vil mu RACTING QUEENS 129 
laid down: three of these were occupied by apzdarius 
and one by ¢ervestrzs. All these proved successful, 
except one in a terra-cotta domicile, and, without 
any aid from me, developed into colonies, although 
the colonies in the tin and terra-cotta domiciles did 
not flourish so well as those under the wooden covers. 
As already stated, no damage was done by ants. 
Neither did mice or shrews destroy any nests. To 
keep these out, as soon as I saw a queen going in 
and out of a nest, I pressed into the ground around 
the mouth of the tunnel a tin cylinder about 5 inches 
high and 5 inches in diameter, and open at both ends. 
These cylinders, I believe, afford greater protection, 
with less risk, than the mouse-excluders used in 
1910; also, they probably give a certain amount of 
protection against ants. The queens soon learnt to 
fly in and out of them. 
The fine dead grass used as nest material in 1911 
was of two kinds, both of them superior to and more 
easily obtained than that of rg10. The better kind 
was scratched up by fowls which were confined to 
a run which was moved once every day about a 
pasture of fine grass in February, before the new 
season’s growth had begun: this material had the 
great advantage of being curly, and it required 
very little picking over and snipping up to prepare 
it for use. The other kind was raked out of tufts 
of fine grass in March and April. 
Reviewing all my attempts at making domiciles 
for queens to occupy, I consider that much the most 
promising were those under the wooden covers, and 
K 
