French v. Buckeye Iron & Brass Works, 10 F.2d 257 (6th Cir. 1926)

Barton A. Bean, Jr., of Buffalo, N. Y. (Charles W. Parker, of Buffalo, N. Y., on the brief), for appellants.

H. A. Toulmin, Jr., of Dayton, Ohio (H. A. Toulmin, of Dayton, Ohio, on the brief), for appellee.

Before DENISON, MACK, and DONAHUE, Circuit Judges.

DENISON, Circuit Judge.

The court below dismissed, for lack of infringement, the bill brought by the French Company, based upon patent 1,140,808, issued to A. W. French, May 25, 1915, upon an application filed April 17, 1911, for a method of and apparatus for cooking oil-bearing material. Cotton seed meal required cooking to prepare it for extracting by pressure its contained oil. It had been found desirable to cook a batch of the meal in successive kettles at differing temperatures. This required manual, or manually assisted, removal from one kettle to the next. This in turn required either the skill of the expert cooker, in determining the proper time for each removal or, at least, the constant attention of the operator, if the time of cooking in a particular kettle had been predetermined. Prior to the developments to be mentioned, the pertinent art had progressed to the point where, for the saving of floor space and to get the aid of gravity, the several receptacles or kettles had been placed one above another, with a door-closed opening in the bottom of each, so that, when the door was opened by the operator, the contents would fall into the kettle below. This was known as a stack cooker; but the element of personal attention by the operator, and usually the dependence upon his skill, were not eliminated. As the art has now developed, it has very generally adopted an automatic or semiautomatic stack cooker, in which when the lowermost receptacle containing the completely cooked meal is emptied, or sufficiently emptied, mechanism is set in motion which, without further attention or manual aid, successively passes each one of the cooking batches along one step from kettle to kettle. The record indicates several *258 great commercial utilities in this automatic process, and so general an adoption of it and discarding of the old manual methods that it must be recognized as having worked a substantial revolution in the meal-cooking art.

In determining the place which the patent in suit should take in this development, we must examine both the patent art and the commercial art. Looking into the former, it seems that French was the first ever to express the conception of an automatically continuous step by step stack cooker. Some years before the actual solution of the problem, he stated it and the great benefits to be had from answering it, though the answer which he then gave was not the right one. For convenience of reference, we will speak of the successive kettles, numbered from the bottom up, as 1, 2, 3, and 4. French's first patent was 852,058, filed in 1905. In this patent he described the advantage in decreased expense and increased quality which would be had by automatically controlling the progress of the meal from one kettle to another, and he said: "The cooker is automatic in its action and the services of skilled meal cookers are unnecessary." He placed his kettles in staggered vertical relation, so that the discharge was from the lower outer edge of 4 into the upper outer edge of 3, through a depending spout. When kettle 1, for example, was emptied, there was no obstacle to the downward flow of the meal through the spout from 2 into 1, and this would continue until the meal accumulating in 1 rose to the level where it closed the spout inlet; and the corresponding transfers were taking place in the upper kettles. It is at once apparent that this was not a step by step progresssion, but that, as soon as, for any reason, the level of meal in 1 fell below the spout inlet, meal would begin to pass from 2 into 1, and from 3 into 2, and from 4 into 3, and the transfers between the different kettles would be simultaneous, and mixtures of the batches with each other would result. French did not, in this patent, obtain any claims upon the automatic process in any broad way, but confined himself to these spouts, extending into the kettle below, so as to be choked by the rising material. Clearly he did not disclose what is now said to be the essential element of the successful process, viz. that the openings from 2 to 1, and from 3 to 2, etc., should remain closed until the kettle below is emptied, and that the bottom door of each kettle, after being opened, should be closed again before the one at the top is opened, so that in this way there can be no mixture of the batches or transfer of any material through one kettle until it has remained therein the predetermined time.

The next step in the patent art was also taken by French (patent 909,778, filed April 23, 1908), and he provided a baffle plate or dam, adjacent to the opening and depending into the lower kettle, which aided the tendency of the meal in the lower kettle to pile up and choke the opening. Still the necessity of preserving the integrity of the batches was not observed, or means therefor provided.

The next patent in this branch of the art is that of Faherty (patent 909,773, filed May 2, 1908). Faherty seems to have been commercially associated with French. The Faherty patent recited that it was an improvement upon French's patent 852,058, and it was issued to French as Faherty's assignee. It provided, for the first time, controlling gates which positively closed the bottom openings in the kettles. The gates were pivoted at one side of the opening, depended into the kettle below, and it was the action of the rising meal accumulating in the lower kettle that pushed up the gates and held them closed hence "float gates"; but, as soon as the level of the meal in the lower kettle fell even a very little, the gate dropped by gravity into a partially open position, and the progressive operation and the confusion of batches were substantially the same as French's. French tried again, by patent 1,000,675, filed November 17, 1908. This, however, was a division of his application, which resulted in patent 909,778, and shows only the means shown in that patent for controlling the flow of meal, although it says that other suitable means, like those of the Faherty application of May, 1908, may be used instead.

This was the state of the patent art in April, 1911, when the application for the patent in suit was filed. There had been no attempt by any one at any time, except by French and Faherty, to produce an automatically progressive step by step batch meal cooker.

In the commercial field, French or his company evidently regarded the above-described Faherty form, having the swinging gates, called "float gates," as being for that or some other reason most suitable for sale, and the French Company began to manufacture that form. The first sales do not appear, but in the summer of 1911 several of these stack cookers, built by the French Company *259 in the specific form of this Faherty patent, were in use in cotton seed oil mills in the South. They were in a sense automatic, and were doubtless so named to the trade; but they had the inherent defect above stated. At that time, the French Company and the Buckeye Company, defendant herein, had been long-time rivals in the manufacture and sale of the older forms of nonautomatic, but manually operated, cookers, in stack or other group form, and we find during that summer a succession of letters from the defendant's salesman in this territory to it to the effect that, while the automatic cooker which was being sold by the French Company was only partly successful, and would not do nearly as much as was claimed for it, yet it was better than the old forms, and the Buckeye Company must get up an automatic cooker that would be as good, or better, or else go out of business. This is the earliest suggestion of defendant's interest in the subject-matter.

After French and Faherty had thus made four unsuccessful attempts to devise a really automatic batch cooker, French, by the patent in suit, stated the precise difficulty which had so far interfered, and also disclosed the remedy. In addition to the same general advantages which he had recited in his previous specifications of structures, in which his expectations had not been fully realized, he stated that the object of the present invention was "to provide an improved method of and apparatus for cooking oil-bearing meals, by which all of the meal remains in the cooker for a predetermined period of time and is cooked uniformly." After thus generally describing his purpose, he then particularly stresses the completeness with which he preserves the integrity of each batch as it passes through. He says:

"When the bottom kettle has been emptied, or a predetermined quantity of meal drawn therefrom, the other kettles are discharged periodically and in succession by mechanism which first opens the gate of the second kettle to discharge its contents into the bottom kettle and closes this gate, then opens and closes the gate of the third kettle to refill the second kettle therefrom, and closes this gate before opening the gate of the next kettle above, which operation is continued until the meal is delivered from each kettle of the series, except the bottom kettle, to the kettle next below. In this way substantially uniform quantities or charges of meal pass from each kettle to the next at regular intervals, and a definite period of time, depending upon the rate at which the meal is drawn from the bottom kettle, is required for a particular batch of meal to pass through the cooker, with the result that all of the meal is uniformly cooked."

He shows, for the first time in the art, mechanism for positively operating the gates which close the bottom openings, instead of having the gates merely floating upon and rising with the accumulating meal in the lower kettle. He then describes the mechanism shown in the drawings for this purpose, stating that this shows his preferred form of construction, but that "other suitable mechanism for operating the gates to discharge the kettles successively in the described manner could be employed." In the form shown in the drawings, each gate was tilted by a horizontal rock shaft extending outside of the casing. Each rock shaft was positively operated by a lifting rod, and each of the three lifting rods was operated by variant cams on a horizontal shaft at the bottom. As this shaft was slowly rotated, the appropriate cam would elevate the lifting rod which would open the gate from 2 to 1, but the lifting rod governing the gate from 3 to 2 would not yet be affected by its cam. The cams were so proportioned and placed that the gate from 2 to 1 would be opened wide and held there a substantial time, enabling the contents of kettle 2 entirely to discharge; then, as this cam finished its passage beneath its lifting rod, that rod dropped sharply by gravity and closed the gate. Immediately thereafter the next cam opened the gate from 3 to 2 and held it open, and then, in turn, permitted it to close, before the third cam opened the gate from 4 to 3. The cam shaft at the bottom was driven by a gear wheel, and this, in turn, driven by a constantly moving worm shaft connected with suitable driving power. The gear wheel was mutilated by the removal of enough teeth to span the space of contact with the worm. The result was that, when the mutilated part of the gear was opposite the worm, there was no engagement, but when the gear was rotated very slightly it made contact with the worm, and then was carried around for one nearly complete rotation. The cams were so adjusted and arranged that one complete rotation of the cam shaft first opened and then closed each of the three gates in succession.

As French expressed it: "The operation of the cooker is thus made automatic, the gates being actuated in succession to deliver the meal from one kettle to the next one below." *260 In his preferred form he also contemplated that there should be an automatic starting of the cycle, and he accomplished this by providing a float on top of the meal in the bottom kettle, which float was connected to a pivoted lever arm extending out near the edge of the gear wheel, so that, when the float fell to a predetermined point and the outer end of the arm rose, it actuated a dog, which rotated the gear wheel slightly and put it in contact with the rotating worm. Of the starting device he said: "This can be done by hand, but it is preferably accomplished automatically," etc. Other portions of the specifications, referring to further features, are not now important.

It may be said that the cooking art, or even the art of cooking meal for oil extraction, was a crowded art; but in that part of the field devoted to automatic batch progressive cooking French was practically the only occupant for several years, and through several efforts. He was the first to observe the final difficulty, and to correct it by providing positive mechanism for automatically successive gate operation. Before that, no one had done this. Since, no one has failed to do it. His advance involved meritorious and substantial invention. Giving it that relatively pioneer scope to which it is without dispute entitled, if the field is confined to the automatic progression of integral batches, it is apparent that French was entitled to four groups of claims: First, those relating somewhat broadly to the conception that each bottom gate should be opened, the kettle emptied, and this gate closed, all before the gate next above was opened, and this conception might rightly be expressed both in terms of method and in terms of means; second, those based on this broad conception, but covering further the subordinate idea that the gate-operating mechanism should come to rest after each cycle of operation, and remain at rest until again started; third, those resting on the same basis, but contemplating, also, that the chain of mechanism should be automatically started by the lowering of the meal in the bottom kettle; fourth, those more specifically relating to the selected mechanisms by which the thought of the other groups was carried out.

Claims were finally allowed and issued representing these four classes, and some of the first, second, and third groups and one of the fourth are here sued upon. We classify them thus: In the first group claims 2, 7, 11, 12, 18, 19; in the second, 15, 16, 21; in the third, 3, 6, 17; and in the fourth 20. Claim 18 is given in the margin.^[1]

Pending the application and before any cookers had been put on the market, and instead of the specific gate-operating mechanism shown in French's drawings, Faherty devised another form therefor. It is shown in his application, filed June 8, 1914, which resulted in patent 1,112,128, issued to the French Company as Faherty's assignee. When this was first put out does not appear, but it was the form which the French Company adopted as most suitable for exploitation in carrying out the step by step, integral batch, automatically progressive conception. It is by plaintiff characterized as adopting the broadest conception of the French invention, as we have stated it, and uniting therewith assistance from the older conception that the accumulating meal rising in a kettle should operate a float gate. Plaintiff insists that it infringes both the process claims and some of the mechanical claims of what we call the first group. In this Faherty-French device, if we should assume kettle 1 to be emptied and the others full, the cycle was started by tripping open the gate from 2 to 1, the meal in the former being discharged into the latter, and as it accumulated raising a float which actuated a lever arm, which finally closed the gate and locked it in the shut position. The chain of mechanism thus started then proceeded to open the gate from 3 to 2, and the operation continued successively and automatically until the shift of batches was complete. In this form of structure, marked as patented under both the patent in suit and the second Faherty and other patents, the device sold by the French Company has had a large acceptance, and, apparently with no persistent competition, excepting from the defendant company, has occupied the whole of this particular field.

As against the validity of the patent in suit, many earlier patents are cited. We see no occasion for referring to those in the meal-cooking art, for no one of them disclosed or suggested the patented subject-matter. The nearest approach to the idea, in any art suggesting analogy, is found in some grain driers, *261 best illustrated by the Rhodes patent, 92,096, of June 29, 1869. This is a "device to operate the tilting floors of malt kilns." The floors, one above another, consisted of a succession of over-lapping slats, all of which could be tilted at once, and would thereby drop the malt onto the floor next below. A sprocket chain carrying a trip successively engaged corresponding trips on the tilting devices. In this broad way, it resembled defendant's Trace form of cooker; but the resemblance goes no further. Whatever analogy there is in the arts is so remote that the thought of transfer is not natural, and, in order to be useful with a meal-cooking process, the device must be reconstructed in almost every particular. Potts-Creager Case, 15 S. Ct. 194, 155 U.S. 597, 39 L. Ed. 275. The claims in suit were allowed by the Patent Office over Rhodes, when it was pointed out that the essential principle bottom closure before top opening was not taught by Rhodes, nor shown, unless obscurely. Claim 18, in suit, is not fairly capable of an interpretation broad enough to cover Rhodes.

In our judgment, the most forceful attack upon the validity of the broad mechanism claims of the patent lies in the proposition that there is no invention merely in doing by automatic machinery what had formerly been done manually. While in a limited way this is often a right conclusion, it may not be safely taken as a general rule. Though, after French had the complete conception of carrying out this process automatically, so as to have the necessary time relation between the opening and closing of the doors, there may have been no great difficulty in devising the specific means, yet this cannot serve to invalidate his patent, where the conception was new, the means not obvious, and the utility great.

Attack is made upon the patent, also, because it is said the patented device has never been manufactured. This proposition, if true, is of so little importance (Paper Bag Case, 28 S. Ct. 748, 210 U.S. 405, 422-430, 52 L. Ed. 1122) that it does not justify the time which has been spent upon it; but it is not true. It is based on the fallacy that the patent is confined to the particular form shown in the drawings as the preferred one. This specific form has not been manufactured; but, if the patent is confined thereto, that is the end of the case, for the defendant has never made that form. We do not need to decide whether the second Faherty form, which the French Company adopted for manufacture, embodies the fundamental idea of the patent in suit, infringes some of the claims, and constitutes commercial use and adoption of the patented invention, nor whether, even if this were not so, and even if it were important that there should have been some other manufacture and sale of the patented structure, that would be found in the use by Procter & Gamble, and on a considerable scale, of the Carr form, built under a patent, some claims of which were later adjudged to French and were included in the patent in suit.

In view of the scope of the invention, as we interpret it, and the rule of strict proof in such cases, the alleged anticipating uses do not require serious consideration. Neither is there force in the claim of abandonment of claims 13-18. These claims were allowed in January, 1912, in an application filed in 1910, disclosing a somewhat cruder apparatus; but in the meantime the 1911 application had been filed, with a disclosure ample to support the old claims, as well as its own, and it was decided to let the 1910 application lapse, and transfer its worth-while claims to the other case. In this we see no symptom of abandonment; nor were there intervening rights, such as might sometimes arise by way of estoppel.

We do not overlook that the specification sets out an additional particular method and advantage which were perhaps impossible of attainment; but that should not prevent due protection for the results which were accomplished.

Coming to infringement: Inspired seemingly by the necessities stated by its salesman, the defendant, in the summer or fall of 1911, got out its first device of this class. Then, or later, it settled down for its commercial form on that shown in the Trace patents, 1,115,133 and 1,254,745, assigned to the defendant. This structure is not to be distinguished from that of the patent in suit, save as to the specific means by which the successive gate opening was accomplished. Instead of a series of cams on the shaft, which, through lifting rods, successively engaged the rock arms of the gates, Trace used the same shaft to drive a sprocket chain, which chain carried cams or trips which successively engaged with the rock arms. Trace also provided that this sprocket chain should come to rest after each complete revolution, and so remain until it was again started. It is this form, save for one change, with which the defendant continued to be on the market in competition with the French Company until this bill was filed.

*262 The fact that defendant is manufacturing the structure shown in the later patents to Trace is of no importance upon the question of infringement. See, e. g., our decisions in Herman v. Youngstown, 191 F. 579, 584, 112 C. C. A. 185, Curry v. Union Co., 230 F. 422, 429, 144 C. C. A. 564, and General Co. v. Electric Co., 243 F. 188, 193, 1007, 156 C. C. A. 54, 664, and that of the Second C. C. A. in Gordon v. Turco Co., 247 F. 487, 490, 159 C. C. A. 541. Indeed, seldom does the fallacy of the theory that manufacture by defendant under a later patent tends to negative infringement appear so clearly as in this case. When Trace filed his application for patent, he made several claims which were probably broad enough to read upon the French patent now in suit. They were rejected on reference to French, and Trace amended them, so as to limit them to the combination of the fundamental French conception with the specific improvements devised by Trace, and in this plainly tributary form the Trace patent issued. In his specification Trace sums up his device, about in French's language, when he says that it is "positive in its operation, and insures that the gates will be actuated successively in a prearranged order, and that each gate will be held open a predetermined length of time."

The contents of the French file wrapper have been carefully examined. Instead of justifying limitations upon the apparent scope of the broad claims, the Patent Office history shows that the theory of the invention, which we have earlier stated, was consistently pressed upon the Patent Office, and was finally accepted by it. Indeed, the Examiner took affirmative part in getting the claims into final shape.^[2]

The effect of the argument made by appellee against infringement is that the claims in suit must be limited to the specific structure wherein a falling float in the bottom kettle automatically trips into action the mechanism for opening and closing the successive gate doors. It is quite impossible to accept this argument. The French specification and drawing showed this construction, but we find that the specification expressly stated that, instead of this full automatic action, the starting gear might be manually operated, if desired, and then we find that the patent contains a group of narrow claims limited to this full automatic action, and a group of broader claims not so limited. Under familiar principles the limitations of the former cannot be imported into the latter.

Claims 18 and 19 of the first group, 17 of the third, and 20 of the fourth are clearly infringed, if given what we think their proper scope. This the defendant's expert admits. Claims 7 and 11, which are in terms of process, are infringed (contributorily); but it may be suggested that they are in form which covers the old hand process, excepting as they contain a general limitation to automatic mechanism. Whether a valid process patent may rest alone upon that distinction would require consideration, if it were a controlling question; but, in view of the broad claims for means, which we sustain, we do not see that it is now important to the parties, from any point of view, whether or not defendant also infringes claims 7 and 11, and this we do not decide.

We take the same course as to those of the second group those which are characterized by the "cycle and stop" action. The Trace patent indicates a construction which would infringe; the testimony indicates that upon most or all of the machines put out by defendant this structure was modified, so that the action would normally be continuous, but so that the operator might adjust it to get the "cycle and stop" effect, and also indicates that the operator would be expected to do this at least occasionally. We do not see that it is worth while to follow this to a conclusion; and, while we are satisfied that these "cycle and stop" claims are valid, the question of infringement is left open. The same course, also, is taken as to claims 2 and 12.

The decree below is reversed, and the case remanded for the entry of the usual interlocutory decree in compliance with this opinion.

[1] The combination of a plurality of heating chambers arranged in a descending series, and each chamber above the lowest having a discharge opening arranged to deliver the material into a lower chamber, gates controlling said discharge openings, and driven mechanism which opens and closes the gates of said chambers one after another at regulated intervals.

[2] He said: "The claims do not seem to bring out with sufficient clearness the fact emphasized in pages 6 and 7 of the last argument, namely, that, when a predetermined quantity of material has been discharged from a kettle, the gate controlling the discharge port is positively closed before a charge begins to feed into this kettle from the one above. In other words, the claims do not seem to clearly recite the positive interruption of the discharge before the filling begins, and, further, this fact seems to distinguish from the patents of record."