SEX DETERMINATION IN CATTLE. 2OQ 



occurred in the State at the time of the inquiry. The statistics 

 do not represent one or even a few large herds, but a large number 

 of small herds scattered all over the State. 



2. The original records were made by person- who knew 

 nothing whatever of what use was to be made of them, or any- 

 thing about what they "ought" to show in order to please or 

 satisfy the person who collected them. 



3. The numbers are sufficiently large to reduce the probable 

 error of the sex ratios to reasonably small magnitudes. Of 

 course, still more figures would be desirable, but the data as 

 they stand are by far the most extensive yet collected on the 

 point at issue. In the present state of knowledge regarding 

 sex-determination it is desirable to submit to careful analysis 

 any collection of statistics sufficiently extensive to be worthy 

 the name. 



Of course, any one considering the sex-determination problem 

 from the statistical standpoint should be clear as to the nature 

 of statistical evidence in general. Nothing can ever be finally 

 "proved" by statistical evidence alone. All that any amount of 

 statistical data can do is to state a particular set of facts, which 

 are a sample, strictly localized in time and space, and restricted 

 in magnitude, out of the whole universe of facts of like kind. 

 If the counting has been correctly done the results from the 

 sample are absolutely and unshakably true, so far as concerns 

 the sample itself. Whether or not the sample fairly represents 

 the relations in the whole universe to which it belongs, can never 

 be absolutely determined, no matter how large the sample may 

 be. All that can be ascertained is a more or less close approxi- 

 mation to the mathematical probability that the sample gives 

 a just representation of the population or universe from which 

 it came. 



The most extensive statistics which have been published on 

 the gross (i. e., unanalyzed) sex ratio in cattle are those of 

 Wilckens. 1 They embrace 4,900 calves and the sex distribution 

 is 2,536 cf : 2,364 9 , giving a sex ratio of 107.3 cf : 100 9 . 



1 Wilckens, M., " Untersuchung iiber das Geschlechtsverh;iltniss und die I'r- 

 sachen der Geschlechtsbildung bei Hausthieren," Laiuhcirtsch. Jahrb., Bd. 15, 

 pp. 611-654, 1886. 



