YOSHIMARO TANAKA 
'35 
Plain (Pq) x Quail (pO) or 
Normal (PQ) x Pale-quail (pq) 
Normal (PpQq) x Pale-quil (ppqq^ 
Norm. PI. Qu. Pal. qu. Norm. PI. Ou. Pal. qu. 
(PQ) (Pq) (pQ) (pq) (PQ) (Pq) (pQ) (pq) 
9:3:3:1 1:1:1:1 
This is exactly parallel to the classical case of fowl-comb inheritance 
investigated by Bateson and Punnett (1905). The most important point 
brought out by the above experiments is that the normal pattern is really not 
a simple character as formerly imagined, but a compound character depend- 
ing upon two allelomorphs P and Q ; this holds true with the normal of any 
strain or of any colour intensity. 
Now we are able to arrange the marking factors in the order of their 
epistasis as follows : 
> P > p 
S > M 
> Q > q 
In one of my previous papers (1914 a), I put forth a question on the 
genetic constitution of the striped and moricaud markings, i. e. which of the 
formulae SN and MN on one hand, and Sn and Mn on the other were correct. 
Substituting N by PQ,, and n by pq according to the just mentioned results, 
the question is no other than the comparison of the formulae SPQ and Spq, 
and MPQ and Mpj. Further experiments done since in this direction solved 
the question ; the formulae SPQ, and MPQ proved to be right, in which Q was 
completely coupled with S and M. 
When striped is crossed with quail, F,s are alike to homozygous striped, 
except somewhat reduced development of the melanin in hybrids. F., families 
consisted of four forms, striped, striped-quail, normal and quail in the ratio 
