446 



SCIENCE. 



[N. R. Vol. XVII. No. 429. 



delian ideas in America were the instruc- 

 tion given by "Webber and others in the 

 Graduate School of Agriculture at Colum- 

 bus last summer, and the prolonged dis- 

 cussion before the International Conference 

 on Plant-Breeding at New York last fall. 

 Lately, several articles on the subject have 

 appeared from our scientific press. 



Mendel's work is important because it 

 cuts across many of the current notions re- 

 specting hybridization. As De Vries's dis- 

 cussions call a halt in the current belief 

 regarding the gradualness and slowness of 

 evolution, so Mendel's call a halt in re- 

 spect to the common opinion that the re- 

 sults of hybridizing are largely chance and 

 that hybridization is necessarily only an 

 empirical subject. Mendel found uni- 

 foi-mity and constancy of action in hybridi- 

 zation; and to explain this uniformity he 

 proposed a theory of heredity. 



One of the most significant points con- 

 nected with Mendel's work is the great 

 pains he took to select plants for his ex- 

 periments. He believed that hybridism 

 is a complex and intricate subject, and 

 that, if we are ever to discover laws, 

 we must begin with the simplest and 

 least complicated problems. He was 

 aware of the general belief that the most 

 diverse and contradictory results are likely 

 to follow any hybridization. He conceived 

 that some of this diversity may be due to 

 instability of parents rather than to the 

 proper results of hybridizing. He also saw 

 that he must exclude all inter-crossing in 

 the progeny. Furthermore, the progeny 

 must be numerous, for, since incidental and 

 aberrant variation may arise in the plants, 

 it is only by a study of averages of large 

 numbers that the true effects of the hy- 

 bridizing are to be discovered. Moreover, 

 the study must be more exact than a mere 

 contrasting and comparing of plants : char- 

 acter must be compared with character. 



The garden pea seemed to fulfill all the 

 requirements. Mendel chose well-marked 

 horticultural races or varieties. These he 

 grew two years before the experiment 

 proper was begun, in order to determine 

 their stability or trueness to type. When 

 the experiments were finally begun, he 

 used only normal plants as parents, throw- 

 ing out such as were weak or aberrant. 

 Peas are self-fertile. It is to be expected 

 that under such conditions the hybrid off- 

 spring would show uniformity of action; 

 and it did. 



In order to study the behavior of the 

 hybrids, it was necessary to choose certain 

 prominent marks or characters for com- 

 parison. Seven of these characters were 

 selected for observation. These marks per- 

 tain to seed, fruit, position of flowers and 

 length of stem, and they may be assumed 

 to be representative of all other characters 

 in the plant. These characters were paired 

 (practically opposites) as long-stem vs. 

 short-stem, round-seed vs. angular-seed, 

 inflated-pod vs. constricted-pod. They 

 were 'constant' and 'differentiating.' Of 

 course every parent plant possessed one or 

 the other of every pair of contrasting char- 

 acters, but in order to facilitate his studies 

 Mendel chose a different set of parents for 

 each character, studying seed-shape in one 

 set of hybrids, seed-color in another, pod- 

 shape in another; in this way he avoided 

 much complication in the results. Since 

 it is not my purpose to discuss Mendel's 

 work in detail, but only the general signif- 

 icance of its results, as they appeal to me, 

 I need not describe these characters here. 

 It will be sufficient for my purpose if I 

 choose only one, the shape of the seed. 



The seed-shape characters were round- 

 ness and angularity— the former being the 

 'smooth ''pea of gardeners, and the latter 

 the 'wrinkled' pea. Let us suppose that 

 twenty-five flowers on round-seeded plants 



