Nothing contributes more to safe and effectual practice, than an acquaintance with those principles on which remedies operate. Hence, inquiries concerning the operation of medicines have been, and ever will be, prosecuted by those who mean to practise on a solid or rational foundation. Many practitioners, indeed, hold every thing that is in the least theoretical, in great contempt. They alledge, that a physician possesses all the knowledge requisite in medicine, who is acquainted with particular remedies which will infallibly cure particular diseases. Could such remedies be discovered, their reasoning would indeed hold good; and medicine, which, of all arts, is in reality the most difficult, would be the easiest. But it is incumbent on those who reason in this manner, to shew, that any such remedies actually do exist. This, however, is a point by no means determined; and, with regard to it, very great doubts may justly be entertained. With certain limitations, some few such remedies may perhaps be allowed to exist. But no one will pretend, that these can, in every case, be employed with equal success; or that, in all circumstances, they are to be used in the same manner. By a knowledge of the principles upon which a medicine operates in curing a disease, we can alone be enabled to accommodate its use to particular circumstances. He therefore who knows, not only that a certain remedy will cure a particular disease, but is likewise acquainted with the manner in which it acts in producing that effect, has at least a double advantage. Errors in theory, indeed, often mislead in practice; and opinions, supported by numerous arguments, after being long believed, have at length been discovered to be without foundation. But this imperfect state of theory, in place of being any objection to its utility, is an additional inducement to continue, with unremitting ardour, a pursuit in which so much still remains to be done. It is by unwearied assiduity alone that improvement can be expected in any art. The imperfection of theory, then, can be no argument for discontinuing farther attempts to render it more perfect.