DEPARTMENT OF GENETICS. 117 



66 or 69 days, records unapproached by any of the control mice, whose 

 modal cycle-length was 6 days. Similar individual differences are shown 

 by the records of the mice treated after their normal cycle-length was tested. 

 Some appear to show that the treatment changed the length of the cycle, 

 and in others there was no change. This is true of both Hght and heavy doses. 

 "Although under the conditions of the experiment alcohol does not seem 

 to have a general or specific influence upon the length of the oestrous cycle, 

 these preliminary findings seem to support the proposition that alcohol 

 influences the length of the oestrous cycle in certain females. This is the 

 situation called for by the data from the rats, a selective action on the mothers 

 themselves. It is to be noted, however, that the rats were given a daily 

 intoxicating dose, and the greater part of the observations so far made upon 

 the mice have been from relatively light dosage." 



Recognizing the possibility that different strains of mice may give different 

 characteristic cycle-lengths, an investigation has been begun into the normal 

 length of the oestrus cycle in 10 of the different strains and families at hand. 

 Already, in preliminary fashion, it may be announced that the different lines 

 do, indeed, show differences in the frequency distributions of their cycle lengths. 

 Combining all distribution curves, the mode is found to lie at 5 to 7 days. 



A separate study was made of the vaginal smears of 5 yellow mice, a race 

 characterized by an enormous deposition of fat. Yellow mice are also slow 

 breeders. Whether or not the failure of the fat yellows to breed is due to the 

 cessation of oestrus is uncertain. Of the 5 mice, 4 gave fairly regular cycles, 

 averaging 10 days; 1 was examined for 80 days and showed only 3 question- 

 able oestrus periods. 



The analysis of the effects of alcohol upon reproduction is being carried 

 further by counts made upon the corpora lutea of pregnancy in living animals. 

 In the case of the rats, formerly studied, the number of young per litter was 

 reduced in the treated animals; but it was not determined whether this was 

 due to a reduction of eggs ovulated or to prenatal mortality. Since the num- 

 ber of corpora lutea formed at one ovulation closely approximates the number 

 of eggs set free, the difference between the number of corpora lutea of any 

 pregnancy and that of young born gives the amount of prenatal mortality. 

 Dr. MacDowell has developed a special technique for finding this difference. 

 During any pregnancy the number of copora lutea belonging to this preg- 

 nancy can be readily counted by exposing the ovaries to view, for at this 

 time the corpora lutea of pregnancy are very large, red, and prominent, 

 while all of the older corpora are small, of an opaque cream-color, and 

 sunken into the body of the ovary. The effects of the operation of ex- 

 posing the ovary are very slight; the young are born normally, even on the 

 following day, with no increase in natal mortality. The operation may be 

 performed repeatedly on the same female without interfering with her 

 subsequent reproductive functions. In normal pregnant mice, 49 operations 

 gave satisfactory counts of the recent corpora lutea in both ovaries; they 

 averaged 8.7 per female. The number of young born to these females after 

 operation averaged 6.4, which may be compared with 5.8, the average 

 given by 1,509 litters in the whole colony. The seven operations on alco- 

 holized mice so far performed indicate that the light dosage does not modify 

 the number of ova liberated. 



In connection with the observations on the number of corpora lutea in 

 alcoholized and unalcoholized mothers, certain studies have been made on 



