260 DIFFERENT HABITS OP MALES AKD FEMALES. 



tubes, which he arranged conveniently for her. From 

 the 1st to the 10th of May she constructed, in one tube, 

 eight cells — first seven female, and then one male. 

 From the 10th to the 17th, in a second tube, she built 

 first three female and then three male cells ; from the 

 17th to the 25th, in a third, three female and thea 

 two male ; on the 26th, in a fourth, one female ; and, 

 finally, from the 26th to the 30th, in a fifth, two female 

 and three male : altogether twenty-five, seventeen 

 female and eight male cells. 



The advantage of this is clear, but the manner in 

 which it is secured is not so obvious. It might be 

 suggested that the quantity of food was not regulated 

 by the sex of the young one, but that the sex depended 

 on the quantity of food. This would be very improb- 

 able, and M. Fabre attempted to disprove it by some 

 very ingenious experiments. He found that if he took 

 some of the food from a female cell, the bee or wasp 

 produced was still a female, though a starveling ; while 

 if he added food to a male cell, the larva still pro- 

 duced a male, though a very large and fine one. 



M. Fabre then made some of his most ingenious 

 experiments. He brought into his room a large number 

 of cocoons of Osraia. When the perfect insects were 

 about to emerge, he arranged for them a number of 

 glass tubes, of which the Osmias gladly availed them- 

 selves, and in which they proceeded to construct their 

 cells. The usual arrangement, as already mentioned, 

 is that the males are placed nearest to, and the female 

 furthest from, the door. But M. Fabre so arrano:ed 

 the tubes that each was in two parts, an outer wider 

 portion having a diameter of eight to twelve milli- 

 metres, which is sufficient for a female cell; and an 



