stem is
less highly developed; (_b_) that certain formations seem to be directly
connected with nerve distribution, while others have been associated with
alternations in neighboring nerve trunks.

(4) _The embryonal theory._--This is known also as Cohnheim's hypothesis.
In early fetal life there occurs a production of cells in excess of those
required for the construction of the various parts of the body, so that a
certain number of them are left over in the fully developed tissue or
become misplaced during the sorting of cells for future development of
tissues and organs. These cells lie dormant until favorable conditions
arise or until some sufficient stimulus is applied, when, released from
their inactivity, they begin to reproduce and grow. Not being normally
related to their site, they lack the controlling and limiting influences of
the part, and, their embryonic character enduing them with a most potent
proliferating power, they develop in a lawless and unrestrained manner.
There are tumors whose existence can be explained only on these grounds.
Still, this theory falls far short of answering the question as to the
origin of tumors.

(5) _The parasitic theory._--This is not only one of the latest, but,
merely as a hypothesis, it is the most attractive and plausible of all. The
serious objections to it, however, are the almost uniform failure that has
met the attempts to transplant these tumors from one animal to another and
the absence of any constant variety of organism in them. Several forms of
parasites have been found in certain tumors, but nothing definite has been
shown with reference to the relation they bear to the causation of the
neoplasm.

CLASSIFICATION OF TUMORS.

In Senn's work on tumors occurs the following: "A uniform system of
classification of tumors is one of the great wants of modern pathology, and
all attempts in this direction have proved failures." It would be folly,
therefore, to burden the pages of a work of this kind with one or several
of the proposed systems which have, admittedly, at some important point,
failed of their purpose. Since the value of this chapter depends chiefly
upon its practical character, which in turn is measured by its aid in
diagnosis, prognosis, and treatment, the old but important clinical
division is here adopted.

Tumors are either malignant or benign. The essential difference between the
two classes is that while _benign tumors depend for their ill eff