DEPARTMENT OF EXPERIMENTAL EVOLUTION. 131 



lead, of course, to more far-reaching results ])y making breeding experiments 

 possible. I shall continue to regard this problem as the most important one 

 upon which to work during the coming year." 



EXPERIMENTAL MODIFICATION OF THE GERM-PLASM. 



This, which may be said to be the loftiest aim of the experimental 

 evolutionist, has been so rarely achieved that all reported successes in 

 this direction are received with critical skepticism. We are now pro- 

 vided with means for carrying out experiments on this difficult topic. 

 In the Year Book for 1914 (p. 121) is mentioned the work, begun by 

 Dr. G. C. Bassett, of inducing changes in the germ-plasm of rats by 

 subjecting them to the action of alcohol vapor. This work is now in 

 the charge of Dr. MacDowell, who reports as follows : 



"Besides having a theoretical scientific interest, this problem has immediate 

 sociological importance. It is frequently asserted that, in man, alcoholism in 

 the parents results in their children having less than normal mentaUty. In 

 order to prove this, much has been said and done. There are countless com- 

 plexities which largely veil a clear solution of such a problem when man is 

 considered directl3^ Since the problem depends largely on a pyschological 

 analysis, it may be that, by employing the psychologists' indirect method of 

 studying lower animals first, the essential relationships may be more accu- 

 rately secured. If one could understand the mental methods of a lower or 

 more primitive organism, those of the higher type might be mastered more 

 easily. For other reasons than this is the use of a more simple form advan- 

 tageous : the germ-plasm to be treated, as well as that of the controls, can be 

 more easily selected in the first instance and bred in subsequent generations, 

 the individuals to be compared can be maintained under like conditions, and 

 the experimental treatment of all animals in the same group can be uniform. 



"Since the white rat is a very satisfactory subject for behavior studies, this 

 animal has been used. Since the beginning of this year the parental generation 

 has been raised. Half of the rats were made to inhale alcohol fumes, for 90 

 minutes a day, following Stockard's method; the rest were saved as controls. 

 The alcoholization was made daily for at least 100 days before mating, and 

 then continued, in the case of the females, till immediately preceding the birth 

 of her young. For every mating of alcoholics there was a mating of normal 

 males and females from the same litters. The offspring from these two sets of 

 rats have all been trained to enter a puzzle-box of a type planned by Watson 

 and modified by Bassett. The comparison between the rats of normal and of 

 alcoholic percentage has been made on the basis of the time required to gain 

 access to the puzzle-box. The sooner a rat learns the trick, the higher his 

 ability is rated. The training of each rat lasts 96 days, during which time 

 the number of seconds required to make each of 225 successful trials is recorded. 

 The niunber of rats included in this training is 145. Since the training of all 

 these rats is not quite completed, no final statement of the results of this work 

 can be made. 



"Whereas the above method gives a fairly accurate comparison of the abili- 

 ties to form a habit, it can not be claimed to measure the general intelligence of 

 the rats. In order to secure data of a higher mental activity than that sup- 

 posedly involved in solving the puzzle-box problem, a method has been 

 adopted that was planned by Yerkes and called by him the multiple-choice 

 method. As adapted for rats, the apparatus consists of a row of nine com- 

 partments with front and back doors. The rats are trained by punitive con- 



