OPINION
SMITH, Chief Judge.Plaintiff Dow Chemical Company sues under 35 U.S.C. §§ 271, 281 (1982), alleging that the United States infringed its patent number 3,817,039 (the ’039 patent or the Stewart patent). Plaintiff also seeks to enforce a licensing agreement with the *625United States Bureau of Mines, an agency of the United States Department of the Interior. Jurisdiction over the infringement issue is lodged in this court pursuant to 28 U.S.C. § 1498 (1982), and jurisdiction over the licensing agreement claim lies under the Tucker Act, 28 U.S.C. § 1491(a)(1) (1982). The court limited trial to the issue of patent infringement, and the defense of invalidity, with each party choosing its best sample project from plaintiff’s list of accused projects, for which plaintiff offered its evidence of infringement. For the reasons given below, the court finds that the ’039 patent is valid and that it has been infringed by the United States.
INTRODUCTION
In section I, the court explains the background of the problem of subsidence control, and how the ’039 patent addressed the problem. Section II addresses the government’s defense that the patent is invalid; the court only reaches the question of infringement on finding the patent valid. In section III, the court addresses plaintiff’s evidence of infringement.
I. FACTS
A. BACKGROUND
The invention that became the patent-in-suit was developed by Jake Stewart and Milton Heslep, employees of Dowell, a division of the Dow Chemical Company. It was developed as a means of controlling subsidence, the main cause of which has been the collapse of underground mine voids and the subsequent deterioration of the overburden (the material between the mine void and the surface) due to the lack of support. Subsidence in populated areas results in buckled streets, ruined foundations, and other types of damage associated with dramatic drops in the surface of the earth. It also can cause the destruction of buildings, and in extreme cases, personal injury.
Subsidence has been a problem since the early days of the mining industry in the United States. See, e.g., Pennsylvania Coal Co. v. Mahon, 260 U.S. 393, 43 S.Ct. 158, 67 L.Ed. 322 (1922) (court notes problem of surface subsidence in areas of heavy coal mining in Pennsylvania, in connection with taking claim before the court). Prior to the invention at issue, subsidence control took one of two forms. The more effective form is known as controlled flushing, and is utilized in mines where the underground environment can be entered safely. In such mines, workers enter the mine void and construct bulkheads around the portion of the mine where additional support is needed. Pipelines are then set up through existing holes in the bulkhead or other areas of the mine and a slurry of water and solid material is piped in until the mine void is filled.
The number of mines that can be entered safely, however, is limited. Most mines are either unsafe to enter or have become inundated with water. In those mines, the most frequently used method of backfill-ing 1 has been blind flushing, which usually requires the drilling of many boreholes into the mine voids from the surface. A slurry of water and solid material is then piped into the mine through the boreholes. Some versions of this method are as simple as pouring sand and water down a large funnel-like device into the mine void through a borehole. Such a procedure usually results in a somewhat conical pile of material directly beneath the borehole, though testimony on this point was not in complete agreement. The obvious shortcomings of this procedure are the large number of boreholes needed, and the relatively small amount of support material which can be placed into each borehole.
The invention which underlies the patent-in-suit was a response to difficulties experienced by the town of Rock Springs, Wyoming. Coal mining began in Rock Springs in 1868. Heavy mining continued until 1920 and gradually declined thereafter until 1953, when a drop in the coal market stopped mining activity. As a result of the mining, substantial empty areas beneath the city were created, leaving large voids, or rooms, supported by pillars of coal. *626Over time these pillars of coal deteriorated, allowing the strata above to collapse into the unsupported voids, causing surface subsidence.
In the early 1970s, the United States Bureau of Mines conducted an investigation of the subsidence experienced at Rock Springs, and evaluated existing methods of subsidence control. That investigation rejected the use of controlled flushing because the mines had become inundated with water and were otherwise inaccessible. Although the study found that blind flushing, the only other option available at the time, was feasible, it concluded that the cost might have been as high as $2.5 million, and would have required the drilling of approximately 3,000 boreholes, many on private residential properties. The study therefore rejected blind flushing, and began a search for another method of subsidence control, a search that led to the invention underlying this dispute.
B. THE INVENTION
Milton Heslep and Jake Stewart, while employed by Dowell, conceived of the idea which lead to the ’039 patent during their work in the unrelated field of oil well stimulation during the late 1960s. The invention requires pumping a slurry of water and solids2 through a closed pressurized system of conduits leading into an underground void.3 (A depiction of the system is reproduced in an appendix to the court’s opinion.) The rate at which the slurry is pumped must be at least equal to the minimum linear velocity (MLV)4 of the suspension. MLV is defined in the patent as “the minimum velocity at which a suspension of particles must be conducted through a conduit so that any substantial deposition of particles from the suspension to the lower portion of the conduit to form an essentially stable layer thereon is prevented.” U.S. Patent 3,817,039, col. 2, lines 60-64.
Once the suspension is transported at the MLV through the conduit to the open void, the difference in the cross-sectional areas between the conduit and the open void results in the velocity of the suspension dropping below the MLV. The change in velocity causes the solids to settle from the suspension. The effect of this process is the formation of a stable mound on the floor of the void. As the height of the mound increases and the cross-sectional area between the roof of the void and the top of the mound decreases, the velocity of the suspension again reaches the MLV. At that point, the suspension then continues to move over the mound through the path of least resistance. The process is intended to continue until the void is substantially filled with solids.
The benefits of the invention are listed in United States Bureau of Mines Information Circular 8665 (1975). The Bureau notes that the advantages over other methods of subsidence control are:
(1) Great reduction in the number of injection boreholes. A single injection hole serves the purpose of many injection holes in the gravity-feed method.
(2) More complete vertical filling of mine openings.
(3) More complete area coverage. Areas inaccessible because of surface improvements can be filled.
(4) Less disruption of the community in the form of noise, dust, and traffic interference by drilling operations and trucking of fill material.
The invention was given its first field test in Rock Springs, Wyoming. The Bu*627reau of Mines and Dowell, pursuant to a contract acknowledging Dowell ownership of the invention, conducted a demonstration project in which 20,000 cubic yards of sand was pumped into mine voids covering about 2.8 acres of land over a ten-day period. Enos Carlson testified that the Bureau concluded that the process held “sufficient promise to justify further experimentation.” See also plaintiffs exhibit 271 (1973 memorandum of understanding), and plaintiff’s exhibit 31 (“Hydraulic Model Studies,” by E. Carlson). At about the same time, Dow applied for a patent for the new process.
After completion of the Rock Springs project, on May 11, 1972, the Bureau and Dowell entered into a contract for a second project located in Scranton, Pennsylvania. The contract also provided for the licensing to the United States government of the invention disclosed in Dow’s patent application and any patent issued thereon.
In Scranton, a major goal of the parties was the use of existing mine refuse taken from local refuse piles, rather than the use of sand, as fill material. The project was very successful in that it alleviated two problems: the backfilling of inaccessible mine voids and the removal of large piles of mine refuse. The project filled voids beneath a 30-acre residential area with approximately 450,000 cubic yards of crushed mine refuse. A total of five boreholes were used, with one borehole accepting nearly 200,000 cubic yards of fill material. This was hailed as a phenomenal achievement in a paper by A. Allen & R. Whaite, Pumped-Slurry Backfilling of Inaccessible Mine Workings, as well as in The Moving Earth, a film produced by the United States government.
After the completion of the Scranton project, the government undertook an extensive program of mine backfilling projects, conducted through independent contractors. Meanwhile, on June 18, 1974, the ’039 patent was issued. Dow requested an accounting for royalties due under the licensing agreement, which request was initially honored. Later the government refused to pay royalties, alleging that the ’039 patent was invalid, and that in any event, the government did not infringe.
Dow then filed for a reissue of the ’039 patent. The Department of the Interior opposed the reissue, and took an active part in the reissue proceedings. After considering prior art, and the government’s assertions of invalidity, the Patent and Trademark Office found the ’039 patent to be valid, and the patent was thus reissued. Dow again requested royalties from the government, but was refused. The present litigation ensued.
Plaintiff alleges that the United States Bureau of Mines and its contractors have been utilizing a pressurized pumped slurry system of filling mine voids, in violation of the licensing agreement between Dow and the United States Bureau of Mines; plaintiff further alleges that such unauthorized use constitutes patent infringement. Prior to trial, plaintiff agreed to supply defendant with a list of all of the projects it believed infringed on the ’039 patent. The parties then agreed that each side would offer evidence as to one project from the list of accused projects, with plaintiff choosing its best sample case for infringement, and defendant choosing its best case for non-infringement. Determining whether any of the other accused projects infringed was left for a later stage of the proceedings, as was the issue of whether any royalties were due under the licensing agreement.
C. CLAIMS IN ISSUE
Claims one through three of the ’039 patent are in issue. They read as follows:
1. In the method of emplacing a layer of solid particles in a subterranean void wherein the layer occupies at least a major portion of the height of said void, comprising injecting a mixture of a carrier liquid and particulate solids through a conduit connecting a work surface and said void, the improvement which comprises:
a. providing a closed system between injection equipment for said mixture, said conduit and said void;
*628b. providing a suspension of said carrier liquid and said solids; and
c. injecting said suspension into said void through said conduit at an injection rate which is sufficiently low such that initially upon entrance into said void from said conduit the velocity of the suspension is below its minimum linear velocity and at least a portion of said solid particles are deposited to form a mound which decreases the cross-sectional area of said void, and sufficiently high to propel the suspension over said mound at a velocity at least equal to its minimum linear velocity to carry particles over said mound whereby they are deposited to increase the length and height thereof, to form a layer of solid particles in said void.
2. The improved method as defined in claim 1 wherein the height of the void is substantially completely filled with said particulate solids.
3. The improved method as defined in claim 1 wherein said conduit comprises a substantially vertical borehole connecting the void with a work surface.
U.S. Patent 3,817,039, column 7, line 7 to column 8, line 4.
D. SCOPE OF THE PATENT
As will be discussed in more detail in the infringement portion of this opinion, the court holds that the patent teaches a pressurized system closed by a series of sealed conduits extending from the injection pump to the top of the void. In addition, the patent’s definition of the term “suspension” includes the particle transportation modes known to hydrology as bedload and saltation, as well as that technically called suspension.
II. VALIDITY
The statutory scheme established by Congress allows an alleged infringer to defend on the ground that the patent is invalid. Defendant argues that the ’039 patent is invalid as obvious under the standards of 35 U.S.C. § 103, anticipated under 35 U.S.C. § 102, and fails to disclose the best mode under 35 U.S.C. § 112. In addition to these arguments, defendant contends that the patent is invalid because the feature which distinguishes the Stewart patent from the prior art is not critical to the patent, and that the formula for MLV was known in the prior art.
By statute, patents are presumed valid. 35 U.S.C. § 282. Alleged infringers must present clear and convincing evidence of invalidity in order to overcome this presumption. Polaroid Corp. v. Eastman Kodak Co., 789 F.2d 1556 (Fed.Cir.), cert. denied, 479 U.S. 850, 107 S.Ct. 178, 93 L.Ed.2d 114 (1986); SSIH Equipment S.A. v. International Trade Comm’n, 718 F.2d 365 (Fed.Cir.1983). The party attacking a patent’s validity bears an even heavier burden when a patent has gone through a reissue proceeding. Fromson v. Advance Offset Plate, Inc., 755 F.2d 1549 (Fed.Cir.1985). The burden on the party attacking validity becomes heavier still when that party was a protester in the reissue proceeding. Windsurfing International, Inc. v. AMF, Inc., 782 F.2d 995 (Fed.Cir.), cert. denied sub nom. BIC Leisure Products v. Windsurfing International, Inc., 477 U.S. 905, 106 S.Ct. 3275, 91 L.Ed.2d 565 (1986). The government has failed to meet this stringent burden.
A. OBVIOUSNESS
Defendant seeks to show that the ’039 patent is obvious under 35 U.S.C. § 103, which provides in part:
Conditions for patentability; non-obvious subject matter
A patent may not be obtained though the invention is not identically disclosed or described as set forth in section 102 of this title, 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.
Section 103 obviousness determinations require an examination of: “(1) the scope and content of the prior art, (2) the differences between the prior art and the *629claimed invention, (3) the level of ordinary skill in the pertinent art, and (4) additional evidence [commercial success, long felt but unsolved needs, failure of others, etc.], which may serve as indicia of non-obviousness.” Environmental Designs v. Union Oil Co., 713 F.2d 693 (Fed.Cir.1983), cert. denied, 464 U.S. 1043, 104 S.Ct. 709, 79 L.Ed.2d 173 (1984) (relying on Graham v. John Deere Co., 383 U.S. 1, 17-18, 86 S.Ct. 684, 693-94, 15 L.Ed.2d 545 (1966)). These inquiries are factual in nature even though the ultimate question of patent validity is a legal one. Graham, 383 U.S. at 17, 86 S.Ct. at 693-94.
1. Scope and content of the prior art
a. Scope
As an initial matter, defendant contends that the scope of the relevant prior art extends beyond the limited area of subsidence control to the more general area of particle transport. The patent defines the problem as finding an efficient method of filling subterranean voids with a particulate material. See generally U.S. Patent 3,817,039, columns 1 & 2. Clearly, particle transport is an element of the relevant prior art, but only as it relates to methods of subsidence control. Thus, the relevant prior art is the underground deposition of material. The invention itself is aimed at increasing the efficiency of placing material in underground voids for the purpose of subsidence control. The problem addressed was not how to transport material in a pipeline, but rather how to place large amounts of material in voids in an efficient manner.
Both defendant’s and plaintiff’s characterizations of the prior art address the problem of filling underground voids rather than increasing the efficiency of particle transport in general. Of course, increased efficiency of particle transport has a bearing on increased deposition of material, but it is not the underlying technology; moreover, overly efficient particle transport is undesirable in this technology. Defendant suggests that the nature of the problem addressed was an asserted particle transportation deficiency in the prior art blind flushing process, and makes several references to the original patent application in support of that assertion. However, the court’s examination of the application discloses no reference to inadequate particle transportation. The only references to inadequacies in the prior art discuss the inadequacies in the two broad types of subsidence control systems. The court notes that plaintiff did not limit its invention to the art of subsidence control but generally addressed the problem as the “method of filling subterranean voids with a particulate material.” U.S. Patent 3,817,039, column 1, lines 2-3. The court holds that such is the scope of the relevant prior art.
b. Content
Several prior art patents, references, and projects were used at trial to attack the validity of the ’039 patent. The government focused upon four patents: Schmidt, et al. (Patent No. 2,710,232), Doolin (Patent No. 3,440,824), O’Neal (Patent No. 3,459,-003), and Magnuson (Patent No. 3,500,934), with specific attention paid to the Schmidt and Doolin patents. In addition to the foregoing patents, defendant relies upon the Francois patent (Patent No. 1,391,678), the Colliery Guardian, Hanson, and McElroy articles, and the prior art projects at Upper Tyrone, Peterman’s Corners, Eynon Street, Empire Street, and Spring Street. Defendant also appears to rely upon the fracturing models presented to Dowell customers in the 1960s. Each prior art item will be discussed in turn, and differences between the prior art and the invention will be analyzed.
(1) The Schmidt, et al. patent
The Schmidt patent, number 2,710,232, discloses a pneumatic (gas) system for backfilling underground voids. The patent provides for a suspension of solids in a gas stream to be conducted into the cavity to be filled. This fluid-like suspension of gas and solids operates under pressure and is directed to the void. As such, the system appears to be closed as contemplated by the '039 patent. The differences between the patent-in-suit and the Schmidt patent are substantial enough, however, for the court to conclude that the ’039 patent is *630neither obvious from nor anticipated by the Schmidt patent.
The biggest difference is that the carrying medium in the Stewart patent is liquid rather than gas. Dr. David T. Kao, the government’s primary expert witness,5 testified that there are two differences between gas and liquid carriers. First, air does not have the capability to erode material as does water. Air works only where pressure forces it to work, while water works with both gravity and pressure. Second, Dr. Kao testified that gas as a carrier would not work properly in situations where a mine was inundated, analogizing such a situation to blowing air into a fish tank.
Although Dr. Kao testified that the operation of gas and liquid would be substantially identical in non-inundated voids, the court does not find this to be a convincing observation. The gravitational erosive power of water is significant, and would provide the additional impetus to clear the way for continuous mound formation through breakout channels as described in the ’039 patent. This is consistent with representations in the Schmidt patent and in Dr. Kao’s testimony concerning that patent. The Schmidt patent noted that initially in its dense-phase injection of material the borehole would be blocked by a steeply banked conical pile of material.
Dr. Kao agreed with this. The Schmidt patent provides a larger mound of material by later diluting the mixture of solids and gas and in essence blowing away the top of the cone with the increased pressure caused by the diminished amount of solids carried by the gas. This in itself is a convincing difference between the two patents. Another problem with the government’s contention follows from Dr. Kao’s testimony concerning the differences between gas and liquid systems. Gas sys-terns require pressure to operate where liquid does not, therefore, a pressurized gas system in the prior art does not obviously lead to a pressurized liquid system when it is thought that a liquid system may work adequately on gravity alone.
(2) The Doolin patent
The Doolin patent, number 3,440,824, discloses a backfilling system whereby a slurry of water and solids is pumped to a nozzle suspended over a delivery pipe extending to the void. Doolin shows a directional nozzle attached to the delivery pipe at the void and a second borehole and pipe designed to withdraw water from the void in order to create a current between the injection borehole and the extraction borehole. Defendant argues that the Doo-lin patent has every limitation of the three claims in dispute here except that the suspension-conveying system in Doolin is open between the pipe carrying the suspension and the delivery pipe extending to the void. Defendant’s point misses the major innovation of the '039 patent. As will be seen below, the major innovation of the ’039 patent is that a pressurized system of conduits closed to the void is used.
Dr. Kao suggested that it would be obvious to modify Doolin to conform to the limitations of the ’039 patent. While it may appear obvious to the defendant, with the benefit of hindsight and a trial, it is not an obvious change to the court. Closing the system to the void with the outlet pointing straight down is not obvious. Doolin must not have thought so and in fact taught that a vertical borehole would become plugged by a conical pile of material. Thus, Doolin added a directional flow orifice to compensate for this perceived problem. Plaintiff suggests that this is a classic example of “teaching away” from the invention, and the court agrees. See Ray-*631theon Co. v. Roper Corp., 724 F.2d 951, 961 (Fed.Cir.1983) (teaching away from the prior art supports a conclusion of nonob-viousness).
Mr. Stewart’s testimony also indicates that in an open system such as Doolin, hydrostatic head is the upper limit of pressure which can be directed to the void. This is true because once the limit of hydrostatic head is reached the slurry will begin to back up out of the open area of the conduit. Plaintiff’s exhibit 279, an internal Department of the Interior document, reflects this position. Furthermore, once the pipe becomes full, velocity does not increase. Thus, open systems such as Doolin lack the ability to put continuous pressure into the void as the ’039 system does. Therefore, Doolin cannot be the source of an obviousness objection.
(3) The O’Neal patent
The O’Neal patent, number 3,459,003, discloses an aqueous slurry method of initially pumping waste-spent shale into a mined out area, and converting a portion of the slurry into cement, which then is used to fill in voids and to compact already deposited waste-spent shale. Defendant contends that because the patent uses the word “pumped” to describe the process of placement of the slurry into the void, the O’Neal system inherently contemplated a closed system. However, the presence of a pumped slurry does not necessarily indicate a pressurized system closed to the void, and nowhere in O’Neal is there a requirement that the system be closed to the void. The O'Neal specification merely suggests that the slurry be pumped into a mined out area through lines or wells, and as Mr. Bruce Kanuch testified, there is nothing in O’Neal which suggests a closed system, or even that a suspension is used. See U.S. Patent 3,459,003. Thus, the ’039 patent is not rendered obvious by O’Neal.
(4) The Magnuson patent
The Magnuson patent, number 3,500,934, discloses a pneumatic system for injecting fly ash into a mine void. The patent calls for the insertion of a capped length of pipe with a total length of about 10 feet into the borehole. The pipe therefore does not extend fully to the top of the void and thus does not physically close the system as called for in the Stewart patent. Mr. Malcolm Magnuson testified that the injection pipes in his system occasionally jumped out of the ground. Clearly, the Magnuson patent does not include the closed system of the Stewart patent.
Dr. Kao’s testimony that Magnuson was a closed pressurized system does not undermine plaintiff’s position. Dr. Kao’s definition of a closed system is any system where back pressure can be established, whereas the ’039 patent defines “closed” as a conduit extending from the pump to the top of the mine void without a break. The definition in the patent is controlling. Studiengesellschaft Kohle MbH v. Eastman Kodak Co., 616 F.2d 1315 (5th Cir.), cert. denied, 449 U.S. 1014, 101 S.Ct. 573, 66 L.Ed.2d 473 (1980) (in construing patent, court looks to claims of patent itself and to intention of patentees).
The merit of the Stewart system is that the slurry enters the mine void under pressure. Under Dr. Kao’s definition, the objective of the Stewart patent is not met. Dr. Kao’s version of “closed” would allow pressure to dissipate at points other than the top of the void. Therefore, this portion of Dr. Kao’s testimony is not supportive of defendant’s position that the Magnuson patent teaches the invention.
(5) The Francois patent
The Francois patent, number 1,391,678, discloses a system of closed pressurized pipes inserted in an open mine shaft. The pipes include a set of control valves placed within the open shaft at the bottom of the insertion pipe near the void to be filled. Francois also teaches that a non-settling suspension such as concrete or cement is used to fill the void. Although Francois teaches the use of a closed system, it does so in the context of a controlled flushing operation utilizing a non-settling suspension. It does not teach, nor even claim, the use of a vertical borehole down into an inaccessible or inundated void. The ’039 patent addresses a different problem than does the Francois patent, and it proposes a *632solution to subsidence in areas where only blind flushing techniques could be used.
The differences between the Stewart patent and the Francois patent are substantial. In Stewart, the process utilizes available materials such as fly ash, sand or other waste material. Cement or concrete is much tougher to work with and is generally used only in systems where controlled flushing is appropriate. Furthermore, the use of a pipe in areas where only blind flushing can be used is not obvious. As is seen in Magnuson and other patents relied upon by defendant, the borehole itself generally has served as the conduit to the void. The crucial innovation of the Stewart invention was the lining of the borehole to the void with a closed pipe acting as a conduit. In Francois, the pipe is necessary because the open mine shaft could not properly act as a conduit. One skilled in the art in early 1970 would have had no reason to line a borehole with pipe, and there is no evidence that such lining actually was done prior to the time of the invention at suit.
(6) The Hanson article
“Underground Mine Fires at Tintic Standard Mine,” by Earl F. Hanson, in 26 Transactions oftheAmer. Inst, of Mining and Metallurgical Eng. 249 (1937), discloses a method of sealing off an area of the Tintic Standard underground mine in order to gain control over ongoing mine fires. The article shows that a series of pipes with control valves at various locations were put down existing mine shafts and through bulkheads built around the fire zone. Slime, a mixture of water and fine rock, was “poured” into the mine and created a seal around the fire. The Hanson article contains references from two mines, the Tintic Standard mine, and an Anaconda Copper Mining Co. mine in Butte, Montana. The article suggests that at the Butte mine, boreholes were made and a suspension pumped down them under pressure. There is no indication that the boreholes were closed to the void.
(7) The Colliery Guardian article
“Hydraulic Stowing,” Colliery Guardian, Vol. 207, No. 5337, pp. 185-87 (Aug. 8, 1963), discloses the pumping under pressure of a sand/water mixture down a vertical borehole to empty mine voids below. It does not indicate that the conduits were closed to the void, nor even that pipes were used in the boreholes, nor is it clear whether the article discloses a controlled or blind flushing technique.
(8) The Upper Tyrone and Peterman’s Comers projects
Defendant additionally relies upon the projects completed at Upper Tyrone and Peterman’s Comers in Allegheny County, Pennsylvania, to support obviousness and anticipation. Defendant argues that the Upper Tyrone and Peterman’s Corners projects used systems closed by fluid-tight couplings from a pump through a hose to a borehole casing, extending vertically down into rock. There is no allegation that these projects, which were public and took place more than one year prior to the invention, extended the borehole casing to the void. In fact, Mr. Magnuson testified that when a three-inch borehole was used, the two-inch casing extended only twenty feet into the ground. When a six-inch borehole was used, a four-inch casing extended only ten feet into the borehole. Mr. Magnuson further testified that the ground in the vicinity of the injection borehole heaved but that there was no leakage from the borehole. However, when asked by the court how the ground in the vicinity came to be moving, Mr. Magnuson testified that the fly ash slurry would proceed through various crevices extending from the borehole rather than the mine. This testimony indicates to the court that the Upper Tyrone and Peter-man’s Corners projects were not closed as contemplated by the '039 patent.
(9) The Eynon, Empire & Spring Street projects
Defendant alleges that three projects, at Eynon, Empire, and Spring Streets, in Scranton, Pennsylvania, included all the limitations of claims one, two and three of the ’039 patent. The court finds two faults with this contention. First, while the project specifications required casing all the way to the void, testimony indicated that this was only done once among the *633many boreholes that were filled during the projects in question. Second, the testimony indicated a haphazard attempt at sealing the pipes that did exist. Mr. Michael Naples, defendant’s main witness concerning these projects, testified that he used a pipe with an opening at the top. He testified that he would then put a closing weld all around the pipe to stop the leakage from around the outer pipe even though the specifications did not call for such a weld, and he preferred the leakage. Mr. Naples’ further testimony on this point is also confusing. The following exchange is typical of Mr. Naples’ responses on cross examination.
Q. [Mr. Lieberman] Is your testimony that in every single case you welded all the way around?
A. [Mr. Naples] Correct.
Q. You put in the ring.
A. Correct.
Q. Not in the specs. Didn’t you also testify that we didn’t deviate from the specs?
A. I didn’t understand the question.
Q. Didn’t you also testify that you didn’t deviate from the specs in Eynon.
A. No, we didn’t.
Q. So if its in the spec, you welded it. If its not in the specs, you didn’t weld.
A. That’s correct.
Q. Okay, I accept that. Now find it in the spec.
A. Find it in the specs?
Q. Yes.
[Mr. Buchan] May it please the court.
[The Court] Yes.
[Mr. Buchan] Which exhibit is the witness looking at?
[The witness] 118.
[The witness] I can’t find anything in here.
The court finds this testimony to be at odds with the contention that the system was closed by an ongoing process of welding, unwelding, and stuffing various connections with rags. Mr. Naples also changed his testimony following a post-deposition lawsuit against his company. At deposition, he stated that the coupling between the five-inch hopper pipe and the six-inch borehole casing was “more or less seal[ed,]” yet during testimony at trial he stated that it was “definitely sealed.”
Further testimony indicates that Mr. Naples agreed that the ’039 patent’s pumped slurry system was new and an improvement over prior blind flushing systems. Mr. Naples said that he had never seen anything like the pumped slurry system before. He also testified that, in an interview with Alice Allen, he had said that the pumped slurry method allowed much more slurry to be fed into a hole. Although he did attempt to qualify that statement at trial with testimony that with all things being equal the blind flushing techniques would allow either an equal amount of slurry or just a bit less, this qualifying testimony is inconsistent with his testimony that only once did he recall a blind flushing technique exceeding more than 12,000 cubic yards per borehole. This is particularly striking when considered in light of his testimony that the pumped slurry process used at Green Ridge I resulted in 500,000 cubic yards of material being flushed down five to ten boreholes. Second, there is no testimony or evidence indicating that the purported prior art projects were done under pressure as is called for in the ’039 patent.
(10) The McElroy article
The McElroy article refers only briefly to the subject matter relevant here. The article describes a procedure for controlling mine fires by pumping water and slush through cased boreholes. Nowhere does the article explicitly suggest a closed pressurized system which would form a mound as contemplated by step (c) of claim one.
(11) The Dowell fracturing process
As the court understands it, defendant contends that Dowell’s fracturing models shown to potential customers in Montana in 1966, along with the actual Dowell fracturing process, constitutes prior art which renders the ’039 patent obvious and anticipated. In short, fracturing is the use of pres*634sure to inject a slurry of water and sand or other solid material into an underground fracture, created by hydraulic pressure, in the rock surrounding an oil or gas well. Sand is then injected into the fracture in order to keep it open after the hydraulic pressure is released. The purpose of fracturing is to stimulate production of oil or gas. Thus, the bigger grained and the more porous the sand or other material is, the greater the flow from the well.
The court’s first concern here is whether fracturing is even prior art. As previously stated, the scope of the relevant prior art is limited to questions of how to develop better methods of filling underground voids with particulate material. In fracturing, no voids initially exist, but, rather, fractures or breaks in the underground rock or strata temporarily are created through the use of hydraulic pressure. Sand or other particulate material then is forced into the fracture to prop it open. The technology does not bear upon how best to fill a currently existing void. The only relationship is the limited one of particle transport to the underground area. Therefore, the technology of fracturing is not prior art.
Even if the fracturing method could be construed to be prior art, it neither anticipates nor renders obvious the ’039 patent. There is, for instance, no mound formation as required by claim three of the ’039 patent. Even though fracturing requires injection of the slurry into a closed system at least at MLV, this does not render the ’039 patent obvious or anticipated, since the fracturing process does not include the main elements of the ’039 patent’s closed system. Testimony indicated that the ’039 patent works in a void because the closed system creates what Mr. Stewart called a roofing effect, where breakout channels are created as the mound approaches the roof of the void. This is simply not related to the fracturing process.
2. The differences between the prior art and the ’039 patent
As seen from the examination of the prior art, the major and significant advance in the ’039 patent is the addition of a system of conduits closed to the void and under pressure. Mound formation is directly related to the use of a pumped slurry system. Mound formation, of course, is difficult to verify because- the voids where the pumped slurry system is used mostly are inaccessible. However, there are many reports, such as Bureau of Mines Information Circular 8667 (1975), prepared prior to the lawsuit, which suggest that mound formation does take place. In addition, physical tests done by the drilling of boreholes in the filled area indicate mound formation, as does the series of tests done with models. Plaintiff’s exhibit 30, a report prepared for the Bureau of Mines entitled “Hydraulic Model Studies for Backfilling Mine Cavities,” indicated that mound formation as described in the patent does take place in the laboratory.
Mound formation under a pumped slurry system is quite different than cone formation under the prior art systems. All of the pre-lawsuit literature on the topic prepared by the government’s own employees charged with subsidence control responsibilities recognized this. It was not until trial that the government’s employees changed their position on cone formation. Dr. Kao testified that cone formation does not take place under the old system. The court believes that both parties are, to some extent, correct on the cone formation question. The evidence shows that in most boreholes an amount of solids equal to more than just the volume of a geometric cone the height of the void was inserted. This evidence, however, does not completely do away with the cone theory, because the evidence indicates that something more than a basic cone but considerably less than the type of mound formed by the ’039 patent forms under the “prior art” methods. Mr. Stewart testified on cross examination that “[if] you pour a suspension into the mine it’s going to form in a cone, and the size and shape of that cone is going to be a function of how much head you put on it.” The Stewart system leads to dynamic mound formation over a greater area than traditional cone-type formations in blind flushing, and therein lies its departure from the prior art.
*635Mr. Stewart also testified that with increased head, under the prior art, some material from the cone may wash away. Mr. Stewart’s testimony here was based on his use of models rather than actual mine voids, but the court believes that the behavior of slurry in models is substantially the same as it is in actual use. Of particular interest is the consistency of Mr. Stewart’s position when compared with the government’s contention that more than just the volume of a cone is placed in mine voids through the prior art blind flushing techniques. In actual use, the static head developed by the height of the surface above the void would, it appears, cause breakout channels to form on the top of the cone. The extent of additional material injected would be limited to an amount far short of that possible under a system using a dynamic head. Irregular walls and variations in the shape of actual mine voids could cause solids somewhat beyond the volume of a geometric cone to be injected. These phenomena explain the additional material without leading to mound formation as contemplated by the ’039 patent. Thus, the differences as pointed out in the previous discussion are significant, and are a major break from the prior art.
3. The level of ordinary skill in the pertinent art
The person of “ordinary skill” in the art is not the inventor, of course, but rather an imaginary person. Kimberly-Clark Corp. v. Johnson & Johnson, 745 F.2d 1437, 1454 (Fed.Cir.1984). This imaginary person having ordinary skill in the art is determined by reference to the factors set out by the Federal Circuit in Environmental Designs v. Union Oil Co., 713 F.2d 693, 696 (Fed. Cir.1983), cert. denied, 464 U.S. 1043, 104 S.Ct. 709, 79 L.Ed.2d 173 (1984). Those factors in a particular field are: (a) the educational level of inventors and active workers; (b) the types of problems encountered in the art; (c) the prior art solutions to those problems; (d) the rapidity with which innovations are made; and (e) the sophistication of technology.
a. Educational level of inventors and active workers in the field
John Stewart, one of the inventors, received a Bachelor of Science degree in engineering from the University of Wyoming in 1960. Since 1960, he has worked for either Dowell or Dowell Schlumberger in the design and evaluation of oil well stimulation treatments such as fracturing. Mr. Stewart’s co-inventor, Milton Heslep, does not have a college degree, but has many years of practical experience in the field. At the time of trial, Mr. Heslep was a product manager in Mining Services at Dowell.
Others practicing in the field include mostly government employees and private contractors. Donald Donner, a mining engineer, mine safety inspector, and at the time of trial, chief of the Federal Reclamations Project Branch of the Bureau of Mines, received a Bachelor of Science degree in geology from Iowa State University. The degree was awarded in 1953, and since that time Mr. Donner has worked in the field of mining. Another Bureau of Mines employee, Ralph Whaite, also was skilled in the area at the time of the invention. Although he did not have a college degree, Mr. Whaite was well learned in the areas of mining and mine backfilling. Mr. Whaite began his career in mining in 1935, and began his employment with the Bureau of Mines in 1948. Mr. Whaite retired from the Bureau in 1974, and has remained in the employ of the Bureau on an annuitant basis since 1975. Alice Allen, a geologist who joined the Bureau’s subsidence group in 1971, described Mr. Wbaite as the Bureau employee with the most knowledge of subsidence control. Ms. Allen published a 1975 Bureau of Mines information circular along with Ralph Whaite, entitled “Pumped-Slurry Backfilling of Inaccessible Mine Workings for Subsidence Control.”
Another person skilled in the area at the time of the invention was Michael Naples. Although Mr. Naples has no college degree, he has years of practical experience in subsidence control, and was the contractor for several Bureau of Mines backfilling projects.
*636b. Types of problems encountered in the art
The art itself is the art of subsidence control, accomplished through various methods including the transportation of particles to underground voids in either a gas or a liquid. The problem encountered in the art as testified to by Dr. Kao was the prevention of surface, subsidence. The pri- or art and the ’039 patent are methods of transporting as much solid material as possible into an underground void. The major problem in the prior art continually has been, at least until the ’039 patent, the relatively small amount of solids being transported through each borehole into the void. Therefore, the goal has been to find a method to increase the amount of material inserted into each borehole, thereby reducing the number of boreholes and, as a result, reducing the amount of time, labor, and inconvenience involved, c. Prior art solutions to those problems
Plaintiff suggests that its exhibit 201, a 1974 summary of the state of the art of subsidence control prepared for the Appalachian Regional Commission by General An-alytics, a consulting firm in Monroeville, Pennsylvania, provides a proper summary of the state of the prior art of subsidence control. That exhibit lists and explains various types of solutions, including grout columns, piers constructed within the mine, deep foundations, groutcasing, hydraulic flushing, pneumatic filling, fly ash injection, grouting, overexcavation and backfill-ing, blasting, and the Dowell pumped slurry process. Based upon all of the testimony, the court finds this report to be a good summary. The basic solutions are set out below.
(i) Grout columns. Grout columns consist of grout — usually a compound of port-land cement and fly ash — injected down a borehole lined with a perforated ground injection pipe inserted to the base of the mine. The pipe slowly is withdrawn as the grout forms a conical column.
(ii) Piers constructed within the mine. This solution, like controlled flushing, requires an accessible mine void. As its name implies, supporting piers made out of timber, masonry, concrete, or a like material are constructed within the mine voids.
(iii) Deep foundations. Deep foundations are used when the bottom of the mine is relatively shallow, within 100 feet of the surface. Using this system, builders anchor their buildings directly to the mine floor through the use of drilled piers, as they would when building on any unstable ground.
(iv) Groutcases. Groutcase supports are drilled piers extending from the roof to the floor of the mine void. The intent is to support the overburden, in a manner similar to grout columns. Groutcasing is used where high voids exist. In such high voids, grout columns would require a large amount of material, and thus, would be relatively expensive.
(v) Hydraulic flushing. This technique, as has been discussed previously, is the backfilling of mine voids with solids transported within a water slurry. It encompasses both controlled and blind flushing. The technique requires the mixing of water and fill material and the transport of the resulting slurry to the void through either a borehole, as is generally done in blind flushing, or through a series of pipes, as is done in controlled flushing. Examples of the various hydraulic flushing systems are the Doolin, O’Neal, Colliery Guardian, Francois, and Hanson prior art systems, and the Eynon, Spring, and Empire Street prior art projects, all of which the court has noted differ significantly from the pumped slurry system.
(vi) Pneumatic filling. These systems involve backfilling by air pressure. The report on the state of the art of subsidence control separates pneumatic systems from the fly ash systems, which are also pneumatic in nature. Because the report differentiates between the systems the court also will do so. However, none of the prior art systems before the court were non-fly ash pneumatic systems.
(vii) Fly ash injection. Fly ash systems, because of the fineness of the fill material, can operate well with either water or air. The system is differentiated solely on the *637nature of the fill material, as opposed to the nature of the system. Thus, fly ash has been used in both hydraulic and pneumatic systems. Examples of these systems before the court are the Upper Tyrone and Peterman’s Comers prior art projects using a hydraulic fly ash system, and the Schmidt and Magnuson prior art systems using pneumatic fly ash injection.
(viii) Grouting. Grouting is a hydraulic injection with a cementing material. Grouting can be accomplished using either gravity or pump-generated pressure.
(ix) Overexcavation and backfilling. This method attacks the problem by elimination of the abandoned mine voids themselves. Essentially the mine is excavated completely and backfilled with overburden.
(x) Blasting. In theory, subsidence could be avoided by blasting the roof down and the floor up, thereby filling the mine with crushed rock, under U.S. Patent No. 1,004,419.
d. Rapidity with which innovations are made
Subsidence control as an art has been continuing for well over 100 years. Defendant fails to point to many novel innovations made during this time. It appears from the testing and evidence that the pri- or art was only barely adequate to perform backfilling tasks. The current innovation is the first system to draw rave reviews in the industry, and appears to perform significantly better than prior art systems,
e. Sophistication of the technology
As is clear from the record and the evidence, the backfilling of mines does not require highly sophisticated technology. The equipment used, and all parts necessary, are commercially available. It is the combination of these previously known and used items and systems which distinguishes the ’039 patent from the prior art. As stated by several witnesses, there is nothing new about the use of minimum linear velocity to transport solids hydraulically; it is the use of the MLV in the closed pressurized ’039 system which makes the patent different from the prior art.
It seems appropriate here to discuss the methods of determining MLV, since the question bears upon whether a non-engineer can operate the system. While the patent provides a sophisticated formula, which defendant contends is the only method provided for determining MLV, the testimony and the patent itself indicate that MLV can be determined in any number of ways. The easiest way, which would be known to anyone practicing in the art, is by experimentation. Testimony shows that the absolute MLV can be determined by pumping a slurry through a clear pipe at a velocity slow enough to cause a stable layer of solids to form. Velocity is then increased until solids begin moving in bed-load, at which point MLV is reached. Once the velocity and the area of the pipe are known, it is easy to calculate MLV by the well-known formula, Q = AV, where Q is the amount of material injected per unit of time, V is the velocity at which the material is injected, and A is the cross-sectional area of the pipe.
Testimony by Mr. Carlson also indicated that MLV can he determined by certain publications reporting the results of experiments already performed. In Mr. Carlson’s opinion the formula is irrelevant, and the court agrees. The formula provides a velocity which will work, but it is not the absolute MLV. Mr. Stewart called it the “brute strength” formula, and analogized using the formula to driving a car through a brick wall at 100 MPH when one knows that 10 MPH will work.
The level of technology and education in this field is not esoteric. The individuals in this field, while requiring specialized knowledge and experience, need no advanced degrees or extraordinary training. The court concludes that the hypothetical person of ordinary skill in the art is best represented by a person with an undergraduate degree in engineering, geology or the equivalent, or who has practical experience equal to the knowledge that such a degree would provide. Mr. Whaite and Mr. Donner are good examples of persons of ordinary skill in the art.
*6384. Additional evidence (secondary-considerations)
The additional evidence is heavily skewed in plaintiff’s favor. The evidence shows that prior to the development of the pumped slurry process, the amount of backfill which could be inserted in a borehole was generally limited. For instance, as noted earlier, Mr. Naples testified that he had seen only one borehole under the prior art take more than 12,000 cubic yards of backfill, and that hole took only 15,000 cubic yards. The testimony has shown that the pumped slurry process can place an amount of backfill limited only by the size of the void. At the Green Ridge project, for example, the pumped slurry method inserted approximately 450,000 to 500,000 cubic yards down five boreholes, with one borehole taking nearly 200,000 cubic yards.
Other evidence is even more telling. The government’s praise for the new pumped slurry process was almost never-ending during the 1970s. The government made a movie on the topic, clearly stating that the pumped slurry process was a new and useful invention. Bureau of Mines Information Circular 8667 states:
In undermined areas, new solutions to subsidence problems are being sought. The Bureau of Mines is investigating a hydraulic backfilling technique whereby fill material is pumped as a slurry through a closed system and widely distributed in inaccessible mine workings from a single borehole. A full scale demonstration was completed in 1973 in Scranton, Pennsylvania, which is underlain by abandoned anthracite mines in several coalbeds superimposed one above another.
A 30-acre residential area was stabilized by injecting about 450,000 cubic yards of crushed mine refuse into two coalbeds through five injection boreholes. Nearly 200,000 cubic yards were injected through one borehole from which material moved into the mine workings on all sides; the injected material reached a maximum lateral distance of 640 feet and filled mine openings from floor to roof. In the gravity feed method formerly used to backfill inaccessible mine workings, quantities of fill injected per borehole averaged about 300 cubic yards. The gravity-feed method requires many injection holes and provided incomplete filling.
Distribution of slurry through a buried pipeline minimized disturbance to the community.
In a more recent pronouncement, (Bureau of Mines Information Circular No. 8846 (1981)), the Bureau of Mines said:
[I]n 1970 a new technique, the pumped-slurry injection process, was tested for the first time in a site adjacent to [the Rock Springs] area of severe surface subsidence. Success of this initial testing program, and of a large scale project in Scranton, Pennsylvania, led to further large-scale projects, funded by Congress, that resulted in the successful backfilling not only of all 15 target areas of potential subsidence in Rock Springs, but also of several areas in other states____ The pumped-slurry method was proved to be much superior to the gravity blind flushing method in terms of the amount of solids that could be injected underground through a single borehole.
The government’s pre-lawsuit publications show, notwithstanding the government’s contentions at trial, that there was a need for a better way to prevent surface subsidence, and that the invention met that need with a successful system.
Upon review of all the evidence relating to the issue of obviousness, the court concludes that defendant has not met the burden of proving invalidity by clear and convincing evidence. The court does not believe that plaintiff’s invention was obvious to a person with ordinary skill in the art, and the court finds highly significant the fact that those skilled in the art praised the invention prior to the filing of this lawsuit. The court therefore holds that the ’039 patent is not invalid on the ground of obviousness under 35 U.S.C. § 103 (1982).
*639B. ANTICIPATION
The Court of Appeals for the Federal Circuit has held that anticipation occurs when a claim reads on the prior art. The party asserting the anticipation defense must show that each element of the claim in issue is found, either literally or through principles of inherency, in a single prior art patent, reference, or project. Kalman v. Kimberly-Clark Corp., 713 F.2d 760 (Fed. Cir.1983), cert. denied, 465 U.S. 1026, 104 S.Ct. 1284, 79 L.Ed.2d 687 (1984), overruled on other gds., SRI Intern’l v. Matsushita Elec. Corp., 775 F.2d 1107 (Fed.Cir.1985). Plaintiff claims that none of the prior art meets the requirement of containing each element of any claim in issue. Defendant, on the other hand, argues that claims one, two, and three are all anticipated, although defendant’s major focus is on the heart of the invention, claim one. Claim one sets out the three elements of the invention: a closed system, a suspension, and mound formation. As observed earlier in section II.A., the court has found that the prior art does not contain all of the limitations of claim one.
The court cannot find a closed system within the prior art, notwithstanding defendant's vigorous argument. A closed system, upon a reasonable reading of the specifications, is defined as a pressurized system sealed from the hopper to the top of the void. Defendant specifically points to the prior art projects as examples of the elements of claim one. The former discussion of each of the projects sets out the court’s reason for disagreeing with this conclusion. Mr. Magnuson clearly indicated that at the Peterman’s Corners and Upper Tyrone projects the conduits did not extend completely to the top of the void. His testimony that the slurry would proceed through various crevices shows that the system was not closed. With respect to other prior art projects relied upon by defendant, Mr. Naples’ inconsistent testimony does not show that the system ever was closed by a pressurized conduit extending all the way to the top of the void. These projects were no more than prior art gravity-feed systems. Apparently, in only one borehole was a pipe extended completely to the top of the void, and no testimony indicated that the pipe was sealed or that pressure was exerted upon the system.
C. VALIDITY UNDER 35 U.S.C. § 112 35 U.S.C. § 112 provides:
Specification
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention. The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
A claim may be written in independent or, if the nature of the case admits, in dependent or multiple dependent form. Subject to the following paragraph, a claim in dependent form shall contain a reference to a claim previously set forth and then specify a further limitation of the subject matter claimed. A claim in dependent form shall be construed to incorporate by reference all the limitations of the claim to which it refers.
A claim in multiple dependent form shall contain a reference, in the alternative only, to more than one claim previously set forth and then specify a further limitation of the subject matter claimed. A multiple dependent claim shall not serve as a basis for any other multiple dependent claim. A multiple dependent claim shall be construed to incorporate by reference all the limitations of the particular claim in relation to which it is being considered.
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding *640structure, material, or acts described in the specification and equivalents thereof.
1. Impossibility of performance
Defendant’s first argument under section 112 is that the disclosure of the ’039 patent does not enable one skilled in the art to use step (c) of claim one. This is so, argues defendant, because of a fundamental misconception concerning underground particle transport. According to defendant, the disclosure assumes that the injected particles must be propelled in suspension over the mound formed by the previously deposited particles at least at MLV. The problem with the disclosure, says defendant, is that it ignores bedload and saltation as means of underground particle transport. Another problem, according to Dr. Kao, is that MLV cannot be reached in an area with erodible boundaries because the top of the mound would be eroded by particles moving in bedload and saltation.
The problem with these contentions is that they ignore the patent’s definitions of suspension and MLV.6 Suspension, as used in the patent and as the court understands it, includes both bedload and saltation. The test for enablement allows such a definition. The specification only needs to be of sufficient clarity to enable someone skilled in the art to make and use the invention. Atlas Powder Co. v. E.I. Du Pont de Nemours & Co., 750 F.2d 1569 (Fed.Cir.1984). The claims clarified in the specifications clearly enable one to practice the invention. Although it may be true as a technical matter that particles transported in a liquid reach states known as bed-load and saltation prior to settling out of a suspension, this does not make the patent impossible to perform, nor does it effect what was intended by the inventors. To divide up suspension into various phenomena may well be useful to the physics of particle transport, but here, such classifications have little relevance to the invention or its theoretical foundation. In this patent the inventors meant for suspension to include both bedload and saltation, as well as classic suspension, since the technology of particle transport required those states to be passed through either on the way to creating a suspension or on the way to depositing the particles within the void. Furthermore, testimony indicates that bed-load, saltation, and suspension can and do coexist.
Defendant’s argument is a little like saying a chair is not really used for sitting because not all of the time is one’s posteri- or located fully on the chair. As one sits down or gets up one’s legs are bent and some weight is on them also!
The patent defines MLV as “the minimum velocity at which a suspension of particles must be conducted through a conduit so that any substantial deposition of particles from the suspension onto the lower portion of the conduit to form an essentially stable layer thereon is prevented.” This definition of MLV must assume a definition of suspension which also includes saltation and bedload. The definition of bedload as put forth in the testimony is that state of particle transport which is the lowest velocity where the particles just bounce along the bottom of the conduit. Saltation is an intermediate stage between bedload and suspension. These definitions must be included within the patent definitions of suspension and MLV, if MLV is the minimum velocity necessary to prevent the particles from settling. In other words, MLV is defined as that velocity necessary to keep the particles in at least bedload. Such a definition follows necessarily from the language of the patent specification.
2. Disclosure not enabling
The defendant contends here that the plaintiff is estopped by the prosecution history from asserting that the patent contained any enabling embodiment other than use of the formula. This argument relates to the contentions that the patent is invalid because plaintiff did not invent the formula, and that if the patent is valid, there was *641no infringement because the formula was not used on the accused projects.
The court does not agree with defendant. The original patent application clearly states that the:
“Minimum linear velocity” for any particular suspension can readily be determined experimentally or by employing known formulae developed by investigations such as those taught in the references hereinbefore cited. The minimum linear velocity for any particular suspension can be readily determined by simple laboratory procedures, for example, as described hereinafter.
To the court, upon a complete reading of the original application, the patent itself and a general review of the talents of one with ordinary skill in the art, which is the one toward whom the patent is directed, In Re Naquin, 55 CCPA 1428, 398 F.2d 863 (1968), the application certainly provides a method to operate the invention without use of the formula. As pointed out by plaintiff, the person experienced in the art at the time of the invention would calculate the amount of material to be contained in a cone with the height of the void, and if the filling process backed up at the insertion of that amount of material, the person with ordinary skill in the art would increase the flow velocity until the MLV necessary to form breakout channels over the mound was achieved. This strikes the court as a fairly obvious method that one trained in the art would undertake in attempting to use the invention. As pointed out in the earlier sections of this opinion, the formula is not critical. It merely gives a velocity that will always work. It does not provide the absolute MLV, and MLV can be determined by experimentation.
3. Best mode not disclosed
The defendant additionally argues under section 112 that the best mode known at the time of the original application was not disclosed. Defendant’s suggestion is that to the extent claim one can be construed to cover injection through a vertical conduit at a rate lower than that given by the formula, plaintiff failed to disclose the best mode as represented by actual injection rates experienced at the initial Rock Springs project. The formula, according to table 1 of the ’039 patent (as corrected), required a minimum injection rate of 17,084 gallons per minute for injection of 4 to 6 mesh sand into a six foot high void. The actual injection rates averaged about 6,300 gallons per minute and were not determined through use of the formula.
The flaw in the government’s argument, to reiterate, is that use of the formula is not critical to practice of the invention. As the court has previously held, MLV can be determined experimentally and is, in fact, that injection rate minimally necessary for the invention to work. What is required by the patent is an injection rate fast enough to prevent plugging of the pipe and slow enough to allow some material to settle out to form a mound in the void. This is what was done at Rock Springs and the other projects employing the pumped slurry method. Also, testimony from Mr. Kanuch indicated that the Rock Springs Project was not complete at the time the patent application was filed. There is no indication that the actual injection rates at Rock Springs were the best mode. As the court understands the technology, injection rates and MLV will vary with the size of the conduit and the particular material to be injected. If this is true then the actual rates used at Rock Springs have no relevance to the practice of the invention under other conditions.
D. DISTINGUISHING FEATURE NOT IN FACT CRITICAL
The government argues that because the injection rate is not critical to the performance of the invention, the patent is invalid for two reasons. First, the patent is anticipated, since the only distinguishing feature from the prior art is the previously suggested element of the injection rate. Second, the patent becomes invalid under 35 U.S.C. § 103 because if the claim one, step (c) rate is not critical, then each of claims one, two, and three becomes obvious. This argument is flawed because of the court’s previous holdings on the issues of obviousness and anticipation. The court has held *642that the invention was not obvious in light of the prior art as a whole, because the prior art did not indicate the use of a closed pressurized system as required by step (a) of claim one. The same is true for the anticipation argument. No single prior art reference, patent, or project contains a closed pressurized system as required by step (a). Thus, defendant’s argument here need not be reached because the absence of step (a) in the prior art requires the anticipation and obviousness arguments to fail.
E. FORMULA KNOWN IN THE PRIOR ART
Defendant’s final argument for invalidity asks the court to hold that the named inventors invented nothing, because the formula was known in the prior art. In addition, defendant contends that the only enabling embodiment is the formula. Because the formula was invented by unknowns who cannot now be brought into court to correct inventorship, the defendant argues that it cannot be shown that the incorrect inventorship arose without deceptive intent. See 35 U.S.C. § 256 (commissioner may correct patent that failed to name inventor in original application, where the omission occurred without deceptive intent on the part of the applicant; otherwise, patent is void).
These arguments are related to the others rejected above, and are answered by the court’s previous holdings that the claims do contain an enabling embodiment in addition to the formula. The main innovation of the ’039 patent is the closed pressurized system, not the formula. As far as the section 256 argument is concerned, defendant again asserts that the formula is critical to the operation of the patent. As noted, the formula is just one method of determining the MLV.
III. INFRINGEMENT
28 U.S.C. § 1498 (1982) provides this court with jurisdiction over cases where the United States is alleged to have made unauthorized use of an invention “covered by a patent.” The scope of the analysis parallels that used in infringement cases brought against private parties under the infringement statute, 35 U.S.C. § 271(a). Lemelson v. United States, 752 F.2d 1538 (Fed.Cir.1985); Motorola, Inc. v. United States, 729 F.2d 765 (Fed.Cir.1984). Under either statute, infringement analysis requires two steps. “[T]he meaning of the claims must be learned from a study of all relevant patent documents; and the claims must be applied to the accused [system].” Caterpillar Tractor Co. v. Berco, S.p.A., 714 F.2d 1110, 1114 (Fed.Cir.1983) (citation omitted). The first issue, construction of the claims, is a legal matter, while the second issue — whether the accused system reads upon the patented claims — is a question of fact. SSIH Equipment S.A. v. Intern’l Trade Commission, 718 F.2d 365 (Fed.Cir.1983). Plaintiff must prove infringement by a preponderance of the evidence. Lemelson, 752 F.2d at 1548. In this case, plaintiff has met its burden.
It is well-settled that an inventor is his own lexicographer. “[A] patentee’s verbal license ‘augments the difficulty of understanding the claims,’ and to understand their meaning, they must be construed ‘in connection with the other parts of the patent instrument and with the circumstances surrounding the inception of the patent application.’ ” Fromson v. Advance Offset Plate, Inc., 720 F.2d 1565, 1569 (Fed.Cir.1983) (quoting Autogiro Inc. v. United States, 384 F.2d 391, 181 Ct.Cl. 55 (1967)). Thus, the patentee “is free to define the claim terms in any way he or she chooses.” Cosler v. United States, 15 Cl.Ct. 717, 733 (1988), aff'd mem., 883 F.2d 1026 (Fed.Cir.1989). When deciding the scope of the various claims in dispute, the question generally will be: what did the inventor mean? As stated by Judge Andewelt of this court:
In resolving this fundamental issue, the court is obliged to consider three sources of information: (1) the patent specification; (2) the other claims in the patent; and (3) the patent’s prosecution history. Howes v. Medical Components, Inc., 814 F.2d 638, 643, 2 USPQ2d 1271,1273 (Fed.Cir.1987); Moeller v. Ionetics, Inc., 794 F.2d 653, 656, 229 USPQ 992, 994 (Fed.Cir.1986). The specification is “the pri*643mary basis for construing the claims,” Standard Oil Co. v. American Cyanamid Co., 774 F.2d 448, 452, 227 USPQ 293, 295-96 (Fed.Cir.1985), because 35 U.S.C. § 112 provides that the specification shall contain a “written description of the invention ... in full, clear, concise, and exact terms.” The other claims in the patent are potentially important because differences between those claims and the claim in issue can provide crucial clues to claim interpretation. For example, the doctrine of claim differentiation raises a rebuttable presumption that differences in claims are significant and, therefore, that narrow limitations contained in one claim should not be read into other claims in which there is no such limitation. See Tandon Corp. v. United States Int’l. Trade Comm’n, 831 F.2d 1017, 1023-24, 4 USPQ2d 1283, 1288-89 (Fed.Cir.1987); D.M.I., Inc. v. Deere & Co., 755 F.2d 1570, 1574, 225 USPQ 236, 239 (Fed.Cir.1985). Consideration of the prosecution history of the patent is required because “the prosecution history ... limits the interpretation of claims so as to exclude any interpretation that may have been disclaimed or disavowed during prosecution in order to obtain claim allowance.” Standard Oil, 774 F.2d at 452, 227 USPQ at 296.[7] In addition to the specification, other claims, and prosecution history, when interpreting claim terms, the court may consider other forms of extrinsic evidence such as expert testimony, see, e.g., Howes, 814 F.2d at 643, 2 USPQ2d at 1273; Moeller, 794 F.2d at 656-57, 229 USPQ at 994-95; Lemelson, 14 Cl.Ct. at 326, 6 USPQ2d at 1663, or texts such as dictionaries which explain common usage of the disputed claim terms. See, e.g., Fromson, 720 F.2d at 1571, 219 USPQ at 1141.
Casler, 15 Cl.Ct. at 731-32.
In this case, as discussed in previous sections of this opinion, the meaning of the patent claims in issue turns initially upon the meaning of the term “suspension” as used in claim one, step (b), and whether the “closed system” of claim one, step (a) is a closed pressurized system. As seen previously, the court is of the opinion that “closed system” means a closed pressurized system. The specifications clearly call for a “closed pressurized system” between the void and the injection equipment. Testimony of the inventors supports this as their intention, as does a reasonable reading of the patent.
The court believes the patent definition of “suspension” includes both bedload and saltation, despite much testimony to the contrary from Dr. Kao. Although his points concerning bedload and saltation may be logical and scientifically true, they bear little upon the meaning of the term suspension as used in the patent. Testimony from the inventors and others clearly indicates that “suspension” as used in the patent includes all modes of particle transport sufficient to prevent a stable layer of solids from forming in and blocking the transport conduit. In fact, a close reading of the patent specifications says just that. Although suspension is defined in the specifications as “a liquid medium having dispersed therethrough solid particles[,]” that definition must be read in light of the patent definition of MLV, which is “the minimum velocity at which a suspension of particles must be conducted through a conduit so that any substantial deposition of particles from the suspension onto the lower portion of the conduit to form an essentially stable layer thereon is prevented.” U.S. Patent 3,817,039, column 2, lines 60-64. Suspension as defined in the patent must include both bedload and saltation, since MLV is defined as that rate which will prevent a stable layer of solids from forming. In other words, using Dr. Kao’s nomenclature, MLV is that rate which would at a minimum keep the solids moving at bedload. This is clearly what the inventors meant. Dr. Kao’s categories of bedload, saltation, and suspension are all *644covered by the patent language, and do not serve as a valid defense for, the government.
The other requirement of the patent in addition to the “closed system” of step (a) and the “suspension” of step (b) is that of mound formation under claim one, step (c), which provides:
injecting said suspension into said void through said conduit at an injection rate which is sufficiently low such that initially upon entrance into said void from said conduit the velocity of the suspension is below its minimum linear velocity and at least a portion of said solid particles are deposited to form a mound which decreases the cross-sectional area of said void, and sufficiently high to propel the suspension over said mound at a velocity at least equal to its minimum linear velocity to carry particles over said mound whereby they are deposited to increase the length and weight thereof, to form a layer of solid particles in said void.
The majority of testimony and evidence concerning step (c) indicates that once the other parameters of the invention are met, mound formation is inevitable. The testimony indicated that what happened in the void was not controllable. This conclusion appears to the court to be valid. Once a slurry is injected into the void at MLV or above through a closed pressurized system no more can be done. Step (c) of claim one describes what happens in the void and is part of the patent, but it is not a physical act within the control of anyone once steps (a) and (b) have been accomplished. For purposes of this case, mound formation under step (c) is to be distinguished from cone formation under the prior art blind flushing systems. Cone formation results from the gravity feed system because pressure cannot be brought to bear at the top of the void. Once the height of the cone reaches the top of the void and a back pressure equal to the static head is created, the slurry begins to back up and the point of refusal is reached.
The court, therefore, holds that steps (a) through (c) of claim one require use of a closed pressurized system and a suspension, as previously defined. The void must be substantially filled with solids, as required in claim two. Claim three requires substantially vertical boreholes.
The next step in the infringement analysis requires examination of the accused projects to determine whether they have infringed upon the claims in issue. Pursuant to court order, the parties were each directed to present only one project from plaintiffs list of accused projects, for which plaintiff was to prove infringement. Plaintiff chose to present the Pittston Avenue/Hickory Street project as its best case for infringement, while defendant chose to present the Dunmore/Throop Street project as its best case of non-infringement. Both projects were performed in Scranton, Pennsylvania. For the reasons set forth below, the court concludes that both the Pittston Avenue/Hickory Street project and the Dunmore/Throop Street project infringe on plaintiffs patent. Each project will be analyzed in turn.
A. THE PITTSTON AVENUE/HICKORY STREET PROJECT
1. Use of a closed pressurized system
The weight of the evidence presented shows that defendant used a closed pressurized system at the Pittston Avenue/Hickory Street project (the Pittston project). The specifications for the project, found in plaintiffs’ exhibit 55, indicate that the defendant’s intent was to close the system:
A slurry pump should be of a design to deliver the slurry at 50 PSI at the top of each injection borehole.
******
Drilling shall accommodate a standpipe adequate in size for the installation of 14-inch O.D., 8-inch I.D. and 6-inch I.D. casing pipe. Standpipe shall be installed from the surface to 6 feet below the top of rock. The Bureau [of Mines] representative will determine if the standpipe will be installed before commencing to drill to accommodate a casing pipe.
*645Drilling to accommodate 8-inch and 6-inch I.D. casing pipe shall be from 6 feet below the top of rock to a point 10 feet above the roof of the designated coal bed or a point directed by the Bureau representative. Casing shall not be installed until the drilling to an opening in the designated coalbed is completed.
Drilling an uncased hole at a maximum diameter below the casing shall be from a point designated by the Bureau representative to a point 2 feet in the solid rock strata below the designated coalbed. The size of the uncased borehole shall be the maximum diameter that can be drilled through the designated 8-inch and 6-inch I.D. casing.
Drilling to accommodate 14-inch O.D. casing pipe for supply shall be from 6 feet below the top of rock to the designated coalbed. (Monitoring boreholes will be enlarged for designated supply boreholes).
The contractor shall furnish and install new or approved-used casing pipe in all boreholes, where required.
All boreholes shall require a standpipe through the unconsolidated material and embedded tightly 6 feet below the top of rock, under the direction of the Bureau representative.
Injection borehole casing shall be installed from the surface to a point approximately 10 feet from the designated coalbed opening or a point directed by the Bureau representative, where the casing pipe shall rest on a ledge. In the event it is necessary to suspend the casing pipe from the surface to prevent its possible drop into the mine bed opening, the method of suspension is the contractor’s responsibility.
Supply borehole casing shall be installed from the surface to the roof of the designated coalbed.
The contractor shall weld a metal collar on only the injection borehole casing 4 feet below the bottom of the standpipe. This collar shall have a size equivalent to the difference between the drill hole diameter and the outside diameter of the casing, i.e., the space between the rock and casing.
Mr. Stewart testified that the weight of the stand pipe casing and the injection pipe casing sitting on their respective ledges, coupled with the airtight seal between the standpipe casing and the injection pipe casing, would constitute closure. The additional requirement that any holes cut in the standpipe or casing pipe must be closed further indicates contemplation of a closed system. Also, the requirement that the slurry pump be capable of delivering the slurry to the top of the borehole at 50 P.S.I. shows that a closed system was intended. Mr. Stewart testified that a pressure of 50 P.S.I. was incompatible with an open system as it would cause the slurry to run over. The specifications also require the slurry transport lines to “be connected directly to at least two injection boreholes.” Plaintiff’s exhibit 56, the completion report for the Pittston project, noted that the distance that material flows away from the borehole is dependent upon, among other things, the pressure and velocity at which it is being injected. For these reasons the court holds that a closed pressurized system was intended and was in fact put in place at the Pittston project.
Figure 5 of the specifications for the Pittston project depicts a “Typical Injection or Monitoring Borehole.” The typical borehole shown is vertical.
2. Use of a suspension
The specifications for the Pittston project provide a list of equipment to be used on the project. Included in this list is a “Mixing tank constructed and arranged in such a manner that it shall produce a uniform slurry of refuse material and water____” The list also included a “slurry pump to withdraw the slurried crushed product from the mixing tank and impel it into the injection hole and into the mine voids.” The specifications use the term “slurry” in several places and leave no doubt but that a suspension was contemplated. This is evident in the requirement that the mixing tank provide sufficient turbulence. Therefore, the court holds that a suspension was contemplated, under claim one, step (b) of *646the patent, and actually was used at the Pittston site.
3. Void substantially filled with solids
In the completion report for the Pittston project, prepared by the Bureau of Mines, the Bureau stated:
This project was unique in that it afforded the Bureau an opportunity to actually observe the material being injected into the mine voids. Constant underground inspections indicated no appreciable difference of slurry distribution pattern; the water drained rapidly from the slurry mix leaving a well-compacted filling of material in each mine void visited. The action of the slurry mixture being injected into a mine void from the lower end of a borehole can best be described as diffu-sible — the material spreads out into the openings in all directions at first but as the pile begins to grow and fill the area around the borehole opening the direction of flow follows a pattern of rivulets (channels).
The direction of flow in the rivulets changes when restrictions create a resistance to flow. These restrictions are natural and are caused when the affluxion of material begins to build up away from the hole. Thus, the distance the material can flow away from a borehole depends upon (1) the pressure and velocity with which it is being injected, (2) the amount of water being used in the slurry mix, (3) the extent and condition of the mine voids being filled, (4) the presence of natural restrictions to flow (such as roof falls, man-made packwalls or ventilation stoppings), and (5) injecting periods (constant “round-the-clock” injection seems preferable — though not always practical — to periodic injection because flow channels are kept well-lubricated which, in practice, keep solid materials in suspension and thus aid dispersion away from the injection point). On one inspection, y2-inch material was observed to travel about 700 feet from the injection borehole.
Mr. Martinelli, who observed the Pittston Avenue mines following completion of the project, testified that he saw filling to the mine roof with material sloping from the mine roof to its floor. Thus, the completion report and Mr. Martinelli’s testimony clearly show that the void was substantially filled with solids, as predicted in the ’039 patent.
Defendant, as seen in preceding sections of this opinion, argues that there can be no infringement because MLV is not attainable in the voids, as required under defendant’s reading of the ’039 patent. However, as discussed previously, MLV is defined in the patent as that velocity which will not form a stable layer in the conduit, as well as the velocity at which the slurry travels over the mound through breakout channels. It is not MLV as calculated in the formula, nor is it such a velocity that would require a suspension. Rather, MLV is that velocity which results in at least the transport mode known as bedload, which provides for just enough movement so that a stable layer is not formed. Because mound formation distinguishes the Stewart system from the prior art, the mound formation at the Pittston project is final strong evidence of infringement.
B. THE DUNMORE/THROOP STREET PROJECT
1. Use of a closed pressurized system
The specifications for the Dun-more/Throop Street project (the Dunmore project), contained in plaintiff’s exhibit 266, indicate the use of a closed pressurized system. The specifications provide in part:
This contract stipulates the use of minus-1^ inch breaker refuse material in the pumped slurry system for backfilling mine voids in the Marcy and Clark beds. * * * # * #
A slurry pump should be of a design to furnish a maximum of 30 psi at the top of the designated boreholes.
* * * * % $
[Performance shall require] piping to connect and make operational the water supply, mixing tank, slurry pump, and injection borehole.
* * * * $ *
*647The slurry pump shall be capable of pumping the required amount of slurry ... through up to 3000 feet of pipeline to the various injection boreholes. ******
The contractor shall operate and maintain a slurry pump of sufficient capacity to impel the required amount of material ... with the equivalent ratio of water, at a velocity of not less than 17 feet per second. All deliberate changes in pressure shall be at the direction of the Bureau representative.
******
Each injection borehole will be provided with a compound pressure gauge while in use.
******
Drilling [shall] accommodate 28-inch O.D. and 8- and 6-inch I.D. casing pipe....
******
Injection borehole casing shall be installed from the surface to a point approximately 10 feet from the designated coalbed opening or a point directed by the Bureau representative.
The specifications make it clear that the Dunmore project used a closed pressurized system. Further buttressing this conclusion is the testimony of John Schimmel, a government witness, who stated that all pumped slurry projects were conducted in the same manner by the government. The government introduced no evidence that would lead the court to disregard Mr. Schimmel’s testimony, and thus, concludes that just as a closed pressurized system and vertical boreholes were used at the Pittston project, so they were used at the Dunmore project.
2. Use of a suspension
In addition to the provisions cited above which suggest the use of a suspension, the specifications for the Dunmore project further provide:
The contractor ... shall operate and maintain ... a slurry plant of sufficient size, stability, and capacity to produce a maximum of 1,400 gpm of water and 150 tons per hour of material to form and pump a suspension of about 5 to 8 parts water to 1 part material by volume. The mixing tank shall be constructed to control and keep agitated the slurry mixture, providing adequate room and depth for the immersion of the slurry suction pipe.
Thus, it is clear that use of a suspension was contemplated at the Dunmore project.
3. Void substantially filled with solids
Unlike the mines involved in the Pittston project, the Dunmore mines were not accessible. However, as stated above, mound formation is inevitable when using a closed pressurized pumped slurry system and thus, the court can infer that more solid than the volume of a cone the height of the void was injected at the Dunmore project. A total of 111,114 cubic yards of mine refuse was injected through 35 boreholes at the Dunmore project. In addition, 15,-078 cubic yards of silt were injected through 17 boreholes.
CONCLUSION
For the reasons given above, the court finds that the ’039 patent is valid, and that it has been infringed by the United States or its contractors at its direction. The parties are directed to file a joint status report no later than August 10, 1990, proposing the future course of this litigation.
*648APPENDIX
[[Image here]]
. Backfilling is the filling of voids, or underground cavities, with solid material.
. The patent limits the size of the particles to minus-3 to 300 mesh under the United States standard sieve series. This is because particles smaller in size than minus-3 are lower in density than water and therefore would become permanently suspended.
. The court uses the terms “pumped slurry system,” "the Stewart system,” and “the '039 patent" interchangeably.
. Mr. Stewart described MLV in this manner: You do as it is suggested you throw the slurry through the pipeline [a clear pipe] and you watch for the point where particles drop out and eventually that part will become stable, they don’t move any more and then you increase the velocity and when it starts to pick up particles, you just see movement begin to develop, that is what we are calling the minimum linear velocity.
. Dr. Kao holds Master of Science and Doctor of Philosophy degrees in civil engineering from Duke University; his doctoral dissertation was on the subject of hydraulic transport of solids. He is clearly a brilliant scientist, well-versed in his field of expertise. He received an award as one of the "Top 100 Innovators of 1985” for his work in the field of particle transport, and he has published numerous papers on that topic. However, Dr. Kao testified that he has never observed slurry being pumped into mine voids. Furthermore, Dr. Kao stated that other than observing surface subsidence in a mining area of China in 1979 (some nine years after the invention at issue), he had no experience in the field of subsidence control until he was consulted by government counsel in connection with the present lawsuit.
. They are also circular arguments in that they set out theoretical definitions, foreign to the patent and to the actual practice of subsidence control, and then use those definitions to attempt to show that the invention does not work.
. Prosecution history estoppel also applies when criticality of a particular feature of a claim over the prior art is alleged. An applicant arguing that a particular limitation distinguishes his claims from the prior art is later estopped from denying the criticality of the limitation. Hughes Aircraft Co. v. United States, 717 F.2d 1351 (Fed. Cir.1983).