BARRETT, Circuit Judge.

Maloney-Crawford Tank Corporation and The Fish Investment Corporation, brought this patent infringement action against Sauder Tank Company, Inc., to obtain damages for, and injunctive relief against, alleged infringement of United States Letters Patent Nos. 2,804,935, 2,804,940, and 2,804,941, hereinafter called the '935, '940, and '941 patents. These patents were issued to A.J.L. Hutchinson on September 3, 1957.

The Fish Investment Corporation, (Fish), is a Delaware corporation operating out of Houston, Texas, and the present owner of the patents. Maloney-Crawford Tank Corporation, (Maloney), is a Delaware corporation operating out of Tulsa, Oklahoma, and the exclusive licensee of the patents. Sauder Tank Company, Inc., (Sauder), is a Kansas corporation doing business out of Emporia, Kansas.

On January 29, 1971 the District Court filed its Findings of Fact and Conclusions of Law holding the '940 patent invalid and therefore not infringed, and the '935 and '941 patents valid and infringed.

Sauder appeals that portion of the decision which held the '935 and '941 patent valid and infringed. Maloney and Fish filed a cross-appeal with respect to the Court's holding that the '940 patent was invalid and not infringed.

The patents all resulted from the necessity of dehydrating natural gas which was being piped into remote, sparsely populated mountainous areas. They were designed to prevent the gas from reacting with the water in the gas, thereby forming solid hydrates which plugged the flow of gas in the pipeline.

The '935 and '941 Patents

The trial court held that the '935 and '941 patents were valid and that "any apparatus containing all the elements of the '935 patent which operates in a manner whereby the flow of liquid brine is metered or limited, and is unrelated to the rate of gas flow, infringes the '935 and '941 patents." Although the Court expressed its reluctance to uphold the validity of the patents merely because "they teach metering a liquid at a predetermined rate independent of rate of gas flow, rather than a rate which is directly related to the gas flow as taught by the prior art", it did hold that Sauder had failed to overcome the presumption of validity of the '935 and '941 patents; that the '935 and '941 patents were valid as a matter of law; and that Sauder's apparatus and process infringed upon them.

The '935 and the '941 patents are inter-related. '935 sets forth an apparatus. '941 sets forth the process that implements the '935 apparatus. The '935 patent is entitled "Vapor-Liquid Contacting Apparatus". This invention relates to improvements in apparatus for contacting vapors or gases with liquids to effect component interchange. It specifically relates to contacting vapors with liquids under conditions wherein the composition of both the gas and the liquid within a given contacting unit are constant over the entire cross-section of the unit. The amount of liquid contacted on a given unit is controlled independently of the vapor flow and independently of the rate of down-flow of liquid in a column. The independent control of the amount of liquid on a given contacting unit is made possible by the utilization of a weir, or metering device. This device also determines the amount of liquid which can be recycled by a contacting unit. Different weirs with different openings or ports can be used in an apparatus to achieve effective dehydration under a variety of conditions.

The '941 patent is entitled "Vapor-Liquid Contacting Method". It is directly related to the '935 patent in that the '941 is directed to a method of operating a component interchange. '941 relates to a method for contacting vapors with liquid wherein the composition of both the gas and the liquid on a given contacting unit are constant over the entire cross-section of the unit. The amount of liquid contacted on a given unit is controlled independently of the rate of vapor flow and independently of the rate of down-flow of liquid in a column. The trial court noted that:

"The method comprises separately introducing liquid and vapor into the unit, collecting and maintaining a first relatively quiescent body of liquid in a liquid collecting zone of said unit, and flowing a metered portion of the collected liquid into a second body of liquid maintained in a liquid entraining zone of the unit, the amount of liquid thus metered being controlled independently of the rate of vapor flow. The method further requires entraining liquid from the second body by vapors flowing upwardly through the entraining zone at a velocity sufficient to entrain an amount of liquid substantially the same as the amount of liquid metered to the second body of said liquid, passing the vapor and entrained liquid into a disengaging zone of the unit above the collecting zone, separating the entrained liquid from the vapor and passing the separated liquid to the first body of liquid in the collecting zone. Finally the method compromises returning to the second body of liquid a metered portion of the separated and collected liquid for entrained as described, and removing a portion of such liquid from the interchange unit."

The '941 method therefore separately introduces liquid and vapor into a unit, collects a portion of the liquid in a liquid collecting zone, and flows a metered portion of the collected liquid into a second body of liquid which is maintained in a liquid entraining zone of the unit. Through the utilization of weirs the amount of liquid metered into the second liquid entraining zone can be controlled independently of the rate of vapor flow.

Sauder's appeal is predicated on its belief that structural and operational differences between its apparatus and process and the apparatus and process of Fish and Maloney is of such a magnitude that there was no infringement, or, in the alternative, that the claims of the '935 and '941 patents have been construed so as to include the prior art and are therefore invalid. Specifically, Sauder argues that: (1) the '935 and '941 patents are invalid; (2) its device does not have a weir, while the Maloney and Fish device does; (3) its device does not have a separate liquid-collecting means while the Fish and Maloney device does; (4) its apparatus has only one body of liquid on a tray, whereas Maloney's and Fish's apparatus has two; and (5) it had not infringed upon Maloney's and Fish's patent, especially since the patentee cannot recapture that which he surrendered to obtain a patent because "[h]e must abide by the restrictions to which he submitted as a condition to the issuance of the patent." Jones v. Bodaness, 189 F.2d 838, 841 (10th Cir. 1951).

Sauder sets forth in its brief the prior art cited by the Patent Office during the prosecution of the '935 patent in support of its contention that the '935 and '941 patents were invalid. We are also directed to references cited against the '941 patent and a number of Sauder's exhibits. Sauder contends that the most pertinent references cited against both patents are the Kobernik Patent, No. 1,613,532, the Huff Patent, No. 1,741,519, the Lichtenthaeler Patent, No. 1,857,816, and the Kraft Patent, No. 2,222,565. However, during the consideration of the applications for patents ('935 and '941) Fish and Maloney distinguished these prior inventions from their own to the satisfaction of the patent examiner.

Since the record is void of clear error, we must affirm the trial court's determination that the presumption of validity was not overcome. Scaramucci v. Dresser Industries, Inc., 427 F.2d 1309 (10th Cir. 1970). In so doing we are not unmindful of the authority set forth by Sauder in its brief for the proposition that claims reading on the prior art are not patentable and are therefore invalid. However, absent a clear and convincing showing that '935 and '941 read upon the prior art, the presumption of their validity must stand since we find nothing in the prior art, the contradict the trial court's determination that "they ('935 and '941) teach metering a liquid at a pre-determined rate independent of rate of gas flow, rather than at a rate which is directly related to gas flow as taught by the prior art."

Sauder also contends that even if the '935 and '941 claims are valid, its apparatus did not infringe because of the substantial structural and operational differences between its apparatus and process and the Maloney and Fish apparatus. Sauder initially alleges that since its device does not have a weir or metering device, it has not infringed on the Maloney and Fish patents. Hutchinson's inventions were distinguished from the prior art at the time the patents were issued by their utilization of a weir or metering device which would allow for the metering of a constant volume of liquid independent of the rate of vapor flow. Under the prior art certain patents utilized an aspirator or educator effect. Aspirator was defined during the proceedings to mean "a device for drawing a stream of air or liquid through an apparatus by suction". It was further agreed by the experts for both sides, that the terms, "aspirator", "educator", and "ejector", were equivalents for purposes of the case. The basic allegation of Sauder is that its apparatus read on the prior art because it utilized the aspirator effect rather than a metering device. Maloney's and Fish's expert witness admitted on cross-examination that the absence of the metering means and the constant flow through an apparatus would constitute an avoidance of the '935 patent.

Evidence was received in support of Fish's and Maloney's allegations that although Sauder's apparatus did not have a physical component in the form of its weir, the opening at the bottom of Sauder's riser was the equivalent of its weir. Dr. Manning testified that this port, in fact, did act as a metering means, allowing liquid to flow from the first body to another at a pre-determined rate independent of the rate of vapor flow. Sauder's expert, Mr. Reid, testified that Sauder's apparatus had neither a weir nor metering means — but that there was a doubt in his mind. Oliver Fowler, the man who designed the Sauder apparatus, testified that the port in the Sauder riser was not intended to act as a metering weir, but that there was a possibility that it did.

Sauder further alleges that there was no infringement because its apparatus, by not having a metering means, utilized an aspirator effect to effectuate liquid-vapor contact in its apparatus. There was evidence indicating that an aspirator system required smooth walls. There was also evidence showing that Sauder's apparatus did not have smooth walls; that it had a "floating plate"; that the gas flow within Sauder's apparatus included passing through right angles; and that the liquid flow within Sauder's apparatus was substantially constant and independent of the gas flow. Sauder's experts did not negate this evidence although they questioned it. Based upon this evidence the Court found that Sauder's apparatus did not read on the prior art, but rather on the metering claim of the Maloney and Fish patents.

The trial court also found that "the flow rate of the liquid in the defendant's apparatus is substantially constant independent of the gas flow rate." This finding was evidently based on the report of Dr. Francis S. Manning, Maloney's and Fish's expert witness. Sauder questioned Dr. Manning's report. Table III of the report indicates that increased gas flow gives rise to decreased liquid flow rates. Conclusion III of the report notes that "[o]ver the reported design range — the liquid re-circulation rate remains essentially constant. It changes less than 3%." Since the trial court was apparently relying on this conclusion, and Dr. Manning's expertise was not attacked, we find no error in the trial court's reliance on his report.

Sauder also argues that infringement was not present because whereas the Maloney and Fish contacting trays used two bodies of liquid, its contacting trays used only one body of liquid. Sauder contends that since its apparatus has only one body of liquid it is following the teachings of the prior art of Huff and Kobernik. A review of the Court's findings indicates that this point was not considered substantial, and that infringement was established prior to considering it. Sauder also contends that infringement was not present because Claim I of the Maloney and Fish '935 patent calls for a "liquid collecting means" and a "means for maintaining a first body of liquid" and that "in attempting to read these two different means on defendant's device, plaintiffs are compelled to read both means on the same body of liquid on defendant's Plate 12___." (Sauder's emphasis).

The trial court held that "we further find that while defendant's Plate 12 of Exhibit 12A acts both as a collecting means and a means for maintaining the first body of liquid, such does not defeat plaintiffs' claim of infringement. These two elements, represented as being separate facts by the patent claims, were merely combined into one member which performs the duties of both in the same way." The District Court properly found that infringement exists "when a device incorporates in its structure and operation the substance of the patent and produces substantially the same effect in substantially the same way as that taught in the patent." Jamco, Incorporated v. Carlson, 274 F.2d 338, 343 (10th Cir. 1959).

In Graver Tank Mfg. Co., Inc. v. Linde Air Products Co., 339 U.S. 605, 70 S.Ct. 854, 94 L.Ed. 1097 (1950), the Supreme Court stated:

"In determining whether an accused device or composition infringes a valid patent, resort must be had in the first instance to the words of the claim. If the accused matter falls clearly within the claim, infringement is made out and that is the end of it." 339 U.S. at 607, 70 S.Ct. at 855.

Also, infringement is not avoided merely by a change in form. Eimco Corporation v. Peterson Filters and Engineering Company, 406 F.2d 431 (10th Cir. 1968), cert. denied 395 U.S. 963, 89 S.Ct. 2105, 23 L.Ed.2d 749 (1969); Jamco, Inc. v. Carlson, supra. Then, too, whereas every element of a combination patent is conclusively presumed to be essential, every element, or its functional equivalent, must be found in the accused device or there is no infringement. McCullough Tool Company v. Well Surveys, Inc., 343 F.2d 381 (10th Cir. 1965). The trial court was correct in holding that infringement was present because "every element, or its functional equivalent, was shown."

The '940 Patent

The trial court found that the '940 patent failed to meet the requirements of non-obviousness and novelty, because the prior art "discloses all elements of the claims in issue, and their combination constituting the vapor-liquid contacting method would have been obvious at the time of the invention to a person having ordinary skill in the art." The Court also held that Sauder had overcome the presumed validity of the '940 patent by "clear and convincing evidence" and that since the '940 patent was invalid there could be no infringement. We must affirm unless these findings are clearly erroneous.

The '940 patent is entitled "Process for Dehydrating Gas". Claim I states:

"I. A process for dehydrating a moisture containing hydrocarbon gas which compromises passing the gas at a pressure in excess of 100 p.s.i.g. into a pre-drying zone, said zone compromising of a plurality of serially arranged vapor-liquid contacting zones, passing a hydroscopic liquid successfully through said contacting zones, passing all of said wet gas through each of said contacting zones countercurrent to the flow of said liquid, removing spent liquid from the first of said contacting zones, passing the partially dehydrated gas from the last of said zones upwardly through a compact bed of deliquescent solid particles wherein all of said gas contacts said deliquescent particles, the dehydrating power of said particles increasing in the direction of flow of said gas so that the last particles contacted by said gas are substantially anhydrous, removing the dehydrated gas from the system, passing the hydroscopic liquid found in the bottom portion of said bed to the last of said contacting zones through which said gas passes, as the hydroscopic liquid employed in said pre-drying zone, the concentration of dehydrating agent in said liquid decreasing progressively from the last contacting zone to the first contacting zone through which said gas is passed."

At the time patent '940 was issued it was known that solid calcium chloride would absorb water from gas and turn it into a liquid brine which would in turn absorb water from gas. However, the efficacy of this process had been limited because often there was too little water in the gas to create sufficient brine relative to the large volumes of gas to effect sufficient water removal. In the prior art, the utilization of solid calcium chloride alone would result in its dissolution and need of constant replacement. Also, the utilization of brine alone required its regeneration and re-circulation, necessitating the use of pumps, coolers, etc. The prior art also recognized dehydrating natural gas with glycol, but this process required a fire to regenerate the glycol. Surveillance, maintenance and manpower supervision was obviously required.

Under the '940 process water vapor is removed from gases at pressures above 100 p.s.i.g. by first passing the gas to be dehydrated into a tower that has a predrying zone composed of a plurality of serially arranged special liquid-brine-gas contacting zones. Above these zones is a bed of solid calcium chloride pellets which accomplish the final drying. The liquid formed by the dehydration of gas in the second stage acts as a source of brine for the initial stage. As water vapor remaining in the gas passes into the tower of calcium chloride pellets, it converts to brine and drops into the predrying zones. As the brine within a contacting zone builds up, it overflows and drops to a succeeding and lower contacting zone. By this process the concentration of the brine decreases progressively as it flows downward within the serially arranged contacting zones, and the water vapor within the gas is progressively reduced as it flows upward. The efficacy of the '940 process is found in its ability to keep a high ratio of liquid to gas in each of the contacting zones in substantial excess of the net flow of liquid dropping from zone to zone and establishing an equilibrium in each of the zones. The specific gravity of the brine at each contacting zone removes only so much water vapor from the gas.

Maloney and Fish contend that the District Court erred in: (1) concluding that the patented process of Claims V and VI (of '940) failed to meet the requirements of non-obviousness and novelty, in view of the uncontroverted evidence showing that men skilled in the art questioned the usage of calcium chloride in this capacity; (2) in holding that Claims V and VI lacked novelty and were obvious in the prior art even though the prior art was inoperable in the environment, circumstances and usages of the applicant's invention; and (3) in holding that Claims V and VI of the invention lacked novelty and unobviousness because the Court considered them one by one rather than in the aggregate wherein they achieved new and surprising objectives.

Although Maloney and Fish have set forth three rather specific allegations of error, their inter-relationship is such that we shall consider them in aggregate. Claims V and VI of '940 state:

"V. A process for substantially completely dehydrating gas which comprises continuously passing all said gas at a pressure in excess of 100 p.s.i.g. through a plurality of serially arranged liquid gas contacting zones, passing a solution of a dehydrating agent through said zones countercurrent to said gas, the concentration of the solution progressively increasing in each zone in the direction of flow of said gas, removing spent solution from the first of said zones through which the gas has passed, passing the partially dehydrated gas from the last contacting zone upwardly through a compact bed of solid particulate dehydrating agent wherein all of said gas contacts said solid dehydrating agent thereby converting said dehydrating agent from an initially substantially anhydrous form to a hydrated form and converting part of said agent to a substantially saturated solution, said bed being progressively less hydrated from the bottom to the top thereof, passing the saturated solution of dehydrating agent to the last of said contacting zones through which the gas has passed, and removing the dehydrated gas from the system, the volume of liquid maintained in each of said contacting zones being in substantial excess of the net flow of liquid to each of said zones, thereby maintaining a high ratio of liquid to gas in each contacting zone."

"VI. The process of claim 5 wherein substantial equilibrium is established in each of the contacting zones."

Maloney and Fish acknowledge that at the time the '940 patent was issued a person ordinarily skilled in the art would have known that: (1) calcium chloride as a bed of pellets alone had been used to dehydrate low pressure air or gases; (2) calcium chloride as a pumped liquid brine over bubble-cap trays or through packing had been used to dehydrate high pressure gas; (3) calcium chloride as a solid bed where the gas contacts the brine drippings before contacting the solid bed had been taught; (4) gases (usually low pressure air) had been dehumidified in apparatus where a solid bed of calcium chloride is above a bed of packing material (stones, rings, screens, excelsior, etc.), the gas contacting with the downflowing drippings in the packing then dried by contact with the solid calcium chloride; (5) low-pressure high-humidity air had been dehumidified using pellets of calcium chloride above cascading trays of brine over which the air first passed horizontally, thence across the pellets, but have required pumps and coolers to handle the large volume of liquid; (6) gases at atmospheric pressure and 70° F. will contain 1,140 pounds of water for each one million standard cubic feet (MMSCF), whereas gases at higher pressure, e.g., 500 p.s.i.g. at 70° F. will contain only 42 pounds of water per MMSCF; (7) glycol for gas dehydration was a recognized and usually recommended practice; (8) glycol dehydration processes require fired heat to regenerate the glycol by boiling the water out; and (9) dehydration systems dealing with calcium chloride were considered objectional.

Maloney and Fish contend, however, that prior to the '940 patent "no one in the prior art had ever proposed a process for dehydration high pressure (hence little water content) gases using a plurality of contacting zones or trays below a bed of calcium chloride particles wherein: `the volume of liquid maintained in each of said contacting zones being in substantial excess of the net flow of liquid to each of said zones thereby maintaining a high ratio of liquid to gas in each contacting zone.'" (Claim V of '940 Patent) (E.7a). They also allege that the art did not suggest "substantial equilibrium is established in each of said contacting zones." (Claim VI).

Maloney and Fish also contend that the trial court erroneously equated "contacting zones" to a "packed column", and that the prior art was clear that when dehydrating high-pressure natural gas, a packed column cannot be equated to a plate or tray column. They contend that the evidence did not warrant a conclusion that the '940 patent lacked novelty because the subject matter of Claims V and VI were not old. They cited Scaramucci v. Dresser Industries, Inc., supra:

"When attacking the validity of a patent under 35 U.S.C.A. § 102, it is necessary for the challenging party to prove by clear and convincing evidence that all the elements of the invention, or their equivalents are found in a single structure or description where they do substantially the same work in substantially the same way." 427 F.2d at 1313, 1314.

Maloney and Fish conclude that the ultimate question of patentability turns on whether the invention met the nonobviousness requirement of 35 U.S.C.A. § 103, under which a patent will not issue "if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains. Patentability shall not be negatived by the manner in which the invention was made." They allege that if the trial court's decision was based on non-obviousness, it failed to follow the mandate of Graham v. John Deere Co. of Kansas City, 383 U.S. 1, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966), which requires that under Section 103, "differences between the prior art and the claim at issue are to be ascertained". 383 U.S. at 17, 86 S.Ct. at 694. Maloney and Fish contend that the difference between the prior art and the invention was that the prior art did not contemplate an integrated combination of a plurality of contacting zones as defined in Claims V and VI with a solid bed of calcium chloride pellets.

Maloney and Fish further argue that the Court disregarded secondary tests suggested in United States v. Adams, 383 U.S. 39, 86 S.Ct. 708, 15 L.Ed.2d 572 (1966), wherein the Court opined that:

". . . known disadvantages in old devices which would naturally discourage the search for new inventions may be taken into account in determining obviousness." 383 U.S. at 52, 86 S.Ct. at 714.

They contend that these "known disadvantages" included the "questionable results" and "costly installations" of earlier calcium chloride gas dehydration units and the need for adequate dehydration units in remote and virtually inaccessible locations.

Sauder contends that the Court properly considered the prior art and that the '940 patent was invalid because it lacked novelty and non-obviousness. Sauder noted that the prior art included: (1) Smith Patent No. 1,772,089 (Exhibit E), a process of dehydrating a moisture-containing gas by passing the gas into a pre-drying zone composed of a plurality of gas liquid contacting elements, passing a hydroscopic liquid downwardly through the elements, passing the partially dehydrated gas upwardly through the gas liquid contacting elements, into a second drying unit of deliquescent solid particles, passing the hydroscopic liquid formed in the bed (second drying unit) back into the first pre-drying zone; (2) an article, "The Drying of Gases by Calcium Chloride" (Exhibit O), which described an apparatus utilizing Raschig rings as a pre-drying zone, into which a hydroscopic solution has flowed, being obtained by the gas passing through the rings and up into and through solid calcium chloride; (3) an A.G.A. article (Exhibit M), showing the dehydration of moisture-containing natural gas by means of calcium chloride; (4) two articles, (Exhibits N and P), disclosing the dehydration of moisture-containing hydrocarbon gas under pressure in excess of 100 p.s.i.g. by passing it through a plurality of serially arranged vapor-liquid zones concurrent to the flow of hydroscopic calcium chloride brine; (5) an article (Exhibit U), setting forth a dehydration process using a two-stage dehydration system in which the gas is first contacted with a calcium chloride solution and thereafter contacts with a drying bed of calcium chloride, with the calcium chloride solution remaining concentrated by drippings from the dry bed; and (6) a book (Exhibit AF), which teaches the equivalence between packed towers and plate columns.

The trial court noted that Claims I, II, III, V and VI of the '940 patent were in issue, and found that the elements of the claims were all found in the prior art. The Court specifically found that: (1) when the inventor applied for the '940 patent he was not the first to conceive of drying a moisture-containing gas with a hydroscopic material such as calcium chloride; (2) such a process had been used to dry moisture-containing hydrocarbon gas under pressures in excess of 100 p.s.i.g.; (3) the inventor was not the first to conceive of dehydrating moisture-containing gas by either passing it through serially arranged vapor-liquid contacting zones countercurrent to the flow of hydroscopic liquid, or by passing the gas through a bed of deliquescent solid particles; (4) moisture-containing gas had previously been dehydrated in two stages utilizing an initial stage of calcium chloride brine, and a second stage of solid calcium chloride; (5) liquid formed by the dehydration in the second stage had been the source of brine for the first stage; and (6) the prior art recognized that the relative humidity of gas decreases rapidly when pressures of 100 p.s.i.g. are exceeded.

The trial court also held "that when Exhibits E, O, U, N, and AF are read together, the combination of all the elements of the claims in issue of the '940 patent would have been obvious to one skilled in the art at the time of the claimed invention." The Court relied upon the Smith Patent No. 1,772,089, the article, "The Drying of Gases by Calcium Chloride", a portion of the proceedings of the natural gas section of the American Gas Association, 1952, an article, "The Dehydration of High Pressure Natural Gas", and excerpts from a book, Absorption and Extraction. Sherwood and Pickford, 2 Ed., McGraw-Hill (1957).

As we noted, we must affirm the decision of the trial court if the findings are not clearly erroneous. Boutell v. Volk, 449 F.2d 673 (10th Cir. 1971). We must, accordingly, ascertain if the findings with respect to the '940 patent are supported by substantial evidence.

Maloney and Fish have repeatedly contended that the critical claims within the '940 patent are Claims V and VI. They stress that the effectiveness of the '940 patent and its difference between the prior art are within Claims V and VI. By using a small amount of water and by achieving substantial equilibrium within a tray, an apparatus could be effective in cold, remote mountainous areas, and offshore locations; it could be effective with very little equipment, personnel or maintenance; it could be effective without heat, flame, or potential explosives; it could be effective on what were heretofore considered very marginal wells; and it could effectively operate after a well had been shut down and the brine frozen.

Sauder alleges that the prior art cited during the prosecution of the '940 patent was not adequate because the prior art before the hearing examiner was devoid of materials showing the dehydration of moisture containing hydrocarbon gas at pressures in excess of 100 p.s.i.g. and the passing of gas vertically upward and through a bed of solid calcium chloride pellets. Sauder further alleges that the prior art submitted by it to the Court was more relevant than the prior art considered by the hearing examiner during the prosecution of the '940 patent. The trial court declared the '940 patent invalid because it failed "to meet the requirement of non-obviousness and novelty." We must respectfully disagree.

Under Title 35 U.S.C.A. § 282, "A patent shall be presumed valid. The burden of establishing invalidity of a patent shall rest on a party asserting it." Sauder had the burden of establishing the invalidity of the '940 patent. It failed to do so with respect to Claims V and VI.

Much of the prior art discovered by Sauder was more pertinent than some of the prior art discovered by Maloney and Fish. We are aware of cases such as Roller Bearing Company of America v. Bearings, Inc., 328 F. Supp. 923 (E.D. Pa. 1971), which held that the presumption of validity weakens when much of the prior art is not cited. Although this prior art did undermine the validity of Claims I, II, and III, it did not vitiate the efficacy of Claims V and VI.

Separate claims in the same patent are in effect separate inventions. Smith Industries International v. Hughes Tool Company, 396 F.2d 735 (5th Cir. 1968). Each claim of a patent therefore grants the patentee an exclusive right. Stukenborg v. United States, 372 F.2d 498, 178 Ct.Cl. 738 (1967). The fact that a particular claim within a patent is invalid, without more, cannot invalidate the entire patent. Each claim of a patent must be considered separately and each must stand or fall alone. Automated Building Components, Inc. v. Hydro-Air Engineering, Inc., 237 F. Supp. 247 (E.D.Mo. 1964). aff'd 362 F.2d 989 (8th Cir. 1966). Claims should also be read narrowly to avoid the prior art whenever possible. Beckman Instruments, Inc. v. Chemtronics, Inc., 439 F.2d 1369 (5th Cir. 1970), cert. denied 400 U.S. 956, 91 S.Ct. 353, 354, 27 L.Ed.2d 264 (1970).

The record does not warrant holding the '940 invalid. We hold that Claims V and VI are valid. Although the prior art recognized contacting trays, it did not recognize contacting trays with a substantial excess of liquid in comparison to the net flow of liquid to each tray (zone) which would maintain a high ratio of liquid to gas. The prior art did recognize Raschig rings, but the undisputed evidence in the record indicates that such rings require a much greater amount of water and pumps to achieve the liquid-gas contact achieved by the '940 patent which is achieved without pumps (which require maintenance) and large amounts of water. Also, the prior art did not recognize the maintenance of an equilibrium within each contacting zone where each zone (tray) would remove a given amount of liquid from the gas. Although the prior art did recognize a system wherein the liquid (brine) was self-created within the system, the prior art did not include a system within which large volumes of gas could be dehydrated with very little liquid formed, nor did the prior art envision a system in which the liquid down flow was independent of the vapor upflow.

We are aware that mere improvements are not patentable. Greening Nursery Company v. J and R Tool and Manufacturing Company, 376 F.2d 738 (8th Cir. 1967); Shelco, Inc. v. Dow Chemical Company, 322 F. Supp. 485 (N.D.Ill. 1970). However, new and useful manufacture as well as new and useful improvement is patentable. Plant Economy, Inc. v. Mirror Insulation Co., 202 F. Supp. 873 (D.N.J. 1962); Waldes Kohinoor, Inc. v. Industrial Retaining Ring Co., 198 F. Supp. 755 (D.N.J. 1961). Old elements may be combined so as to be patentable. Eimco Corporation v. Peterson Filters and Engineering Company, supra; McCullough Tool Company v. Well Surveys, Inc., supra; Oliver United Filters, Inc. v. Silver, 206 F.2d 658 (10th Cir. 1953). The combination of old elements does not preclude patentability whenever novel results are produced. A.E. Staley Manufacturing Company v. Harvest Brand, Inc., 452 F.2d 735 (10th Cir. 1971). Under the circumstances, the '940 patent was valid even though old elements were used because they did produce novel results.

Maloney and Fish allege that the Court erred in holding that Claims V and VI lacked novelty and were obvious because the prior art was inoperable in the environment, and because the trial court considered each of the elements of the '940 patent one by one rather than considering them in the aggregate, where, when taken as a whole, they achieved a new and surprising result. We agree. There was no question as to the utility and ability of the '940 process to fulfill Claims V and VI. When the '935, '941 and '940 patents are considered in the aggregate their inter-relationship and inter-related validity cannot be disputed or considered invalid. When we consider the lack of maintenance, lack of pumps, lack of personnel, together with the equilibrium achieved within the trays and the downflow of liquid independent of the upflow of gas that resulted from them — all of which was unknown in the prior art, they are not obvious and they are novel. A patent that fits a long felt need should have the benefit of any doubt. Shaffer v. Armer, 184 F.2d 303 (10th Cir. 1950). A primary patent should be given broad and liberal construction. McCullough Tool Company v. Well Surveys, Inc., supra. We therefore hold the the '940 patent is, along with its sister patents, valid.

We affirm the trial court's judgment upholding the validity of the '935 and '941 patents. We reverse the trial court's judgment holding that the '940 patent was invalid and not infringed, and therefore we remand in part for further proceedings determinate of Maloney's and Fish's prayer for injunctive relief and damage award.