i9io.] THROUGH HYBRIDIZATION. 289 



for only when this is done can the full scope and significance of 

 the Mendelian phenomena be understood. 



It need scarcely be pointed out that the complete tracing of the 

 germinal analysis which takes place in Mendelian hybrids, is at- 

 tended with many difficulties. The unit-characters represent 

 capacities for reaction in a certain, very specific way to given condi- 

 tions of environment. Individuals having the same unit-composi- 

 tion may react in a totally different way to a different environmental 

 complex. Some unit-characters are so sensitive to slight differ- 

 ences of environment that they offer a wide range of fluctuation, or 

 they may represent such a slight differential as to be readily dis- 

 tinguishable only in their plus-fluctuations. Two or more unit- 

 characters may even be indistinguishable from one another as 

 Nilsson-Ehle 5 has shown to be the case in certain unit-characters of 

 wheat and oats, and East 6 in endosperm colors of corn. Many 

 unit-characters are quite invisible except when occurring in com- 

 bination with some one or more other characters, and this fact has 

 led to what is called the " factor hypothesis." That the factors are 

 real unit-characters, differing in no essential way from ordinarily 

 visible unit-characters, is now in a fair way to be demonstrated by 

 such work as that of Miss Wheldale, and others who are working 

 along similar lines. The implication by some writers that the 

 factor hypothesis is a late development of Mendelism is not correct, 

 as Mendel himself suggested it tentatively. The difficulty of tracing 

 invisible characters necessarily made the development of knowledge 

 regarding them slower than that regarding the easily visible char- 

 acters, but the essential correctness of Mendel's suggestion has been 

 abundantly substantiated. 



All of the foregoing difficulties can be overcome, and are con- 

 tinually being overcome by careful analysis and patient, long-con- 

 tinued breeding tests. 



Finally, since we are examining the Mendelian process as one of 

 germinal analysis it is appropriate to discuss for a moment the 



5 Nilsson-Ehle, H., " Kreuzungsimtersuchungen an Hafer und Weizen," 

 4to, pp. 122, 1909, Lund: Hakan Ohlsson. 



6 East, E. M., " A Mendelian Interpretation of Variation that is Appar- 

 ently Continuous," Amer. Nat., 44, 65-82, Feb., 1910. 



