le us to picture to
ourselves the development of the whole range of living organisms. Such
a representation will, of course, have only a subjective value.

No doubt, it is logically unjustifiable to argue from the variable
concept to the variability of the species. Still there is something
real in plants and animals which corresponds to our specific concepts.
In some cases the corresponding reality may be so well defined that it
is not difficult to form the concept accurately; whereas in other cases
where the task is more difficult, the difficulty must be due to the
object. Under these circumstances we may safely conclude from the lack
of definiteness in our concepts to a certain lack of rigid delimitation
in the organic forms.

This blending of certain forms suggests the idea of transformation, but
does not furnish definite proof of it. Such proof can be had only by
the direct observation of a transformation. And no doubt in certain
cases a transformation may occur. As regards animals, I may call
attention, for instance, to the experiments made with butterflies by
Standfuss, and as regards plants, to the experiments of Haberlandt, of
which I treated in Chapter III. The limits within which these
transformations take place are indeed very narrow as are also the
limits of those indisputable varieties which naturally arise within an
otherwise rigidly defined species. I am aware that the transformation
of one species into another has not yet been effected, but the
above-mentioned attempts at transformation have nevertheless
demonstrated that certain organic forms when subjected to changed
conditions of life, display certain mutations which clearly show that
variability is to be attributed, not, certainly, to the specific
concepts, but to the corresponding reality. This observation and
reflexion, joined with the fact that organisms form a progressive
series from the simple to the more complex, and with the observed
phenomena of individual development, lead me to regard the concept of
Descent as admissible, and in a certain sense, even probable. But I
agree with Fleischmann in saying that this is a mere belief, and that
all attempts to give it a higher scientific value by inductive proof
have signally failed.

My standpoint, moreover, requires me to admit the validity of the
hypothesis of Descent as an heuristic maxim of natural science. I
believe that we shall be justified in the future, as we were forty
years ago, in di