus calami recorded by me (antea, p. Ill) 

 as A. schenkella, Per., is the A. niveata, Saund., nee Friese, and like nana 

 is at once distinct from any of our other species by the combined 

 characters of its pubescence and densely sculptured apices of the 

 abdominal segments. It is less common than any of the preceding 

 species and moi-e local, though widely distributed. It is sometimes 

 attacked by 8ttjlo2)s. It is foiuid on the same flowers as nana, and 

 often in company with it. I have found the J* in the first half of 

 May, but it is more usually taken in June. 



A. sj>. (3. This, I think, cannot l)e considered as a second brood 



