Kaymond Pearl and Maud Dewitt Pearl 
421 
a broad Mendelian category. In Mendelian discussion " single comb " is a " unit 
character." All "single" combs arc put together in one category, all " pea" combs 
in another. But nothing is more certain than that all single combs are not alike 
in respect to any feature whatsoever, even including their " singleness." How 
much and in what ways do they vary ? Do the variants within the category 
mendelize ? Are all variants exactly equivalent in crossing with other categories ? 
An answer to these and other easily suggested questions could not fail, it seems 
to us, to throw light on the problem of the constitution and physiology of the 
gametic determinants of " unit characters," assuming that such determinants exist. 
In this paper we have endeavoured to give a clear and, so far as possible, 
quantitative description of the nature and amount of variation normally occurring 
in a homogeneous pure bred strain of Barred Plymouth Rock hens in re.spect to 
the form and size of the comb. The aim of the paper is purely descriptive, and it 
is regarded by the authors as preliminary to the analytical investigation of comb 
inheritance. 
2. Up to this time we have only been able to collect anything approaching 
a statistically adequate amount of material regarding comb variation for single 
combs alone. There is a certain fitness in taking up this comb at the start, 
because it is the primitive form*. In the course of the routine work of the 
laboratory, there came an opportunity in connection with certain autopsy work 
to examine and record the condition of the comb in a series of adult Barred 
Plymouth Rock hens. The hens from which the combs were taken for this work 
had been carefully and closely selected in their breeding for more than 25 years. 
During the last nine years they have been " line-bred." This means that no new 
" blood " had been introduced into the stock during that period. Consequently, 
the material is racially exceedingly homogeneous. Indeed it would be difficult to 
find anywhere material for the study of variation more homogeneous than that 
dealt with here. All the birds whose combs are included in this study were 
females, and adult. In age they varied from eleven months to about five years. 
The great majority of the birds were two years and over in age. Since the comb 
attains its full development within the first year of a hen's life and probably does 
not thereafter change in form — except as the result of accident or mutilation — 
this difference in the age of the different birds used is not significant. Mutilated 
combs were not included in the study. 
As will presently be made clear to the reader, the comb is an exceedingly 
variable structure in fowls. It varies both in size and in form. So great and 
of such a peculiar character is this variation that its biometrical appreciation is 
a rather perplexing problem. A priori one would be inclined to say that fewer 
things would be easier than counting the number of points on a single comb 
or measuring its height. Actually we have found it practically impossible to do 
* Cf. Davenport, C. B., " Inheritance in Poultry." Carnegie Institute of Washinyton. Publ. 
No. 52, p. 65. "The primitive form of the comb is the single comb seen in the wild species of the 
genus Gallus, and in most domestic races." 
