Feb. 17,1923 
Inheritance in Swine 
569 
for red and sandy. It may be protested that the red and sandy pigs 
could as logically be included under sandy, but the writers have not 
accepted this idea because in each case where red and sandy occurred 
on the same hog, the sandy markings were so distributed as to indicate 
a failure of the intensity factor for red, just suggested, to extend to the 
extremities of the animal, in much the same way that the factor for 
restriction of black pigment in the bay horses varies centrifugally. 
Sandies do not appear in the back-cross to permit the testing of this 
hypothesis. 
The data with regard to white spotting are less suggestive. Omitting 
the three individuals that were self white (Tables IV and V), there were 
53 with white markings in the F 2 generation and 92 without. This indi¬ 
cates three possibilities: First, that white markings are not dominant 
unless dependent on the interaction of three or more factors; second, 
that white markings are recessive; or, third, that they are dependent 
on several factors, some of which are dominant and others recessive. 
The data are not complete enough to test any of these hypotheses. 
However, if the first hypothesis is correct, three or four factors are indi¬ 
cated. If the interaction of three factors is necessary, the expectation 
for 145 F 2 ’s would be 61.2 with white markings to 83.8 without. If 
the interaction of four is necessary the expectation would be 45.9 with 
white markings to 99.1 without The actual ratio, 53 to 92, is about 
half way between the two. There is no simple recessive condition that 
will produce this ratio, nor is there any combination of recessive condi¬ 
tions probable that would give the actual results. Hence the third 
hypothesis is most probable. In fact, three distinctly different types of 
white markings are readily recognizable in addition to the red-and-sandy 
“bicolor.” The first is roaning; the second is the broken splotching of 
white, probably related genetically to the Berkshire “six white points,” 
and the third is the light belly marking. The indications are that sandy 
may be substituted for white in many cases. 
Roaning appeared in three forms. Two animals were red-and-white 
roan with black spots, one was sandy-and-white roan with black spots, 
and one was white with black-and-red roan spots. The roaning was not 
evident until after weaning time, and hence many of the pigs which 
died young also may have possessed this character. No attempt is made 
to analyze or explain the roaning condition. 
The second form of bicolor was the irregular spotting of red or deep 
sandy upon a white or light sandy ground color. It always occurred in 
the presence of black and was readily distinguished from other colors 
by the fact that the red spots were always most numerous on the face, 
back of the head, and front part of the back. In nearly every case of 
this type one or both upper eyelids and eyelashes were quite red. The 
distinctness of this type of red spotting was not suggested until late in 
the investigation and consequently exact numbers are not available; 
but since it always occurred in the presence of black and did not appear 
among the back-cross individuals the suggestion is very plausible that 
this red spotting is genetically identical with the red or sandy splash 
which occasionally appears on the pure-bred Berkshire. The fact that 
the sandy color appears on the white splash on the face more commonly 
than on the feet or tail of pure-bred Berkshires supports this suggestion, 
and the fact that a pure-bred Berkshire practically never has all six 
“white points” showing sandy is another argument in favor of it. This 
