Collins: Nature of Mkndelian Units 



429 



hooded character reappears in one- 

 fourth of the progeny in the second 

 generation. The result of the crossing, 

 however, appeared to undo the work of 

 selection. Thus, when a member of the 

 strain in which the amount of white 

 had been increased by selection was 

 crossed with the wild type, the extracted 

 hooded individuals showed less white 

 than the hooded parent; and when a 

 member of the strain in which the white 

 had been reduced was used, the ex- 

 tracted hooded individuals showed more 

 white than the hooded parent. 



That typical Mendelian characters 

 are amenable to change through selec- 

 tion is held to prove that unit characters 

 are variable. The apparent undoing of 

 the work of selection by crossing with 

 the wild type has induced the authors 

 to grant the probable existence of 

 "modifiers," but the "main factor" is 

 still considered as a unit. May not the 

 results of these experiments be brought 

 into closer harmony with those of other 

 investigators if this "main factor" is 

 looked upon as a rather tenacious 

 group of minute factors similar in nature 

 to the detached "modifiers"? 



Paleontological records afford further 

 evidence of the gradual evolution of 

 characters that are discontinuous in 

 inheritance. While the evidence from 

 this source is direct, geological records 

 are seldom sufficiently complete to 

 observe differences as minute as those 

 brought to light by analytical breeding. 

 Professor H. F. Osbom has carefully 

 traced the development of cranial and 

 dental forms in some of the mammals. He 

 shows that a number of the characters 

 that distinguish the horse from the ass 

 afford no evidence of discontinuous 

 origin, but appear to have developed 

 gradually from very small beginnings. 

 Yet there is evidence that the resulting 

 differences, as they now exist, behave as 

 units. The sterility of the hybrid has 

 made it impossible adequately to test 

 the Mendelian behavior, but Professor 

 Osbom shows that these characters 

 exhibit the phenomenon of dominance 

 in the first generation. The behavior is 

 thus shown to be Mendelian as far as it 

 goes. Professor Osborn says: 



^ Osborn, H. F., The Continuous Origin of Certain Unit Characters as observed by a Paleon- 

 tologist. American NaturaHst, 44, 186-7. 



"It is true that the evidence for dis- 

 continuity in the heredity of characters 

 is as convincing as that for discontinuity 

 in the genesis of character is debatable," 

 and some characters "while apparently 

 continuous in origin certainly become 

 discontinuous in heredity; from which 

 it follows that discontinuity in heredity 

 constitutes no proof of discontinuity in 

 origin."^ 



A WARNING TO EUGENISTS. 



Once the complex nature of visible 

 characters is fully appreciated, less 

 confidence will be placed in the purity 

 of the germ cells. Is it safe to assume 

 that an undesirable character intro- 

 duced into a hybrid with one of the 

 parents has left no trace or tendency in 

 those members of the progeny which on 

 Mendelian analysis show the alternative 

 character? Since the rediscovery of 

 Mendel's laws, this question has been 

 acute, but most practical breeders still 

 remain skeptical. With the greatly 

 increased interest in eugenics, the im- 

 portance of deciding this question is. 

 brought home to a large part of the 

 thinking public. In the light of recent 

 research, for one to assert with con- 

 fidence that a heritable human defect, 

 which investigations have shown to be a 

 dominant character, is completely ab- 

 sent from offspring that fail to show 

 the defect, regardless of its prevalence 

 among their ancestors, indicates a 

 complete failure to appreciate a grave 

 responsibility. Mendel's laws may be 

 inexorable, and yet all the elements 

 which go to make up a physiological or 

 mental defect may not always remain 

 associated. 



Complete analysis of characters in 

 man is impossible at present, and expec- 

 tation regarding the behavior of undesir- 

 able characters must be largely inferred 

 from analogy. So many of the char- 

 acters which in other species have been 

 exhaustively studied have been found 

 resolvable that it appears unscientific 

 as well as dangerous to assume that the 

 segregation of gross pathological char- 

 acters is complete in the himian species. 

 Individuals that are pure recessive with 

 respect to what appears as a simple 



