6 



Indiana University Studies 



the later generations they were all of this type. In most cases 

 single pair matings were made, but a few times when the 

 vigor of the strain seemed low, two pairs were placed in the 

 same bottle, or sometimes two or three males were placed with 

 a single female. It was the male which showed lack of vigor. 

 The females have been vigorous thruout the experiment (38 

 generations). I have asked a number of persons whether 

 there were any objections to this method, but so far no one 

 has suggested any. It was suggested, however, that if the 

 strain were heterozygous for a particular character, it might 

 take a little longer to produce a homozygous strain. This is, 

 of course, perfectly obvious. On the other hand, it might pro- 

 duce a homozygous strain more quickly. Certainly, at any 

 rate, there can be no serious objection to the method, especially 

 if the experiment is carried on for a sufficient length of time. 

 This method, I believe, has been followed by Castle in his 

 experiments on hooded rats and has been criticized by Mac- 

 Dowell in his bold denunciation of Castle's interpretation. 

 Castle's starting-point, however, was several individuals 

 instead of a single pair. 



The flies were bred in eight-ounce bottles. In all cases 

 fresh banana was used as food. The bananas were bought 

 when fairly green, before the skin was broken, and allowed 

 to ripen in glass containers protected from flies. When they 

 began to turn black and get soft, a small amount of alcohol 

 v/as poured into the jars to prevent growth of molds. The 

 bananas in this ripe condition were peeled and cut into pieces 

 about three-fourths to an inch in length. One piece was put 

 into each bottle along with such absorbent paper as was 

 needed to take up the extra moisture. If this food became 

 bad because of mold, bacteria, or too much acid, before eggs 

 were laid, the parents were transferred to new bottles. In 

 case the food became too dry or the supply insufficient for 

 the developing larv^, fresh food was added. Extreme care 

 was used to prevent contamination. 



Nothing has been done in the way of a biometrical treat- 

 ment of the data. This has not been done because of any 

 belief that such treatment is valueless in all cases, but because 

 in this particular case the facts stand out clearly without such 

 treatment. If anyone believes he could draw different con- 

 clusions by such a biometrical study, I am willing to turn the 

 data over to him and to assist him in the attempt. 



