402 PROCEEDINGS OF THE AMERICAN ACADEMY. 



even in the extreme cases, are so nearly uniform that we may conclude 

 with reasonable assurance that there is no intimate relation between 

 sex and the size of the litter. 



The thousand litters that we worked upon, then, were composed of 

 from 1 to 15 pigs each, most frequently of 6, and the sexes were dis- 

 tributed almost indiscriminately among them in the ratio of about 

 1026 males to 1000 females. 



The Number and Arrangement of the Nipples. 



Inirodudory. The nipples on pigs form two series situated on either 

 side of the ventral median line and extending from the pectoral to the 

 inguinal regions. In some pigs the nipples in these two series are so 

 exactly paired, right and left, as to give the appearance of extreme 

 segmental regularity. In others they are grouped on the two sides 

 with so much diversity in number and arrangement as to seem to be 

 quite irregular. These relations have been discussed by Bateson 

 (1894, p. 190) and especially by Wentworth (1913), who has developed 

 a somewhat extended classification to include the chief variations. 

 This aspect of the subject is not essential to our problem and we have, 

 therefore, not attempted to deal with it. In collecting our data, we 

 have incidentally recorded the arrangement of the niiDples as either 

 regular or irregular. By a regular arrangement we mean one in which 

 the nipples are grouped in pairs right and left and we have used this 

 term to include cases were certain members of assumed pairs were 

 absent. By an irregular arrangement we mean one in which the 

 members of the right and left series are so disposed as to show no 

 obvious pairing. Most examples fall naturally under one or other of 

 these heads, but the process of classifying is at times somewhat arti- 

 ficial. These two classes do not agree with the two chief classes 

 established by Wentworth (1913) on the basis of what he calls the 

 symmetrical and the asymmetrical patterns. These patterns have 

 not been clearly defined, but apparently they refer to the e\en or odd 

 numbers of nipples possessed by the animal in the sense that when one 

 side exliibits the same number of nipples as the other, the arrangement 

 is called symmetrical; when otherwise, it is designated asymmetrical. 

 The asymmetrical pattern may be due to a "suppressed nipple" or to 

 the presence of a "triangular pattern" and the latter may exliibit 

 several subordinate ^ ariations. Although some such classification as 

 that proposed by Wentworth is necessary for a treatment of the 

 subject from the standpoint of inheritance, our own observations have 



