EXPERIMENTS IN THE HEREDITY OF PEAS. 



485 



it provides an interesting illustration of the vigorous persistence of a race 

 after more than a hundred generations ol close inbreeding by self-fertilisa- 

 tion. 



In order to simplify the experiments still more, I selected sub-races of 

 both ' Eclipse ' and 1 British Queen ' as follows : 



I first selected individual seeds of ' Eclipse ' which were apparently 

 perfectly round and smooth (save, of course, for the slight suture or seam 

 at the base of the seed) ; to be precise, all these seeds were rounder than 

 Professor Weldon's roundest Peas of Lightning (1902, pi. II., fig. 13), and 

 equally smooth. These seeds were sown and produced plants which gave 

 precisely similar seeds, perfectly round and smooth, with no trace of ir- 

 regularity or pitting. This apparently constant sub-race of ' Eclipse ' pro- 

 duced the parent plants used in the experiments. 



In the same way a sub-race of the most wrinkled ' British Queen ' was 

 selected. Individual seeds were chosen which were all angular, flattened, 

 deeply- dented, and broadly wrinkled, without a trace of either roundness or 

 smoothness ; these may be compared to Professor Weldon's most wrinkled 

 Peas of ' Telephone ' (1902, pi. II., fig. 6). These seeds were sown and 

 produced plants which gave similar seeds. This apparently constant 

 sub-race of ' British Queen ' produced the parent plants used in the 

 experiments. 



First Generation (F,). 



Experiment 1. — ' British Queen ' $ x 'Eclipse' 

 (Wrinkled $ x Round j). 



A specimen plant cf 'British Queen' was selected as the sesd parent, 

 and as soon as the first flower-buds appeared, the sepals, standard, wings, 

 keel, and immature stamens were all cleanly removed at an early stage, 

 and at the same time the ripe pollen of a partially-expanded flower of an 

 individual plant of ' Eclipse ' was applied in profusion, completely 

 covering the susceptible surface of the pistil. In a few days the pods 

 began to form and ultimately 19 hybrid seeds were developed, all of 

 which were round when dried, thus demonstrating cnce more the domi- 

 nance of round seeds over wrinkled, and confirming the previous experi- 

 ments of Mendel (1865), Correns (1900), Tschermak (1900-1901), and 

 Bateson (1902). These dominant hybrid Peas, though technically round, 

 were slightly imperfect in their roundness, all being practically the same 

 shape. They may be described as rounded, with slightly elongated apex, 

 and rather flattened sides ; all were practically smooth with faint traces 

 here and there of both slight pitting and feeble wrinkling ; they compare 

 well with Professor Weldon's ' Telegraph ' (1902, pi. II., fig. 2a), except 

 that their surface is much smoother. There is not yet sufficient evidence 

 to show whether these slight imperfections of roundness are due to the 

 conditions of life or to incomplete dominance, or both, but so far the 

 evidence seems to point towards the former rather than the latter, as all 

 the Peas were large and were packed tightly in the pods, and it is indeed 

 difficult to imagine how any of them could have been perfectly round 

 under these conditions, whatever the hereditary tendency might have 

 been. 



