384 Kauffman . — A Contribution to the 
studied. Although the number of antheridia-bearing oogonia seem the 
chief constant distinctions, other minor and less noticeable differences are 
present. Maurizio made out a series of six or more forms, with constant 
differences, of A. hypogyna. It is not certain that his conditions of culture 
were uniform for all his studies, for Maurizio worked under the assumption 
that time was the principal factor in the test for constancy. The fact that 
under certain conditions I was able to obtain various characters which 
distinguished his varieties, seems to indicate the method by which Maurizio’s 
varieties may have been derived ; namely, by mutations from the simplest 
•S’. hypogyna. It is necessary to distinguish clearly between physiological 
variation and mutation, but this need not preclude the possibility of 
a mutation taking place in a direction such as a physiological variation 
takes. Of course the whole problem becomes deeply involved because of 
our complete ignorance as to what conditions a mutation. 
It seems to me conceivable that the variations of any individual may 
extend far in any one direction ; and also that the variations may be such 
as to produce structures which are present in other bona fide species, and 
yet the two species need not necessarily be phylogenetically related. In 
this way we would expect to find overlapping of forms, and it is on this 
supposition that the possibility rests that Maurizio’s plant is not the same 
as that obtained by me in CaHP0 4 solutions and apparently having all the 
characters of his variety. 
What shall we call all these closely related forms? De Vries has offered 
to solve the problem for us in the term { elementary species ’, with all that 
it implies. Because of the presence of parthenogenesis in our plants, and 
in the absence of cultures carried through generation after generation from 
oogonia, no evidence in this paper can be offered to substantiate or to refute 
De Vries’s view. But indirect application of his theory to the facts obtained 
in the foregoing pages is possible, and to that extent only has it been 
applied in the foregoing pages. Klebs’s notion of variations, on the other 
hand, is strictly substantiated in so far as it applies to the individuals 
studied. Furthermore, his idea that we can produce De Vries’s retrogressive 
and degressive mutations by the change in conditions seems also borne out 
by the results. Shall we say, then, that the only mutation which exists 
is progressive mutation, and that the many forms found, and which remain 
constant from each other under the same conditions, were developed from 
some common ancestor by this method ? In so far as the matter of this 
paper is concerned, this would be an apparent explanation of the facts. 
Whether it is the real one cannot be determined by such few and narrowly 
restricted experiments. 
