va AT RACTING QUEENS 107 
I also adopted the plan of examining all the 
domiciles every tenth day or so, replacing the nest 
material if wet, which was often the case, with dry 
material. The sods were lifted with an ordinary 
digging fork. When making these examinations I 
carried in my pocket a note-book, in which was 
recorded anything of interest that was noticed about 
the nests. 
Nine out of the forty domiciles were tenanted by 
queens—three /apzdarius, a terrestris, anda hortorum 
having been attracted in May, and two /apzdarzus, a 
ruderatus, and an unknown queen early in June; but 
only three of the queens, namely, two /aprdarius 
and the vaderatus, remained until their larve span 
their cocoons, and only one, a /afzdarius, succeeded 
in hatching out her first workers and in establishing 
acolony. I formed the conclusion that the majority 
of the queens, after having occupied the nests and 
started to breed in them, found them so unsuited to 
their requirements that they abandoned them. The 
dampness of the nests or the vermin, or both, had 
driven them away. 
During the next three years no further domiciles 
were prepared, but in 1909 two flourishing colonies 
with workers, one of ¢errestris and the other of 
ruderatus were found to be occupying two of the 
artificial domiciles made in 1906. Of course, by 
this time the ground had become perfectly hard, 
and’ the nests resembled natural nests. The nest 
material, though rotten and scanty, proved to be, on 
examination, not that which I had placed in the 
