EXPERIMENTS IN THE HEREDITY OF PEAS 



489 



coat is more or less transparent, the skin being rather thicker in the 

 former than in the latter ; in both cases, too, the white seed-coats them- 

 selves contain varying amounts of green pigment when the seeds are not 

 well ripened, and if the seeds are gathered prematurely this green pigment 

 persists. Great care is, therefore, necessary in dealing with these compli- 

 cations in seed-coat colour, and the whole question bristles with difficulties 

 owing to the changing conditions of life. When we come to deal with the 

 heredity of cotyledon colour, the necessity of carefully eliminating the ques- 

 tion of seed-coat colour becomes even more evident. There is a vital differ 

 ence between the hereditary nature of cotyledon colour and that of seed-coat 

 colour, inasmuch as the cotyledons, representing as they do the first two 

 leaves of the hybrid plant, are hybrid in their nature, partaking of the 

 qualities of both parents, while on the other hand the seed-coat is purely 

 a maternal structure, so that it is possible to have two distinct generations 

 present in one seed. For example, in Experiment 1, where a plant of ' British 

 Queen ' was fertilised with pollen of ' Eclipse ' in the summer of 1902, the 

 ripe seeds resulting from that cross, gathered in the autumn of the same 

 year, were true hybrids of the first generation, as far as the shape and 

 colour of the cotyledons were concerned, but the seed-coats of the hybrid 

 Peas were purely maternal, i.e. ' British Queen,' being as much a part of 

 the mother plant as the pods. When the hybrid seeds were sown in 1903, 

 and grew into hybrid plants of the first generation in the same summer, 

 they were allowed to self-fertilise, and the resulting seeds, gathered in the 

 same autumn, were, as far as seed shape and cotyledon colour were 

 concerned, hybrid seeds of the second generation (Experiments 2-13) ; but 

 the seed-coats of these same seeds were maternal, i.e. hybrids of the first 

 generation ; and so on with all the generations. It is true that, occasionally, 

 the foreign pollen does apparently influence the maternal seed-coat and even 

 the pod, and Darwin (1868, p. 428) has collected a number of such cases 

 from the experiments of Gartner, Berkeley, and Laxton. This " infection," 

 or, as Darwin terms it, " the direct or immediate action of the male element 

 on the mother form," is now generally known as Xenia, a term suggested 

 by Dr. Focke. Curiously enough, in all my experiments, I have never 

 yet met with a clear case, though I have often suspected it. In some cases, 

 however, I quite appreciate that even if it did occur it would be a difficult 

 matter to establish it clearly. Professor Correns goes so far as to doubt 

 it altogether (1901 B.), but Mr. Bateson regards it as a substantial fact in 

 certain races (1902, p. 139). In view of all the complications it will be 

 seen how absolutely necessary it is to regard the cotyledon colour alto- 

 gether apart from the seed-coat colour. Ordinary garden Peas have as a 

 rale cotyledons either yellow or green. 



The precise shades of these colours vary in different races, in different 

 Peas of the same race, and even in the cotyledons of the same Pea. 

 Generally speaking, however, I have found many races quite constant, 

 much more so, indeed, than in seed shape. 



With regard to the nature of the two colours, Mr. Bateson admirably 

 puts it : " In the green certain pigmentary matters persist in the ripe seed 

 which disappear or are decomposed in the yellow as the seed ripens." 

 (1902, p. 120.) Thus it may be said that all Peas are green, but that 

 some remain green while others change into yellow as they ripen. It has 



N 



