1906.] Certain Invisible Characters in Peas. 31 



others again contain both m. and p. — ABc. Finally, one white out of 16 

 contains neither — abc. 



It seemed so certain that this was the true explanation that, in spite of the 

 fact that no case of the kind had then been worked out, I ventured to predict, 

 in the summer of 1904, that, when crossed with the simple grey type, some 

 of these whites would yield plants bearing purple spotted seeds— p.#. ; others 

 seeds with the full brown marbling — m.g. ; others again with both — m.p.g. ; 

 whilst some I expected to give only grey, like the grey parent used to 

 render visible the hidden factor, if present. 



The actual crosses were carried out in 1905 upon plants of the fourth 

 generation (derived from heterozygotes in F 2 ) with the result set down in 

 Table I. 



One or more plants were raised successfully from each of 62 crosses, one of 

 the parents having in each case exhibited grey testas and the other white 

 (with or without the "ghost" maple). Each cross is entered twice, once 

 according to the presence or absence of the purple character among the off- 

 spring, and once according to the presence or absence of the full maple 

 character. 



Considering first the purple character, nine of the white-flowered plants 

 examined are found to have been heterozygotes in respect of the hidden 

 factor, whilst 51 plants gave rise to offspring showing only the dominant or 

 only the recessive character. The offspring were too few in number to allow 

 of exactly determining how many of the parents were truly homozygous, but 

 it is clear that a considerable majority must have been of this nature. The 

 Mendelian expectation in the second generation of descendants from a 

 heterozygote is a ratio of 3 homozygotes to 1 heterozygote. 



In the case of the presence and absence of the maple character the result 

 was closely similar. Thus the latency of m. and of p. in white flowered 

 plants is demonstrated on a considerable scale, and the proportions in which 

 they appear is not inconsistent with precise Mendelian segregation. 



A similar result was obtained in the case of the offspring of a cross made 

 by Mr. Bateson between Maple (m.g.) and Victoria Marrow (w.). On learning 

 that I was engaged upon the experiments already described, Mr. Bateson very 

 kindly handed over to me the crop obtained from this cross in F 2 . 



In Fi the maple character was described as being simply dominant. In F 2 

 there appeared, m.g. 38, g. 12, w. 19, which is very close to the expected ratio 

 of 9:3:4. Some of the seeds produced by two of the white flowered plants 

 showed just the faintest possible trace of " ghost " maple. 



In this particular instance the " ghost " maple made its appearance, even in 

 later generations, in only a small number of those plants which could be 



