at is to say, in the development of the embryo.

Th. Boveri, studying the development of a threadworm, made the
interesting discovery that this differentiation began at the first
division. Of the two daughter-cells produced from the zygote, one
continued dividing at a very slow rate, and without showing any
specialization. Its "line of descent" produced only germ-cells. The
products of division of the other daughter-cell began to differentiate,
and soon formed all the necessary kinds of cells to make up the body of
the mature worm. In this body, the cells from the first daughter-cell
mentioned were inclosed, still undifferentiated: they formed the
germ-cells of the next generation, and after maturity were ready to be
ejected from the body, and to form new threadworms.

Imagine this process taking place through generation after generation of
threadworms, and one will realize that the germ-plasm was passed on
directly from one generation to the next; that in each generation it
gave rise to body-plasm, but that it did not at any time lose its
identity or continuity, a part of the germ-plasm being always set aside,
undifferentiated, to be handed on to the next generation.

In the light of this example, one can better understand the definition
of germ-plasm as "that part of the substance of the parents which does
not die with them, but perpetuates itself in their offspring." By
bringing his imagination into play, the reader will realize that there
is no limit to the backward continuity of this germ-plasm in the
threadworm. Granted that each species has arisen by evolution from some
other, this germ-cell which is observed in the body of the threadworm,
must be regarded as part of what may well be called a stream of
germ-plasm, that reaches back to the beginning of life in the world. It
will be equally evident that these is no foreordained limit to the
forward extension of the stream. It will continue in some branch, as
long as there are any threadworms or descendants of threadworms in the
world.

The reader may well express doubt as to whether what has been
demonstrated for the threadworm can be demonstrated for the higher
animals, including man. It must be admitted that in many of these
animals conditions are too unfavorable, and the process of embryology
too complicated, or too difficult to observe, to permit as distinct a
demonstration of this continuity of the germ-plasm, wherever it is
sought. But it has been demonst