When the chemist mixes substances together, the composition of which he knows, he arrives at results which may be strictly denominated certain and invariable. If he be not able to do this at once, he can do so ultimately, by a series of experiments, varied to test each doubtful point. The results which he thus obtains are so exact, that they can be expressed by numbers and definite proportions. The physician can imitate the chemist, it is true, in the application of tests in the investigation of disease; but it is necessarily a very humble and distant imitation, and no approach to the certainty and definiteness of chemical analysis and synthesis can be expected in medical practice. When the chemist mixes substances together, he knows what they are; and when he sees their effect upon each other, he has a right to expect the same effect to follow, with absolute certainty, whenever he shall make the same mixture again. But the physician cannot infer from the effect of a remedy in one case, that the same result will certainly occur in another case which appears to be precisely similar. For he cannot know enough of the circumstances of the two cases, to determine beyond a doubt that they are exactly alike. There are often causes, utterly undiscoverable by human wisdom, which essentially modify the effects of remedies. If you suppose that the chemist knows the nature of only a part of the substances which he puts into his retort,—that the retort itself is made of materials which will act upon these substances, and be acted upon by them, and that in the midst of his experiment some other substance is introduced accidentally or by stealth, producing an entire change in the process; you will then make the chemist to resemble the physician in the uncertainty of his results. He would then be obliged, as the physician is, to go through with a great many observations to establish any one fact; and instead of making, as he now does, a well-defined line of separation between what is known and what is not known, he would, like the physician, have a wide middle ground of probability and supposition. The causes which make disease complicated, and prevent uniformity in the effects of remedies, are principally these, viz.: 1. The sympathy which exists between the different organs of the body. 2. The influence of unseen causes or agents. 3. Natural changes, arising from the tendency which exists in the system to throw off disease, appropriately called the _vis medicatrix naturæ_, or restoring power of nature; and in connection with this the tendency to a definite limit manifest in many diseases; for example, small pox, whooping cough, measles, scarlet fever, &c. 4. Mental influences. 5. Idiosyncrasies, or individual peculiarities. We will examine in a familiar way each class of these causes separately. 1. The sympathy which exists between the different organs of the body.