PART IL EXPERIMENTAL DATA. 

 EXPLANATION OF SYMBOLS. 



Before giving in detail the matings made, a few general facts may be 

 stated. In the course of the experiments noted here, more than 10,500 animals 

 have been raised. The independence of all the factors given in this paper has 

 been made certain from breeding tests. The results may first be considered in 

 a general way which may serve, for the present, to show that the factors in- 

 cluded in this paper exist as units in inheritance. We may best consider the 

 yellow and non-yellow forms under different headings. For non-yellow mice, 

 the results given here are largely corroborative of evidence previously given by 

 Cuenot, Bateson, Durham, and other investigators. 



In expressing the gametic or zygotic formuliie of the various types we 

 employ a slight modification of the two systems most frequently used. Thus, 

 when a given factor is known to be absent, its symbol is o/?2?7ferf from the formula. 

 For example, the gametic formula of homozygous brown-pigmented animals is 

 given as YBrDP, the zygotic formula Y-iBr^DiP^- Albinos, which are gameti- 

 cally homozygous brown animals, would differ merely in that the color pro- 

 ducer Y is absent; they would, therefore, have the gametic formula BrDP and 

 the zygotic formula Br^DoP-i. If the original brown race had been hetero- 

 zygous in the color factor Y, its zygotic formula would be YBriDoP^ and the 

 gpjnetes that could be formed would be obviously YBrDP and BrDP. 



If, however, we are dealing with such a pair of characters as "dark-eye" 

 P and ''pink-eye" p, we are not justified in supposing an absence of P in the 

 pink-eyed forms. We merely know that an hypostatic modification of P has 

 occurred, and this is best designated by p. If now we have a homozygous 

 brown animal, with the zygotic formula l^/^r^Do-Po, we may designate the homo- 

 zygous pink-eyed brown as Y^BriDiPi. The dark-eyed brown heterozygous in 

 P would then be Y-iBriDiPj) and would form two sorts of gametes YBrDP and 

 YBrDp. 



It is interesting to note that the system employed by the foremost expo- 

 nents of the presence-and-absence hypothesis really hints at the presence of two 

 conditions, the one (B) epistatic to the other (b). On the other hand. Castle, who 

 does not follow so implicitly the limits of the presence-and-absence h;j/pothesis, 

 adopts a system of notation which, in the case of certain characters, indicates 

 the hypostatic condition by the absence of any symbol to designate it. 



The excellent comparative study by Cuenot (1911) has served to make 



clear many debated points due to variations in systems of notation, and it is 



therefore with some doubt as to its advisability that the writer advances still a 



different one. It seems, however, exceeding present knowledge to denote by 



the same system characters which may obviously be absent and those whose 



absence is far from proved. 



47 



