mental investigation it is essential that observations
be made on a pure species, or, as Johannsen (Johannsen, "Ueber
Erblichkeit in Populationen und reinen Linien", Jena, 1903.) says, on
a pure "line." What has long been recognised as necessary in the
investigation of fungi, bacteria and algae must also be insisted on in
the case of flowering plants; we must start with a single individual
which is reproduced vegetatively or by strict self-fertilisation.
In dioecious plants we must aim at the reproduction of brothers and
sisters.

We may at the outset take it for granted that a pure species remains the
same under similar external conditions; it varies as these vary. IT IS
CHARACTERISTIC OF A SPECIES THAT IT ALWAYS EXHIBITS A CONSTANT RELATION
TO A PARTICULAR ENVIRONMENT. In the case of two different species, e.g.
the hay and anthrax bacilli or two varieties of Campanula with blue and
white flowers respectively, a similar environment produces a constant
difference. The cause of this is a mystery.

According to the modern standpoint, the living cell is a complex
chemico-physical system which is regarded as a dynamical system of
equilibrium, a conception suggested by Herbert Spencer and which has
acquired a constantly increasing importance in the light of modern
developments in physical chemistry. The various chemical compounds,
proteids, carbohydrates, fats, the whole series of different ferments,
etc. occur in the cell in a definite physical arrangement. The two
systems of two species must as a matter of fact possess a constant
difference, which it is necessary to define by a special term. We say,
therefore, that the SPECIFIC STRUCTURE is different.

By way of illustrating this provisionally, we may assume that the
proteids of the two species possess a constant chemical difference. This
conception of specific structure is specially important in its bearing
on a further treatment of the subject. In the original cell, eventually
also in every cell of a plant, the characters which afterwards
become apparent must exist somewhere; they are integral parts of the
capabilities or potentialities of specific structure. Thus not only the
characters which are exhibited under ordinary conditions in nature, but
also many others which become apparent only under special conditions (In
this connection I leave out of account, as before, the idea of material
carriers of heredity which since the publication of Darwin's Pangenesis
hypothe