20 



INTRODUCTION. 



(Science, 1914, xxxix, 488; Bot. Gaz., 1914, LVIII, 328; 

 Amer. Nat., 1915, XLIX, 5) asserts that there seems to be 

 absolutely no doubt upon morphological grounds and 

 sterility that the Oenothera mutants are really hybrids. 

 He records that an examination of a large amount of 

 material of recognized wild species of Oenothera led 

 him to the conclusion that spontaneous hybridism is 

 extremely common in the genus ; that in general it repre- 

 sents a condition of high genetical impurity; and that 

 in orders such as Bosaceae and Ornagracese there is 

 grading of recognized species and hybrids into each 

 other, having in common the character of partial or com- 

 plete sterility. Such literature would make volumes. 



6. GENETIC PURITY IN RELATION TO INTERMEDI- 

 ATENESS OF THE HYBRID. 



It may be held that intermediateness of the hybrids 

 depends upon the existence of purity of the parents and 

 that, as a corollary, absence of intermediateness is diag- 

 nostic of parental impurity. It will be noted, however, 

 that while Davis (loc. tit.) with carefully selected, pre- 

 sumably pure stock recorded intermediateness in the 

 hybrid, Jeffrey refers to Oenothera lamarckiana as a 

 hybrid having a similar intermediateness, yet being the 

 offspring of spontaneous hybridism that represents a 

 high degree of genetical impurity. In fact, there is no 

 conclusive evidence in any of the investigations referred 

 to that the parents were pure. The term pure is an 

 arbitrary conception. The only test of purity we have 

 at present is in the constancy of characters of the off- 

 spring through successive generations. Nor are purity 

 and typicalness by any means synonymous terms. A 

 typical specimen of a species or hybrid is one having 

 characters which in their sum total are nearest the mean 

 of the species or hybrids, but a typical specimen may be 

 far from being pure inasmuch as there may be latent 

 or undeveloped characters that may not appear except 

 under some peculiar condition. In the investigations of 

 Macfarlane and others quoted by him, the parent species 

 examined may have been typical, yet there is no evidence 

 of purity. Darbishire used for the preparation of the 

 starch only two seeds from crosses of garden varieties of 

 peas the round pea " Eclipse " and the wrinkled pea 

 " British Queen " (hardy variety) being crossed. The 

 parents referred to in Focke's work may or may not have 

 been pure, but there is no satisfactory evidence in either 

 direction. Mendel was extremely careful to select speci- 

 mens belonging to groups that possess constant differen- 

 tiating characters, and in both of his papers he makes 

 notes of only certain selected differentiating characters. 

 He found, as already stated, that the hybrids, as a rule, 

 are not exactly intermediate between their parents, and 

 that while in the case of some of the more striking charac- 

 ters intermediateness is always present, in other cases 

 one of the two parental characters is so preponderant 

 that the corresponding character of the other parent is 

 almost or wholly absent. He also notes in Hieracium 

 hybrids there may be three types, one being almost ex- 

 actly intermediate, a second nearer to the seed parent, 



and a third nearer the pollen parent. In all of these 

 instances the parents may have been typical, yet not pure, 

 and in Mendel's experiments they might be regarded 

 as being both typical and pure pure, because of the 

 constancy of Mendelian inheritance in succeeding gener- 

 ations. But even here purity is questionable. Thus, in 

 the second generation the dominants which breed true 

 to the dominant character are looked upon as being pure, 

 yet they may have latent or undeveloped characters that 

 can be demonstrated only under peculiar conditions. 

 This has been shown by Darbishire (Breeding and the 

 Mendelian Discovery, 912, 218) in crosses of the common 

 albino and the Japanese waltzing mice. In the second 

 generation he found two types of albinos, one to all ap- 

 pearances identical with the pure albinos and the other 

 with waltzers. When these apparently pure albinos are 

 mated with each other they breed true, but when mated 

 with waltzers they were found to be very different from 

 pure albinos, " for among the offspring of extracted 

 albinos mated with waltzers there appeared pink-eyed 

 and even albino mice, forms which are never produced 

 when pure albinos are mated with waltzers." 



7. THEORETICAL REQUIREMENTS IN THE PROPERTIES 

 OF STARCHES TO CONDITIONS IN THE HYBRID 

 CORRESPONDING TO THOSE OF ANATOMIC CHAR- 

 ACTERS. 



It is evident from the literature quoted that the doc- 

 trine of intermediateness of the hybrid and the doctrine 

 of Mendel are expressions of rules that have many ex- 

 ceptions and hence are only of limited applicability. 

 The success of the plant and animal breeder depends 

 upon the elimination of undesirable characters; the 

 redistribution of characters ; the variation, modification, 

 and recombination of characters; the development of 

 some particular characters to a degree beyond parental 

 extremes, together with their perpetuation and even 

 further exaggeration in subsequent generations ; and the 

 development of new and perpetuation of desirable char- 

 acters. Neither the doctrine of intermediateness nor 

 the doctrine of Mendel admits of the possibility of gen- 

 erating ideal organisms by crossing and selection ; nor 

 are they consistent with the development of parental 

 characters in the hybrid beyond parental extremes ; nor 

 are they compatible with the appearance of new charac- 

 ters except upon the untenable assumption of such char- 

 acters being latent in the parents. Both are doctrines 

 of non-plasticity, yet the most significant phenomenon 

 of successful breeding and the genesis of elementary 

 species is plasticity which is manifested to a pre-eminent 

 degree of importance in development in the offspring of 

 characters beyond the extremes of the parents, new com- 

 binations of characters, and the appearance of new char- 

 acters. No investigations on record have shown more 

 forcefully the utter inadoquatcness of these doctrines 

 and their limitations than their application to the ex- 

 planation of the building up of ideal forms and the 

 appearance of elementary species by hybridization and, 

 on the other hand, none has better set forth the great pos- 



