The patent in issue relates to wire-mattress beds, supported by frames made of tubular metallic rails, and it bears date of November 5, 1895. The application therefor was filed on August 12,1895. It is maintained that the invention somewhat antedated the application. The claims in controversy are 2 and 4, as follows:

“(2) The combination, with a tubular metallic mattress frame and the wire fabric therefor, of the flattened metallic strips, J, mounted longitudinally on the end rails, the ends of the wire fabric passing over the upper surface of said strips and around their outer edges, and being then turned under, and clamped between the flat under-surfaces of said strips and the surfaces of the end rails, said strips being secured to the end rails by the screws, K, substantially as set forth.” “(4) In a metallic bed bottom, tubular metallic side rails and end rails and rigid metallic brackets uniting said rails into a rectangular frame, in combination with a woven-wire fabric secured to the end rails by flattened metallic strips held thereon by screws passing through said strips and fabric, the ends of such fabrics being hooked over the outer edges of said strips, and the screwheads, strips, and wire ends being covered by the body of wire fabric, to prevent contact of the bedding therewith, substantially as set forth.”

Claim 2 seems to be broader than claim 4, and is the only one which need be especially considered. The specification states that the invention relates to the method of connecting together the rails, and to the means of securing the ends of the wire fabric. The claims in issue cover the second subject-matter. The specification further states that the invention is an improvement on that of letters ¡latent No. 510,541, granted on December 12,1893, to the patentee and W. H. Taber, for mattress frames, and on that of letters ¡latent No. 515,801, granted to the patentee on September 3, 1895, for metallic mattress frames. It also states that in both prior patents the frame was provided with adjusting screws, but that the patentee now omits all adjusting devices, and it adds: “I give the wire fabric at the outset such permanent tension as is desirable, by attaching it under suitable strain.” Several prior patents relating to wire mattresses for beds, shown in the record, contained special devices for adjusting the *338mattress as it sagged by use; and it appears by the patent to Thompson and Wells, of March 22, 1881, put in by the respondent below, that the ordinary methods of adjusting had proved ineffectual. Apparently, therefore, a construction which would dispense with adjusting devices, and hold the wire fabric in position by merely giving it at the outset a suitable strain, would be useful; but nothing of this kind is claimed in the patent, and the fact is referred to by us for a purpose which we will make evident. It is plain that, as a new proposition, the proper securing of the ends of wires between metallic surfaces so as to resist great strain would involve difficulty. This is, of course, increased in the patentee’s method of construction, which gives the mattress, at the outset, a special strain. It is this difficulty which, in the case at bar, the patentee claims to have overcome. The advantage of the patentee’s construction is stated by an expert as follows:

“The wire fabric is pinched and firmly held along the line where the flat bottom of the strip comes nearest to the convex surface of the tubular end rail.' Any tilting or slight tipping movement of the strip tends to press it more closely upon the fabric held between the rail and the strip, thus preventing the ends of the wires from being pulled toward the outer edges of the strip.”

There is no evidence, and it is probably not the fact, that, when the construction is in its perfect form, there is any tilting, as spoken of by this witness; and such tilting, if it exists, occurs after the screws or other devices for securing the parts together work loose in use. But this fact alone would not defeat invention, because it may sometimes be quite as useful to secure the efficient continuous working of a device as to obtain its successful working in its new and perfect condition. The tilting in the patented device in issue operates with the advantages of leverage. The fulcrum is at the point at which the curved surface of the rail comes in contact with the flat under-surface of the metallic strip around which the wire fabric passes. The lever is a transverse section ¡of that flat surface. The point at which the power is applied is the exterior edge of the transverse section. The point through which it is communicated, so as to bear upon the wire and give a pinch, is, of course, very close to the point of contact, or fulcrum, so that the ratio of leverage may be very great. The complainant below, in his prior construction, shown by his patent of September 3, 1895, secured the ends of the wire fabric between two metallic surfaces, one of which was concave and the other convex. With such a construction, the point at which the power is applied to the lever and the point at which it is delivered for the purpose of holding the fabric are ordinarily equally distant from the fulcrum, so that there is no effective ratio.

Before proceeding further, it is necessary to understand what was the date of the alleged invention. It appears by the record that a corporation in which the respondent below had an interest purchased from the complainant below mattresses constructed as shown in the patent in suit as early as March 12, 1891, although the record fails to show that the respondent himself had any knowledge of the • transactions. The complainant, in his patent of September 8, 1895, *339to which we hare referred, and which was applied for on March 16, 1894, used the concave-convex surfaces which we have described. The specification in that patent contained the following:

“I make no present claim to the described means oí securing the ends of the wire fabric, since such fastening, in more perfect form, is fully set forth and claimed in my application for patent on metallic mattress frames, filed Aug. 12, 1895, serial No. 559,049.”

Thus we have evidence of the patentee’s invention of the device in suit as early as March 12, 1834. We also have a patent applied for by him in March, 1894, showing the concave-convex surfaces to which we have referred; but we have no precise proof as to which form, namely, the concave-convex or the convex-flat, was first devised. The pith of the concave-convex form, so far as the patent law is concerned, is found in the patent to Segar, of March 19,1889, although the precise form there shown consists of flat surfaces against flat surfaces, and this must be held to be a part of the art at the time of the alleged invention in suit.

The difficulty' with the case at bar is that the advantages of the patentee’s form of construction over those which must be taken to be old are fanciful, or a minimum to that extent that the law cannot take cognizance of them. The only testimony on this point is that of Livermore, who says, with reference to the form of construction covered by the patent in suit, that, if the strain were continued to the breaking point, he should expect the wire to break where it passes around the outer edge of the strip, before it would pull out from between the strip and the pipe. But he also testifies that there would be no substantial difference in effectiveness whether the wire were clamped between two flat surfaces, or between curved surfaces, or between a flat and a curved surface, and that any of these fastenings, if properly applied, would hold the wire fabric securely. Therefore, if we look at the matter of any claimed advantages of the complainant’s form of construction, there are none such that the law can recognize them as the basis of a patent. It is true that, while the patentee claims that his construction contains advantages arising from the facts which we have described, he at the same time maintains that it is the simplest and least expensive known. Under some circumstances it may happen that there is invention in substituting for a complex or expensive form something simple or inexpensive. If, therefore, the concave-convex form had been a long time in use, and the patentee had discovered that the convex-flat form was less expensive, and more simple, and yet would answer the purpose, and if he had been the first to put that discovery into practice, there might be invention in so doing, notwithstanding he had made a mistaken claim that his construction had other advantages; but the case does not show snch a condition of things.

On the question of invention the complainant below urges the usual argument of a large demand for his particular structure, but there is no evidence that this arose from the peculiar feature in issue here. From aught that appears, it may have come from his superior construction, or the other claims of the patent, and especially from the fact that he omitted all adjusting devices. In like *340manner, the use of the complainant’s form of construction by the respondent does not, in this particular case, bear on the question of invention, because the record shows that this form was used by Bent before, the application for the patent in suit was filed, and without any evidence that its use was in any way suggested to him. Apparently, Bent used it as most nearly at hand, as shown by the fact that he could thus avail himself of a flat bar without going to the expense of forming a curved surface. On the whole, both parties made use of this form of construction within a few months of each other, and in the early stages of the art of making beds with metallic rails and ends, cylindrical^ formed, and each of them used what was first at hand, namely, a bar with a flat surface; and also the patented construction, so far as it differs from anything shown 'in the prior art, yields such unimportant results as not to furnish the basis of a patent, in the absence of all proofs of any peculiar circumstances appertaining thereto. The decree of the court below is affirmed, and the costs of appeal are awarded to the appellee.