﻿STARVATION AND SEX-RATIO 



17 



desirable. Drosophila impelophila furnishes such conditions ad- 

 mirably, and the experiments to be detailed below have been per- 

 formed on this species. 



Method. The method employed was to starve, through a series 

 of generations, the parental organisms, and then obtain the sex- 

 ratio of the offspring. This was done as follows: On June 17, 

 1909, five pairs of these fruit-flies were caught on the banana stems 

 in each of five different grocery stores in Bloomington, Indiana. 

 Each of the five lots was placed in a vial one inch in diameter 

 and three inches in length, sealed at one end, the other being cov- 

 ered by a piece of cheese-cloth secured by means of a rubber band. 

 The vials were allowed to lie on one side in long narrow trays. 

 Water was given the flies by soaking a piece of blotting paper and 

 laying it in the vials. This paper was easily torn up by the larvae 

 and so made an excellent place for their pupation. A supply of 

 food, dipped in yeast solution to prevent the growth of moulds, was 

 given them in pieces of from one-fourth to one-half cubic centi- 

 meter of banana on the blotting-paper. 



It is evident that the first to hatch had a decided advantage 

 in securing plenty of food, and so pupated early, thus producing 

 large, fully-developed imagos. Then, as the food-supply became 

 low, the larvae were forced to pupate early, and were hatched as 

 undersized imagos because there was not sufficient food for their 

 complete larval growth. In this process, it is clear, also, that a 

 very great number of larvas of all ages died, and some that had 

 made sufficient growth to pupate were unable to emerge as adults. 

 Each generation, thus, was characterized by a number of weU-fed, 

 fully-developed imagos, a few under-sized starved adults, a few 

 pupae unable to emerge, and a very great number of under-fed, un- 

 successful larvae of all sizes. 



In starting a new generation great care was used in mating only 

 the starved flies. These were easily picked out from the fed flies, 

 since they were always the smallest of the hatch, due account being 

 taken of the fact that well-fed males are smaller than well-fed 

 females. Normally the males of these flies measure, from the anterior 

 part of the head to the tip of the wings, 2.94 mm., on the average, 

 and the females, 3.35 mm. ; but from the extreme starvation to 

 which these flies were subjected, the males were often as short as 

 2.5 mm., while the females were occasionally as short as 2.3 mm. 

 Since the five strains were unrelated, it was possible to avoid ex- 

 cessive inbreeding by crossing, at times, one strain with another. 



