240 



THE AMERICAN NATURALIST [Vol. XLV11 



stood. To view these hybrids as a typical case of Mendelism is 

 to overlook some of the distinctions which need to be recognized 

 before correct applications can be expected. 



It seems to be taken for granted by Dr. Nabours, as by many 

 other recent writers, that all forms of alternative inheritance 

 represent the so-called Mendelian principle of heredity, that is, 

 the alternative transmission of unit-characters in pure germ- 

 cells. In reality the facts of alternative inheritance extend far 

 beyond the field of Mendelism into regions where the Mendelian 

 conception of alternative transmission can not be applied. 1 



The old assumption regarding hybrids and mixed races was 

 that they represented intermediate combinations or averages be- 

 tween the contrasted characters of the parental stocks. A gen- 

 eration ago belief in "the swamping effects of intercrossing" was 

 even more general among biologists than acceptance of Mendel- 

 ism is now. But we have learned that the "swamping effects" 

 were largely imaginary. The scientific world has its history of 

 easily forgotten fads, no less than the world of politics or 

 fashion. 



The usual result of crossing is not the formation of an inter- 

 mediate average, but the reappearance of the parental characters 

 in the later generations of the hybrids, if not in the first. Recog- 

 nition of the principle of alternative inheritance is having a 

 revolutionary effect upon the science of heredity. The facts of 

 Mendelism are of special interest because they represent extreme 

 cases of alternative inheritance, but the interest is in no way de- 

 pendent upon the Mendelian theory of heredity. Indeed, the 

 theory often interferes with appreciation of the facts. 



Mexdelism a Theory of Alternative Transmission 

 Mendelism, as a theory of heredity, is an assumption that alter- 

 native inheritance is due to alternative transmission of independ- 

 ent particles or "units," which are supposed to represent the 

 characters in the protoplasm at the time when the germ-cells are 

 formed. If sufficient magnification could be applied, so that the 

 structure of the protoplasm in the germ-cells could be fully 

 ^ook, O. P., "Dimorphic Leaves of Cotton and Allied Plants in Eelation 

 to Heredity," Bulletin 221, Bureau of Plant Industry, U. S. Department 

 of Agriculture, pp. 36-50. 



