300 
BRITISH BEES. 
insects which I have captured in the course of my ento¬ 
mological career. The fourth and the ninth species, the 
0. bicolor and 0 . aurulenta, have very much the same 
habits, both usually burrowing in sandbanks, sometimes 
however in wood, in which case the perforation, contrary 
to the mode of wood-drilling bees, is made upwards, a 
sagacity or instinct which saves it much trouble, for the 
particles as they are removed by the mandibles are passed 
beneath the insect, and their own gravity carries them 
downwards, and thus the insect saves itself the labour of 
conveying them out as they accumulate in inconvenient 
quantities. The cells in this case are placed end to end. 
When they burrow in the earth, the latter species often 
associate gregariously in large numbers, and if they select 
a cavity, instead of tunnelling it themselves, and it be 
too large to take one cell upon the others, they form them 
side by side, and thus fill the space. This is the case when 
they adopt snail-shells as the receptacle for their incu¬ 
nabula, and this is done by both these species, and the 
shells they select are the empty ones of Helix nemoralis, 
hortensis, and adspersa. The capacity of the latter shell 
being much greater than that of the others, and too wide 
for a single succession, she fills the interval by placing 
them side by side, and with the increase of the whorl 
of the shell towards its orifice she places them across the 
space, and thus completes her task. In the former 
shells, the cavity at first admits of the succession of but 
one upon the other, but with its enlargement she places 
them side by side, and this repeated fills the hollow. Its 
aperture is then closed with earth and pebbles or sticks 
agglutinated together, as described at the commence¬ 
ment. The 0 . fulviventris burrows in wood, and upon 
this species the Stelis phceopjtera is parasitical; and that 
