Voh. 7, 1921 
GENETICS: T. ELLINGER 
135 
with the question of the influence of age of the sow on the size of her litters. 
Although a number of records on this and similar subjects have been pub- 
lished from time to time, the writer is of the opinion that the special qual- 
ities of his material, and the mathematical treatment of it that it is pos- 
sible to employ, justify this further publication. 
In going over the literature on the influence of age on fertility in mam- 
mals, a number of records bearing on the question can be collected. Sarah 
V. Jones and James E. Rouse 10 have in a recent paper give an extensive 
summary of the literature, to which paper I only need to refer for further 
information. 
The records published are of two different kinds. First: results of ex- 
periments chiefly with rats (King 11 ), rabbits (Hammond 8 ), and guinea- 
pigs (Minot 14 ). The experiments on swine by Mumford 15 (1917) are of 
further interest in this connection. 
The second group consists of statistical investigations generally using 
as material the numerous records given in the herd books of the different 
breeds of domestic animals. 
Carlyle and McConnell, 6 Pearl, 16 Bell, 3 ' 4 Humphrey and Kleinheinz, 9 
Jones and Rouse 10 give records showing that the percentage of multiple 
births in sheep increases with age up to five years and then again declines. 
Pearl 16 and Jones and Rouse 10 give similar data for cattle showing a steady 
increase in the percentage of twin births with age. 
Swine have often been subject to investigations of this kind. One of 
two sources of error have always been present in these. 1. The number of 
individuals involved have been so limited as to make dependable re- 
sults unobtainable, or 2. The method of computing has been so rough as to 
give room for the effect of selection. The usual way of approaching the 
problem has been to take from herd book records of all available litters 
and to refer them to the age of their mothers at their birth. In this way 
it is possible to get a great number of litters from young sows but rather 
few from older ones as the less desirable animals are disposed of early. 
The older sows represent only a selected group of the best of those recorded 
at an early age, and the average size of litters at different ages of the sows 
are for that reason not directly comparable. Owing to these difficulties 
the results are very contradictory and uncertain. Rommel 18 found a 
steady increase in size of litter from one to five years. Frolich and Georgs 7 
report a maximum in the second litter and then a decrease in size of the 
the litters. Machens 12 claims a maximum at the fourth litter. Car- 
michael and Rice 5 give a table showing an increase in fertility to the third 
year followed by a decrease. 
2. The material used in this investigation consists of 134 sows of native 
Danish breed 2 which all have records from all of their first ten litters. The 
