EVANS, Circuit Judge.

The District Court held claims 1, 2, and 4 of Reissue Patent No. 18,667 for a "Vermifuge and Insecticide" valid and infringed, and granted an injunction and accounting. Defendants are aggrieved. Hence this appeal.

The patent in suit, a reissue to William J. Dennis, one of the plaintiffs, was originally issued March 15, 1927, on an application filed April 16, 1923. It covered an insecticide in extract or powdered form, made from the root of the cube plant found in Peru or other South American countries. It is effective in killing insects, parasites, and other vermin which may be exterminated by contact poison.

Claims 1, 2, and 4 read as follows:

"1. An insecticide and vermifuge comprising ground cube root with the fibrous element removed.

"2. An insecticide and vermifuge comprising an extract of cube.

"4. An insecticide and vermifuge comprising concentrated extract of cube root and a carrying agent."

Coplaintiffs Stamberg and American Cube Syndicate are exclusive licensees of the patent. Stamberg was, prior to July 29, 1936, associated with defendants, who conduct their insecticide business under the name of Agicide Laboratories and sell among other products, cube insecticide compounds.

The District Court found on the controverted fact issues:

(1) No anticipation by prior U.S. and foreign patents.

(2) No prior discovery.

(3) Cube is sufficiently identified botanically by the patent.

(4) The reissue patent is not broader than the original.

(5) No showing of prior discovery in Peru.

(6) The product is of patentable subject-matter.

(7) No laches established in filing application.

(8) The reissue patent is valid, (claims 1, 2, and 4).

(9) No estoppel between plaintiff Stamberg and defendants.

(10) Defendants purchased and used cube root for their product, and not timbo root, and their insecticide infringed the claims of the patent.

The record discloses a prior art which among others included a United States patent to McDougall, for an insecticide, with derris as a base, applied for September 8, 1911, and issued June 17, 1913; two patents to Lemmens were for an insecticide, based on tephrosia of the order of Leguminosae, applied for January 16, 1917, and issued October 16, 1917. The specifications of a United States patent to Dow, filed March 31, 1920, and issued November 4, 1924, refer to the use of "timboin, nicotine and derris (or `fish poison')."

Defendants contend:

(1) The patented article is a product of nature, namely the powdered cube root, either as dust, extract, or concentrate — the latter being merely modified forms of the natural product.

(2) Prior publications, particularly one published in Peru in 1910 and received in the Library of the United States Department of Agriculture in June, 1910, described the use of cube root, in powdered or extract form for fish poisoning, and also as a "great" insecticide — for sheep ticks, etc.

(3) Anticipation by several United States patents wherein powdered forms of roots of other members of the same botanical family, leguminosae, namely, derris and tephrosia, were used in the same manner and process and with the same results.

(4) The phraseology of the patent is too broad, too indefinite, and too vague to support a valid patent.

Defendants also contend that Dennis' testimony of invention-date as early as 1917 is uncorroborated, and is rebutted by the correspondence between him and Dr. McIndoo, subsequently carried on. Defendants also contend that plaintiff Stamberg, exclusive licensee of the patent, in several states, is estopped from recovering because he promoted defendants' business, and was in partnership with these defendants, and upon dissolution of the partnership he received compensation for his share of the profits of the business based on what he now alleges to be illegal or infringing use of a product covered by plaintiff's patent.

Defendants also deny infringement, first, because they do not follow the teachings of the patent in that they do not remove the fibre as is specified in claim 1, and they do not make an extract as taught in claim 2. They also claim non-infringement on the ground that they do not use cube at all, but use timbo, which is another plant, though of the same species as cube. They assert the material which they used was mislabeled as cube when it was imported, and it was in fact timbo roots which they used to make their insecticide.

We must reject defendants' first defense, namely, that the cube roots utilized as an insecticide may not involve patentable invention under any circumstances, because such discovery is merely a revelation of an existing law of nature. We believe the importance of this conclusion justifies a somewhat extended discussion.

We believe it to be a sound pronouncement to say — The discovery of a natural phenomenon, or of a quality or attribute of a well-known article, which discovery is of value to mankind, may be entitled to patent protection. The objection frequently offered to the patentability of such a discovery is that it is a law of nature or a principle of nature and for that reason not patentable. Section 31, Title 35, U.S.C.A., authorizes the issuance of a patent to

"Any person who has invented or discovered any new and useful art, machine, manufacture, or composition of matter, or any new and useful improvements thereof * * * not known or used by others in this country, before his invention or discovery thereof, and not patented or described in any printed publication in this or any foreign country, before his invention or discovery thereof * * *."

There would seem to be no valid reason or sound support for a position which would deny to discoveries by researchers in the field of science the protection of our patent laws when such discovery is that an old, or at least well-known chemical product, will, acting in a given state, alone, or combined with other elements or physical elements, produce new, unknown, and unexpected results, whereas one who puts together at least two old and well-known chemical substances in certain prescribed proportions and gets new results helpful to man may receive patent protection. In the latter case, patent protection is universally accorded to the discovery.

In both cases it must be and is assumed, of course, that the discoverer is first, that he timely files his application, and that his said discovery is "new and useful," and was not the result of the exercise of ordinary mechanical skill or the fruit of the reasoning of the normal individual skilled in the art wherein the invention lies.

It would seem to be an unjustifiable distinction to recognize as patentable a machine or other product which puts into novel combination a plurality of old elements producing a new and useful article, and deny to one working in a research field the protection of his discovery, because in his search he finds that a certain article, which we will call a chemical, will act, when taken into the human system or applied to the earth or upon some mineral found therein, in a beneficial way hitherto unknown to man. Clearly the patent law was enacted to benefit society by encouraging discoveries and inventions. That is accomplished by the Government's giving to the inventor or discoverer a patent — a legal monopoly for a term of years.

An analysis of the essential parts of the statute is illuminating. Congress provided for the granting of patents to (a) any person, (b) who has invented or discovered, (c) any new and useful, (d) art, machine, manufacture, or composition of matter, (e) or any new and useful improvements thereof, (f) not known or used by others in this country, (g) before his invention or discovery, (h) and not patented or described in any printed publication in this or any foreign country, (i) before his invention or discovery thereof, etc.

Emphasis must be placed upon the words "invented or discovered" — "new and useful" — "art, machine, manufacture, or composition of matter."

It is true that an old substance with newly discovered qualities possessed those qualities before the discovery was made. But it is a refinement of distinction both illogical and unjustifiable and destructive of the laudable object of the statute to award a patent to one who puts old ingredient A with old ingredient B and produces a cure for ailment C, and deny patent protection to one who discovers that a simple and unadulterated or unmodified root or herb or a chemical has ingredients or health-giving qualities, hitherto unknown and unforeseen.

Let us take an illustration, in another field, of oil cracking, so valuable in gasoline production. Without pretending to give the precise figures, it might be said the problem was to refine crude oil. A chemist who had spent much time and experimented long, found that high heat caused the molecules of oil to crack or divide and a larger amount of gasoline from a given amount of crude oil was obtainable. Another researcher was not satisfied with the percentage or quality. He finally experimented more and discovered that by heating 250 degrees higher the remaining crude oil would crack again. Is he to be denied the protection which the Congress intended he should receive because his discovery was merely that of a principle of nature? Such was not the purpose or the plan of the Congress that enacted the patent law in 1790.

The statements, "The laws of nature," "the principles of nature," "the fundamental truths," etc., are not patentable, have been oft repeated but seldom understandingly used. They have led to misunderstanding and much confusion, not limited to members of the bar. In fact, the words "laws of nature," "principles of nature," and "fundamental truths" are all words of broad and also elastic meaning and are frequently used carelessly and without any attempt at refined distinctions.

In fact, they have not been legally defined or defined with such accuracy as to permit of wide or safe application. Like the term "aggregation" they have been invoked indiscriminately by members of the patent bar to cover many kinds of defenses and often when counsel are desperately searching for even a paper defense. They are thus used as somewhat synonymous with "general principles."

Obviously, there are operations of the law of nature which are not patentable. On the other hand, there have been discoveries made in the scientific field which in a sense are phenomena of old and well-known products which cannot, have not, and never will be properly described as laws or principles of nature.

The discoveries that a cube root would act as an insecticide, that copper and iron when properly mixed in predetermined proportions (Wisconsin Alumni Research Foundation v. Breon Co., 8 Cir., 85 F.2d 166) would increase the red corpuscles in the human blood, are not properly understood, nor are decisions which uphold them sound, if we assume that they are discoveries of principles of nature. They are not.

Each of said discoveries necessitated the co-acting of two or more things. The insecticide needed the breath of the insect upon which the powdered cube root could act before it became an effective insecticide. The copper and iron mixture needed contact with the human blood before a change in red corpuscle count occurred. Seldom is there any discovery of a new phenomenon of an old chemical product that does not call for the old product's contact with a material to which it must be applied by human agencies before the phenomenon occurs. In all such cases the discoverer is well outside of the rule which excludes the issuance of patents to those who have merely discovered a law or principle of nature or fundamental truth.

The statute above quoted, which has remained for 150 years almost in the exact language of its first adoption, evidently intended to include all worthwhile discoveries, for it includes inventions and discoveries. The words "invented" and "discovered" are not synonymous nor were they intended to be. However, all patentable inventions or discoveries must be of a new and a useful thing. The subjects which must be discovered or invented were then defined in the comprehensive words — "art, machine, manufacture, or composition of matter or any new and useful improvement thereof."

How are we to distinguish between "art," "machine," "manufacture," or "composition of matter" if we do not make the four of them as broad and as inclusive as the mind can conceive or the words justify? And this conclusion is strengthened by the use of the words "inventions and discoveries."

One might readily apply a more restricted meaning to these words if only inventors thereof were being considered. We know what it is to invent a machine, but what about the "discoverer" of an "art," or a discoverer of a "composition of matter"? The word "machine" is not used in a restricted sense and yet it must be distinguished from "art," from "manufacture," and from "composition of matter."

True, it is that patents do not cover results or functions. Yet, the test of usefulness must often be found in results and functions. They can not be entirely ignored, though not the valid subject of a patent.

The conclusion on this phase of the case seems unavoidable. It is that Congress meant to be comprehensive and inclusive in patent coverage and liberal in patent protection, provided "any person" either invent or discover a new and useful art, machine, manufacture, or composition of matter. It meant to cover all inventions and discoveries which were new and useful. Both the intent and the words of the statute negative limitations. Comprehensive rather was the patent protection to be, provided the inventor or discoverer made a new and useful contribution to society. So construed, the expressions "laws of nature," "principles of nature" and "fundamental truths," must be limited to results or functions and not include discoveries such as Dennis here asserts he made, — of the phenomenon of cube roots as an insecticide, as possessing poisonous qualities which will destroy insects if man contacts the two — the insect and the powdered cube roots. A discovery in the field of science of a new quality or phenomenon of an old product may be (other necessary facts such as being first, timely application, etc., existing) the proper subject of a patent. It does not fall within the term "law of nature" as that expression is used in patent law.

Dealing with actualities as they exist to-day, it is apparent that most of the outstanding discoveries are in the field of pure science, chemistry and electricity. Industries founded thereon, or health promoted thereby, result. "Machines" or "composition of matter" are designed and constructed to make the scientific discovery a commercial success. Absurd and illogical is an interpretation of the statute which gives patent protection to the instrumentality used in the distribution of the discovery and yet deny to the discovery itself the reward which the Congress intended to grant to those who serve man through enriching discoveries or through the birth of an invention.

The serious question on this appeal arises not, however, over the patentability of a discovery of benefit to man, which was achieved by a researcher who found that a certain root, herb, or vegetable would, under certain conditions, destroy insects, a much-sought and highly-desired end, but whether the applicant for this patent, Dennis, was in fact the first discoverer of the new product, an insecticide derived from a plant grown in South America.

Equally important is the language of the patent in passing upon the defendants' alleged infringement and also the conduct of one of the plaintiffs, who was at one time engaged with the alleged infringers in producing and selling an insecticide in which cube root was used and wherein he, for valuable consideration, accepted the proceeds of such alleged infringing business.

The specifications of plaintiff's patent disclose the following process for obtaining his patented compound: First, the cube roots are pulverized, and the resultant powder drawn off by an aspirator (a fan) and delivered to a concentrator, and from there delivered to a container where it is subjected to a suitable solvent such as benzol alcohol; the solution is then passed through a filter to remove the woody fibre. The resultant product is the extract in solution. This is dried or evaporated, and the concentrated extract may be used as such. It may be used as an insecticide with the addition of talcum powder, cornstarch or other carrier, or it can be used with water or with soap and water or other emulsifying agent, as a spray. For aphids he specifically recommends a solution made by using the extract from one-third pound of powder in one hundred gallons of soap solution.

Defendant Pitner described defendants' product to be made as follows: The cube powder as shipped to them was first screened (sifted) into a crutcher, through ordinary screen wire to catch debris such as nails, that might have fallen into the powder. Sulphonated oil, sulphurized oil, dipentine and water were put in the crutcher and mixed with the powder. The cube (alleged to be timbo) powder was then mixed with definite proportions of other ingredients, principally walnut shell flour and talc to increase the quantity of bulk, and blended and mixed and put in bags and sold to the market as insecticide dust. He said they never removed fibrous elements and never used alcohol or benzine in preparing the extract.

In the specifications of the McDougall patent for making an extract of derris for an insecticide, he states he mixes the compound (organic salt obtained from roots) soap, sulfur or other ingredients, or an emulsifying or carrying agent. Lemmens in his specifications for tephrosia insecticide, mixed the organic salt compound with a carrying agent or a diluent, which might be soap, sulfur or other substance. He says that compound is extractable from the root by moisture or immersing the material in water. In his second patent he says the material is preferably acted upon by a suitable solvent such as benzine which removes the compound from the plant.

Dow teaches the use of a halohydrin extract of derris admixed with benzyl alcohol. He says the chemical serves as an excellent solvent for various so-called deadly insecticides, such as timboin, nicotine and derris (or "fish poison"). "In utilizing such combination of ingredients in this connection, we preferably extract the poison first with the halohydrin, and admix the extract with benzyl alcohol in desired proportions, depending upon the specific use to be made of insecticide. Furthermore, such combination of ingredients may readily be mixed with inert dry materials such as lime, lead calcium or magnesium arsenate, and the like, capable of absorbing the same, and can then be applied in the form of a powder or dust with highly satisfactory results."

There is much conflicting testimony as to the botanical classifications of the various roots covered by the instant and prior patents. Cube belongs to the species Lonchocarpus, of which there seems to be 60 species, and to the family Leguminosae. It is described in a recent publication as Lonchocarpus utilis. Derris is also a species belonging to the Leguminosae genus.

Derris and cube have certain toxic elements, among which rotenone, is claimed to be extremely effective as an insecticidal agent. Both derris and cube also contain other toxicants in greatly varying amounts, depending on age and condition of specimen, — deguellin, tephrosia. Toxicaria is found in derris but has never been found in cube.

A recent publication states that rotenone is obtained from the substance latex from the roots. The latex can be squeezed from the roots into cotton, and extraction carried on in the same way as though the root had been pulverized.

This publication deals with Rotenone-Yielding Plants of South America, giving the botanical and common names of several varieties of Lonchocarpus species of the Leguminosae family (genus).

Dennis' story of his discovery of what was known as a fish poison which could be advantageously used as an insecticide is interesting. In 1917 he was a teacher in Peru where he saw natives grind the root of a plant called cube with their feet and use the product to kill fish. It was seemingly a poisonous concoction that they used, although the edible quality of the fish was apparently not affected by the poisoning or asphyxiation.

The first step in Dennis' discovery of cube as an insecticide occurred when he had the happy thought, so he says, of grinding the cube root (as one might grind corn) in a stone bowl, and distributing the dust or powder over the floor of his home to rid the place of insects. As grinding involved laborious effort, his thoughts turned to boiling the roots, which he did, and he used the extract. He claims to have imparted the knowledge of his discovery to Dr. McIndoo of the United States Department of Agriculture, and the commercial value of cube root was soon established. The plant grows wild in Peru and since this insecticide industry has been developed, raising cube as a crop has been promoted and substantial exportations to the United States have followed.

Dennis denies (but defendants assert) there is a difference between cube and timbo, the latter being what the defendants claim to use for insecticide. Timbo and cube are the same plant with different tribal names. This is the contention of Dennis. We are convinced from our study of the record that derris, tephrosia, as well as cube, belong to the Leguminosae genus, and it is reasonable to believe that chemical or poisonous properties found in the roots of one member of that family may be found in other members of the family, although possibly in different quantities and intensities. The use of one member of the family in place of another member containing the same chemical properties, both plants being native to a country like Peru, would, we think, be the use of a chemical equivalent. Both plants have the same asphyxiating properties and had all been theretofore used by the natives as fish poison. The same narcotic or poisonous qualities are present in both plants.

In determining whether they are equivalents in a patent sense, we must consider the qualities for which they are used in the patent. Neither cube nor timbo were to be used as bouquets. Nor were they sought for their fragrance. It was their roots — aye, the poison in their roots — that were sought. Of the same species, native of the same part of a South American country, confusion as to the names by which they were called was not surprising. The botanical vocabulary of the native of Peru did not include either leguminosae or lonchocarpus or even derris or tephrosia. When one tribe spoke of cube and another of timbo, they were describing a plant, the root of which was used to asphyxiate the fish, whereby a meal could be obtained. In other words, great etymological accuracy was not to be expected.

Every patentee may be his own lexicographer. Advance Rumley Co. v. John Lauson Mfg. Co., 7 Cir., 275 F. 249, 251; Northwest Engineering Corp. v. Keystone Driller Co., 7 Cir., 70 F.2d 13, 15.

All the courts ask of the patent applicant is that when acting as lexicographer he make his definitions clear. It is not uncommon for one to use a term other than in its strict or scientific meaning. Purely, as an illustration, it might be said that fishermen often describe a northern pike as a pickerel. They even speak of a rod as a pole. However, if a user is careful to be consistent in the use (or misuse) of his word, no misunderstanding arises. It is true, a discriminating plant specialist, trained particularly in South American plant life, an educated and discriminating botanist who had specialized in Peruvian plant life prior to 1926 (if any such person can be found) might point out a legitimate distinction between timbo and cube. A copy of an article by Krukoff and Smith on "Rotenone-Yielding Plants of South America," published in 1937, appears in the record and justifies the making of two varieties of the common species out of cube and timbo.

The important fact is that cube and timbo were the same kind of plant in the respect that interested Dennis — to-wit, in their roots. Poison, destructive of fish — poison, destructive of insects — was the common property of the roots of both plants. It was the poison in the roots of a Peruvian plant that the patentee, the Department of Agriculture, and the defendants were interested in. There was no mistake, no confusion. It would be unconscionable for defendants, if they secured their information from a valid discovery made by Dennis, to avoid the consequences of their infringements through a technical and, in this instance, an unimportant distinction of plant names.

We are, however, not persuaded that Dennis was the first to discover that the powder from the cube root could be used as an insecticide. His story of being the first discoverer is hardly consistent with his letter to the Department of Agriculture. Nor can it be reconciled with the other patent applications above enumerated, which were of record and which antedated his application.

It will be noted that Dennis claimed to have made his observation in 1917. It was in 1921 that Dr. McIndoo, an official long in the U.S. Department of Agriculture, learning from one Prof. Allen of Indiana University about the possibility of the use of fish poison called cube or barbasco, asked Mr. Dennis for information. He asked for the scientific name, and the poisonous property it contained. He further inquired whether the leaves and stems were poisonous and whether the dry roots could be had in large quantities and at reasonable prices. He also added, " We want them for killing insects." In November, 1921, four years after his alleged discovery, Mr. Dennis replied:

"* * * I rather doubt however if it would serve the purpose you wish. Only the roots are used for fishing, and in my opinion they are not very poisonous but rather narcotic. * * *"

Dr. McIndoo wrote Mr. Dennis, on December 13, 1921, stating he found the specimen of cube furnished by Dennis to be "very promising as an insecticide. * * * In fact, the extracts from it seem to be a little better than those from any other plant. * * *"

Mr. Dennis replied to Dr. McIndoo, March 13, 1922, stating:

"(I am real interested in the commercial possibilities of the Cube and shall appreciate any suggestion you may make toward getting it in the form of an insecticide at a reasonable price.) * * *"

Dr. McIndoo replied (March 22, 1922):

"All I can say about its preparation now is that the roots should be powdered as finely as possible and then it can be used as a dust like Pyrethrum powder, although the extract in all probability will be the best for commercial purposes. * * *"

Mr. Dennis wrote McIndoo, August 23, 1922:

"* * * It might interest you to know that I have a friend here who experiments with insecticides for his company, The Iowa Seed Company, and he is anxious to get a copy of your findings as soon as they are ready for distribution. He says if it is as good as I say he wants to go in with me to develop the sale of it. If you have any suggestions about the matter I should appreciate it very much." * * * "If it should prove to be of commercial value I am in a position to get at the supply easily."

Dr. McIndoo wrote Mr. Dennis, Sept. 18, 1922:

"I used extracts of it (cube) this summer in La. on the cotton aphid and found it satisfactory and hope it still proves to keep its good reputation. * * *

"Derris in its various forms is now becoming quite a promising insecticide and I hope that we can also prove that Cube is even better. * * *."

The record also showed that prior to this date the knowledge which Mr. Dennis acquired was known and the use of Cube roots for insect poison was rather common. There was enclosed in a letter from Allen to Dr. McIndoo, dated June 15, 1921, a photograph, beneath which was the inscription "Washing cow with barbasco `milk', which has been crushed into the bowl with water. * * * To kill tick. Near Iquitos, E. Peru, Sept. 1920."

The record contains several publications which appeared years ago, which were informative on this subject of the use of lonchocarpus and leguminosae plants as insecticides.

An agricultural bulletin of the Malay Peninsula, of 1898, received in the Library of the U.S. Department of Agriculture, stated:

"Dirris Elliptica. * * * (Leguminous). — `Tuba', * * * often cultivated for use as a fish-poison. * * * The roots are the parts used * * * being pounded and mixed with the lime and water. * * * A similar poisonous resin occurs in many other Leguminous plants used for fish poisons and has been named timboine, nicouline and pachyrhizin, * * * will kill fish in half an hour * * * The extract has long been used by the Chinese and other gardeners here as an insecticide * * *."

A Dutch article was received by the same library, in Sept., 1920. It stated:

"Derris-root is not only poisonous for fish, but also affects other animals, but to a different degree. * * * in Perak the plant is used extensively as insecticide by the Chinese gardeners; the fresh root, for this purpose is crushed fine with water and the milk-like juice is spread over the plants with a feather-duster. Wray, furthermore, reported in full detail on the efficient action of Derris-root as insecticide. He recommends the use of fresh root crushed with water or an alcohol infusion of dried roots mixed with soap water.

"In the literature, there is very little mention of Derris as insecticide. In a short article * * * published * * * in 1912 * * * it is reported that it has been used successfully for caterpillars on cabbage and for ants which uprooted celery patches. * * *

"Our knowledge of the insecticidal properties of Derris, however, reached a somewhat sound basis only when a systematic investigation was made by the Amer. Dept. of Agr. a short time ago. (MC. McIndoo. Derris as an Insecticide * * * 1919) * * * Tests with different extraction agents show that alcohol is most suitable and that the extraction can be effected without heating.

"The powder was without effect on three kinds of * * * insects * * *. However it was effective on two kinds of plant lice (aphis)."

The Peruvian publication, received by the Department of Agriculture, June 8, 1910, stated:

"At present, I am studying and testing a liquid for the combatting of sheep ticks, to which I referred in one of my previous letters. There is here a plant called, `Cube', * * * which is used for fishing * * * by poisoning the water with the root. From tests made by me, I have noted that the maceration produces a great insecticide." Non-Infringement. While we might have disposed of this appeal by considering but one of the three defenses ((a) invalidity of patent; (b) non-infringement of the patent; (c) estoppel), two of these defenses, (a) and (b), were so dependent for solution upon a fact situation which bore directly upon and affected them both, we have found the consideration of both defenses unavoidable.

Having ascertained the fact background to better appreciate the scope of the patent, we are also in a better position to fairly determine the range of equivalents which patentee may rightly assert to the defense of non-infringement.

Defendants urge non-infringement on the ground that their product does not answer to the language of any of the claims. For instance, they point out that claim one describes the product as an insecticide and vermifuge comprising ground cube root with the fibrous element removed. Defendants do not remove the fibrous element from the roots they use. They pulverize the root, make a powder of it, and in doing so use the fibrous element. In short, they argue unanswerably that an essential element of claim one is in their product, absent.

Defendants also argue that their product is not an extract, — a requirement of both claims 2 and 4 of the patent. To better understand the language of these two claims we must turn to the specifications. They instruct one working in this field how the product is made. They define what is meant by the term "extract."

Patentee uses benzol-alcohol. Giving the term "extract" the best and most liberal construction possible, we can not say that defendants' product is an extract, — certainly not a concentrated extract. In reaching this conclusion we have, as in the forepart of the opinion, allowed the applicant to choose his own words and to define them.

Defendants powder their product. They add oil to facilitate the spraying of their product. This, however, is clearly not in accordance with the teachings of the specifications nor does it make an extract as that word is defined in claims 2 or 4 of the patent and in the specifications. Our conclusions are that Dennis was not the first to discover a new insecticide consisting of the roots of the cube plant. He doubtless made a discovery of an insecticide, but he was not the first to make it. He helped to develop the industry but patent protection goes only to the first discoverer. If the first does not see fit to apply for a patent, it goes to the public. The discoverer who is second cannot rely on failure of the first to apply for a patent. Only the first may receive a valid patent.

Moreover, the claims, as we are impelled to construe them, were not infringed by defendants' product. The claims measure and define the limit of the patent. What may have been discovered but not claimed by the inventor is dedicated to the public. Dennis saw fit to limit his claim to roots "with the fibrous element removed." These words restricted the claim and defendants avoided infringement by leaving the fibrous element in their product.

The decree is reversed with directions to dismiss the suit.

SPARKS, Circuit Judge.

I approve the opinion, except as to that portion of it which rejects what is referred to as appellants' first defense, "namely, that the cube roots utilized as an insecticide may not involve patentable invention under any circumstances." The question presented by this record does not involve all conceivable circumstances, but it involves only those uses disclosed by the claims in issue. I think the defense referred to is a valid one, and that the claims here involved are invalid for the reasons therein stated.

The statute which authorizes the issuance of a patent does not contemplate that a patent shall issue for every new and useful thing discovered, but only for any new useful art, machine, manufacture, or composition of matter, or any new or useful improvements thereof. It is not obvious that the disclosures of this patent can be properly classified under any of these subjects. It is certainly not an art or a machine, and I fail to observe how it can be called a manufacture or a composition. The word "composition" in my mind presupposes a combination of elements. The patent before us is neither a combination nor a process. It is a product and consists only of one natural element. True, appellee uses it as a powder, or a tincture, or a fluid extract, or a saturated solution, or again, he may reduce it to a powder by means of evaporation. It is not claimed, however, that any of these processes add in value to the cube root, except as a matter of convenience in its use. All of these processes are quite old in the art and patentee is making no claim as to them. In any way he uses it, it is nothing more than the pure cube root which is admitted to be an element of nature. We cannot conceive that Congress intended that a patent should issue for the discovery of a mere natural element or its properties. If Congress had so intended, it would have been easier to have placed no limitations in the statute.