April 24, 1914] 



SCIENCE 



611 



to appear to the general public as having the 

 weight of authority behind them, there is no 

 telling where we shall be before the inevitable 

 reaction sets in. 



Personally I believe that every student who 

 takes a four year course in high school should 

 be required to study the English language 

 during those four years, and at the same time 

 he should be getting some definite knowledge 

 of either mathematics or the classical lan- 

 guages, preferably of both. Distinct voca- 

 tional training might well be left in the back- 

 ground until the student has had an oppor- 

 tunity to get some real mental training. I 

 know this is the opinion of the great majority 

 of my colleagues at the University of Cincin- 

 nati, including Dean Schneider, of the college 

 of engineering, who is a recognized expert on 

 vocational training. And I do not doubt that 

 it is the opinion of the great majority of col- 

 lege teachers throughout the country. 



Charles N. Moore 

 University or Cincinnati 



SEX IN MULTIPLE BIRTHS 



In a copy of lectures delivered by Dr. Ray- 

 mond Pearl at the 1912 Graduate School of 

 Agriculture at Lansing, Mich., I find the fol- 

 lowing tables given, indicating that in mul- 

 tiple mammalian births as the numbers per 

 birth increase, the ratio of males to females 

 decreases. 



Man 



No. Young per Birth Males per 1,000 Females 



1 1,057 



2 1,043 



3 548 

 Sheep 



3 Males ; 2 Females 



3 Males 1 Female 1 Male 3 Females 



16 39 22 38 



In the sheep there are 215 females to 130 

 males. 



It is worthy of note that these data are 

 from normally uniparous species. In swine 

 where the number at a birth may vary from 

 •one to twenty- three (in an exceptional in- 

 stance) this excess of females is not apparent. 

 In 174 litters the number of males per litter 



and the expectation based on chance, using the 

 relative frequency of the different-sized litters 

 (fourteen per litter being the largest) was as 

 follows : 



No. Males 



per litter 1 2 3 4 5 



Expectation . . 3.4 12.8 24.6 33.4 34.7 28.5 



Actual 2 13 26 28 31 28 



No. Males 



per litter 6 7 8 9 10 11 



Expectation ..19.0 10.4 4.6 1.7 0.48 0.11 



Actual 21 12 8 2 2 1 



This shows but a slight departure from 

 expectation and is within the limits of error 

 for such small numbers. It seems doubtful if 

 there is a tendency toward increased propor- 

 tions of females in multiparous births. In 

 fact the excess is slightly on the male side 

 here. 



In 126 births from various private collie, 

 fox terrier, Scottish terrier and Boston bull 

 terrier records, the following results appear: 



No. male pups ... 1 23 4 56 



Expectation 15.1 35.75 37 24.5 10.86 2.8 .3 



Actual 14 36 39 22 11 4 



These statistics give qualitatively the same 

 result. That this accordance with expecta- 

 tion on the basis of chance is not necessarily 

 a property of normal multiple births, is shown 

 by the following statistics on sheep triplets 

 from the Iowa State College flock and two 

 farmers' flocks located near there. The total 

 number of lambings is 146. 



This gives 226 males and 212 females. The 

 smallness of these numbers does not conclu- 

 sively indicate that influences other than mere 

 chance do not operate, but they are interesting 

 since they give opposing evidence on the point 

 discussed by Dr. Pearl. 



Edward K. Wentworth 

 Chicago Veterinary College, 

 July 26, 1913 



