shield, over which those two knights of old story wrangled so long, an inspection of the subject from two points of observation decidedly simplifies matters. The first of these methods of studying the subject I would designate the negative, and the second the positive method. By the negative method we deduce certain principles in regard to cases which have at first sight a color of novelty, but are not novel, so as to be patentable. By the positive method we arrive at certain conclusions in regard to what actually constitutes novelty in a patentable sense.
I propose to discuss, first: What cases, at first sight possessing novelty, do not actually possess novelty? A study of cases warrants us in accepting, as a first principle, that every change or mere substitution of a mechanical equivalent is not necessarily a patentable novelty; for it may not be substantially unlike some prior thing. We may have, for instance, a machine, comprising, say, three distinct parts. A man, not the inventor, substitutes for each of these parts other equivalents, producing the same results. This is not a patentable invention. Even if the products be better or cheaper, it is at most only an improvement upon a former invention, and can be used only with the permission of the former patentee.
I spoke above of a "mechanical equivalent." This needs to be defined in order that we may have a clear comprehension of the above principle. To define it, however, is not so easy. Mr. Parsons has said that "he would be a very acute man who could certainly discern, or a very bold man who would certainly assert, what is meant by a mechanical equivalent." At the risk of being considered bold, certain judges have, nevertheless, ventured to attack the Gordian knot. We find one definition in Smith vs. Downing, 1 Fisher's Patent Cases, 87: "By equivalents in machinery is usually meant merely the substitution of one mechanical power for another, or one obvious and customary mode for another, of effecting a like result." This definition is not sufficiently explicit. A better one is to be found in Carter vs. Baker, 4 Fisher's Patent Cases, 409: "When, in mechanics, one device does a particular thing, or accomplishes a particular result, every other device known and used in mechanics, which skillful and experienced workmen know will produce the same result, or do the same particular thing, is a known mechanical substitute for the first device mentioned for doing the same thing, or accomplishing the same result. It is sufficient to constitute a known mechanical substitute that, when a skillful mechanic sees one device doing a particular thing, he knows the other device, whose uses he is acquainted with, will do the same thing."
This definition not only covers those elements which come strictly under the head of mechanics, but is also our guide in determining what constitutes an equivalent in an "art," or a "manufacture," or a "composition of matter." The definition of an equivalent of any substance in a composition made of several ingredients, for example, is, in
