228 
such where two are necessary but where either is absent and 
so on, because any one quality of an organism is only reached 
by development under the influence of a great number of factors. 
And among these factors it can hardly be said that some are 
more important than others. In the case of our bakery, it is 
just as important that water or yeast are to be had as that the 
baker is sober or the oven heated. 
It is very well possible to study the genetic and the non- 
genetic factors in heredity separately as such, but whenever 
we want to study the qualities of an organism, we must take 
both kinds into account. 
It has long been a question whether a good distinction 
was possible between these two kinds of factors, a question to 
which Mendelism has undoubtedly given a positive answer. We now 
know that genetic factors can only be either present or absent, non- 
genetic ones may each vary in intensity. As selection in a group, 
whose members exhibit a continuous variability for any quality 
has repeatedly been shown to shift the mean of the variation- 
curye in the desired direction, it is clear that either continuous 
Variation can depend upon genotypic differences within the group, 
or eise, that, if this Variation should depend solely on a Variation 
of intensity of the non-genetic factors influencing the quality, that 
these non-genetic factors must have transmittable influence. 
By the experiments of Johannsen and Nilsson-Ehle, 
it has been conclusively shown, that continuous Variation may 
depend on genotypic differences between the members of the 
population, and that in a group of organisms with identical genetic 
factors, the modifications by the non-genetic factors are not trans- 
mitted. Johannsen 's well-known experiments with beans have 
shown that selection within a biotype has absolutely no effect. 
The author recently concluded a series of selection-experiments 
with dandelion, which fully corroborated Johannsen's work, 
The experiments of Nilsson-Ehle on the colour of wheat- 
grains, have conclusively shown that several different genetic 
factors can contribute to the development of the members of one 
biotype, which all of them tend to influence this development in 
the same direction. In the work with the colours of mice, and 
above all with cavies, evidence has been found for the same 
fact. (Miss Sollas). A population of which the members difi'er 
in respect to the possession or non-possession of one or more of 
