of the Sexes among the Todas. 487 



Summary. 



(1) Among the Todas the Js preponderate greatly over the $ s. 



(2) This preponderance is doubtless due to the practice of $ 

 iufanticide, which is probably still to some extent prevalent. 



(3) The numerical preponderance of the ,/s has been steadily 

 sinking during recent years, owing probably to the check which 

 foreign intercourse has imposed upon $ infanticide. 



(4) The fertility of the race has been steadily growing less 

 during the last few generations. 



(5) This is probably due to the immorality of the women 

 incident upon increased intercourse with the outer world. 



Note by W. H. R. Rivers, M.A. 



The genealogical records which form the basis of Mr Punnett's 

 paper were collected by me in order to work out the details of the 

 social organization. There are certain features of the records 

 which must be borne in mind when appraising the value of 

 evidence derived from them. 



People such as the Todas remember the pedigrees of their 

 community because the knowledge is necessary for the proper 

 working of social regulations. In low stages of culture an indi- 

 vidual only becomes important when he marries, and if a child 

 dies before marriage there is some danger that its existence may 

 be forgotten and its name omitted from the pedigree of its family. 

 Many such children are remembered by the Todas, in some cases 

 for three or four generations, but it is almost certain that some 

 must be forgotten, and in so far as this is the case the pedigrees 

 are vitiated as a record of vital, as opposed to social, statistics. If 

 such children have been omitted, the omissions would probably 

 have been more numerous in earlier than in later generations, 

 and if this has been the case the tendency to sterility of which 

 the genealogies bear evidence would therefore come out more 

 strongly. Even if the record should be defective in this respect 

 it is improbable that the main problem of Mr Punnett's paper — 

 the proportion of the sexes — would be affected, for there is no 

 reason to suppose that there would be any special tendency to 

 forget children of one sex rather than of the other. 



A more serious source of error is present in connexion with 

 the latest additions to the population. Owing to the existence of 

 a taboo on the names of dead relatives, I was obliged to obtain my 

 information about any given family from the members of other 

 families. N.o one gave me the names of his own immediate 

 progenitors, and it often happened that I obtained the whole 

 record from a member of another family. Here again the younger 

 members of the community before the marrying age of five or six 

 years may have been omitted. I have little doubt that this is an 



