36 



homo- and heterozygous combinations 

 might, on seriation, give so near an ap- 

 proach to a continuous curve that the 

 purity of the elements would be un- 

 suspected, and their detection prac- 

 tically impossible." This hypothesis, 

 which presents no difficulties from the 

 point of view of the chromosome 

 theory, is sufficient in the present state 

 of our knowledge to bring many cases 

 of apparently continuous variation 

 into definite relation with strictly 

 Mendelian cases; but, on the other 

 hand, it seems probable, as already 

 noted (p. 221), that the individual 

 variation in many characters now 

 thought to be strictly Mendelian may 

 prove to be due to the existence in the 

 species of many variations of what 

 may be regarded as the type allelo- 

 morphs, accompanying similar varia- 

 tions of the homologous chromatin 

 entities representing those types. 



2. First Crosses that Breed Tnie.—\t 

 is obvious that in the germ-cells of 

 true-breeding hybrids ^^ there can be 

 no qualitative reduction. In the nor- 

 mal process synapsis must be ac- 

 counted for by the assumption of an 

 affinity existing between maternal and 

 paternal homologues, and conversely 

 reduction is the disappearance of that 

 affinity or its neutralization by some 

 greater force. Now in Hieracium the 

 characters of the hybrid are frequently 

 intermediate between those of the two 

 parents, showing that both allelo- 

 morphs (or chromatin-entities) are at 

 work, but on self-fertilization there 

 is no resolution of allelomorphs (re- 

 duction division). On the contrary, 

 all the germ-cells are equivalent, as 

 shown by the fact that all combina- 

 tions produce similar offspring which 

 in turn are similar to the parent. The 

 suggestion made by Bateson in an- 

 other connection, that "if one allelo- 



18 Cf. Mendel's experiments on Hieracium. 



SUTTON 



morph were alone produced by the 

 male and the other by the female we 

 should have a species consisting only 

 of heterozygotes," which would come 

 true as long as bred together, at first 

 sight seems logically applicable to 

 these cases. For such an idea, however, 

 we can find no cytological justifica- 

 tion, since if any reduction occurs 

 both chromosomes occur in both male 

 and female germ-cells in equal num- 

 bers; and further, the evidence is in 

 favor of a great variety of combina- 

 tions of maternal and paternal chro- 

 mosomes in the germ-cells so that the 

 exact chromosome group of a hybrid 

 parent could hardly be duplicated ex- 

 cept by fusion of the very pair of cells 

 separated by the reducing division. A 

 more plausible explanation from the 

 cytological standpoint is that the union 

 of the chromosomes in synapsis is so 

 firm that no reduction can take place, 

 /. e., that in each case, a paternal and 

 a maternal chromosome fuse perma- 

 nently to form a new chromosome 

 which subsequently divides only equa- 

 tionally. The result must be germ-cells 

 which are identical with one another 

 and with those of the parents, and 

 hence self-fertilization would produce 

 offspring practically without variation. 

 If this explanation be the correct one 

 the process is distinctly pathological 

 and hence it is not surprising that such 

 cases, as noted by Bateson, should 

 often present "a considerable degree 

 of sterility." 



3. The ''False Hybrids'' of Mil- 

 /^nY^f.— Millardet, de Vries and Bate- 

 son. have all described experiments in 

 which the offspring resulting from ;i 

 cross between dissimilar individuals 

 showed the character of one parent 

 only, those of the other parent being 

 shown by further experiment to be 

 lost permanently. The obvious cyto- 

 logical explanation of such a phenome- 

 non is hinted at by Bateson in the 



