stics exist, there is a necessity for a system of
blood-vessels. It is not enough that there be adequately extensive
surfaces for absorption and aeration; for in the absence of any means of
conveyance, the absorbed elements can be of little or no use to the
organism at large. Evidently there must be channels of communication.
When, as in the _Medusae_, we find these channels of communication
consisting simply of branched canals opening out of the stomach and
spreading through the disk, we may know, _a priori_, that such creatures
are comparatively inactive; seeing that the nutritive liquid thus
partially distributed throughout their bodies is crude and dilute, and
that there is no efficient appliance for keeping it in motion.
Conversely, when we meet with a creature of considerable size which
displays much vivacity, we may know, _a priori_, that it must have an
apparatus for the unceasing supply of concentrated nutriment, and of
oxygen, to every organ--a pulsating vascular system.

It is manifest, then, that setting out from certain known fundamental
conditions to vital activity, we may deduce from them sundry of the
chief characteristics of organized bodies. Doubtless these known
fundamental conditions have been inductively established. But what we
wish to show is that, given these inductively-established primary facts
in physiology, we may with safety draw certain general deductions from
them. And, indeed, the legitimacy of such deductions, though not
formally acknowledged, is practically recognized in the convictions of
every physiologist, as may be readily proved. Thus, were a physiologist
to find a creature exhibiting complex and variously co-ordinated
movements, and yet having no nervous system; he would be less astonished
at the breach of his empirical generalization that all such creatures
have nervous systems, than at the disproof of his unconscious deduction
that all creatures exhibiting complex and variously co-ordinated
movements must have an "internuncial" apparatus by which the
co-ordination may be effected. Or were he to find a creature having
blood rapidly circulated and rapidly aerated, but yet showing a low
temperature, the proof so afforded that active change of matter is not,
as he had inferred from chemical data, the cause of animal heat, would
stagger him more than would the exception to a constantly-observed
relation. Clearly, then, the _a priori_ method already plays a part in
physiological reaso