the same various races. If we want fowls of a particular breed we obtain
eggs of that breed and hatch them with the certainty born of experience
that we shall obtain chickens of that breed which will develop the colour,
comb, size, and qualities proper to it. Similarly, in nature we recognise
that the 'characters' of species or varieties are not due to circumstances
acting on the individual during its development, but to the properties of
the ova or seeds from which the individuals were developed.

Formerly we regarded these congenital or innate characters as derived from
the parents or inherited, and heredity was the transmission of
constitutional characters from parent to offspring. Now that we fix our
attention on the fertilised ovum or the gametes by which it is formed we
see that the characters are determined by some properties in the
constitution of the gametes. What, then, is heredity? Clearly, it is
merely the development in the offspring of the same characters which were
present in the ova from which the parents developed. When the characters
persist unchanged from generation to generation, we call the process by
which they are continued heredity. When new characters appear, _i.e._ new
characters determined in the ovum not due to changes in the environment,
we call them variations. When a fertilised ovum develops into a new
individual, it divides repeatedly to form a very large number of cells
united into a single mass. Gradually the parts of this mass are
differentiated to form the tissues and organs of the body or soma, but
some of the cells remain in their original condition and become the
reproductive cells which will give rise to the next generation. The
reproductive cells also undergo division and increase in number, and when
they separate from the new individual and unite in fertilisation they
still possess all the determinants of the fertilised ovum from which they
are descended. Heredity thus continues from gamete to gamete, not from
zygote to soma, and then from soma to gamete.