rked modifications, "variations which seem to us in our
ignorance to arise spontaneously." ("Origin of Species" (6th edition),
page 421.) The selection theory demands the further postulate that such
changes, "whether extremely slight or strongly marked," are inherited.
Darwin was no nearer to an experimental proof of this assumption than to
the discovery of the actual cause of variability. It was not until the
later years of his life that Darwin was occupied with the "perplexing
problem... what causes almost every cultivated plant to vary" ("Life
and Letters", Vol. III. page 342.): he began to make experiments on the
influence of the soil, but these were soon given up.

In the course of the violent controversy which was the outcome of
Darwin's work the fundamental principles of his teaching were not
advanced by any decisive observations. Among the supporters and
opponents, Nageli (Nageli, "Theorie der Abstammungslehre", Munich, 1884;
cf. Chapter III.) was one of the few who sought to obtain proofs by
experimental methods. His extensive cultural experiments with alpine
Hieracia led him to form the opinion that the changes which are induced
by an alteration in the food-supply, in climate or in habitat, are not
inherited and are therefore of no importance from the point of view of
the production of species. And yet Nageli did attribute an important
influence to the external world; he believed that adaptations of plants
arise as reactions to continuous stimuli, which supply a need and are
therefore useful. These opinions, which recall the teleological
aspect of Lamarckism, are entirely unsupported by proof. While other
far-reaching attempts at an explanation of the theory of descent were
formulated both in Nageli's time and afterwards, some in support of,
others in opposition to Darwin, the necessity of investigating, from
different standpoints, the underlying causes, variability and heredity,
was more and more realised. To this category belong the statistical
investigations undertaken by Quetelet and Galton, the researches into
hybridisation, to which an impetus was given by the re-discovery of
the Mendelian law of segregation, as also by the culture experiments
on mutating species following the work of de Vries, and lastly the
consideration of the question how far variation and heredity are
governed by external influences. These latter problems, which
are concerned in general with the causes of form-production and
form-mo