220 E. C. MACDOWELL AND E. M. VICARI 



g. Criteria employed 



The evidence about to be presented is derived from three 

 sources: the time for each trial, the distance covered on each 

 trial, the number of departures from the true path, or errors. 

 The time and distance data make it possible to calculate the 

 speed or the distance covered per second; from the error data 

 the occurrence of perfect or errorless trials is obtained. The 

 number of perfect trials and the number of days before the first 

 perfect trial can be used as criteria, since all the rats received 

 the same number of trials. Under these main headings the test 

 and control rats have been compared in various groupings of 

 the strains and sexes, for various periods of the training. 



The question of the differences between the behavior of the 

 males and females presents a difficulty. Although we have made 

 extended studies of this question on the animals so far sum- 

 marized, we offer no conclusion. Since we have a much larger 

 mass of data still awaiting analysis, namely, the records of the 

 rats that were actually treated with alcohol and their immediate 

 offspring, we withhold judgment for the present and give sum- 

 maries with the males and females separately and together. 



RESULTS 



1. Morphological characters 



Although this paper is on behavior, it may be of interest to 

 note here that we have found, as the result of alcoholizing the 

 rats, no signs of such morphological abnormalities as Stockard 

 and Papanicolaou ('16 and '18) found in the descendants of 

 alcoholized guinea-pigs. The few abnormalities that have 

 appeared have been limited to one strain; they have appeared 

 both in control and test rats. Very obviously, these abnor- 

 malities were carried in the strain and were not induced by 

 alcohol. The immediate effect of the alcohol upon the treated 

 animals was to render the hair exceptionally fine and attractive 

 by keeping off all skin parasites. 



