This is a bill in equity filed by the Stirling Company against the Rust Boiler Company, charging infringement of patent No. 526,947', granted October 2, 1894, to John E. Schlieper for a steam boiler. The defenses thereto are invalidity and noniniringement. The boiler in question is of the water-tube type, and the proofs largely concern the current circulation of the water therein. The testimony of a number of scientists has been taken, and as their views are divergent, if not, indeed, conflicting, we feel all the more impelled to go back to the patent as the foundation and measure of the rights here involved. While the helpful value of expert scientific proof to throw light on many subjects before them is recognized by courts, yet it must be remembered that such testimony is admissible to remove and not create uncertainty. The first and primary duty of a court is to examine the patent and see whether its statements are in themselves plain and self-explanatory, and to resort to extrinsic, aid only where it is required to aid in construing it by making plain something not in itself clear and intelligible. Fet us therefore turn to the patent itself. Now, in the first place, a perusal of that instrument discloses the somewhat significant fact that the patentee nowhere alleges any defect or difficulty in prior boiler practice which he sought to remove or overcome. In the absence of any such statement, and, indeed, of proof that any such recognized defect or difficulty did exist, which was overcome by the subsequent use of this device, or one in substantial conformity thereto, we are justified in assuming the patent was not one of that broad pioneer character which overcame some obstacles that had long blocked advance in its art. And such, indeed, would seem to have been the view of the patentee for he states: “The object of my invention is to improve the Stirling boiler,” and that his invention consists “in the features and combinations hereinafter described and claimed.” if, then, we ascertain what the Stirling boiler, which he sought to improve was, and ascertain the “features and combinations” used in making such improvement, we know just what the patentee disclosed, what he claimed, and to what extent of patent monopoly he is entitled. *850In the single figure of his specification sheet herewith given is shown the type of Stirling boiler,which he sought to improve:
Discarding the rear upper and lower drums, which are merely auxiliary feed water connections, we have four drums or boiler shells, two above and two below, forming a fan-shaped structure. The two lower ones, known as the “Forward and Rear Mud Drums,” have water connections with each other, and each one is also connected by four rows of tubes to the forward and rear upper drums, respectively, which are termed “Steam and Water Drums.” These latter drums are connected to each other below their water levels by water pipes, and above by steam pipes. This water connection was a device which Stirling added to fan-shaped vertical water tube boilers, and thereby made possible what was called the Stirling cycle of circulation and gave his name (although his patent showed but a single mud drum) to boilers of the general type shown in this specification. Now it will be noted that the Stirling type which Schlieper sought to improve was not itself a broad, pioneer addition to the boiler art. That device was considered by this court in the case of the Stirling Company v. The Pierpont Company, 72 Fed. 786, and it was there .said:
*851“This brief review of the art. which by no means embraces all the patents pertinent thereto and the satisfactory character of the results attained in i he same general lines which Stirling followed some years later, shows the iieid was so fully occupied that the advance made by him was the gradual step of the improver, not the stride of the pioneer. Singly considered, the (dements of his combination were old. He did not discover the principle of circulation nor was he the first to devise means to effect it; circulation through a mud drum was not original with him; bafflers and means for s-ifeetually distributing the heat to the water tubes were known before; compactness of structure and facility Of access for cleaning had been attained, and the deposit of scale secured. That he united all these desirable points in. a structure combining simplicity, economy, and effectiveness Is true; this his combination, showed novelty and patentability is, for present purposes, assumed; and to the extent of his specified combinations and to others using substantially the same elements, or their equivalents, to accomplish the same result in substantially the same way, his rights will be enforced. Further ihan this we cannot go, nor are his claims entitled to a broader construction.”
The Stirling boiler had a strong, main, or general cycle of circulation which was well known, and which indeed is described by Schlieper in his patent. After referring to the passage of the water through the feed water system, which, as stated above, is not here material, Schlieper says:
“The water then passes on into the forward mud drum or drums, up into ihe elevated drams, across from one drum to another, down again to the mud drums, and so on continuously — allowance being of course made for evaporation.”
Now a Stirling cycle such as thus described by the patentee is caused by subjecting the different banks of vertical tubes to different heat conditions by dividing the furnace into different heat zones by means of a division or baffle wall. It is this main circulation — which alone the patentee mentions — that his device, which “aids in regulating the circulation,” is intended to affect. What then'is the mode in which he regulates that circulation and what the means he employs to do so? The means consists in a smaller, or what Schlieper terms a secondary water drum placed midway between the upper stream and water drums and which taps the water pipes connecting them. This drum is connected to the rear mud drum by tubes located in the rear or cooler heat zone. The secondary water drum and the pipes thus connecting it with the rear mud drum are the only features Schlieper added to the Stirling boiler. What function did they add thereto? it will be noted that the steam raising capacity of a boiler depends on the maximum of water it can subject to a maximum of heat. It is therefore desirable a boiler should, up to its working capacity, carry as much water as possible. If there is too much, or the water line of the upper drums is too high, foaming water is carried into the steam, an operation called priming, and wet steam produces pounding in an engine. In tubular boilers, it should also be observed, the most effective heat'transference is when the water is circulating through the vertical tubes. In view of these facts the purpose of Schlieper’s two added features is clear. The dram he places below the water level of the steam and water drums in order that it will take it to its lower level and full capacity water from the higher stream and water drums. This is set forth by die patentee who states:
*852“The secondary water drum, which may be of any desired size, Is preferably placed at a lower level than the main steam and water drums, so as to reduce the water level therein and thus prevent .priming or foaming. It also provides additional heating surface and aids in regulating the circulation.”
That it adds to heating surface is apparent for its contents instead of being in the upper drums situate at corners of the surface and off from the path of heat travel, are now placed in a drum in the very center of the furnace crown, lower fhan the corner drums and in the direct pathway of heat travel. It also adds an extra row of tubes in the center of the furnace. Here, then, we have a plain, unambiguous statement of certain definite advantages, resulting from laws of physics, well recognized and produced by the two agencies, viz., water drum and tubes, placed in locations functionally effective, and forming working parts of a definite combination. These are the “features and combinations” in which, as the patentee states, “the invention consists.” The advantages resulting therefrom, as stated by the pat-entee, are four in number, viz., reduction of water level; preventing foaming and priming; increase in heating surface; and aid in regulating circulation. The first three of such advantages result from the location of the secondary drum below the water level of the steam and water drums, and from its capacity to hold a large quantity of water. The effect of the location below water level is stated thus:
“Tbe secondary water drum, which may be of any desired size, is preferably placed at a lower level than the main steam and water drums, so as to reduce the water level therein and thus prevent priming and foaming.”
And the capacity of such drum thus:
“The secondary drum, being so ■ arranged that a large part of the water, passes through it, performs the functions above stated.”
It will thus be seen the submerged location and the substantial reservoir capacity of this secondary water drum are avowed factors and functional elements in the invention disclosed; for by its use a given quantity of water can be used and a lower level had in the steam and water drums than would otherwise be the'case. The mode in which the added features, “aid in regulating the circulation” is not stated. As we have shown, “the circulation,” which it aids in regulating is the main, quadrilateral Stirling circulation which alone is referred to in the patent. In the proofs in this case substantially the whole controversy has hinged and the important feature of this patent is alleged to consist in the tubes connecting the secondary water drums with one of the lower mud drums. If the gist of this invention is the function of those tubes the patentee was singularh successful in ignoring it in his specification, for the only reference to it is where he says, “the secondary elevated water drum is connected with one of the lower mud drums, preferably the middle one, by means of a series of tubes, etc.” Such omission is not without significance, for, as was said in MacColl v. Knowles Loom Works, 95 Fed. 986, 37 C. C. A. 350:
“In the construction of a patent, the omission of the patentee to point out or refer in his specification or claims to the special feature which he subsequently maintains is the most important part of his invention is very significant, and should be carefully scrutinized. ‘If this feature be an advan-*853lage, as now claimed, it is strange that no allusion is made to it in the specified ion.’ Fastner Co. v. Kraetzer, 150 U. S. 111, 14 Sup. Ct. 48, 37 L. Ed. 1019.”
It is evident to us that these connecting tubes were regarded as a mere incident to the more important secondary water drum and that from their location in the second heat zone, it was simply intended they should perform the office which a cool, row of tubes fulfill with reference to a hotter row, that is, to draw downward; in other words, they co-operated with the banks of tubes leading from the rear upper to the rear lower drum in drawing downward, and, in so doing, shouldered a part of the work of those tubes, and in that way aided in carrying the main circulation. And it is certain that there is no suggestion in the patent that they created any new or additional form of circulation, nor was their situation, to wit, in the second heat zone, one that tended to create any constant current movement, other than to shunt a small part of the main Stirling cycle, whose existence was not dependent upon them, and which, at most, they could simply aid to regulate. And this view of the patent, drawn from its own terms, is not unsettled by anything that became evident as the device was used; for as no boilers on the general lines disclosed by this device were built and commercially used, we have not the qualifying fact which often affects the estimate of a patent that the use of a device disclosed merits which the patentee himself did not expect.
Upon this device three claims were allowed as follows:
“(J) In a water tube boiler, a secondary drum baying water communication with elevated steam and water drums and a lower mud drum, substantially as described.
“(2) Tn a water tube boiler, the combination of two elevated steam and water drums having communication with each other, a secondary drum arranged at a lower level than the elevated steam and water drums and having water communication therewith, a lower mud drum or drums having water communication with the elevated steam and water drums, and means for supplying and feeding water into the lower mud drum or drums, substantially as described.
“(3) In a water tube boiler, the combination of the two elevated steam and water drums having communication with each other, a secondary drum ¡(laced at a lower level than the elevated steam and water drums and having water communication therewith, an elevated feed water drum, a front lower mud drum or drums having water communication with the elevated, steam and water drums, and a rear lower mud drum having water communication with the elevated feed drum and with the front mud drum or drums, substantially as described.”
Assuming, for present purposes, the validity of the patent, we next inquire whether respondent infringes, hi the Stirling type of boilers the lower drums were coupled closely together and the drums above spaced apart. 'Phis resulted in an approximately fan-shaped construction in which the front bank of tubes inclined forward and overhung the furnace. The baffle walls, following the inclination of the tubes, were also inclined forward. Owing to the forces of contraction and expansion tubes should pierce the boiler shell at approximately- right angles. It appears from the proofs that four rows of tubes to a bank is about the limit of proper construction on that ac*854count. Such right angle connection with the curved boiler surface was effected in the Stirling type by curving the tube ends. It was to this type of boiler where the unused space between the upper and lower drums permitted the introduction of a secondary drum that Schlieper added that feature. Now while the Schlieper device expressly stated it was an improvement on this type of boilers we do not regard it as limited to such boilers, fan-shaped and curved tubes; but, in respondent’s device, shown in the accompanying cut we have
*855another and wholly different type. There we find the two upper drums directly over the two lower drums; the spacing between the same, above and below, equal; and their vertical connections made by banks of straight, instead of curved tubes. The boiler shells are pierced at right angles, through shoulders pressed outward on the shell. By staggering them, four banks of tubes may thus be located without extending the shoulders to an undue width. Such construction of vertical tubes between the square-placed drums leaves a central space of equal width at top and bottom. A baffler being placed between these banks of tubes puts them in unlike heat zones and creates a strong, quadrilateral cycle of circulation. Now, at the lower part of this space, the respondent puts a secondary • drum from which two hanks of tubes extend upward, and, by means of a header, connect with the pipe uniting the water space of the two upper drums. A bank of tubes also connects the secondary drum to the rear mud drum. The two banks leading upward by supporting and making possible the use of a vertical shell baffle of three inches thickness, save valuable space for access within and cleaning the boiler. This baffle being placed intermediate the rows of central tubes, throws one set into the forward or hottest heat zone, and the other iñto the rear or coolest heat zone. The effect of this is to make the current flow of these pipes in opposite directions; that in front, up, that in the rear, down. Such being the case, it will be noted we have between the secondary drum and the upper drum-connecting pipes a minor quadrilateral circulation within the outer main current cycle. This is obtained by locating the secondary drum in the central lower cham - ber space, and by placing the baffler between the tube rows leading from it. This not only adds to the heating surface an entire bank more than Schiieper had. but places the added bank in the hottest zone of the furnace where Schiieper added none. Almost the. entire shell of the secondary drum in Rust’s construction is subjected to heat and a large part of it at the initial heat stage. So, also, the secondary drum of Schiieper is not found in respondent’s construction, and instead of the whole bank of tubes entering into such a reservoir, each pipe of Rust and its fellow in the row opposite form an individual or sectional system of their own, by means of a header •coupling. We do not understand it is contended that the Rust secondary water drum below is the equivalent of the Schiieper secondary drum above but that the several elements of Rust, viz., the double row of tubes, the header, the lower drum, and the connecting row of pipes between the upper drums perform the same function as the Schiieper drum and its connecting pipes or to use the language of complainant’s expert:
•‘The double row of tubes, together with flie herders, lower drum and their connecting pipes, perform the same function in this boiler as do the secondary drum and its connecting pipes in the Schiieper boiler.”
We cannot accede to this view. What object, as affects circulation, had Schiieper in view in his use of the secondary drum and its pipes connecting with the rear mud drum. Clearly his purpose was to draw water from the forward upper drum to the rear mud drum *856and to that extent avoid carrying it through the rear upper drum and down the rear banks of tubes. He provides a device for diverting the water from the main .current path béfore it enters the rear upper drum, and provides no method to return it to that current, so as to afterwards pass through the rear upper drum. And his device is functionally constructed to attain that result. In the first place, he causes the water from the forward steam and water drum to strike a solid volume of water which completely fills the submerged drum. Now it is clear that to a certain extent the larger the volume of water thus interposed before the outlet to the rear drum is reached, the .more difficult it is for the water to follow that course, for. it serves to check the velocity and' change the course of the incoming stream. For that reason the purpose of the patent is affected by a structure that provides for such a deflecting body of water, and in ’the same way its purpose is defeated by one which Closely approximates the facility of transmission incident to the use of a mere pipe connection. Moreover in Schlieper’s device diversion to the rear mud drum is intensified by a pipe leading thereto whose withdrawing force is not only aided bv gravity, but is increased by the relative thermal relation of the pipe itself.- Consequently, Schlieper places his row of tubes back of the baffle wall, where, as compared with the main forward bank of tubes, its tendency to draw down would be pronounced; but, even as respects the rear banks of tubes, an auxiliary baffler deflects heat from it to the rear bank, and so tends to make Schlieper’s added row the coolest bank in the whole system. Nor is this all, for the tendency of this bank of tubes to draw down is increased by the fact that the least tendency to- create upward counteracting currents is found in the rear mud drum to which it leads by the fact that the cooler feed water is there introduced. It will thus be seen that the whole device is constructed in order to divert water from the point where it might possibly be drawn into the rear upper drum, and pass down through the rear banks where, as complainant’s expert says, “I do not think there would even be a general upward circulation through the tubes.”
On the other hand, a study of the Rust system satisfies us that its whole purpose is in an opposite direction, and while it may temporarily divert water from the main quadrilateral circulation, it eventually returns it to the same point where it may again join in that circulation. It should be noted also that during its temporary diversion the water has been subjected to a heat transference which Schlieper neither contemplated or indeed was possible by his device. How is this done? In the first place, the Rust header which is interposed in the. water connection leading from the front to the rear upper drum contains, no such checking volume of water as the Schlieper drum. Respondent’s estimate makes the header’s cubical capacity twice that of the pipe sections it connects, while complainant’s only makes it three times.’ While there are no degrees in infringement, yet the mere fact of increased capacity of a four-part header over the pipes it connects does not necessarily evidence an infringing intent, for the use of headers in boilers was common enough to lead us in Stirling v. Pierpoint, supra, to say-:
*857“Water tube boilers include two classes — those whose tubes are horizontal, or substantially so, and connected at the ends by headers, and those whose tubes ar<¡ vertical and connected at the ends to cylindrical drums.”
The use of screw pipe connections for tubes is • not feasible in boilers, and the form of header here used is of the smallest capacity-compatible with safety. It will therefore be seen that in reducing- the pipe connection between the upper drum to a safety minimum of size, and in using an individual header for each individual connecting pipe, the current passing from the front upper drum to the rear upper drum, through each individual pipe and header, is subjected to a minimum of current checking and diversion. And in this connection it is proper to note that whatever dispute there may be as to the current flow in other parts of the system, no question is raised, but that the water ejected into the rear upper drum of the Rust system is delivered in such a strong, positive, and continuous stream, that it would seem as though its propulsive force was the current passing through the section of the pipe between the forward drum and the header.
In addition to this there are other features, which tend to a continuance of this main circulation without obstruction. While Rust has led down from the header a pipe through the coolest section of the rear heat zone, and which therefore tends to draw water downward from the header, yet it will be observed that this tube does not deliver its contents into the lower mud drum and thus return it to the main current cycle by which it would again be eventually carried to the upper forward drum, but the delivery is made into a secondary drum, the effect of the water volume of which drum, as we have seen in the reservoir of Schlieper, is to retard or break up the downward current at that point. Now if this reservoir had no other outlet than the tubes leading to the rear mud drum it would correspond in that regard with Schlieper’s pipes, but Rust makes effective provision not alone for preventing the continuance of such a downward current course, but for positively ensuring the return of the water to the upper cross water connection from which it came. This is done by the forward row of tubes which he places in the front heat zone where its tendency is to draw upward from the reservoir the water the rear row has carried down — a step utterly at variance with Schlieper’s object and practice. But not only is it the case that Rust's device is constructed to return to the water-connecting pipes after a temporary absence the water which Schlieper designed to permanently withdraw therefrom, but Rust’s combination of a header, double bank of tubes, lower drum and vertical baffle wall effect other and different results from any obtained by Schlieper. In the first place, it adds a new element in circulation in an interior quadrilateral cycle, which, in addition to its novelty in location, adds one-ninth to the boiler’s tube capacity for circulation, and 25 per dent, of additional heating surface in the forward bank of tubes, and that, too, in the most favorable zone for heat transference. Its construction is such also that it permits the substitution for a much wider, inclined baffler of a narrow, or shell, vertical baffler, a saving of valuable space. By the use of the lower secondary water drum as a base of *858superheated supply from which this minor cycle of circulation draws its supply, and by the heat of which the upward current is stimulated and the downward lessened, it will be noted that Rust draws upon a very different' supply than that of the rear, mud drum, since in any theory of circulation, it must be hotter than either of the lower mud drums; a result due, not only to the pipes which supply it, but to the fact that half of its surface is subjected to the initial heat entering the chamber.
It will be seen that such a drum interposed at the lower end of Schlieper’s tube, and in the rear of the baffle would be a mere negligible factor, but in the Rust construction, through combination with tubes placed in front of the baffler, it becomes a functional factor in producing a very different result. It will also be noted that this drum, its location centrally and directly under the header, makes «possible the use of two banks of central tubes which enter the boiler shell at right angles. It will thus be seen that Rust’s lower drum is a functional element in combination, and that no such element is found in combination in any of the Schlieper claims. The same thing may also be said of the header. We are also of opinion that no such ■element as a vertical baffler, or of the use of such a baffler, between two pipes leading from a lower secondary mud drum is found in Schlieper’s combination. Such being the case, we are of opinion that interpreting this patent, as we are able to do without the aid of intrinsic proof, and applying thereto and to the respondent’s structure certain known and recognized laws of physics, we are warranted in saying that the combination which Rust has used in his boiler is neither the mechanical or functional equivalent of that disclosed by Schlieper. ‘In reaching this conclusion we would by no means imply we have ignored the engineering testimony in this case. On the contrary, we have given the same careful study and due consideration. This testimony would make the case center on the fact whether the tubes passing from Rust’s secondary drum carry water from the secondary drum into the rear mud drum as contended by complainant, or in the opposite direction as contended by respondent. We cannot regard the case as governed by that factor. Assuming for present purposes that Schlieper was the first to draw water from the forward upper drums to the rear lower drum, he acquired no monopoly to that current course, but at most only to a combination of old elements placed in hovel relation to each other, accomplishing that result. But he did not thereby block the way against all others who sought by. combination of other and different elements or to those using other and different combinations of the same elements, to accomplish the same result and especially he did not preclude the accidental and occasional use of such a current path in a device whether other and different elements were employed in other combinations to accomplish wholly different results. Now the proofs in this case unquestionably show that where the Rust boiler was used under such moderate working conditions as permitted the use of plate glass sections and ocular demonstration, the current course was from the rear mud drum upward into the secondary drum. We go further *859and say that under the laws of physics to which we have referred, this was the current course we had reason to expect. This shows that under such conditions the Rust device does not accomplish the same result as Schlieper’s.
It is, however, contended that under actual working conditions this upward flow in the Rust boiler reverses itself to a downward one. We are free to say that, if such is the case, it is at variance with the results to be expected from recognized physical and hydraulic laws. The burden of establishing by a fair preponderance of proof such a circulation exists assuredly rests on those asserting it. As we said in reference to proof of circulation to establish infringement, in Stirling v. Pierpoint Boiler Co., supra:
“In this connection we cannot too strongly emphasize the law that the burden of affirmatively establishing the fact of its actual existence — not of its mere possibility or probability — rests on the complainant.”
Unfortunately, high working conditions preclude the use of plate glass heads, so we can have no ocular proof. It is alleged an electrical indicator, which was devised during this trial by one of the experts, affords reliable means of determining current circulation. Its reliability is as strongly questioned by other scientific men. We do not. discredit or distrust the accuracy of that instrument when we say that its use has been for so short a time that it has not yet attained such an indorsement and recognition by scientific men as a standard instrument as would warrant a' court in holding that its findings met the burden of proof resting on those who rely wholly upon it in that regard, viz., to show a downward current into the rear mud drum.
After patient study and careful consideration of the case, we are clearly of opinion the charge of infringement is not established, and the bill must be dismissed.