EFFECTS OF INBREEDING AND CROSSBREEDING. 7 
duration) we obtain the desired estimate for inbreds producing litters 
simultaneously with Experiment Cl. 
In dealing with other characteristics, the number of matings present 
in each 3-month period does not necessarily provide the proper 
weights to be applied 
to the records of the 
inbreds. Litters per 
year, young per year, 
and young raised per 
year are properly 
weighted by the num- 
ber of matings, but in 
the case of size of lit- 
ter the number of lit- 
ters produced by the 
given experiment in 
each 3-month period = gu 
(Table 23) should be i a re 
used. Thepercentage Fic. 3.—The average gain between birth and weaning (33 days) in 
of young born alive, the inbred stock during successive 3-month periods, 1916-1919. 
the percentage raised of all born, and the birth weight of all born should 
be weighted by the number of young born in each 3-month period as 
given in Table 24. There is a slight impropriety here, owing to 
the use of indices for these characters, but it is of no practical impor- 
tance. Similarly the percentage raised of the young born alive is to 
be weighted by the numbers born alive, shown in Table 25. The 
birth weight of the 
young raised, the rate 
of gain to weaning, and 
the weight at weaning 
are to be weighted by 
| N / STN the numbers weaned 
| K I\ it in each period, shown 
| | NEA V in Table 26. The 
Pees} estimated record of 
the inbreds, simul- 
taneous with each ex- 
"1326 FOWRIZ¢5 TIWRII 46 PINE 45730 periment, is shown 
 Eiee ee Hike ge for each character in 
Fic. 4.—The average number of litters produced per year by mature 
matings in the inbred stock during successive 3-month periods, Table 28 
aa This method of cor- 
recting for the seasonal fluctuations rests on the assumption that 
the condition of the total inbreds was genetically constant during 
the four years. Theoretically one would expect practical constancy 
m each family after a dozen generations of brother-sister mat 
