Warren Steam Pump Co. v. Blake & Knowles Steam Pump Works

COLT, Circuit Judge. This bill was brought for the infringement of two patents for improvements in pumping engines, the Whiting & Wheeler patent, No. 526,913, dated October 2, 1834, and the Hall & Gage patent, No. 522,938, dated July 10, 1894. The application for the Whiting & Wheeler patent was filed January 18, 1892, and the application for the Hall & Gage patent was filed June 10, 1892. The Hall & Gage patent is for an "improvement on the Whiting & Wheeler structure. The Circuit Court held the patents valid, and that the defendant’s pumping engine infringed. The fundamental question upon this appeal is whether the patents are void for want of invention in view of the prior art. The patents relate to the type of pumping engine known as the “air pump,” which is used for withdrawing the water and air from a steam — 264 — ■engine condenser for the purpose of maintaining a vacuum in the condenser. ' The Whiting & Wheeler specification begins by stating that it is advantageous in steam vessels to malee use of an independent engine to drive the air pump, and then follows a statement that the object of the invention is to apply a direct-acting engine to the air pump or combined air and circulating pump in such a way as will secure economy of space and accessibility of the different parts: “With engines employed, upon steam vessels it is advantageous to make use of a separate engine to drive the ‘air pump,’ in order that the vacuum in the condenser may be maintained independently of the movement of the main engine, and in war vessels especially it is of great importance to have the engines and machinery in as compact a space and as light as possible. The circulating water that passes through the condenser can also be advantageously pumped by the same independent engine that operates the air pump. “The object of our invention is to apply a direct-acting engine to the air pump or combined air and circulating pump in such a manner that the whole will occupy very -little space and at the same time to furnish facility for access to the valves and other parts of the pumping engine.” The specification then refers to the drawings: “Figure 1 is an elevation partially in section of our improvement applied to a vacuum pump or as it is usually termed in marine engine practice an ‘air pump.’ * * * Figure 3 is an end elevation.”. [[Image here]] — 265 — The specification then proceeds to describe the general structure •shown in the drawings, comprising the air cylinders, frame, buckets, valves, walking-beam, links, etc., referring to the features of accessibility, the permanent and rigid connection between the engine and the pumps, the direct action of the steam cylinders upon the pump buckets, and the unifying action of the walking-beam. The specification then proceeds: “Henee wo are able to make Ibis pumping engine very compact and uniform in its operation and entirely Independent of the main engine.” The specification further says: “Single and compound engines being well known are available with the present Improvements, and in instances where only a vacuum is required, both cylinders, A and A', are fitted with the proper buckets and valves for maintaining the proper vacuum, and in cases where a water circulating pump is required, both cylinders, A and A', may be used for pumping water, and where it is desired to combine in one engine an air pump and a water circulating pump, one cylinder, A', may be for water, and the other cylinder, A, for air. “In pnmps that have heretofore been made with an engine elevated above the walking-beam and pump, the piston or plunger has been adapted to forcing water. In pumps that are used with a condenser, especially those fitted into vessels, it is important to have the buckets move vertically, and there is usually but little head room for the engine. “By our present improvements tho lifting buckets and valves work vertically, and .the inlet water way being below and between the pumps, but little space is occupied and Lhe engine piston rods being coupled direct to the rods of the lifting buckets, the vacuum pump and engine are brought into the smallest possible space, and the walking-beam being between the engine and pump, renders the action uniform because the buckets are hung by tension from the ends of the beam and their actions balanced as nearly as possible by the two engines acting simultaneously and directly on tho rods of the buckets, but at the two ends of the beam and controlled by it. Hence the operations of the vacuum pump are more reliable than heretofore and the conditions of uso are the reverse of those in the ordinary direct-acting force pump.” The claims are as follows: “1. The combination with the two vertical pump cylinders and their lifting buckets and valves, of direct-acting steam cylinders and pistons with their piston rods in line and connected to the rods of the buckets, a frame and columns connecting with the pump cylinders and on which frame the steam cylinders are directly connected, a walking-beam between the pump and the engine and a pivot for the same upon a bearing extending down from the frame, and link connections at the respective ends of tho walking-beam to the cross-heads of the piston rods for unifying the action of the engine directly upon the lifting buckets substantially as set forth. “2. In a direct-acting pumping engine, the combination with a pair of pump cylinders having lifting buckets and valves and valves in the lower part of tho pump cylinders, of steam cylinders and piston rods connecting the buckets of the pump and the pistons of the engine, a walking-beam between the pump cylinders and tho steam cylinders, links connecting the ends of the walking-beam with the piston rods, and a frame extending up from the pumps and to which the steam cylinders are directly connected and which frame also supports the pivots of tho walking-beam substantially as set forth. “3. Tho combination in a pumping engine of two vertical pumping cylinders, buckets and rods for the same, a frame upon the pump cylinders, and steam cylinders directly connected to and supported by the frame, the piston rods of the engine being in line with and directly connected to the rods of the pumps, a walking-beam between the pump and the engine and links connecting' the — 266 — walking-beam and the. cross-heads of the piston rods, the parts being arranged so that the links are in the same plane or nearly so as the piston rods at the termination of the respective strokes substantially as set forth.” It may be noticed that claim 2 is substantially like claim 1, while claim 3 is somewhat more limited. Prior to the Whiting & Wheeler invention, vertical bucket air pumps were controlled by the main engine or by a crank and fly-wheel or by a water circulating pump. Looking now at the drawings of the Whiting & Wheeler patent in connection with the specification and claims, it may be observed: First. These drawings show a vertical twin bucket air pump. Second. This pump is independent of the main engine; in other words, it is not attached to or controlled by the main engine. Third. This pump is actuated by a separate direct-acting engine. Fourth. This pump is not controlled by a crank and fly-wheel. Fifth. This pump is not controlled by a water circulating pump. Sixth. The action of this pump is controlled by the pressure of steam in the steam cylinders. Seventh. The walking-beam unifies the action of the pump buckets. Eighth. The engine is located directly above the walking-beam and the pump cylinders. Ninth. The piston rods of the steam cylinders are in line and connected with the rods of the buckets. Tenth. There is a rigid frame with columns connecting with the pump cylinders, upon which frame the steam cylinders are directly connected. Eleventh. There is a walking-beam between the pump and the engine, which is pivoted upon a bearing extending down from the frame, and there are link connections at the ends of the walking-beam to the cross-heads of the piston rods. To summarize: The structure shown in the drawings is a vertical twin bucket air pump actuated by a separate direct-acting steam engine. It is an absolutely independent air pump, in that it is not controlled by the main engine, or by a crank and fly-wheel, or by a water circulating pump, but is controlled by the'pressure of steam in the steam cylinders; and, further, it is a structure which is compact, accessible, and of comparative simplicity. The Whiting & Wheeler patent does not show or describe any special form of valves or valve-controlling mechanism for the steam engine. The specification says: “We have represented a steam chest at F and valve chests at G, and we remark that the valves and the devices for actuating the steam valves may ¡be of any desired character, so long as they are positive in action, i. e.. with■out ‘dead center,’ and do not require to be further represented or described.” The Hall & Gage patent in suit covers these valves and valve-controlling mechanism in combination with the Whiting & Wheeler air pump. The specification says: “This invention has for its object to simplify and improve the construction and operation of what is known as a twin air pump, but what is herein designated a pumping engine. “Figure 1 of the drawings represents in elevation, partial section, an air pump embodying this invention.” — 267 — [[Image here]] — 268 — The claims of the Hall & Gage patent in issue are as follows: "’1. The combination of two pump cylinders, two steam cylinders arranged in line respectively with said pump cylinders, buckets in said pump cylinders, pistons in said steam cylinders, and piston rods, connecting said pistons and buckets, valves for said steam cylinders to control the admission of steam thereto and the movements of the pistons therein, a valve-driving engine connected with and to positively move said valves, a valve mechanism for said valve-driving engine, a rocking beam pivoted at its middle and connected at its ends with and to be rocked by said piston rods, and a connection between said beam and valve mechanism whereby the movements of the valve-driving engine are positively controlled by the vibration of the beam and the movements of the main steam valves for the steam cylinders by the valve-driving engine, substantially as described. “2. The combination of two pump cylinders, an entablature supported above the same and sustaining thereupon two steam cylinders in line respectively with said pump cylinders and having their pistons connected with buckets in said pump cylinders by common piston rods, a bracket on the under side of said entablature, a beam, D, comprising two members pivoted in said bracket and having its opposite ends connected with and at opposite sides of the said piston rods, a valve-driving engine having its valve mechanism connected with and operated by said beam, and valves to control the admission of steam to the steam cylinders respectively actuated by said valve-driving engine, substantially as described.” It may be observed, with respect to the valves and valve-controlling mechanism, that claim 1 includes the following elements: (1) Valves for said steam cylinders to control the admission of steam thereto and the movements of the pistons therein. (2) A valve-driving engine connected with and to positively move said valves. (3) A valve mechanism for said valve-driving engine. (4) A connection between the beam and valve mechanism whereby the movements of the valve-driving engine are positively controlled by the vibration of the beam and the movements of the main steam valves for the steam cylinders by the valve-driving engine. . The complainant’s pump, constructed in accordance with these patents, is known as the “Blake Twin Bucket Air Pump.” The evidence shows the immediate acceptance and extensive use of this pump as a high-grade pump for condenser purposes. It further shows that the high-grade air pumps in use at this time were unsatisfactory, that a need was felt among engineers for a more efficient vacuum pump especially for marine use, and that the patented pump at once met the need.' In the official report of the Board of Trial on the trial trip of the Minneapolis, published in November, 1894, the year the patents in suit were issued, Commander Willits said respecting these pumps: “The operation of the main air pumps simply emphasized the Justice of the claim of this class of pumps for highest honors. Not only did they do remarkably efficient duty at the smallest cost in power, but the .regularity and certainty of their action and their low speeds conduced to other efficiencies by reducing to a minimum all anxiety on the part of those in charge of the running of the machinery regarding their possible stoppage or breakdown, or of sudden and excessive change of speed. Those who have had experience with the crank air pumps (however well they may be made, in theory, to operate) cannot fail of comfort in the knowledge of the adoption of the Blaise — 269 — vertical twin cylinder air pump for our latest cruisers. Here, at about only 15 double strokes per minute, these pumps maintained a steady vacuum of over 25 inches, at a cost of but little more than 30 horse power for all three double pumps, and with the main engines aggregating over 20,000 horse power. This power of air pump is only about one sixth (i/e) of one per cent, of the horse power of the main engines.” Commander Willits also testified in this case as follows: “The opinion expressed in this article regarding the efficiency and excellence of the twin air pump noted was based upon its very superior performance in comparison with anything we had ever had before the introduction of this typo in other naval vessels. For many years since the beginning of the adoption of modern machinery, prior to which the air pumps were generally of the attached character on horizontal engines (frequently horizontal piston air pumps, double-acting), we had been trying to get an efficient independent air pump, in the sense of a pump detached from the main engine entirely. As far as my knowledge goes from the time L graduated and up to the trial of the New York, which was just prior to the Minneapolis, we had been wrestling with the matter, and had made several attempts to secure a desirable jiump, resulting in the production of both the crank and fly-wheel combined air and circulating pump and crank and fly-wheel independent pump, but nothing had been at all satisfactory. It was in this comparison and in noting its superiority, both in power economy and in' every other respect, that the opinion therein was based.” Upon this point Mr. Bourne, chief engineer of the complainant, testified as follows: “Tlie twin air pumps referred to have come into general use on shipboard, as well ns in power plants on land, wherever high-grade machines and high vacuum are required. These pumps have been very generally used in the vessels of ihe United States navy; a very large number of the most important battleships and cruisers, as well as revenue cutters, torpedo boat destroyers, being so fitted. Some of the most important of those vessels are the following battleships and cruisers: Massachusetts, Indiana, Iowa. Georgia, Nebraska, Virginia. Louisiana, Minnesota, Kansas, New York, Brooklyn,» Columbia, Minneapolis, Colorado, Pennsylvania, West Virginia, Maryland, Tennessee, Charleston, and others. In addition to being used in the navy, these pumps are largely used in the merchant marino in many of the largest ocean-going and lake steamships and in a very large number of important ferryboats and in many large tugboats and dredges, and in very many well-known steam yachts. Some of the best known ocean-going steamships fitted with these twin air pumps are the Kaiser Wilhelm der Grosso, Hamburg, Princess Irene, Koenig Albert, Deutschland, New York, Philadelphia, La Lorraine, La Savoie. In addition to the extensive use of these pumps on shipboard, as above outlined, many of the largest electric-light and electric-traction power plants throughout the Uni fed States are fitted with these pumps, on account of the extreme reliability of these pumps and the avoidance thereby of any danger of loss of power, which is of great importance in these plants. Altogether, the George F. Blake Manufacturing Company and its successors, the complainant, have built and sold over $1,000,000 worth of these twin air pumps. “Int. 51. Can you give the-court any idea of how rapidly these pumps were introduced, or what favor they met with when they were first introduced by the Blake Company? “Ans. These pumps, when first introduced by the Blake Company, sprung at once into popularity; a large number of sales being made even in the first few years after the first pump was built. These pumps met with especial favor in naval vessels, as the government engineers and designers had been striving for a long time to design or get hold of an independent air pump which would satisfactorily meet their needs. The earlier cruisers and naval vessels, built prior to the introduction of the Blake twin air pump, were — 270 — •fitted with various kinds of air pumps; a large number of them being of the fly-wheel type. None of these earlier air pumps were satisfactory, and many might be called failures. The introduction of the twin air pump completely met the demand of the naval engineers for a satisfactory independent air pump, which is evidenced by the large number of this type of pump sold for naval vessels during the first few years after «the pump was introduced; the larger vessels built at that time being almost universally supplied with this pump. “This pump has become so widely known and given such general satisfaction as to have become the standard type where a high-grade air pump is desired and high vacuum is to be maintained.” To the same effect is the testimony of complainant’s expert, Freeman: “They early secured the attention of engineers, and such favorable attention that they have largely replaced all other forms of air or vacuum pumps for marine service, and have been largely adopted by the United States and foreign navies for war vessels, and they are also in extensive use in stationary .power plants where a high vacuum and efficient service are required.” The evidence further shows that among the practical results secured by this pump were a high vacuum, efficiency, reduction in power consumed, reliability, slow speed, lightness, and compactness of structure. The evidence further shows that these pumps possessed several special advantages: First. The air pump automatically adjusts itself to the varying conditions of the condenser; the speed varying with the amount of air and water coming from the condenser, and the pump slowing up in case of gulps of water coming over, so as to avoid shock and strain. This automatic adjustment aids greatly in maintaining a high vacuum and preventing damage to the pump. Second. The speed of the pump, apart from this automatic adjustment, mayo be readily regulated to the speed which is found best suited to the running conditions of the condenser and engine with which the pump is used. The speed is not dependent upon crank and fly-wheel operation, nor upon the speed of a circulating pump. Third. The movement of the buckets is controlled by the pressure -in the air pump cylinders, so that, the pressure being small at the beginning of the stroke, on account of the air in the cylinders, a jump of each bucket during the first part of the stroke takes place, this jump being very quick and extending through a large portion of the stroke, until the bucket strikes the water on the down stroke, or the water strikes the head valves, if these be used, on the up stroke, when the bucket slows down and moves very slowly during the rest of the stroke. This jump is of importance in securing a high vacuum, by acting to open the valves in the bucket quickly, and thus getting the air out of the condenser quickly. Respecting the advantages of this jump action, Mr. Freeman testifies as follows: “When in action, the pump operates differently from the ordinary pump; the peculiarity of the operation being that during the.first part and usually the greater part of its stroke it moves very quickly, and during the latter part of its stroke it moves very slowly. This peculiar action results, in part, from the fact that the pump is not operating upon a fluid of uniform density like — 271 — water, but upon two fluids, air and water, which differ greatly in their physical properties. Not only is the air much less dense than the water, but it is compressible, and very much rarefied, as it comes from the condenser together with the water, and where the condenser is a surface condenser the volume of the water is very small as compared with the volume of air, which in its rarefied condition is to be pumped out of the condenser so as to maintain a proper vacuum therein. By reason of the quick movement of the pump buckets during the first part of their strokes, an especially efficient withdrawal of the air and water from the condenser is effected, thus enabling a materially better vacuum to be maintained in the condenser with a given size pump than would otherwise be the case.” Upon this point Mr. Bourne testifies: “In addition, the two buckets connected by a beam take advantage of the fact that during the first part of the stroke of an air pump the pump piston or bucket is compressing the air which has come in from the condenser with the water, and during the latter part of the stroke is pumping nearly solid water. On account of this condition existing in the air cylinder, a direct acting air pump makes the first part of its stroke at a much higher velocity than it can make the latter part, as during the latter part it is handling water, and the moving parts meet with a considerable resistance, whereas, during the first part of the stroke, comparatively little resistance is met. At the instant the pump iracket brings up on the water the speed of the moving parts is materially and suddenly checked. Now, on account of this high velocity or jump action during the first part of the stroke, the pump valves open more quickly and more widely than would be the case if the pump piston made its stroke at a comparatively uniform velocity throughout. Pump valves, as is stated, opening more quickly and wider, the water and air, particularly the latter, are taken out ofjiie condenser more quickly, due to this jump action, than would be the caséÑif this did not occur. The air coming into the condenser with the steam from the low pressure cylinder tends to become thoroughly mixed with same and to expand to many times its original volume. It thus becomes essential to get this air out of the condenser before this expansion has taken place or has progressed to an appreciable extent.” Upon the same point Commander Willits testifies: “The ‘jump,’ as you call it, at the beginning of each upward stroke of a direct-aofing vertical single-acting bucket pump, is due to the vapor and space unfilled by solid water between the bucket and the head valves, and the moment the water strikes the head valves or vapor is compressed to the point of lifting those valves, the action of the pump is steadied. The advantage also gained by this jump is tiiat it seats the bucket valves quickly and sharply, preventing any back drain which might occur with a slow-seating valve, and opens the suction valve in the suction chamber, or foot valves, as they are called, at once, and gives ample time for the refilling of the lower part of the pump barrel with water from the condenser. I consider this quick initial motion as of very considerable advantage in the above respect.” The defendant’s witnesses do not contradict in any substantial way the foregoing testimony as to the commercial use, practical results, and special advantages, of the Blake twin bucket air pump,' nor do they deny that these advantages are due in large measure to the embodiment therein of the Whiting & Wheeler invention. At this time, or before the introduction of the complainant’s pump, the prior art disclosed two general types of vertical bucket air pumps; those which are attached to and controlled by the main engine, and — 272 — those which are controlled by a separate engine, and which are known as independent air pumps. The prior art also disclosed two types of independent vertical bucket air pumps; the crank and fly-wheel type, and the combined air and water circulating pump. There also existed in the prior art wholly independent horizontal double-acting piston air pumps. A brief consideration of these several prior types of air pumps is necessary in order to understand their bearing upon the patents in suit. 1. Attached vertical bucket air pumps controlled by the main engine: This seems to have been the common type of high-grade air pump before the Blake twin bucket pump. It is admitted that there were serious objections to this class of pumps, among which it is only necessary to mention that the movements of the pump are controlled by the movements of the main engine. As the Whiting •& Wheeler patent says: “It is advantageous to make use of a separate engine to drive the air pump, in order that the vacuum in the condenser may be maintained independently of the movement of the main engine.” 2. Independent crank and fly-wheel vertical bucket air pumps: In this type of pumps the steam cylinder .acted on the pump through a crank and fly-wheel. In other words, the movements of the buckets were controlled by a crank and fly-wheel. The evidence shows that there were objections to this class of pumps, and that comparatively few were made and sold. Among the objections were high and constant speed, inability to adapt itself to varying conditions of load, and irregularity in the vacuum action. 8. The Aries combined air, water circulating, and feed pump: — 273 — [[Image here]] — 274 — The Aries type of pump is illustrated in the above cut. In this organization the movements of the bucket are controlled by the circulating and feed pumps. The evidence shows that there were many objections to this type of pump, and that not more than 10 were placed upon the market. 4. The Coryell combined air and water circulating pump: [[Image here]] The Coryell pump is illustrated in the above cut. This pump is a twin bucket air pump combined with a double-acting horizontal water circulating pump. The circulating pump operates the air pump. The evidence shows that, although the Coryell pump was superior to other air pumps, it had important defects, and that not over 12 of these pumps were sold in 10 years; the last sale being prior to 1894. 5. Absolutely independent horizontal double-acting piston air pumps: An example of this type of pump is found in the Knowles independent air pump and condenser. This pump is illustrated in the following cut: — 275 — [[Image here]] The evidence shows that this was a comparatively cheap pump, and that it did not meet the demand lor a high vacuum pump; also, that it contained many objectionable features, such as the change of direction and churning of the fluid, the lack of valve sealing, the location of the valves outside the cylinder, etc. The evidence further shows that the vertical bucket air pump was known as the best type of air pump for condenser purposes. The advantages of the vertical bucket air pump, as compared with the horizontal air pump, are stated as follows: “First. The inlet valves can be placed much lower, thus decreasing the distance that the water from the condenser has to be lifted, and frequently permitting this water to drain into the suction chamber of the pump, thus contributing to a higher vacuum. “Second. The direction of flow of the air and water through the pump is never changed. “Third. For the same reason there is no churning and mixing together of the air and W'ater by which the air might be prevented from rising to the top. “Fourth. All the vapors tend to rise, thus helping the vacuum. “Fifth. If head valves are employed, the air does not have to be discharged against atmospheric pressure. ‘This is very important in assisting to obtain a high vacuum.’ “Sixth. The valves are all well sealed, being covered with water, which prevents the leakage of air back to the condenser. “Seventh. The valves can be made more accessible.” Prom this review of the air pump art it appears that at the time of the inventions covered by the patents in suit the advantages of an — 276 — independent air pump driven by a separate engine were recognized. It was also known that the vertical bucket pump, as compared with the horizontal air pump, was the best type of pump .for high-vacuum work. The art, however, had advanced no further than the production of independent vertical bucket air pumps controlled either by a •crank and fly-wheel or by a circulating pump, and these pumps had proved unsatisfactory. The Blake twin bucket air pump therefore marks a distinct advance in the art, in that it was the first vertical bucket air pump which was not only independent of the main engine, but which was also independent of control by a crank and fly-wheel •or by a circulating pump. The fundamental defense in this case is that the patents in suit are void for want of invention in view of the prior art. In the able and exhaustive brief of defendant’s counsel, this defense is stated as follows : “First. The pumps of the patents in suit differ merely in minor details from prior pumps and involve no invention whatever. The patents do not suggest any new result or method of operation, and none such is involved in the use of the patented structure. The patents purport to be only for rearrangements of old features; the merit of these rearrangements, according to the patents themselves and in fact, residing only in shape, strength, and proportion of parts. The pump is said to be compact and to occupy little space. The various parts are said to be convenient of access. Even in these particulars nothing was done by the patentees except to select and to arrange from and in accordance with the prior art. The precise vertical bucket positively driven twin air pumps of the patents in suit, connected by a walking-beam as in those patents, were disclosed in the prior Coryell patent, where they are found organized and operating exactly as in the ease of the patented pumps. In the patented construction the Coryell vertical bucket positively driven twin air pumps are simply driven in a different way from that shown in Coryell, namely, in exactly the way in which the ‘Aries’ pump, the Chiswick pump, and all the many direct-acting self-supporting pumps in the record were driven. Being self-supporting vertical pumps, the patented pumps were, of course, incorporated ip a suitable frame, which was that almost invariably employed in the art for a self-supporting pump. “Second. The patents in suit are absolutely anticipated by the Chiswick pump. The pumps of the patents in suit are the same as the Chiswick pump, except for insignificant details of frame construction which the defendant does not use. Certain additional features appearing in the Chiswick pump do not affect its complete identity with the patented pumps for all the purposes ■of this case. “Third. The only possible way of saving the patents is to limit them to minute details of construction which the defendant does not use. The file-wrapper and contents of the earlier patent show that it was the intention of the applicants and of the Patent Office that that patent should be so limited. The prior art requires such limitation as to both patents. The defendant’s pump dqes not have these details of construction, and therefore does not infringe.” There is the further defense that, if the alleged jump action in the patented pumps when used as air pumps had any advantage which was not present in the prior art, the patents cannot be so construed as to give them any benefit from this consideration, in view of the fact that the patents describe the invention and structure as equally involved when the pump is used as a combined air and water pump, or as a water pump; in other words, that patentability cannot be supported by the alleged merits of the patented pumps when used only as air pumps. — 277 — The defendant also contends that the same advantages which are claimed for the patented pump are found in the pumps of the prior art, and that the alleged jump action has no beneficial effect, and is based upon an erronepus theory. These several grounds of defense call for the consideration of the patents in suit in four aspects: (1) The proper interpretation of the patents; (2) anticipations in the prior air-pump art; (3) anticipations in the prior pump art generally; (4) the merits of the patented pump. 1. It is true the patents on their face, especially the Whiting & Wheeler patent, relate largely to structural details and the arrangement of the different parts .of the structure. It is also true that the patents do not mention any new mode of operation or any new results. In construing these patents, however, these considerations are immaterial, provided the pump structure disclosed in the patents and covered by the claims does in fact contain a new mode of operation, and does in fact produce new results. The failure of the patents to state these merits does not prohibit the court from taking them into consideration in determining the question of patentable novelty, neither does it limit the scope of the inventions. These patents are for a structure, and the patentees are clearly entitled to the benefit of all the advantages which that structure possesses over prior structures intended to accomplish a similar purpose. It is also true that the Whiting & Wheeler patent states that the object of the invention is “to apply a direct-acting engine to the air pump or combined air and circulating pump,” also that the pump “cylinders, A and A',” may be used as a combined air and water circulating pump, or as a water circulating pump, and the claims cover these several uses of the structure shown in the drawings. At the same time there is no doubt that the primary object that Whiting & Wheeler had in mind was the use of this structure as an air pump. This being so, the fact that the pump may also be used as a combined air and water pump, or as a water pump, cannot deprive the patentees, in determining the question of invention, of the weight which should be given to the merits which attach to the use of the pump as an air pump. This is not the case of a preferred structure, but only of the use to which the structure shown in the drawings and covered by the claims may be put. It would, indeed, be an unjust and novel doctrine to hold that a patentee is to be deprived of the benefits which may characterize the use for which the patented structure was mainly designed. 2. In the prior air-pump art the pumps relied upon as anticipations are the Coryell pump and the Aries pump. As may be readily seen from the foregoing cuts, there is a wide difference in construction between the patented pump and these prior pumps. There is also an essential difference in mode of operation, since in the Coryell pump the air pump is controlled by the water circulating pump, and in the Aries pump the air pump is controlled by the water circulating and feed pumps. 3. In the general pump aft the defendant relies mainly on the Chis-wick pump as an anticipation. This pump is illustrated in the following cut: — 278 — The Chiswick pump is a large bucket and plunger pump used for pumping sewage, and it belongs to an entirely different type of pump from the air pump. The construction of- a sewage pump and the construction of an air pump for the condenser of a steam engine manifestly present very different problems. The Chis-wick sewage, pump, in our opinion, would not have afforded any suggestion as to the way to solve the problem of a high-grade independent air pump for vaccuum purposes. Further than this, viewed merely as a structure, the Chiswick pump differs in important particulars from the patented pump, as may be seen by a comparison of the drawings which are given above. While it may appear that the separate elements of the three claims of the Whiting & Wheeler patent and the first two claims of the Hall & Gage patent, broadly speaking, are found in the Chiswick pump, yet it is also true that these claims, when read in connection with the drawings and specifications of the patents, cover a pump structure which is not found in the Chiswick pump or in any of the numerous prior pumps shown in this voluminous record. Upon a comparison of the Chiswick and other pri- or pumps with the Whiting & Wheeler structure, it may be said that the real essence oi the Whiting & Wheeler invention does not reside in the separate elements which make up the subject-matter of the claims; in other words, it does not reside in the two — 279 — vertical pump cylinders willi their buckets and valves, nor in the direct-acting engine, nor in the frame, nor in the walking-beam pivoted upon a bearing extending downward, nor in the link connections, nor does it reside in the features of compactness and accessibility; but the real essence of the invention is found in the fact that Whiting & Wheeler were the first to embody in a structure a vertical twin bucket air pump in combination with a direct-acting engine, in which structure the movements of the pump buckets are not controlled by the main engine, nor by a crank and fly-wheel, nor by a water circulating pump. This is the really important novelty of the Whiting & Wheeler patent, when viewed in the light of the prior art. And with respect to the Hall & Gage patent, it may be said that the essence of the invention consists, first, in a system of valves and valve-controlling mechanism for two steam cylinders placed side by side, which is different from anything shown in the prior art, and second, in the combination of this valve mechanism with the Whiting & Wheeler twin bucket air pump, whereby certainty of action and reliability in the operation of the pump are secured. 4. As to the merits of the patented pump, the defendant, as we have already said, has failed to meet' the complainant’s proofs. The immediate acceptance and extensive commercial use of these pumps are not questioned. Nor do the defendant’s witnesses undertake to contradict by satisfactory testimony the evidence with respect to the high vacuum and other practical results secured by these pumps, and the special advantages they possess in the way of automatic adjustment, speed regulation, and the utilization of the jump action. If it be true, as contended in the defendant’s brief, that the same practical results and the same advantages characterize the pumps of the prior art, these facts should have been established by clear and satisfactory proofs. The inherent weakness of the defense is that the merits of the complainant’s pump remain Substantially unchallenged upon the record. The defendant’s brief discusses at length the jump action. It is said that the jump action was always present in independent air pumps, that the full jump action was undesirable and was always prevented, and that the patents in suit seek to control the jump action. As to these contentions, the defendant is clearly right. In prior independent air pumps it was deemed necessary to control the full jump action, and the patents in suit seek to control it. There is, however, this important difference: Among the means employed in prior independent vertical bucket air pumps to control the jump action were the crank and fly-wheel and the water circulating pump, while in the pump of the patents in suit these means of control have been discarded, so that we have a greater degree of jump action in the patented pump than in these prior pumps. The whole question therefore of the jump action, in this case, resolves itself into the inquiry whether the freer jump action, or the greater degree of jump action, in the complainant’s pump, is in fact an advantage. The counsel for defendant in their brief undertake to show that this jump action cannot have any beneficial effect. The difficulty with this argument is that the defendant has introduced little or no evi — 280 — dence to support it. In other words, the defendant has introduced no-substantial testimony to contradict the full and positive evidence of complainant’s witnesses to the effect that in the patented pump the initial quick movement of the buckets, or the first part of the jump-action, opens the valves quickly, and thereby secures a higher vacuum.. In the absence of testimony contradicting this action of the complainant’s pump, it is not for the court to disregard this evidence, and to-hold that this feature is based upon an unsound theory. We must decide this case upon the record which is before us. In this connection, however, we may say that there seems to be much weight in the reasons given by complainant’s witnesses in support of the theory of the advantages derived from this jump action in the patented pump- In looking at the prior pump art as illustrated in this record, it may seem a comparatively simple thing to take the old twin bucket air pump of the Coryell patent and combine with it a direct-acting-engine located above it. It may also seem a comparatively simple-thing, in view of the prior art, to supply a reliable and efficient valve movement for this structure. But it must be remembered that Whit•ng & Wheeler were the first to devise such a structure, and that Hall' & Gage were the first to devise a valve mechanism specially adapted to that structure, and that it was by these means that the problem of a high-grade air pump was successfully solved. In an art so highly advanced as the pump art, where all the elements which enter into the construction of a pump may be said to be-old, where most conceivable conditions of use have been presented to engineers, and where the art exhibits the greatest variety of form and structure, it is impossible, in many cases, as an abstract proposition, to draw the line between invention and the skill of the 'designer. There is, however, strong evidence of invention, where we have presented the circumstances such as exist with respect to the patents in. suit, namely, a demand for a more efficient air pump, the failure of' previous efforts to meet this demand, the immediate success of the-patented device, and its great utility. It may also be observed that, if the construction of a successful independent vertical bucket air pump by the elimination of the crank and fly-wheel and the circulating púmp as means of control was within the knowledge of any skilled pump designer, it is remarkable that such a pump was not built as. soon as it appeared that the existing pumps were unsatisfactory. Again,. it may be said that these patents belong to the class which the patent law was designed to protect, inasmuch as they cover a device which ■ is new and useful, and which immediately met a public want. For - these reasons we are of the opinion that these patents involved the exercise of the inventive faculty, as distinguished from mere mechanical: skill. If, however, in view of the prior art, there is any doubt upon ■ this question of invention, that doubt must be resolved in favor of-the patents by reason of the utility, efficiency, and extensive commercial use of this air pump. Upon full consideration of this case, we agree with the conclusions. of the Circuit Court that the claims of the Whiting & Wheeler patent.. — 281 — and the first two claims of the Hall & Gage patent cover valid combinations, and that the defendant’s pump infringes these claims. The decree of the Circuit Court is affirmed, and the appellee recovers its costs of appeal.