LATENCY. 



65 



of species of plants, found one species {Oenothera lamarclciana) 

 which occasionally produced offspring that were unlike the parent. 

 Some of the new individuals were capable of reproduction true to 

 type. These forms he called mutations. The work of Dr. R. R. 

 Gates, of the University of Chicago, who is making a careful study 

 of the chromosome behavior in these mutants, indicates strongly 

 that these mutations are due to accidents occurring in the reduction 

 division by which chromosomes are either lost or gained or ex- 

 changed, so that some of the daughter cells are provided with a set 

 of chromosomes differing from that of the parent species. A good 

 deal of work must yet be done before this matter is settled, and 

 until we know more about it we can not assign mutations of this 

 class to their proper place in heredity and in evolutionary progress. 



The term '^mutation" has also been applied to any permanent 

 evolutionary change of whatever magnitude or whatever its cause. 

 It is hardly probable that all evolutionary changes are due to acci- 

 dents in cell division. It would seem rather that most of such 

 changes are due to permanent changes in whatever material is 

 responsible for the development of hereditary characters, and it is 

 probable that we shall ultimately have to make a distinction between 

 these two types of so-called mutations. 



LATENCY. 



Characters sometimes fail to develop, although present. The 

 reason for such failure may fall into any one of several categories. 

 One of the most important recent papers dealing with this phase of 

 the subject is that of Dr. G. H. Shull in the American Naturalist, 

 Volume XLII, July, 1908. The following classification of the differ- 

 ent types of latency follows, in the main, Doctor Shull's paper, 

 departures therefrom being noted in the text. 



I. LATENCY DUE TO SEPARATION. 



Bateson and his coworkers crossed a cream-colored strain of 

 gillyflowers with a white strain and secured a hybrid which had red 

 flowers. This is explained by assuming that in the cross two char- 

 acters are brought together which when separated are incapable 

 of producing red flower color, but which when together give rise to 

 this color. Many other breeders have found similar instances, 

 and these characters, which when alone produce no visible effect 

 but when in the presence of other characters give rise to visible 

 manifestations, have been called ' ' cryptomeres " by Prof. E. von 

 Tschermak, a term which is derived from the Greek and which 

 etymologically means ''hidden parts." Both von Tschermak and 

 Bateson have shown that purple color in gillyflowers is due to three 

 such cr}^tomeres. Two of these without the third give rise to the 



165 



