The appellee, by assignment, is the owner of the Gans patent, No. 1,195,923, dated August 22, 1916, for softer ing water. The Refinite Company is the manufacturer of an apparatus which, it is charged, infringes the patent in suit, and the appellant Harvey Laundry Company is a user of the apparatus.
Hardness of water in water supply varies. Water which comes in through the ground may carrj»- with it a high proportion of hardening constituents, and water which runs in on the top of the ground carries lil tie. As the proportion of the water supply from these sources varies, the hardness of the water in the si’pply will vary. Therefore a wide variation may be produced by the rains and seasonal variation. It has been necessary, because of this variation, to watch the water and add chemicals thereto, always in the right proportion, because, if the addition be too much or too little, undesirable results are obtained. It is desirous, in many businesses, where water is necessary, to obtain soft oi zero water. The patent in suit covers a zeolite apparatus for softer ing water.
It has -been found that certain double silicates have the remarkable power of changing their chemical compositions by the exchange and re-exchange of their bases. In the art, these silicates are invariably called zeolites. They have calcium as a base and when brought into contact with a solution containing potassium, the zeolites will give up their calcium to the solution and take up the potassium as a base. By this exchange a new chemical composition is formed. When new zeolites with the potassium bases are brought into contact with a solution containing lime, they will re-exchange their potassium for the calcium of tie lime, and thus resume their original chemical composition; and when zeolites containing sodium as a base are brought into contact with a solution containing lime, or lime and magnesia, they will give up their sodium to the solution, and take up as a new base the lime base or the bases of the lime and magnesia. When this is done, and they are brought in contact with a solution containing sodium — that is, a salt solution — they give up the lime base, or bases of the lime and mag-' nesia, and take back their original sodium base. It has been found that zeolites will exercise this power of exchange and re-exchange of bases p 'actically indefinitely and with little or no wear.
It is a known fact of chemistry that water is rendered hard by lime o • magnesia. The patent is claimed to disclose an apparatus for utilizir g zeolites having a sodium base in the softening of water. The zeoli:es take the hardening constituents, lime or lime and magnesia, competely out of the water, even though there originally were sodium *715compounds iu die water. The process of filtration is commonly known. It is a mechanical process for removing impurities mechanically suspended in the water, and in the process the impurities collect iu the filter bed and assist in its action. In practice, the sand filter must be seasoned, its grains must be coated with the slimy impurities of the water, before it attains its highest utility.
Zeolite water softening is a process of alternate chemical'action, and to operate the zeolites must be clean and kept clean. It is essential therefore, that there be unimpeded surface contact between the zeolites and the water iu order to obtain the exchange, for coaling the zeolites with impurities will prevent the exchange, and it has been found necessary for commercial purposes to filter the water before it reaches the zeolites. Iu practice, it has been found necessary to regenerate zeolite beds after 10 or 12 hours of softening, and, on the other hand, filter beds run for months without cleaning. It has also been found from experience that channeling of the zeolite bed destroys the operation of a zeolite softener. Such channeling is not fatal to filtering, for channels which permit the water to run therethrough permit the impurities to be carried by water, and this will shortly stop the channels.
Claims 1 and 5 of the patent in suit are in issue and are as follows:
“1. A water softening apparatus comprising a casing, a filter bed consisting of a layer of sand or quartz and a layer of zeolites or hydrated ainminoiilieates disposed on ilto layer of sand or quartz, means for permitting the passage of water through the casing, moans for cutting off the supply of water on the exhaustion of the zeolites, and means for passing through the casing a solution of a salt capable of regenerating! the zeolites.”
“5. Water softening apparatus comprising a casing, a filter bed consisting of a layer of zeolites or alumino-silicatos, supporting means for said layer, means for permitting the passage of water through the casing, means for cutting off the supply of water on the exhaustion of the zeolites, means for supplying and passing into the casing a solution of a salt capable of regenerating zeolites and means connected to the lowest point of the casing for removing the salt solution so introduced.”
In the construction of the patent in suit, the zeolite bed f is supported on a layer of gravel or quartz g. The bed is not confined by any structure at its top, so that, when water is admitted, the zeolites are free to and will arrange themselves with the fine zeolites at the top, thus forming this layer of high resistance to the water flow. This the bed will automatically do, because of the relative lightness of the smaller grains of the zeolites. The water enters through a pipe l, and first flows downwardly through a filter e, which takes out the impurities mechanically suspended in the water. The water then meets this layer of fine zeolites, which forces an even distribution of the water throughout the entire bed. The softened water flows out -through.a pipe j, and then through pipe collections to a point where the soft water is utilized- In use, the bed has a top layer of fine zeolites, and the coarser zeolites are below, so that the water flowing downwardly has the advantage that it can be easily cleaned. The inflowing water, although it may be filtered, will contain slimy constituents and.other impurities which will pass the filter. - If such impurities are not washed out of the bed, it has been found that they will coat the zeolites and prevent the necessary intimate contact between the water and the zeolites. So, where the *716hard water flows, downwardly through the bed, and the apparatus is sc constructed that zeolites are free to arrange themselves with the fine zeolites in a layer at the top and the coarser ones at the bottom, these impurities will collect on or in the top layer of the fine zeolites. They cs.n be dislodged and washed out from this position by a current of water coming backward through the bed, and this is referred to as back-washing. 'With this back-washing through the bed, the slimy and mechanical inpurities will pass through the coarser bottom grains and will collect in the center of the bed, from which position they can only be dislodged with great difficulty.
In the structure of the patent in suit, the apparatus is provided with an arrangement of pipes and valves by which a back-washing current of yrater can be sent upward through the bed; that is, through the pipe / md up through the bed, the water passing out through the hard water pipe l. This current of water readily dislodges and carries off any impurities which are on the top of the bed or in the top layer. The effect also is to stir up the entire zeolite bed, and this adds to a proper grading automatically, resulting in the heavier, coarser particles settling first, leaving the finer zeolites at the top. In regenerating the bed after exhaustion, the zeolites must disgorge the lime and magnesia which they take up and then take back their sodium. The apparatus in the patent in suit accomplishes this by running the salt solution into the bed through the pipe controlled by the valve; the hard water being slut off and the salt solution following the course of the hard water. Through this salt solution, the zeolites exercise their peculiar power of re-exchange, giving up lime and magnesia, and taking back sodium. These beds are regenerated in practice by filling the casing with the se It solution and allowing it to stand for hours. At the end of the regenerating period, the casing contains spent or exhausted salt solution, heavily charged with lime and magnesia. It is important that the spent salt solution be completely and thoroughly washed out. The lime and magnesia is picked up by the water, and, although the water has been softened by the zeolites, it will be rehardened by the lime and magnesia it takes from this salt solution, unless it be removed. The apparatus is constructed so that there are no traps or pockets in which tt e salt solution containing this lime and magnesia can collect, and from which it is not washed.
By the use of the apparatus in suit, for the first time, zero water has been produced. Theretofore, in the art, water softening consisted solely of the use of a precipitation apparatus which did not produce soft water. Men deeply interested in the art have been turning their inventive minds toward this problem for half a century. The appellee was the first to produce the result. We think that both the structure and the result obtained are new. Unless it has been anticipated by what the prior art disclosed,, the appellee is entitled to the protection o: its patent.
[ 1 ] The appellants urge as a defense that what the appellee did was kiown to the art by reason of the experiments and use of zeolites by Bumpier and Harm in the sugar industry, and by the foreign publications of Gans, Feldhoff, and Siedler, who wrote on the subject while *717the art was in search of an invention to accomplish the result which the Gans patent has accomplished.' It also relies on the Gans German patent 197,111 (applied for in 1906), covering artificial zeolites, also the Aquaril filter. It declares that there is no invention because of the patents taken out in the filter art and on the Gans United States patent relating to the manufacture of artificial zeolites, and this because of an alleged prior use by Hird & Sons. It appears that about 1903 Harm and Rumpler, German chemists, tried to use zeolites for purifying sugar juices, and the latter discussed his experiments before the luternationaler Kongress fuer Angewandte Chemie. These two chemists thought that they could employ zeolites containing calcium bases, and cause the calcium to be exchanged for potassium, and in this way could increase the sugar yield from the juice. Sugar juices containing potassium interfere with crystallization and increase the residual molasses. It appears, also, that lime assists crystalization.
Very substantial testimony was given by a chemist of one of the largest sugar manufacturers, who was thoroughly familiar with the industry in all its branches, and who stated that zeolites or base exchange silicates have never been used in sugar refining. It appears clearly that the experiments of these German chemists failed. Another witness testified tlsat he was familiar with zeolite installations which had been made in Germany by the appellee. His testimony is to the effect that zeolites have never been commercially employed in the sugar industry for purifying sugar juices. It also appears that, when Rumpler attempted to use zeolites, he filtered upward. This failure in the sugar art did not point the way to success in the softening water art. A reference to the investigation fails to show any commrcial utility made of zeolites in the water softening art.
The chief of the foreign publications relied on is the report in the Centralblatt of Feldhoff’s lecture before a meeting of sugar technologists, in -which he described the first zeolite water softening apparatus built by Riedel & Co. Here, too, it is clearly established that the apparatus described by Feldhoff was a failure, and was abandoned, and a successful apparatus was not built until two years later (1909). Describing an experiment which failed cannot be said to anticipate a patent for a later successful apparatus. In the apparatus described in the Centralblatt, the hard water runs upwardly into the gravel, through the bed of zeolites and through the layer of excelsior provided. It then flows out through a nozzle and a horizontal pipe, which is connected with this nozzle and is controlled by a two-way valve. The water which issues from the apparatus flows out from one of the branches of this pipe, and the other is used for the disposal of the spent salt solution. In regenerating, the two-way cock of the hard water pipe is closed, and the salt solution is allowed to flow upwardly through the bed and oat through the nozzle.
Riedel’s chief chemist recognized its failure. Not only was it a failure, but the builders did not know why, and to determine the reasons a thorough physical and chemical examination was made. It was discovered that the excelsior layer between the two plates became rotten, and the bed of zeolite was divided into horizontal layers about six or *718eight inches thick, and samples were taken from each layer working from the center toward the circumference. Analysis showed that the samples from at or near the center of the bed contained large amounts of lime and were exhausted, while the samples taken near the circumference of the bed contained only small amounts; that is, they were not exhausted. Cracks were found in the bed, and the zeolites around these cracks were very dirty, and they contained much more lime than those at some distance away from the cracks. As soon as the cracks or channels formed, the water mainly flowed through them. The apparatus then no longer produced soft water, because immediately around the channels the zeolites had become exhausted, and those further away were getting little or no water. The inspection further showed that the dissolved air, which the water carried into the casing when it went out of the solution, was trapped in the bed of zeolites by the superposed layer of excelsior between the plates, which rested directly on top of the zeolite bed, and this trapped air prevented the intimate contact of water with the zeolites.
[2] In the Centralblatt apparatus, the water flowed through tire casing from the bottom upward, and the lighter zeolites floated up from the bed in a current of water, and the fine material of the zeolites penetrated into the excelsior layer itself. The layer of excelsior between the plates was put in to prevent this loss of zeolites, and to prevent channeling. The zeolite bed could not be cleaned, and it did not prevent channeling. It is well established that complete failure by an inoperat ve and abandoned apparatus cannot defeat a later meritorious and successful invention. If the apparatus did not, when constructed, anticipate the inventive thought, a description of it coul'd do no more. Cimiotti Unhairing Co. v. Am. Unhairing Co., 115 Fed. 498, 53 C. C. A. 230; Schmertz Wire Glass Co. v. Western Glass Co. (C. C.) 128 Fed. 990. Adding a new element, or bringing about a result by a lew arrangement, in the combination which produces the new result, entitled its originator to protection. Loom Co. v. Higgins, 105 U. S. 580, 26 L. Ed. 1177; Topliff v. Topliff, 145 U. S. 156, 12 Sup. Ct. 825, 36 L. Ed. 658.
[3] A final step which has turned a failure into success, such as is exhibited in the case at bar, should be protected. Washburn v. Beat ’Em All Barbed Wire Co., 143 U. S. 275,12 Sup. Ct. 443, 36 L. Ed. 154. So, also, an invention, patented here, will not be defeated by a prior foreign patent, unless its descriptions and drawings contain or exhibit a substantial representation of the patented invention in full, clear, and exact terms, so as to enable a person skilled in the art or science to which it appertains, without the necessity of making experiments, to practice the invention. Seymour v. Osborne, 11 Wall. 516, 20 L. Ed. 33; Gen. Ed. Co. v. Hoskins, 224 Fed. 464, 140 C. C. A. 150.
Dr. Siedler, Riedel’s chief chemist, delivered a lecture on artificial zeolites before the London Congress of Applied Chemistry in May, IS 09. It is argued tha.t this was a prior publication. The apparatus of the patent in suit was not constructed until the fall of 1909, and at the time Siedler delivered this lecture he did not know how zeolites cculd be practically used for softening water. His knowledge on this *719subject could only be the experiments had in attempting to purify sugar juices, but none of the requirements of a successful water softening apparatus are referred to by Siedler. He states that “construction of the filter is determined by local conditions according to circumstances,” and that “the filteration may be upwards or downwards.” Experience has proven that, if the'apparatus is constructed as the appellee's it may be used anywhere, and does not vary with local conditions, and it is also found that filtration was one of the causes of the failure of the Ceutralblatt apparatus, and that the mere reversal of the flow in that apparatus could not have made it successful.
[4] Siedlcr’s Zeitschrift article of May 28, 1909, does not disclose the apparatus of the patent in suit. We are satisfied that the description in this printed publication is insufficient for one skilled in the art to build what appellee has. If prior patents and publications can be reconstructed by extraneous efforts to fit the exigency of the case, it would, as was said in Badische Anilin & Soda Fabrik v. Kalle & Co., 104 Fed. 802, 44 C. C. A. 201, require an inquiry, not only as to what the publication communicates to the public, but “it will be transferred to an endeavor to ascertain what its author intended to communicate.” We cannot read into this article what is not there,- and which it would be necessary to obtain from the later development of the art in order to meet with appellee’s success.
In 1906 Cans applied for the German patent 197,311. This patent covers the invention made in artificial zeolites. It deals with the precipitation system of softening water, but is devoted to the method of manufacturing zeolites. The patent fails to anticipate the patent in suit. It was the same inventive mind at work. When he applied for his German patent, lie did not disclose the subject-matter of the patent in suit, because it was three years later when he disclosed his present inventive thought and applied for the patent in suit. He does not describe the same idea of means in the same stage of development as that which his later invention embodies. A reading of the statement of the German patent fails to disclose a description of a complete and operative art or instrument ready for immediate employment by the public. Westinghouse v. Great Northern, 88 Fed. 258, 31 C. C. A. 525. Gans applied and was granted patents in the United States, Nos. 943,535 and 960,887. They are process patents. They do not disclose an apparatus, and cannot be relied upon as anticipations. The first of these patents was filed February 18, 1907, and reissued October 13, 1911, three months after the application of the patent in suit.
[5] The appellants rely upon the reissue patent as a method of regenerating zeolites, but an examination of the original patent discloses that it is silent as to the regeneration. It does not aid the appellants. The second patent is for the process of softening water mixed with “hydroxide of lime” by the use of “permutit.” This mixing of hydroxide of lime with hard water before treatment is the old precipitation or lime-sodium process. It is referred to in the patent as “advantageous also to employ upward filtration”; but it is unimportant in this consideration, because the patent in suit is not for a process of treating water mixed with hydroxide of lime, but is for an apparatus *720and this patent does not disclose either that apparatus or any other. Tae process and mechanical inventions under the statute are distinct arid separate. Miller v. Eagle Manufacturing Co., 151 U. S. 198, 14 Sup. Ct. 310, 38 L. Ed. 121; Steinmetz v. Allen, 192 U. S. 561, 24 Sup. Ct. 416, 48 L. Ed. 555.
Patents Nos. 632,091 and 519,565 were granted to Bommarius, and are now relied upon to support the defense of noninvention. The chief features of the first patent are the slitted tube structure at the bottom of the filter, with pot and pipes for feéding alum, and a stirrer located immediately above the slitted tubes; but the evidence indicates d early that the alum pot structure would be useless in a zeolite water sc ftener. It would not hold enough salt solution for regenerating purposes. It feeds the alum into the water ,to be filtered, and it is constructed to feed it in very minute quantities. The slitted tube construction would provide a trap at the filter bottom for the spent salt solution. In a zeolite water softener, suda impurities as pass the filter collect at the top of the bed, and not at the bottom, and therefore the stirrer is located in the wrong place to be of value in a zeolite water softener. To use it as a water softener, it would be necessary to discard the features which Bommarius regarded as important, and then would require the substitution of zeolite for the sand bed, and connections for introducing regenerating salt solution and taking care of the spent salt solution.
In the latter patent, the chief feature is the revolving screen, which is designed to break up the filtering material and the “germs” during back-washing. • Filtering material mentioned in the patent, such as ssnd, cannot be used, because it would fall through the openings in tire revolving screen and choke the pipe. Solid filtering material could not be used, because the screen could not break it up, and in this the stirrer screen is located at the extreme bottom of the filter bed. The revolving screen was found to be inoperative and discarded, and later was provided with a horizontal stirrer differently located; but, as so located, it would not be of advantage to the zeolite water softener. We tl'ink tliat what was disclosed by these patents did not anticipate the appellee’s invention.
[6] Three circulars issued by the Permutit Company of Berlin'are re ferred to as a defense. The date that they were issued is not proven, but it is proven that a trade-mark Aquaril was registered and used in a trade circular. The contents of this circular are used with the argument that the Siedler article in the Zeitschrift of May, 1909, refers tc the Aquaril filterer. Therefore it is argued that' it can be used to help out the Siedler article. But, as we have pointed out, the article itself is not an anticipation, and a foreign publication like the Siedler article cannot be helped out by reference to another publication, particularly one for which no date is proven. We conclude that none of the prior publications or the patents of the prior art anticipate the patent in suit.
[7] None of the prior filterers is adapted for use as a zeolite water softener. They all require' reconstruction and change to.be used for tlis purpose. The apparatus of the patent in suit produces a new re-' suit in the art. It is one for which the chemists had searched in vain *721for half a century. This inventive thought has been applied industrially, and has resulted in the production of zero water for the first time. Its commercial success has been great. It is a new and useful result obtained by the use of a new apparatus, although in the construction old elements have been used. This entitles the inventor to the protection of a patent. Loom Co. v. Higgins, 105 U. S. 580, 26 L. Ed. 1177.
[8] A disclaimer was filed as to claim 1. The petitioner disclaimed, from the scope of claim 1 of the patent in suit, any water-softening apparatus including a layer of zeolite or hydrated alumino-silicates disposed on a layer of sand or quartz in which the water to be softened is so introduced into the casing that it passes upwardly through said layer of zeolite. It was intended by this to exclude from the scope of claim 1 the apparatus in which the water to be softened passes upwardly through the layer of zeolite. By this nothing was injected into the patent as argued by the appellants. This action was taken for the purpose of limiting its effect. The power to disclaim is a beneficial one, and should not be denied, except where a fraudulent and deceptive purpose is apparent. Sessions v. Romadka, 145 U. S. 29, 12 Sup. Ct. 799, 36 L. Ed. 609. We find no evidence of such purpose here.
As we have pointed out, from a somewhat exhaustive examination of what was submitted below on the question of anticipation, it was apparent that appellants’ apparatus was not constructed from either the description set forth in the various publications or from that which is indicated by the prior art. The appellants’ construction has a bed of zeolite resting on a layer of gravel, and this in turn rests on a concrete filling. There is a free space above the zeolite, which permits water to be introduced and the zeolites to arrange themselves with a layer of fine zeolite at the top, thus providing an even distribution of water throughout the bed. The water is first filtered by a filter located outside the casing, instead of within it, and it then enters through a pipe at the top of the apparatus, which flows downward through the zeolite bed. The water is soft, and then flows through the nozzle secured to a nest of pipes imbedded it) the concrete filling. A tank is located above the apparatus, and the salt solution enters through the hard water pipe. '1'lie hard water is shut off during regeneration. The salt solution flows downwardly through the bed of zeolite, and is discharged through a pipe. When regeneration is complete, the salt solution laden with lime and magnesia, and the washed water is carried away from the casing through the nozzles into the pipes at the bottom. The exit pipe-are attached to the casing on a level with the cement bottom. Tims claims 1 and 5 of the patent in suit are infringed.
This invention has been used by power plants, laundries, tanneries, canneries, and various other manufacturers. It is a great advance in the art of softening -water. Its savings to manufacturers has been enormous. It is a new and widely used apparatus in a useful art. It is that class of inventive thought which it was the intention of Congress to protect as a valuable contribution to a useful art. The inventor is entitled to his reward, and the appellee to its protection.
Decree affirmed.