In this connection, we exhibit Figs. 1 and 2 of the drawings:

Fig. 2 is a cross-section of Fig. 1 on the line 2, 2. In Fig. 2 it is seen that the onions rest in the trough, and the tops of the onions are presented vertically downward therefrom into the space between the rollers and are engaged by the latter in that position; while in Fig. 1 the onions are resting partly on roller 21, which is the right-hand roller in Fig. 2. The left-hand roller, 20 of Fig. 2, is not shown in Fig. 1. It could not be without impairing the view of roller 21. The significance of what is shown by Fig. I will appear later. The defendants contend that claim 5 is limited to a trough which performs the whole function of supporting the onions and delivering the onion tops to the rollers, with which the onion bulbs do not come in contact. This is a vital matter for the defendants. For the defendants' machine is substantially a facsimile of the complainants' in all other respects than this. A cross-section of it is shown by Fig. 2 of an exhibit which is admitted to be a correct representation:

One round roller is shown, 4. 12 is not round, but is a revolving bar with sharp plates set out from the corners. The sharp plates almost touch the opposite roller. Both roll over toward each other, and the onion top is detached in like manner as in Vroomans' 1897 patent, except that it is cut off instead of being pinched off. This may be an improvement; but it is only an improvement, tal-n from the Vrooman patent of 1901, as we shall presently see. It is also seen

from the above Fig. 2, just shown, that a part of the surface of the roller, 4, supplies the place of the bottom of the trough. The onions coming down the trough plates 6, 7, and on the roller, 4, protrude their tops through the open space between the roller, 4, and the lower edge of the plate, 7, and become engaged between the roller, 4, and the sharp-edged plates of 12 and are cut off. These conditions establish the necessity upon the defendants of showing that the complainants' patent of 1897 does not cover a form of construction in which the onions come in contact with the rollers. And to aid in this contention they urge that the word "engage," in the next to the last line of claim 5 of the patent, includes also the idea of "coming in contact." But this conclusion is refuted by several reasons. The words themselves are not synonymous. There may be contact without engagement which means in its mechanical sense a seizure, a laying hold of, an active prehension, and does not include mere contiguity. And in this patent the words are used to signify different things. In the following claim (6) the word "contact" is used to signify another meaning than the word "engage" in the fifth claim. Stress is laid upon the following paragraph in the specifications:

"It will be seen that the purpose of the trough is to prevent the rolls from engaging the onions. If the rolls engage the onions, the onions will be bruised and peeled, but by resting them in the trough the tops only are permitted to engage the rolls."

But we think this does not aid the argument. It is the engagement of the onions between the rollers which is here meant. It is that which would bruise and peel the onions, and not the being carried along on a smooth roller so as to protrude their tops and be turned over into the opening, that would be likely to bruise and peel them. They have the same treatment in the Vrooman machine of 1901 and in the defendants'; and there does not appear to have been any trouble on that score. The gist of the controversy over the infringement of this patent depends upon this contention as to whether the patentees are limited to a form of construction which does not expose the onions to contact with the rollers or either of them. It is important to observe, in this connection, what has already been shown, that one of the forms shown in the drawings discloses the onions resting on, in contact with, a portion of the surface of one of the rollers.

We turn next to see whether the proceedings in the Patent Office require this limitation. In 1895 the Vroomans filed a caveat on their invention in the Patent Office in which they described their invention with a drawing to illustrate the form in which they proposed to embody it. They described a trough with an opening at the bottom and a pair of rollers "adapted to engage the tops of onions or other vegetables, so as to pull the tops from the vegetables and deposit the said tops beneath the machine, while the vegetables themselves roll down. and are delivered at the lower end of the machine." And they showed one form in which the bottom of the trough was wide and a roller supplied the bottom of the trough, and apparently conveyed the onions to one side for engaging the tops between the rollers. The application for the patent was filed the following year. We have already refer

red to the drawings as illustrating the forms which their invention might take, namely, one in which the onions were held out of contact with the rollers, and another in which they were in contact with a limited part of the surface of one of the rollers and the cutting off of the tops was done after the onion tops passed out of one side at the bottom of the trough, and there is nothing in the specifications to indicate that the inventors intended to limit themselves to one of these forms which varied only in a matter of detail; the main purpose of the invention being present in both. On presentation of the application, original claim 5 was rejected by the examiner on reference to corn huskers and threshers. It was canceled, and a substitute was tendered in the following form:

"5. In a machine for topping vegetables, parallel topping rollers, a trough located over the said topping rollers and having an opening over the space between said rollers, and mechanism for driving the rollers, substantially as described."

And we start with this to see what limitations the examiner proposed. He said "claim 5 is rejected upon reference to Fig. 3 of Partelow." He did not specify what his reference was intended to develop. But on turning to Fig. 3 of the Partelow patent, which is in the record, we find it indicates two parallel converging rollers and the sides of an open-bottomed trough the lower edges of which are just above the tops of the respective rollers. There is nothing to prevent the ears of corn falling between the rollers and being engaged by them. Looking at substituted claim 5, we see that in this respect it was like Partelow's Fig. 3. It had no provision for keeping the onions from being engaged by the rollers. The object of the reference is thus readily perceived. Thereupon the applicant canceled this substitute, and proposed another, which was allowed, and stands as claim 5 in the patent. If Fig. 2 of this patent had been the only diagrammatic representation of the invention, and even if the specifications themselves had only indicated this form, the result would be the same. The inventor is required by the statute to point out the best mode in which he has contemplated the application of the principle of his invention. But this does not preclude him from claiming any other mode which embodies his principle. Mr. Walker, in his work on Patents (4th Ed. § 115), in commenting on this provision of the statute, says:

"The second provision cannot mean that every inventor must infallibly judge which of several forms of his machine will eventually be found to work best, for, if it means that, it requires what is often impossible; requires the inventor to foresee the ultimate effects of new and comparative untried causes."

This is elementary. It is the foundation of the doctrine concerning equivalents. The case of Winans v. Denmead, 15 How. 330, 14 L. Ed. 717, is a pertinent illustration. There the applicant had described his invention as being one of a body for a car and gave, as a form representing his invention, "the frustum of a cone," an inverted cone, and he assigned his reasons for adopting this form. He said that thereby "the force exerted by the weight of the load presses equally in all directions, and does not tend to change the form thereof, so that

every part resists its equal proportion," etc. Notwithstanding he had represented his invention in this form and assigned his reasons for it, the court held that he was not restricted to this form, and that his patent covered any other form constructed substantially upon the principle of his invention, and upon this ground further held that car bodies constructed upon his principle, but varying therefrom in that, instead of employing the circular form of a cone, the car bodies were built in an octagonal or quadrilateral form, evidently did not possess in like extent the peculiar advantages which the inventor had contemplated. We postpone further consideration of this subject until we have examined the Vrooman patent of 1901. The description of the invention for which that patent was granted is facilitated by reference to our description of the invention in the patent of 1897. The "vegetable topper" of the later patent was constructed on the same lines in general as those of the former patent. It was an improvement upon that, and the improvement consisted in this: Instead of employing two circular rollers, the inventor inserted in place of one of them a revolving square bar having comparatively sharp right angled corners. Both revolved in the same way as the rollers of the former patent, and the only difference worth noting is that, in place of the round surface of a roller, the corners of a revolving bar co-operated in detaching the tops of the onions. As in the former patent, provision was made for withholding the onions themselves from being engaged by the detaching apparatus. Fig. IV of the drawings, here inserted, shows all that is necessary for the present purpose:

The onions come along down the trough, H and H'. The roller, G, carries them over and feeds the tops through the opening at the base of the side H' of the trough. The roller, G, and the bar, L, rolling over toward each other, gripe the tops of the onions and detach them.

Counsel for the parties devoted their attention to claim 3, and we will do likewise. It reads as follows:

H H

"The combination in a machine for topping vegetables, of a revoluble feeding cylinder, a revoluble bar presenting cutting edges parallel to and in close proximity to the perimeter of the cylinder, but of much smaller diameter; said bar being provided with collars having a diameter nearly equal to the greatest diameter of said bar, and means whereby the cylinder and cutter are caused to revolve toward each other."

The validity of this claim, so far as concerns the presence of invention, is not denied. The contention most vigorously pressed is that this claim was so dealt with in the Patent Office by the exaction of the examiner and the concessions of the patentee as to preclude the latter from claiming more for his patent than his original specification particularly described. The application the defendants make of the rule referred to in this case comes to this: That the patentee is restricted to a pair of rollers, one of which, the cutting roller, is of much smaller

diameter than the other. And they point to the fact that their cutting roller is as large as, if not larger, than the other. This is the ground on which they think they avoid the charge of infringement.

When Vrooman made his application for this patent, although the drawing (which is shown on a preceding page) showed a revolvable cutter bar of much smaller diameter than the other roller, the specifications themselves made no suggestion that it should be smaller and assigned no reason why it should be. Nor were the specifications changed in this regard during the pendency of the application. Nor did the original claim make any reference to the relative diameter of the rollers. Following claim 3 through the proceedings, we find that it was rejected on references to two patents to Rosenthal, one a threshing machine, the other a corn husker. There was nothing in the references which related to the relative sizes of the rollers. As the same references were made later on, we will postpone comment thereon to that place. The next occurrence was the presentation of an argument by the applicants' attorney against the relevancy of the references, above mentioned, to claim 3. The argument made no reference to any supposed requirement in respect to the diameter of the rollers, and, of course, no acceptance of a proposal on that subject. But the examiner again rejected claim 3 and other claims for reasons that seem trivial and inadequate, for they suggest nothing which a mechanic organizing the machine would not understand to be required, and would know how to provide for. But, at the close, the examiner says, "claims * ** 3 are rejected in view of the references of record," and particular reference is made to lines 45-48, page 3, of the specification on Rosenthal 621,505, and to lines 76-88, page 2, of 588,184. In reference to his suggestion to "references of record," there had been no other references than those to Rosenthal which he repeats, as above stated. We must suppose that the "references of record" must have been some theretofore made in support of objections made to the other claims. As the object of the references is not disclosed, it is necessary to turn to the references themselves to find the point. The defendants have put in evidence the Rosenthal patent No. 621,505 and we find therein at the place cited this language:

"The sharpened, edge or point of the nippers then severs the stalks from the ears, and said stalks are carried between and below the rollers."

It is obvious that this has nothing to do with the comparative size of rollers. The other patent referred to, No. 588,184, is not found in the record, and we have no knowledge of its relevancy. If it contained anything profitable to defendants, we must suppose it would have been produced. On a later day the appellant amended all his claims, among them claim 3, but not in any respect involving the diameter of the rollers. Then followed the misadventure which the defendants think furnishes them a good defense. On the presentation of these amended claims, the examiner made reply (of which we quote so much as is relevant to claim 3):

"The wording of the claims is found to be faulty in several respects to be exemplified below; and while there is acknowledged to be patentable matter, the structural differences constituting the novelty are not expressed in the claims, notably the first and second.