NON-MENDELIAN CHARACTERS. 



63 



appear that in the cross-fertilized ovule, part of the cytoplasm of 

 the cell carries the tendency to develop green tissue and part the 

 tendency to develop white tissue. If at any cell division one of the 

 daughter cells should happen to receive only cytoplasm of a certain 

 kind then the tissue descended from that daughter cell will be either 

 pure white or pure green, as the case may be. Plants originating 

 from this cross between white and green are thus called mosaics. 

 A leaf or bud originating on the line of contact between the tissues 

 will be white on one side and green on the other. If it originates 

 wholly from white tissue it will be pure white; if wholly from green 

 tissue it will be pure green. Occasionally, however, the white tissue 

 on the stem of such plants may extend as a thin surface layer over 

 the green tissue. A bud coming through such a layer will be com- 

 posed of green tissue within and a thin layer of white tissue without, 

 and this bud gives rise, by division and propagation, to a new white- 

 margined plant. 



Another case which should be mentioned here is that of the in- 

 heritance of ear length in rabbits studied by Prof. W. E. Castle, of 

 Harvard. The cross between long-eared and short-eared rabbits 

 had ears intermediate in length, and their progeny were like the 

 hybrid in this respect. In this case the mechanism of inheritance 

 is not clear; and it is barely possible that it is simply a very complex 

 case of Mendelian inheritance. 



In a great many crosses between very distinct species we do not 

 get strictly Mendelian phenomena and we do not know exactly 

 why. It is highly probable, however, in the writer's opinion, that 

 the reason is to be sought in the following. Most of the organs and 

 parts of an organism are developed as the result of the interaction 

 of a good many factors which are Mendelian in nature. For in- 

 stance, the development of horn tissue in cattle may be the result of 

 the interaction of three or four or even more chemical substances 

 arising from different organs in the cell. Now, if in two races of 

 cattle v/e find a difference in only one of these chemical substances, 

 then the differences between the two races would behave as a simple 

 Mendelian character; but if there were differences in all the sub- 

 stances concerned we should have an exceedingly complex case of 

 Mendelian inheritance, the unraveling of which would require such 

 large numbers of progeny from hybrids that it would be practically 

 impossible to determine the nature of the inheritance in the case. 

 The writer is of the opinion that the lack of simple Mendelian char- 

 acters in species hybrids is partly of the nature here outlined. In 

 some species crosses apparently wide departures from Mendelian 

 principles occur. Take, for instance, the cross made by Burbank, 

 resulting in the Primus berry. Here there was wide diversity in the 

 first-generation hybrids. The one first-generation individual from 

 81599°— Bui. 165—11 5 



