Circuit Court of Appeals, Second Circuit.
Jacob M. Zinaman, of New York City (Samuel E. Darby, Jr., of New York City, of counsel), for appellant.
Alan N. Mann and George T. Bean, both of New York City, for appellee.
Before L. HAND, SWAN, and AUGUSTUS N. HAND, Circuit Judges.
L. HAND, Circuit Judge.
This is the usual suit in equity upon patent 1,854,600, for a product and process; the only claims in suit are for the process and only four of those, 1, 2, 3 and 5. The product is a chemical combination of carbolic acid (phenol), and formaldehyde, whose supposed novelty lies in its physical, as distinguished from its chemical, characteristics. It is in the form of a "gel," a word denoting the hardening of a colloidal solution, or "sol." In a true solution the molecules of the dissolved substance are thought of as being dissociated, separate "little masses" moving freely amid the molecules of the solvent; in a colloid the theory is that they group themselves together and move in clusters which keep their integrity. A well known reaction between *594 phenol and formaldehyde takes place when the two substances are heated in the presence of an alkaline catalyst, and the "condensation product" takes a colloidal form. This colloid may then be neutralized by acid, and distilled until it becomes a "gel," that is, until enough water is driven off to trap the rest as imprisoned droplets in microscopic cellules from which they cannot escape. In this stage the "gel" is still viscous so that it can be poured into moulds where it is hardened by further heating. An opaque stuff results, which looks like ivory, is tough and has high tensile strength, owing to the droplets evenly distributed throughout its reticulated texture; the minuteness of these determine its gross characteristics. The defendant attacks the patent upon several grounds; it says that the patentee is estopped to read the claims as it must read them to succeed; that the specifications are insufficient for the practice of the invention; that the putative invention was anticipated, and that so far as it was not, it was obvious; that the defendant does not infringe; that the judge erred in refusing to admit a later patent of the plaintiff's. Judge Galston has written an opinion (D. C.) 11 F. Supp. 603, which finds the facts fully; and, except in so far as we indicate otherwise, we accept his findings as we do his conclusions.
The first question is as to the supposed estoppel arising from what took place in the prosecution of the application through the Patent Office. What is now claim one previously read as follows: "A process for the manufacture of condensation products of a phenolic body and formaldehyde, which comprises producing a colloidal solution of a condensation product the composition of which is in the proportion of one molecule of phenolic body to about two and a half molecules of formaldehyde, by reducing the dispersity of a highly disperse system and thereafter allowing the sol thus produced to set into a gel which is then hardened by heat." In place of the italicized words the following were substituted: "In the presence of excess of alkali, neutralizing the excess alkali and thereafter removing sufficient water from the sol to give a gel and thereafter hardening the gel by heat." We have often said that we would not look at the arguments used by solicitors to the examiners; specifications are intended to be the measure of the monopoly and of the contribution to the art. In each aspect they should be self-contained; that is the very purpose of their embodiment in a formal grant, which is all that is accessible to the public without much trouble and vastly more uncertainty. If the doctrine of the "integration" of a written instrument has any basis at all, surely it should apply to such a document, for if a patent can be construed only by threading one's way through all the verbal ingenuities which casuistical solicitors develop to circumvent the objections of examiners, a labyrinth results, from which there is no escape. For this reason we have steadily refused to look further than to learn this: whether a patentee who seeks to disavow an element of his claim, was forced to introduce it in order to avoid rejection. Westinghouse Electric & Mfg. Co. v. Condit Electrical Mfg. Co. (C. C. A.) 194 F. 427; Id. (C. C.) 194 F. 430; Auto Pneumatic Action Co. v. Kindler & Collins (C. C. A.) 247 F. 323, 328; A. G. Spalding & Bros. v. John Wanamaker (C. C. A.) 256 F. 530, 533, 534; Baltzley v. Spengler Loomis Mfg. Co. (C. C. A.) 262 F. 423, 426; General Electric Co. v. P. R. Mallory Co. (C. C. A.) 298 F. 579, 586. It is clear that the amendment just set out does not come within this exception; the phrase, "reducing the dispersity of a highly disperse system," was really not descriptive of any steps in the process, but of the result of several of them; formally at least it was therefore objectionable as functional; the substitute described the process and introduced no new element which the plaintiff now asks to disown.
The next challenge is to the sufficiency of the disclosure for practicing the invention. It is quite true that the degree of heat and the length of heating the original mix are nowhere given. The general directions (page 2, lines 37-42), are merely to heat the proper proportion of phenol and formaldehyde with so much "condensing agents having basic properties" that the product will remain in solution. In example one the condensing agent is added when the mix (with a little hydrochloric acid added) has been boiled till the "condensation product" separates. It is not heated at all thereafter, and the acid is added when the exothermic reaction "begins to weaken." In example two the first mix contains very little of the condensing agent and only a third of the formaldehyde; it is boiled till an "oily layer" separates off. Then the rest of the formaldehyde and a little acid are added and it is boiled again for thirty *595 minutes, when the bulk of the condensing agent is added, after which it is boiled a third time till the exothermic reaction has "ceased." In example three the mix is complete at the start, and is "heated" until the exothermic reaction "sets in, after which" the acid is added; that is, during, or after the conclusion of, the reaction. It seems to us that from all this the extent of heating both in time and degree is plain enough, though not quantitatively specified. Whether the process so described will produce the stuff is another question. Again as to distillation, the general direction is to distill until "the sol * * * gelatinizes in the presence of heat" (page 2, lines 18-20); "the resulting sol being caused to gelatinize" (page 2, lines 43, 44); is "thickened to gelatinization" (page 2, lines 119, 120). The examples say that the distillation should go on till "the product is clear" (page 3, line 49), and there is a difference in the testimony as to whether that may not be before the distillation is complete. We need not resolve that dispute, because it is plain that the disclosure means that the "sol" shall be concentrated by distilling (heating in a partial vacuum) until it is thick enough to become a "gel." Though the time is not given in minutes, the heat in degrees, or the vacuum in subatmospheric pressure, there is not the least reason why it should be. If any one found that when "the product is clear," it does not "gel," he would, if he had common sense, distill it further. In practice the gelling point is determined by trial, just as a cook tries out her jellies. Finally, it appears that in the final stage hardening the "gel" the practice is to heat for from five to seven days, according to the size of the batch, rather than for the four days specified in the examples. Once more, if one were bent on making the process fail, this might be an excuse; but people who are familiar with the customary manipulation of these "condensation products" ought not to find difficulties. So much therefore for the formal sufficiency of the directions. As to their adequacy to direct skilled persons, it is enough to say that Judge Galston, who saw the witnesses, believed that they had actually made the stuff by following the recipe; we see no reason to question that finding.
The only stricture that seems to us at all serious is that the specifications do not expressly forbid the use of hydrochloric acid as a neutralizer, though, according to the plaintiff's testimony, this would prevent the formation of any "gel" at all. The general directions merely are that the solution shall be "neutralized" (page 2, line 17; line 119); and hydrochloric acid in very small quantity is part of the initial mix in example one, as we have said. On the other hand, in all the examples the neutralizing acid is monochloracetic, and the "preferred method," though mentioning many others (lines 16-19), does not include hydrochloric acid. The worst that can therefore be said is that a practitioner, not finding hydrochloric acid forbidden, might disregard both the examples and the "preferred method" and fail. As we have just said in connection with times and temperatures, it does not seem to us that such a person would stop at that if he were in earnest; or that, if he were reasonably adept, he would need invention to turn either to the examples, or to the "preferred method." Nobody would be likely to go into commercial production without some experimental trial of the process; if he did try it, his vade mecum was at hand within the four corners of the document.
The next and most important question is of invention. The defendant chiefly relies upon four patents: Knoll (British, 6430 of 1911); Lorival (British 156,675, not accepted); Baekeland, 954,666; Aylsworth, 1,111,288; of which the first is the closest. Its chief purpose was to keep the "condensation product" as colorless as possible; it was an improvement upon Knoll's other patent, No. 27,096 of 1908, which hardened "condensation products" of phenol and formaldehyde by the use of acids; the improvement was by adding more formaldehyde to avoid darkening the stuff. The second example prescribed a ratio of one molecule of phenol to 2.15 of formaldehyde, within the limits of the disclosure of the patent in suit, and near to the proportion used in practice, 2.35. The formaldehyde together with an alkaline condensing agent, was added in two doses, one at the start, the other after the separation by heat of the product as a yellow oil, like example two of the patent in suit. This intermediate product was then boiled for thirty minutes, after which between five and fifty per cent. of 20 per cent. hydrochloric acid solution was added without any distillation. The result was a "tough mass," which could be "pressed into moulds" and hardened "by short heating." McKee, the defendant's *596 expert, a man of exceptionally high attainments and standing, did indeed say that the kind of acid used to neutralize had no effect upon the result. Neville, the plaintiff's expert, something of a specialist in this field, thought otherwise. It might be hard to choose between them, were it not that a product of the process showed very plainly in the microphotograph that it was not physically organized as those made with other acids. It is quite true that Flood, the plaintiff's experimenter, did not say how much hydrochloric acid he had put in; only that he used a 20 per cent. solution. But Knoll's limits were themselves between five and fifty per cent., as we have said, and there is no reason to suppose that Flood did not keep within so wide an option; unhappily nobody seemed to think of this feature at the time. It may indeed be true that between five per cent. and fifty per cent. there is a point where 20 per cent. hydrochloric acid used as a neutralizer will turn out a stuff in which the water is held in microscopic droplets by a colloidal "gel" concentrated out of the "condensation product." The plaintiff need not disprove it, for an anticipation must speak affirmatively and with certainty; must disclose the invention without debate; it is not enough that the proper proportion of that acid might empirically be hit upon. Furthermore, Knoll says nothing of distilling the intermediate stuff; it is merely to be "treated" with acid and "rinsed" with water. As the temperature at which it is to be "treated" is atmospheric, it is difficult to see how any distillation could have been intended. Apparently the acid would at once precipitate the "tough mass."
Baekeland's patent, No. 954,666, was for a varnish, not in the least like the stuff that these inventors and Knoll were after. The molecular proportion of phenol to formaldehyde, properly reduced for comparison with the disclosure in suit, was one to one and a quarter, just one-half the optimum of the patent. To make even a plausible use of the disclosure the defendant is obliged to seize upon the following passage (page 1, lines 65-68): "In fact it is still possible to obtain acceptable results if the amount of formaldehyde indicated above is doubled or reduced to one-half." A varnish at the outer limit of "acceptability" is a far cry from the stuff made by the process in suit. Moreover the disclosure does not prescribe neutralization, though there is a suggestion, no more, that the excess of alkali may be neutralized (page 2, lines 35-39). The claims call for an alkaline stuff, which excludes the disclosure in suit unless we were to accept the defendant's construction of the phrase, "excess of alkali," in claim one, a matter to be dealt with later.
Lorival's unaccepted British patent, 156,675, showed better in microphotograph than any of the others, the mottled surface indicating a small and equal pattern of droplets, which probably, though not necessarily, accounts for its opacity. The plaintiff distinguishes it chiefly for two reasons; the use of hydrochloric acid, and of dry formaldehyde. The second seems to us a meagre distinction; as to the first perhaps the plaintiff may bear as hard as that upon the defendant's burden of proof, but it is not very likely that two and a fifth per cent. of hydrochloric acid could do serious harm. However that may be, the proportion of formaldehyde to phenol was far from the plaintiff's, and that, in part anyway, determines the size of the colloidal groups and of the resulting cellules. The highest molecular ratio disclosed by Lorival is one to one and a half, and the only surprise is that his result as shown in Exhibit 49 was as good as it turned out to be. As for Aylesworth, 1,111,288, it is also enough to say that he prescribed an equal or less proportion of formaldehyde to phenol (page 2, lines 101-104; page 3, lines 8-12). The defendant has mistaken for a part of the recipe a passage (page 2, lines 74-96), which merely describes and disapproves processes in which the proportion of formaldehyde is greater than that recommended; and even that passage does not disclose the proportion of the patent in suit. So much for the four putative anticipations.
The defendant also put in evidence ten or fifteen other patents going back to 1903, twenty-three years before this application was filed in 1926. It is not necessary to discuss them in detail, for they were offered only to show how little of the field was left for occupation by these inventors. It is of course plain the four patents which we have discussed prove it without more that the reaction of phenol and formaldehyde was known and used long before; so too was the acceleration of that reaction by an alkaline catalyst; so too was the addition of an acid to neutralize the resulting product; so too was the distillation of the water and the hardening of *597 the end product by heat. But nobody, so far as we can find, had ever treated a mix of the prescribed proportion by all these steps. Knoll approached that proportion, as we have seen, but he had not treated it similarly. The difference might be of little or of very great, importance; it might be a step which the art was sure to take, or it might demand a high degree of originality; it is impossible to decide such questions à priori, and it is a mistake to belittle the invention because it demanded only a trivial variant. The true question is what new understanding was a condition upon its appearance, however inconspicuous by itself the variant might be. In the case at bar the disclosure was a conscious and thoroughly understood effort to reach a form of matter whose gross characteristics in use, and therefore whose value, depended upon a physical structure that nobody before had observed; that is, the permanent entrapment of microscopic droplets of water by the congealed colloidal clusters of molecules of the old "condensation product." The novelty of the invention rests in that discovery; its validity as an invention in the fact that it did not remain merely a discovery, but that the inventors disclosed a formula, a recipe, by which in ways in general well known, though new as an entirety, that discovery could be made useful to the art at large. Without this information nobody could have made the new stuff; indeed nobody would have comprehended its significance if he had hit upon it by chance. If there are to be any patents at all, it is idle to say in such a setting that the invention only carried forward what was already implicit in what had gone before. The art had been full of experiment for over twenty years, and though no technical obstacle existed and the result would always have been equally acceptable, no one had perceived this correlation between structure and gross characteristic.
As to infringement, it is true that the defendant does not strictly follow the disclosure in its commercial practice; nor indeed does the plaintiff whose practice is the same. Still the only verbal departure from the claims is in the moment at which the acid is added as a neutralizer; all four of the claims in suit describe this as taking place before distillation begins; it has been found possible, and indeed commercially preferable, to neutralize as the distillation goes on. Literally the defendant does not therefore infringe and if there is no room for equivalents, it would escape. We do not see why we should treat the invention so severely, although it was not indeed a pioneer. The extent to which it is ever proper to stretch patent claims depends upon the importance of the invention and the extent of the distortion. Claude Neon Lights v. Machlett & Son, 36 F.(2d) 574, 576 (C. C. A. 2); Directoplate Corporation v. Donaldson Co., 51 F.(2d) 199, 202 (C. C. A. 6). Perhaps there are patents so thin as to deserve no latitude whatever; but, if there be, this is not one, based as it is upon a fruitful discovery. We may read the word, "thereafter," in claim one, and "then," in claims two, three and five, as meaning that the distillation need not follow the addition of all the acid, but that the acid may be added as distillation proceeds. The time and degree of heat of the hardening process give us no trouble in point of infringement; the claims merely say that the "gel" is to be hardened by heat. We need scarcely labor the argument that this phrase did not so incorporate the times and temperatures given in the examples as to emasculate the claims. Finally, the phrase, "excess of alkali," in claim one does not mean so much alkali as is in excess of the prescribed minimum, which in turn is to be sought in claim five; i. e., .32 per cent. It means no more than that the mix as such shall be "strongly alkaline * * * at any stage of the condensation product" (page 2, lines 126-128); and shall be wholly "neutralized" by the acid. The defendant's fanciful interpretation has no support in the specifications and contradicts their natural implication.
There remains only the judge's refusal to admit in evidence a patent to one, Pantke, owned by the plaintiff, but issued later than the patent in suit. During its prosecution the solicitor made declarations which the defendant wished to use as admissions against interest, thus entangling the patent in suit not only in all that that solicitor might say who secured it, but in whatever another solicitor might say in securing another patent for the same client. Certainly it is an answer to this that such declarations should not charge the patentee at all. When an inventor retains a solicitor, the solicitor's declarations no doubt do charge him when relevant, so far as concerns the issuance of that patent; his retainer implies that that solicitor may in reason do *598 what he thinks best to secure it. With equal certainty it does not imply a like authority to affect a patent already granted even to the same inventor. The solicitor has no warrant for assuming that to secure the new patent his client would wish to throw doubt upon the validity of the first, or to limit its scope. The judge was therefore right so far as any declarations in the file wrapper of the Pantke patent went.
But when he ruled that that patent was not formally competent on the general issue of infringement, under the decisions as they stand, he was apparently wrong. There can be no doubt of this if the defendant had itself owned the Pantke patent; for in that event there is a presumption that whoever followed its disclosure did not infringe. Reis v. Rosenfeld (C. C. A.) 204 F. 282; Knowles v. 138 West Forty-Second Street Corporation (C. C. A.) 43 F.(2d) 929, 930. While the Supreme Court has not been altogether consistent on the subject, as we said in Reis v. Rosenfeld, supra, its last declarations have so declared. Moreover, if there be any rational basis whatever for the doctrine, it is impossible to see a valid distinction between a patent owned by the infringer and any other; for the theory is that the Patent Office does not grant a second patent if it infringes the claims of one already issued. But we will not reverse the decree for this reason, because we can see that the patent could not conceivably have changed the result. This appears from a consideration of the duties of the Patent Office whose action is supposed to support the presumption. When it is considering an application for a patent, the only thing before it is whether the proposed claims constitute valid monopolies; that is, whether they are for some "new and useful art, machine, manufacture, or composition of matter" (Rev. St. § 4886, as amended [35 USCA § 31]), and whether the specifications support them. That inquiry certainly involves an examination of the contents of all available earlier disclosures, outside and inside the Office; and it also involves an examination of the claims of such patents as have issued. The first is necessary in order to see that the claims do not cover matter already disclosed; the second to avoid granting the same monopoly to two persons. But the presumption here in question presupposes a supposititious third duty; that is, to examine the disclosure of the pending application, and though that duty does exist, it is only to learn whether the specifications are plain enough to support the claims, and otherwise in accord with the statute and the regulations. But the Office has no occasion whatever to compare the disclosure with the claims of earlier patents; for the validity of the pending claims cannot possibly be involved in the infringement of other claims by the disclosure. Indeed it is apparent that, if the Office troubled itself with infringements, it could never grant any patent for an improvement, whenever there were outstanding claims covering the genus of which the improvement was a species. In all such cases the disclosure must always infringe the generic claims; yet it is one of the commonest functions of the Patent Office to grant improvement patents.
Decree affirmed.
