328 O. E. PLATH. 



described briefly as follows: Each nest-box was provided with a 

 double cover, the lower being made of glass and the upper of 

 wood or tar-paper. At one end of the box, a round hole, \ inch in 

 diameter, which could be closed by means of a cork, served as a 

 flight-hole. A small piece of honeybee foundation (wax), about 

 an inch square, was then firmly pressed to the bottom of the nest- 

 box near the opposite end. Around this piece of wax, a circular 

 layer of cotton was placed as nesting-material. A tin can, about 

 3 inches in diameter and 2 inches high, from which the cover was 

 first removed, was then put upside down over the honeybee 

 foundation and cotton, after a hole, through which the queen 

 could readily pass, had been made in the rim. Every two or 

 three days, a fresh supply of pollen, obtained from two colonies 

 of honeybees which had been especially secured for this purpose, 

 was provided on the layer of wax within the cotton ring. Liquid 

 food, consisting of about half water and half honey, was supplied 

 daily in a porcelain dish, about \ inch high, outside of the tin 

 can. In order to keep the nest-box as sanitary as possible, a 

 small pile of dry sand was put in one of the corners of the box. 



Not being acquainted with the methods employed by Lind- 

 hard ('12), at the time my experiments were carried out, I at 

 first placed two queens (of the same species) in each nest-box, 

 as was done by Sladen ('12) and Prison ('18). However, like 

 Sladen (p. 131), I found that one of the queens invariably killed 

 the other, 1 sometimes within a few minutes after they had been 

 placed together, and that the victorious queen was frequently 

 made useless for further breeding experiments by the loss of one 

 or more antennae, or legs. 2 I therefore only placed one queen in 

 each nest-box in subsequent experiments, and furnished each one 

 with from one to three workers, preferably of her own species. 

 Whenever a worker died a rather frequent occurrence, especially 

 as long as there was no brood another was substituted as soon as 

 possible. Those bumblebee "nuclei" which belonged to species 

 easily obtainable in or near Boston, were permitted to forage out 



1 As already stated, Mr. T. H. Prison ('18) found that two queens of Bremus 

 auricomus behaved differently in this respect, but, judging from Mr. Prison's (p. 45) 

 account, it seems probable that one of the queens was in poor health. 



- These observations were made on about twenty queens of Bremus bimaculatus, 

 fervidus, impatiens, and vagans. 



