BATESON AND PUNNETT 



solved into r. and p. Mr. Edgar Bran- 

 ford, of Woodbridge, who has a long 

 experience of the breed, was kind 

 enough to give much help in this mat- 

 ter, and further assistance was also 

 received from Mr. Wootten, a breeder 

 of White Malays. It appears that 

 Malays bred inter se sometimes pro- 

 duce true peas as well as the normal 

 walnut. Occasionally in certain strains 

 peas come often, but generally speak- 

 ing thev are uncommon. Whether 

 pure Malays ever throw actual roses or 

 singles is not quite certain, but if they 

 do, these occurrences must be exces- 

 sively rare. 



A pure black-red Malay $ and two 

 pullets were obtained from Mr. Bran- 

 ford. The $, had been used by him in 

 1902 with pure hens, and certainly 

 then gave no rose or single, though 

 possibly occasional peas. 



The Malay $ was bred with 5 s.c. 

 hens of various origins and gave (Ex- 

 periments 188-191) — 



r.p. 



22 



r. 

 21 



P- 



20 



s. 

 20 



the equality of the four types of 

 gamete being evident. The Malay hens 

 are notoriously bad layers and only 

 gave small totals. Each was bred with 

 a s.c. cock and the result was 



• r-P- P- 



14 13, no true r.c. or s.c. 



Of the pea-combs, 3 were of the 

 high, intermediate type, inclining to 

 s.c, but none was a true s.c. Of the 

 r.p. group, 1 on hatching was regarded 

 as an irregular rose, but it developed 

 into a fairly ordinary type of walnut 

 comb, though having two rather prom- 

 inent knobs behind. 



In Experiment 171 a Malay 9 is re- 

 corded as giving X artificial r.p. $ , 3 



49 



r.p., 1 r., 1 p., 1 s. The r. was irregular, 

 and the s. was very low. In the light 

 of later experience, it is perhaps pos- 

 sible that both were wrongly classi- 

 fied. Both were dead in the shell, and 

 at that date the distinctions were not 

 familiar to us. 



In these results there are still certain 

 points obscure. The "artificial" r.p. 

 birds are giving off the four types of 

 gamete. Therefore there must be five 

 types of r.p. comb possible, namely r. 

 X p., r.p. X r-p-, r.p. X r., r.p. X p-, 

 r.p. X s.; but owing to the fact stated 

 above, that s. may be created afresh by 

 at least one of the combinations into 

 which it does not directly enter, the 

 subsequent analysis becomes very 

 complex. 



A further difficulty is suggested by 

 the behavior of the natural r.p. 

 Though the $ r.p. with s. 9 's gave 

 the four forms in equal numbers, it is 

 practically certain that in Malay breed- 

 ing, if r.c. and s.c are ever seen, they 

 are excessive rarities; whereas if there 

 were gametic equality on both sides, 

 3 in 16 should be r.c, and 1 in 16 

 should be s.c. For some time the con- 

 clusion seemed irresistible that the ap- 

 pearance of these forms on crossing 

 with s. must be due to some imperfec- 

 tion of fertilisation, leading to a partial 

 monolepsis and this account was pro- 

 visionally adopted. At that time the 

 diflFerence between reciprocal crosses 

 was not fully appreciated; but after a 

 critical study of their material it is 

 more likely that the true explanation 

 lies in a difference of constitution be- 

 tween the $ 's and 9 's. 



It is possible that only $ 's with wal- 

 nut combs produce all four kinds of 

 gamete, r.p., r., p., s., other males giv- 

 ing only r.p., or r.p. and p., or r.p. and 

 r.; while r.p. 9 's give either r.p. and p., 

 or r.p. only. If this conclusion is sub- 

 stantiated, it will be possible to give a 

 complete account of this curious case. 



