Mar. ao, 19x6 Inheritance of Fertility in Swine 1149 
This furnishes opportunity to link together any desired number of 
generations. 
In treating such data, the degree of confidence which can be placed in 
the figure for litter size must be considered. Its accuracy depends on the 
carefulness and honesty of the breeder, the accuracy of the clerks in the 
registry office, and the freedom from typographical errors in the printing 
of the volume. The matter of personal integrity can be accepted to a 
high degree, for fortunately the majority of breeders are quite reliable. 
Whenever falsification wittingly occurs, the tendency is to raise the 
number per litter; but, owing to the publicity involved in pure-bred 
breeding as well as the personality invested in breeding animals due to 
the registry systems, it is doubtful if litter sizes are ever exaggerated by 
more than one or two pigs. 
Investigations in color discrepancies, mistakes in parentage, etc., have 
shown that about 2 per cent of errors are involved in the work of registry- 
office clerks and in printing. Some associations are more careful than 
others, but, of course, none are absolutely free from errors. Unfortu¬ 
nately swine books show a relatively greater number of mistakes than 
do those published by breeders of some of the other classes of live stock. 
However, assuming, as has been done, that the bulk of the records can 
be accepted, there still remains a question as to their genetic value. It 
is doubtful whether a sow will ever exceed her hereditary possibilities in 
number per litter, but there are many forces that may cause her to fall 
short of that number. Tack of proper nutrition, failure to have all ova 
released or fertilized, loss of ova, atrophy of fertilized ova or embryos, 
and disease may all operate against the complete realization of the 
hereditary make-up. The age at which a sow farrows, the number of 
litters she has per year, and certain other environmental conditions .may 
also reduce the litter size. It is interesting to observe that this source of 
error operates in a compensating direction to that of record falsification, 
when such exists, and in the end the two may counterbalance, although 
these physiological and pathological factors operate more often than 
does the misrepresentation of litter numbers. 
After admitting all of these sources of error, but hoping that enough 
records are made under natural conditions to give the figures an investi¬ 
gational value, there still remains the big question of the geneticist, Does 
the somatic expression of the character indicate the germinal (zygotic) 
condition of the individual ? In other words, Does the fact that a pig 
is farrowed in a litter of eight indicate that it will transmit a tendency to 
produce litters of eight? The answer very evidently is No, and the 
greater the degree of outcrossing in the ancestral lines, the less reliable 
an index of heredity the size of litter is. Yet it is the only single index 
obtainable in the study of herdbook records; so for the present it will 
have to be accepted for what it is worth. 
