1928] 
The Bee that Works in Stone 
77 
149 members, over twice as many as were known in 1896. By 
1922 the list had been further enlarged by at least ten more 
species. Up to this time, the nests of but two species had been 
observed: In 1910 Grsenicher had reported that P. maculipennis 
made its nest in sandy soil in Wisconsin. And in 1920 J. Be- 
quaert (’20) had stated that P. octomaculaia was seen nesting in 
sandy soil at Forest Hills. 
By 1896 only two species were found east of the Mississippi 
These were P. octomaculaia Say, a northern type, and P. obscurata 
Cr., a southern species occurring in Georgia and Florida. How- 
ever, by 1914, twelve eastern species were reported. Six of these 
were from Wisconsin. 
According to Professor Cockerell and others, it may be laid 
down as a general rule that each species of Perdita visits, nor- 
mally, but one species of flower. More than half of these flowers 
are Compositce. These bees do not usually frequent the northern 
type of flowers but rather those that extend northward from the 
neotropical region. This, taken with the normal distribution of 
the genus, strongly suggests that in the main we have to do with 
a southern series of types which have spread northward and 
become largely differentiated since the glacial epoch. But we 
must look upon P. octomaculaia as a survivor from pre-glacial 
times. 
The great majority of the members of this genus fly in the 
late summer and autumn. Few have been seen both in spring 
and late summer but Yiereck (CklUll) has collected P. phacelice in 
May and P. pectidis in April in New Mexico, these being pre- 
viously known as September bees. And so in these cases there 
are apparently two broods each year or else the September 
brood hibernates through the winter. At White Rocks, but one 
brood of P. opuntice is found each year. Even by December 
I have been unable to find any cocoons spun by the larvae. The 
cocoons shown in Fig. 1 are probably those of parasites. 
Perdita opuntice is not the only bee that confines its visits 
to the cactus flowers. Professor Cockerell has shown that the 
genus Lithurgus and certain members of Heriades and Ashmead- 
iella are confined to their flowers for their pollen. Professor 
