28. Under Vitalism are included all the hypotheses of a soul, a spirit, an archæus, a vital principle, a vital force, a _nisus formativus_, a plan or divine idea, which have from time to time represented the metaphysical stage of Biology. The characteristic of that stage is the personification of a mystery, accompanied by the persuasion that to name a mystery is to explain it. In all sciences when processes are imperfectly observed, the _theory_ of the processes (which is a systematic survey of the available evidence marshalled in the order of causal dependence) is supplemented by _hypothesis_, which fills up with a guess the gap left by observation. The difference between the metaphysical and the positive stages of a science lies in the kind of guess thus introduced to supplement theory, and the degree of reliance accorded to it. I have more than once insisted on the scientific canon that “to be valid, an explanation must be expressed in terms of phenomena already observed”; now it is quite clear that most of the extra-organic hypotheses do not fulfil this condition; no one having ever _observed_ a spirit, an archæus, or a vital principle; but only _imagined_ these agents to explain the facts observed. As an example of the difference, and a proof that the value of an hypothesis does not rest on the facility with which it connects observations, and seems to explain them, take the three hypotheses of animal spirits, nervous fluid, and electricity, by which neural processes have been explained. The animal spirits are imaginary; the nervous fluid is without a basis in observation, no evidence of such a fluid having been detected; but electricity (or, speaking rigorously, the movements classed as electrical), although not proved to be _the_ agent in nerve-action, is proved to exist in nerves as elsewhere, and its modes of operation are verifiable. It, therefore, and it alone of the three hypotheses, is in conformity with the scientific canon. It may not, on full investigation, meet all requirements; it may be rejected as imperfect; but it is the kind of guess which scientific theory demands. The second difference noticeable between the metaphysical and the positive stages is the degree of reliance accorded to hypothesis; which is very much the same as that noticeable in the uncritical and critical attitudes of untrained and trained intellects. The one accepts a guess as if it were a proof; is fascinated by the facility of linking together isolated observations, and, relying on the guess as truth, proceeds to deduce conclusions from it; the other accepts a guess as an aid in research, trying by its aid to come upon some observation which will reveal the hidden process; but careful never to allow the guess to _supersede observation_, or to form a basis of deductions not immediately verified.