Full opinion text
MEMORANDUM DECISION, FINDINGS OF FACT, CONCLUSIONS OF LAW AND ORDER FOR JUDGMENT HAUK, District Judge. PRELIMINARY FINDINGS AND MEMORANDUM DECISION On June 6, 1963, some two and one-half years after Dr. Garbell became convinced that the Boeing 707 and the Douglas DC-8, on each of which he had taken one flight as a passenger, must incorporate his wing design since each exhibited such excellent flying qualities, the Garbell plaintiffs filed these actions against the defendants Boeing and Douglas. The cases were assigned to Judge Peirson Hall who, on his own motion, dismissed each with prejudice. Subsequently, on plaintiffs’ application, Judge Hall permitted amended complaints to be filed. Defendants’ answers asserted non-infringement of the patent, invalidity of the patent and laches. Discovery was initiated by the Garbell plaintiffs in the form of interrogatories directed to both defendants. The defendant Boeing moved to dismiss for want of venue and this motion- was denied. Thereafter both defendants noticed the deposition of Dr. Garbell, the patentee and president of plaintiff corporations. At that early stage of the cases, the Court was given an indication of the discovery problems which would later ensue when the plaintiffs moved to prevent the taking of Dr. Garbell’s deposition. In 1964, the cases were transferred to Judge Francis Whelan who had been newly appointed. The cases had been started with plaintiffs being represented by Wallace & Parker, Charles Parker, Esq. and Ronald Rattner, Esq. of San Francisco. On the motion of plaintiffs, Wallace and Parker and Charles Parker, Esq. were substituted out and the plaintiffs’ new lawyers-were Morris Lowenthal, Esq. and Jerome Field, Esq. of the firm of Lowenthal & Lowenthal, and Ronald Rattner, Esq. of San Francisco. Defendants were represented by the firm of Older, Cazier, Preston & Hoegh (now Hahn, Cazier, Hoegh & Leff) and Richard B. Hoegh, Esq. of that firm in Los Angeles. In addition to Mr. Hoegh’s firm, defendant Boeing was represented by J. Paul Coie, Esq. of Seattle, Washington, and defendant Douglas was represented by Walter J. Jason, Esq. of Los Angeles. The first affirmative action taken by the plaintiffs’ new lawyers (Messrs. Lowenthal, Field and Rattner) was to move to stop further discovery on behalf of the defendants for a period of several months. In the meantime, the plaintiffs had begun document inspection at the Long Beach plant of the defendant Douglas. These cases were transferred to Judge Irving Hill in 1965. The defendant Boeing filed a motion to transfer the action against it to Seattle, which motion was denied by the Court. Both defendants then moved to have a Special Master appointed to supervise discovery in these actions. At that time the docket entries in the Boeing action alone filled five pages’ of the Court’s docket book. The Court granted the motion stating: “The Court, being duly advised in the premises, finds that an exceptional condition exists requiring that discovery be had under the supervision of a special master.” On September 27, 1965, the Order re Appointment of a Special Master was signed and Robert Henigson, Esq. of the law firm of Lawler, Felix & Hall in Los Angeles was appointed to supervise all pending and future discovery. The plaintiffs moved to vacate this Order and their motion was denied on November 22, 1965. Plaintiffs then moved for leave to file Petition for Writ of Mandamus in the Ninth Circuit asserting, among other things, the bias of Mr. Henigson. The motion was denied by the Ninth Circuit and the discovery matters were from that time presented to Mr. Henigson. The discovery matters presented to Mr. Henigson for scheduling or determination filled practically fourteen pages of the Court's docket. On August 1, 1966, these cases were transferred to Judge A. Andrew Hauk. In October 1966, defendant Boeing filed a motion for an early and separate trial on the issue of infringement. The Garbell plaintiffs vigorously opposed Boeing’s motion. On January 20, 1969, the cases were set for trial in May of the following year. The parties were given until November 1, 1969, in which to complete all discovery in both cases, including document inspection on the DC-9 aircraft at the Douglas Long Beach plant and document inspection at the Boeing plant in Seattle. With respect to the Boeing discovery, apart from a request by letter made to Boeing’s Seattle counsel for inspection of a few limited items, the plaintiffs had not initiated any document inspection at the Boeing plant and had not taken the depositions of any wing designers at Boeing in order to assess their infringement allegations against Boeing in the five and one-half years that the case had been pending. Indeed, the document inspection ordered by the 'Court on January 20, 1969, was done on the Court’s own motion [Page 33 of the January 20, 1969 transcript]. On January 20, 1970, the parties were in Court on plaintiffs’ motion to expand the reference to the Special Master to include the taking of evidence on all issues to be determined at trial. The Court expanded the order of reference to the Special Master to include the so-called geometry issues relating to the accused Boeing and Douglas aircraft and to the aircraft which defendants relied upon in support of any of their invalidity defenses. Trial before the Special Master commenced on June 16, 1970, and lasted thirty days. The Special Master’s report was filed December 31, 1970, and covers certain airfoils of the DC-8s and four prior art aircraft, the CurtissWright Models 21B and 23, the Grumman F6F, and a German sailplane, the D-30 Cirrus. Other DC-8 airfoils and the remaining invalidity aircraft were to be covered in a supplemental report by the Special Master. During the hearing before the Special Master, on cross-examination, plaintiffs, through Dr. Gar-bell, dropped their charge of infringement as to the DC-9 aircraft wing basing their decision to do so upon evidence supplied to them long before they subjected Douglas to a complete document inspection on the DC-9. In mid-January, 1971, the plaintiffs filed two “motions for action” relating to the DC-8 findings in the Special Master’s report and to the findings on the invalidity aircraft. These motions, in effect, were motions for partial summary judgment seeking a determination that the DC-8 wings infringed, based on the Special Master’s findings and that the patent was valid over the prior art aircraft. The motions were denied and, following a hearing on objections to the Special Master’s report, the report was modified by interlineation and filed. On February 1, 1971, trial was scheduled to commence on June 1, 1971. Pretrial was set for May 3, 1971, and discovery was opened up for an additional month. During the weeks before the trial, the plaintiffs filed several motions for continuance which were denied. The Special Master filed his Supplemental Report on May 5, 1971, and on May 14, 1971, plaintiffs moved to remand for further findings by the Special Master on the DC-8 airplane wings on the grounds that Douglas had supplied erroneous data to plaintiffs in 1964. The plaintiffs claimed they were surprised by their own evidence presented to the Special Master and contained in the Supplemental Report. The Court granted the motion. On June 1, 1971, a few minutes before the trial was scheduled to commence, the plaintiffs filed a motion for recusal of the trial judge on the grounds of bias and prejudice. The moving papers were considered and the'motion denied. The parties spent the remainder of the Court time on June 1st attempting to settle the actions. On- June 2, 1971, trial of the actions commenced. The validity or invalidity of the patent was to be tried first, followed by Douglas infringement issues, to be followed by Special Master hearings on Boeing wing geometry and Court trial on Boeing infringement issues. The plaintiffs put their patent in evidence and, without availing themselves of the opportunity to present further evidence as to the invention, rested. The defendants called Dr. Garbell as their first witness under Rule 43. On Friday, June 4,1971, the plaintiffs filed a Petition for Writ of Mandamus seeking review of the denial of their motion for recusal of the trial judge and on June 9, 1971, the Ninth Circuit Court of Appeals stopped the trial pending determination of the Petition for Writ of Mandamus. On August 12, 1971, the Court heard the petition and denied it on the same day. On September 20, 1971, at a status hearing, Mr. Lowenthal, who had represented the plaintiffs since 1965, stated that Mr. Jerome Field, who had been co-counsel with him, had removed himself from the case. A hearing was held on November 1, 1971, substituting Jerome Field and Ronald Rattner out of the case and granting each of the attorneys a lien on any recovery made by plaintiffs in these actions. Plaintiffs appealed from this order, which appeal is now pending in the Ninth Circuit. On November 20, 1971, at a further hearing, Mr. Henigson was ordered to proceed with the remand requested by the plaintiffs and resumption of the trial was set for March 7, 1972. On February 3, 1972, after obtaining a 58 day extension of time in which to file a Petition for Writ of Certiorari in the United States Supreme Court to review the Ninth Circuit’s ruling on the Petition for Writ of Mandamus, the plaintiffs filed the Petition for Writ of Certiorari together with an application for a stay of the trial pending determination of the application for a writ. The defendants filed their opposition to the stay application in the Supreme Court on the same day and the stay was denied by Mr. Justice Douglas on February 14, 1972. The Special Master had set the trial of issues raised by the remand for February 7, 1972. The circumstances surrounding the setting of the trial date were described by Mr. Henigson in his order to show cause as to why thfe remand should not be dissolved on the grounds of lack of communication and cooperation from the plaintiffs in the following language: “IT IS HEREBY ORDERED that the parties show cause before the Special Master, if any they have, on Monday, March 6, 1972, at 10:00 a. m. at Room 800, Standard Oil Building, 605 West Olympic Boulevard, Los Angeles, California, why the Special Master should not submit his final report to the above entitled Court without holding an evdientiary hearing to resolve the factual issues in dispute and without making the findings of fact contemplated by the May 24, 1971 order, for the reasons that all efforts directed to that end are frustrated by plaintiffs and that there is no reasonable prospect that the work can be accomplished in the foreseeable future.” Leading up to the Order to Show Cause, the Special Master observed in Letter No. 95, dated November 18,1971: “I concede also that I think the difficulty imposed upon any attorney or firm required to stand by indefinitely but always at the ready, pending plaintiffs’ obtaining trial counsel they think adequate to the task, is utterly intolerable.” In his Letter No. 100, the Special Master stated: “I cannot refrain from observing that the delays incurred in bringing this phase of the litigation to a close have been intolerable and are in large part attributable to unwarranted demands that counsel and Dr. Garbell be convenieneed in the prosecution of the litigation which plaintiffs commenced.” Dated December 30, 1971. ' On February 1, 1972, the Special Master reported to the Court in Letter No. 102 the following: “. . . We have the astounding situation in which: plaintiffs desire to defer indeterminately any hearing pertaining to the geometry issues which, by their motion and the Court’s order, were made a matter of reference to me as Special Master; and, defendants are anxious to have the matter heard and determined at the earliest possible time. My own view of the matter is that plaintiffs’ counsel are entirely competent to represent plaintiffs in the action, that there are no serious prospects as (if, indeed, there are any serious endeavors to obtain) substitute counsel for plaintiffs and that, if plaintiffs’ counsel have the time and energy to generate the copious paperwork recently filed with the respective clerks of the district court, the court of appeals and the Supreme Court, they can find the time and energy to attend a short hearing before me. Further, as you know, there is outstanding a Court order requiring that we proceed with deliberate speed in the completion of the work remaining to be done under my May 24, 1971 order. While “deliberate speed” has been variously construed, I do not think deferral of the hearing for an indeterminate period of time could possibly fit within any permissibly stretched meaning of the term.” At one Court appearance relating to the Special Master’s proceedings, the plaintiffs were represented by Mr. John R. Jacobson from the Lowenthal & Lowenthal firm and on the day the Order to Show Cause was scheduled to be heard by the Special Master, March 6, 1972, plaintiffs were represented by Mr. George White, who appeared, but was not prepared to proceed and was not accompanied by Dr. Garbell or any other witnesses. On March 7, 1972, trial re-commenced and in due course the Supreme Court denied the Petition for Writ of Certiorari. The Court observes that the actions of the plaintiffs in January and February required the defendants to prepare for trial on March 7, 1972, when the defendants could reasonably expect that plaintiffs would apply for a stay of the trial pending determination of their Petition for Writ of Certiorari and, at the same time they were preparing for trial before the District Court, defendants would be required to prepare an opposition to the Petition for Certiorari. At the same time, defendants were required to prepare for a hearing before the Special Master scheduled to last several days commencing February 7, 1972, for a hearing requested by the plaintiffs in May of 1971, while the plaintiffs had no intention of showing up ready to proceed on the scheduled date. The trial on the validity issues took 23 Court days and concluded June 2, 1972. Broadly, the defendants contended that the patent was invalid as being not “new or useful” under 35 U.S.C. § 101, that the purported invention was anticipated under 35 U.S.C. § 102, that the subject matter of the patent was obvious under 35 U.S.C. § 103, and that the teachings of the patent were insufficient and the claims ambiguous under 35 U.S.C. § 112. The patent in suit entitled “Fluid-Foil Lifting Surface”, No. 2,441,758, names Maurice A. Garbell as the inventor. The application for the patent was filed July 16, 1946, and the patent issued on May 18, 1948. The patent has been assigned to two corporations, Maurice A. Garbell, Inc. and Garbell Research Foundation, whose major asset is the patent. Airplane wings are “lifting surfaces”; and in this action the accused devices are the wings of jet transports manufactured by the two defendants, Boeing and Douglas, later McDonnell Douglas. The principal object of the patent in suit is to design and construct a wing which causes the initiation of stall to occur at or near the mid-semispan of the wing, and to spread inboardly more quickly then outboardly. This object and the patentee’s solution are not new. The patent utilizes what Dr. Garbell, on occasion, called the tri-section wing principle in order to obtain a wing which would not exhibit the dangerous stalling characteristics attributed by Dr. Garbell to the prior art aircraft. In the prior art aircraft construction it was customary for wing designers to choose a cross-section, or airfoil section, shaped for the wing tip and to choose another cross-section, or airfoil section, shaped for the wing root. The root section was typically located at the side of the body where the wing is attached to the fuselage or at the center line of the airplane where the two ends of the airplane wing were bolted together. The wing surface between the wing tip and wing root were then generated by a technique known as straight-line fairing between the wing root and the wing tip. The technique was similar to that of rolling up a piece of paper to form a cone with the tip section being thought of as the small end of the cone and the root section being thought of as the large end of the cone. The claims of the patent here in issue, Claims 1, 2, 3 and 7, utilized a third or interjacent section along the wing semi-span. This third section is different from that which would be found in the wing at the wing station chosen for the interjacent section if the wing were generated by straight-line fairing between the root and tip sections. A tri-section wing shape is generated by drawing straight-line elements or allowing the straight-line fairing technique between the wing root and the interjacent section forming what is called the inboard panel, and by utilizing the straight-line fairing technique between the interjacent and tip sections to define the shape of what is called the outer panel of the wing. As many panels as the designer desires may be generated in this way between the wing root and the wing tip to form the entire wing surface. In the claims of the patent, the airfoil sections for these various stations, the root, the tip and any interjacent section, is defined solely by what is called in the claims “mean line camber”. In the claims here in issue, the mean line camber of the interjacent section must be greater than that which would be found in the old two-section wing generated by straight-line fairing between the root and the tip. The mean line camber of the root section must be smallest in value, and the mean line camber of the tip section must be the greatest in value. Other claims of the patent not here in issue, cover wings in which the camber of the interjacent section is less than that obtained by straight-line fairing between the root and the tip sections. In another patent obtained by Dr. Garbell, the camber of the interjacent section is greater than that at the tip section. The combined coverage of the two patents thus encompasses all wings except those which Dr. Garbell conceded to be old. In 1925, an inventor by the name of Cronstedt obtained a patent on a wing shape. His wing was defined by a tip section and a root section. The tip section had the greatest camber, as is the case in the claims of the patent here in 11 suit, and the root section had the least camber, as is also the case in the claims of the patent here in suit. The Cronstedt patent drawing depicted an interjacent section, but did not disclose whether the interjacent section in the drawing was that which you would find in a wing by straight-line fairing between the root and the tip sections, or something different. If it were something different, it would be covered by one of the claims of Dr. Garbell’s patent. Whether Mr. Cronstedt knew more than he understood about the use of an interjacent section in that early wing is a matter of conjecture. Dr. Garbell’s claims are an obvious variation of what is shown in Cronstedt. By 1929 the National Advisory Cornmittee on Aerodynamics (the NACA) set about systematizing the properties of airfoil sections and the relationship of camber and other parameters affecting the shape of the airfoil and relating such parameters to lift and drag. The earliest of such work systematizing data for the sections developed by the NACA was published in a Technical Report No. 460 in the early ’30s. Professor Pinkerton, who appeared in this action as an expert on behalf of the defendants, was directly involved in the experimental and analytical work as an employee of the NACA. The work was undertaken to give aircraft designers a knowledgeable basis for deriving wing shapes so that lift and drag of the aircraft could be predicted with some accuracy. This and other publications of the NACA were distributed throughout the world and were available to Maurice Garbell when he was a student in Milan and working with others in designing the Pingüino sailplane. Later studies of the NACA, undertaken in the '30s utilized NACA sections in the construction of complete wings which were tested or in study wings approximating the planform of wings found in production airplanes which at the time did not have swept-back wings. One such wing had six interjacent sections which compares with Dr. Garbell’s claims covering wings with one or more interjacent sections. NACA continued its development and publication of airfoil data on through World War II with the publication of NACA Technical Report No. 824 co-authored by Mr. Ira H. Abbott, who appeared as one of the witnesses for the defendants in this action. Independent researchers were also studying the effects of airfoil parameters, such as mean line camber, in order to avoid tip stall on tapered monoplane wings being designed in the ’30s and early '40s; e. g. Dr. Lombard, Exhibit BC. All of these NACA publications included the relationship of the lift across the span of the wing to the maximum lift which could theoretically be generated at each station across the wing, similar to that shown in Figure 2 of the patent and called for by Claims 2, 3 and 7 of the claims in issue in this action. As to whether or not the published NACA work made the subject matter of the patent in suit obvious, Professor Pinkerton said: “I can’t read anything in the patent beyond what we knew and practiced in the ’30s”. [T.R. 1021] In his work and in his published materials in evidence, Dr. Garbell utilized NACA airfoil sections and referenced NACA reports. Dr. Garbell was employed at Consolidated Vultee in San Diego during World War II. There he worked on a tailless aircraft design and proposed at least as early as 1944 using in it the wing configuration he later patented. Dr. Gar-bell gave a description of such wing without restriction to NACA which in turn distributed it to its various research facilities in early 1944. About the same time, Dr. Garbell personally went to see Captain Diehl of the Navy in an attempt to sell the aircraft to the Navy on CVAC’s behalf. Dr. Garbell, while at CVAC and prior to July, 1945, also assisted in preparation of the proposal to the government for the sale of XB-46 aircraft. The proposal included a wing design covered by certain claims of the patent. Before July, 1945, Dr. Garbell was at the NACA wind tunnel at Ames, in California, in connection with wind tunnel work on the XB-46 model CVAC had sold the government; The model incorporated a wing he had proposed and which was covered by claims of the patent. Also while at CVAC and before the critical date of July 16, 1945, Dr. Gar-bell submitted his manuscript for a paper to be published by the Institute of Aeronautical Sciences. The manuscript was seen by a great many people in the aircraft industry and in educational institutions. Dr. Garbell regarded the manuscript as a disclosure of the invention on which he subsequently obtained the patent at issue in this action. More than a year after Dr. Garbell engaged in the foregoing selling activities and submission of his manuscript for publication, he applied for the patent in suit. From the record it appears that Dr. Garbell took the prior art he was familiar with and proceeded to claim everything except that which he conceded to be old in an attempt to exact - tribute from the aircraft industry. Using real property as an analogy, the metes and bounds of what he staked out in his two patents in effect claim everything on both sides of the river. These considerations and others which appear in more detail in the Findings lead to the decision that the patent in suit is invalid for the reasons advanced by defendants. The Court adopts the foregoing preliminary statements as part of its Findings of Fact and makes additional Findings of Fact as follows: FINDINGS OF FACT A. GENERAL FINDINGS — THE PARTIES AND HISTORY OF THE SUIT Al. Plaintiff Maurice A. Garbell, Inc. is a California corporation having its principal place of business at 1714 Lake Street, San Francisco, California. A2. Plaintiff Garbell Research Foundation is a California non-profit corporation having its principal place of business at 1714 Lake Street, San Francisco, California. A3. The president of the plaintiff corporations, Dr. Maurice A. Garbell, is the patentee. A4. The patent in suit, No. 2,441,75s, is entitled “Fluid Foil Lifting Surface”. The patent application which issued as the subject patent was filed on July 16, 1946, and assigned to Maurice A. Garbell, Inc. on April 15, 1948. Said patent issued on May 18, 1948, to Maurice Adolph Garbell, assign- or to Maurice A. Garbell, Inc. An undivided three-fourths (%ths) part of the entire right, title and interest in said patent was assigned to the Garbell Research Foundation on September 15, 1949. A5. Defendant The Boeing Company is a Delaware corporation having a regular and established place of business at 5822 Avion Drive, Los Angeles, California. A6. Defendant Douglas Aircraft' Company, Inc. was a Delaware corporation having a regular and established place of business at 3000 Ocean Park Boulevard, Santa Monica, California. Defendant Douglas was merged with and into McDonnell Company on April 28, 1967. As of that date, McDonnell Company’s corporate name was changed to McDonnell Douglas Corporation. McDonnell Douglas Corporation is a Maryland corporation and has a regular and established place of business at 3000 Ocean Park Boulevard, Santa Monica, California. A7. The complaints in these two consolidated actions were filed in 1963 by the plaintiffs for infringement of their patent No. 2,441,758 by the defendants The Boeing Company and Douglas Aircraft Company, later McDonnell Douglas Corporation. A8. This action arises under the Patent Laws of the United States, 35 U.S.C. Sections 271, 281, and 28 U.S.C. Section 1338 The jurisdiction and venue of this Court was determined by the fact that each of the defendants has a regular and established place of business within this District and their activities charged by plaintiffs to be infringements of the patent in suit were carried out in this District and elsewhere. A9. The patent in suit, No. 2,441,758, is entitled “Fluid Foil Lifting Surface’’ and is directed to the shape of aircraft wings. The accused devices manufactured by the defendant Boeing are the 707-320 B/C aircraft and the accused devices manufactured and sold by the defendant McDonnell Douglas are the DC-8 series aircraft. The claims of the patent at issue are Claims l, 2, 3 and 7. Plaintiffs allege that each of such claims is infringed by each of the defendants. The defense of invalidity is the only issue per the Amended Pretrial Conference Order, dated May 27, 1971. In that connection, defendants allege and have shown to the satisfaction of the Court that the patent and each of said Claims 1, 2, 3 and 7 are invalid because they do not meet the absolutely essential tests of validity: 35 U.S.C. 101, Novelty and Utility; 35 U.S.C. 102, No Prior Use or Sale and No Prior Publication; 35 U.S.C. 103, Non-Obviousness; and 35 U.S.C. 112, Definiteness. A10. The patent was previously litigated in this District in an action by plaintiffs against Consolidated Vultee Aircraft Corporation (CVAC), Civil Action No. 10930-Y. At the conclusion of the four day trial in that action, this Court, by Judge Yankwich, found the patent to be valid and infringed by CVAC and denied the defendants’ claim of a shop-right. 94 F.Supp. 843 (S.D. Cal.1950). The Ninth Circuit Court of Appeals, 204 F.2d 946 (9th Cir. 1953), found that the defendant had established a shop-right and declined to rule on whether the patent, or what it referred to throughout its opinion as the “alleged invention” was valid or infringed. Upon remand, Judge Yankwich dismissed the action with prejudice. All. In the present action, on motion of defendants, a Special Master was appointed in 1965 to supervise discovery. A12. Judge Hill’s order appointing Mr. Robert Henigson to serve as Special Master was challenged as an abuse of discretion and Mr. Henigson was challenged for bias in plaintiffs’ Petition for Writ of Mandamus filed in the Ninth Circuit. The Court denied the Petition. A13. During the discovery proceedings, Mr. Henigson held approximately 37 meetings with counsel and attended numerous depositions to supervise the discovery proceedings. A14. In April, 1970, plaintiffs asked to have Mr. Henigson appointed to hear all evidence in the case. The motion was granted to the extent of the geometry relating to the accused Boeing and Douglas jet transport wings and wings relied upon by defendants in support of invalidity defenses. A15. The Court reserved all remaining issues, including the issues of aerodynamics and interpretation of the patent, for trial by the Court. A16. The trial before the Special Master commenced June 16, 1970, and lasted for thirty days. Eighteen witnesses testified and approximately 700 exhibits were received into evidence. A17. The Report of the Special Master was filed on December 31, 1970, and adopted by the Court. A18. The Special Master filed his Supplemental Report on May 3, 1971. It was adopted by the Court. A19. Plaintiffs moved to remand on the grounds that Douglas had supplied erroneous data. The Court granted this motion on May 24, 1971. A20. After repeated unsuccessful attempts to get plaintiffs to attend a hearing on remand issues, Mr. Henigson issued an order to show cause why remand should not be vacated and thereafter reported to the Court that the remand be vacated. A21. In the spring of 1971, the plaintiffs filed repeated requests for a continuance of the trial. A22. On June 1, 1971, the day scheduled for commencement of trial, plaintiffs filed a motion to disqualify Judge Hauk. This motion was denied and taking of testimony commenced on June 2, 1971. A23. On June 4, 1971, the plaintiffs filed a Petition for Writ of Mandamus seeking review of the ruling on their motion. On June 9, 1971, pursuant to an order of the Court of Appeals, the trial was stopped pending determination of the petition. A24. The petition was argued in the Court of Appeals on August 12, 1971, and denied the same day. Rehearing was denied on September 9,1971. A25. On November 1, 1971 the Trial Court rescheduled trial to commence March 7, 1972. A26. Plaintiffs obtained a 58 day extension for a total of 148 days in which to file a Petition for Certiorari in the Supreme Court and approximately three weeks before trial, on February 3, 1972, filed their Petition for Certiorari on the recusal matter and concurrently filed in the Supreme Court a Petition for a Stay of the trial and on February 4, 1972, plaintiffs filed a like application in the Ninth Circuit. Both were denied and trial recommenced on March 7,1972. A27. Trial extended 23 court days, including that which commenced June 1, 1971. The record comprises approximately 4,000 pages of reporter’s transcript and 50 volumes of papers filed with the clerk in connection with these two cases. A28. At trial, plaintiffs put the patent in evidence and thereupon rested their case in chief. Defendants called Dr. Garbell, the patentee and president of the plaintiff corporations, under Rule 43 and the following witnesses: • Professor Robert M. Pinkerton: Former Research Physicist for National Advisory Committee for Aerodynamics (NACA), Author of NACA Reports; Professor Emeritus, North Carolina State University. Wesley T. Butterworth: Employed by North American Aviation on the Apollo Space vehicle; former employee of Curtiss-Wright and designer of the wings of the Curtiss-Wright Model 21B and Model 23. Ira H. Abbott: Former Director of Research and Advanced Technology, National Aeronautics and Space Administration (NASA), Author of NACA Reports, Co-author of “Theory of Wing Sections”. Al Riedler: Former Chief of Aerodynamics, Convair San Diego, during development of Convair 880 and 990 aircraft. Currently working for the City of San Diego. Dr. Albert E. Lombard, Jr.: Ph.D. Aeronautics and Physics, Cal Tech, 1939; Former Director of Research and Staff Vice-President, McDonnell Douglas Corporation. William T. Hamilton: Formerly with NACA, now employed by The Boeing Company; was Director of Technology on the SST. Orville Dunn: Director of Aerodynamics, McDonnell Douglas Corporation. A29. Plaintiffs’ rebuttal comprised the testimony under Rule 43 of Mr. Glenn Orlob, Patent Administrator of The Boeing Company and Mr. Walter J. Jason, Patent Director of McDonnell Douglas Corporation, and the testimony of the following: Dr. William Bailey Oswald: Ph.D. in Aeronautical Engineering, Cal Tech. Former Chief of Aerodynamics, Douglas Aircraft Company; Santa Monica Division. Dr. Maurice A. Garbell: The patentee and president of plaintiff corporations. A30. The parties jointly designated portions of the depositions of the following named witnesses. The designated portions were read in open court by the trial judge. William E. Nickey: Was a test pilot for Curtiss-Wright. Flew CurtissWright Models 21B and 23. Charles W. Harper: Deputy Associate Administrator for Aeronautics, Office of Advanced Research and Technology, NASA. Max Munk: Formerly with the NACA. Theodore Theodorsen: Consultant in Aerodynamics. James G. McHugh: Senior Staff Engineer, Dynasciences Corporation. Formerly with NACA and NASA. James C. Sivells: Staff Engineer at the Aerodynamics Division of the Von Karman Gas Dynamics Facility, ARO, Incorporated. Loren Facka: Supervisor of Financial Accounting, Convair Division of General Dynamics Corporation. Orville Dunn: Director of Aerodynamics, McDonnell Douglas Corporation. Harold F. Kleckner: Aerodynamicist and wing designer, McDonnell Douglas Corporation. Harold T. Luskin: Former Douglas aerodynamicist, and with Lockheed Missiles and Space Company. Now deceased. Thomas Neill: Chief of Technical Publications Branch of the Office of Advanced Research and Technology, NASA. Hartley A. Soule: Formerly with NACA, author of several NACA reports. B. DEFINITION OF TERMS USED IN THE PATENT. Bl. The patent in suit, United States Letters Patent No. 2,441,758, relates to the design and construction of a fluid-foil lifting surface. The term “fluid-foil lifting surface” is synonymous with “wing”, “fin” or “blade”, which produces lift when moving through a fluid such as air or water. B2. The term “fluid-foil section” is synonymous in the art of aerodynamics with the term “airfoil section”, and typically represents a cross-section of a wing. B3. A controlled fluid-foil section is one which has been pre-selected by the designer to form a cross-section of the wing at a given station. B4. The term “at the root”, as used in the patent, and as interpreted according to standard industry practice at the time of the alleged invention, refers to the location at the plane of symmetry (longitudinal center-line) of the airplane, and possibly to a location at the wing-fuselage intersection. To the aerodynamicist, the term “root” most commonly refers to a location at the plane of symmetry. Figures 2 and 3 of the patent in suit indicate that the “root” (Figure ' 2) is synonymous with the “center line” (Figure 3). B5. The surface of wings is usually formed by fairing between controlled sections, i. e. from the station located at the root to the one located at the tip or from root to interjacent section and from interjacent section to the tip section. B6. Mean line camber is a property of the mean line of an airfoil section, i. e., a line lying midway between the top and bottom of an airfoil section measured perpendicular to the mean line. B7. An airplane in flight is sustained in the air by lift forces generated by the flow of air past the surface of the wings of the airplane. B8. The “stall” of an airplane refers to a phenomenon in which the airflow over the upper surface of the wing separates from the surface, or ceases to flow in a pattern approximately conforming to the upper surface of the wing. An airfoil performs its normal function of generating lift most efficiently when the airflow over the foil is smooth rather than turbulent. The main flow remains “attached” to the surface of the foil. As the angle- of attack increases, however, the airflow over the upper surface encounters difficulty in remaining smooth, “attached”. A separation or stall occurs when the wing is forced to' an angle of attack which is too high relative to the oncoming airflow. B9. All wings will stall at some ascertainable angle of attack. The major factor which determines whether stall characteristics of an airfoil are acceptable is the degree to which the airplane is controllable during the stall. BIO. Stall or air separation causes the wings to lose their lift capability and as stall increases the airplane will eventually begin to drop. If, as a result of the initiation of stall, the airplane experiences a pitch-up moment, causing the angle of attack to increase even further, the pilot, will necessarily have to exert a positive downward control moment, viz, by pushing the control column forward, in order to prevent a worsening of the stall condition. On the other hand, if the natural tendency of the airplane at the initiation of stall is to experience a pitch-down moment, the plane will tend to recover from the stall without attention by the pilot. Pilot control of pitch attitude at stall is an important stall characteristic. Bll. Another aspect of stall which is of importance to the designer, as well as the pilot, is the effect of the stall on the roll behavior of the airplane, which brackets the pilot’s ability to control the plane’s roll behavior. Stall initiation on a typical wing usually means that only a portion of the wing has stalled, that only a portion of the wing has lost its lift. Loss of lift on a portion of the wing results in .a redistribution of the forces operating on the wing and the airplane. If the respective areas of separation on the right wing and on the left wing are not symmetrical, there is an imbalance of the forces which results in a roll moment or roll behavior of the airplane. Changes in net lift on the wings become more significant the further outboard on the wings they occur because greater roll moments are created as the location of the force imbalance moves away from the longitudinal centerline of the airplane. For this reason, it is generally considered desirable to avoid a wing design which results in initiation of stall at or near the wing tip. B12. The principal object of the patent in suit is to design and construct a wing shape which causes the initiation of stall to occur at or near the mid-semi-span of the wing, and to spread inboardly more quickly than outboardly. This object and the patentee’s solution are not new since the desirability and technique of avoiding tip stall were known and understood in the art long prior to the application date of the patent in suit. B13. The maximum attainable section lift coefficient of an airfoil section, or its Cimax, is a quality which was well appreciated in the art of aerodynamics prior to the date of the application which matured into the patent in suit. In theory, the lift coefficient of an airfoil section is directly proportional to the angle of attack of the section with respect to the oncoming airflow. As the angle of attack is increased, there is a proportional increase in the lift coefficient or lift capability of the airfoil section. As the angle of attack is continually increased, however, at some point there will be no further corresponding increase in lift coefficient or lift capability. At this point, the section is said to have reached its maximum lift coefficient, or Cimax,. In fact, further increases in angle" of attack will result in decreases in lift capability. The physical phenomenon which begins to occur at the maximum lift coefficient is referred to as stall. B14. The spanwise distribution of actually prevailing section lift coefficients, or Cj distribution, is also a term and quality well understood in the art of aerodynamics prior to the filing date of the application for the patent in suit. Theoretical approaches to determine the Cj distribution were advanced and published even before the Wright brothers’ first flight. B15. The relationship of the Cj and Ci distributions set forth in the patent is reflected in Figure 2 of the patent, which is commonly referred to as a stall diagram. B16. Stall diagrams, including the stall diagram of Figure 2 of the patent, and the methods for creating stall diagrams were well known in the prior art as early as the 1930’s. For example, methods for creating stall diagrams were taught in NACA Technical Report 572 (Exhibit E, 1936). NACA Technical Report 703 (Exhibit I, 1940) and NACA Technical Note 713 (Exhibit J, 1939) were extensions of the work described in TR 572 and further showed the use of stall diagrams and the limitations in the theories underpinning such diagrams. B17. The so-called envelope feature of Claim 2 declares a relationship between the Cjmax and Cj distributions in the stall diagram which relationship is, in fact, inherent in the creating of the stall diagram. In theory, the Cimax distribution always “envelopes” the Cj distribution because the actually prevailing lift (lower curve in Figure 2 of the patent) at any given spanwise location can never exceed the maximum attainable lift (upper curve in Figure 2 of the patent) at the same spanwise location. Therefore, wherever the Cl at any point on the wing reaches its maximum as shown in the C]max distribution on the stall diagram, stall is said to initiate on the wing. C. PRIOR ART KNOWN TO DR. GARBELL WHEN THE APPLICATION FOR PATENT WAS FILED. Cl. The patent application sought to utilize airfoil section data in a wing which, broadly speaking, was defined by three airfoil sections, one located at the root having the least mean line camber, one at the tip having the greatest camber, and an interjacent section having camber at variance with that obtained by straight-line fairing between root and tip. The wing was claimed to improve stall. C2. At the trial Dr. Garbell conceded: a) It was old in the art at the time of the alleged invention to use three or more controlled sections to define the exterior shape of an aircraft wing; b) It was old in the art as of 1946 to select three or more (as many as eight) controlled airfoil sectiqns to define the shape of a wing for aerodynamic purposes; c) It was old in the art to define the exterior shape of a wing using three or more controlled sections for the purpose of avoiding tip stall; d) It was old in the art, prior to 1946, to obtain an increase in the Cimax of an airfoil section by increasing the camber of that same section for the purpose of avoiding tip stall; e) It was old in the art to have a wing with the least camber at the root; f) It was old in the art to have a wing with the greatest camber at the tip; g) It was old in the art to compute the actually prevailing lift across the semispan in accordance with the teachings of the patent in suit; h) It was old in the- art to have an interjacent section that was different from the section obtainable by straight-line fairing between the root and tip sections; i) It was old in the art to determine the spanwise distribution of maximum attainable section lift coefficients in accordance with the teachings of the patent in suit. C3. In the early development of the art of aerodynamics and wing design, it was recognized that the airflow around three-dimensional bodies such as airplane wings is an extremely complicated phenomenon. Simplifying assumptions about the airflow were made so that some systematic theoretical research could be carried out. C4. One of the first simplifying assumptions made was to substitute a two-dimensional airfoil section for the three-dimensional wing or fluid-foil surface. In the conception of the airfoil section, the section actually represents a cross-section profile or slice through a wing having the same section and dimensions throughout the span. The airfoil section, having only two dimensions, induces airflow in only two dimensions rather than in three dimensions. C5. The concept of the airfoil section proved to be extremely useful for both theoretical and experimental research efforts. The aerodynamic properties of airfoil sections could be investigated in a “laboratory” setting, apart from how any given airfoil section might behave in an actual three-dimensional airplane wing. During the 1920s a great many airfoil section shapes were investigated in this country as well as in Europe, but the most expansive work was begun by the National Advisory Committee on Aeronautics (NACA) in the late 1920s. C6. In 1931, the, NACA published Technical Report 383 by Dr. Theodore Theodorsen (Exhibit N). In this report, Dr. Theodorsen described the thin wing theory, a refinement to the theory of airfoil sections. The thin wing theory applies to the component of flow associated with the relative curvature or mean line camber of the section. C7. NACA’s approach to research on the behavior of airfoil sections was to develop families of related airfoil sections, to determine certain properties of those sections. Such work was done by Prof. Robert M. Pinkerton around 1929. C8. The NACA set about testing two-dimensional models of these sections in their wind tunnel facilities. In 1935, the NACA published Technical Report 460 (Exhibit A) which presents the results of the development and testing of the NACA’s first series of related airfoil sections, called the NACA four-digit airfoils. C9. Research continued, and a five-digit series of airfoil sections was developed, followed by the six-digit series of airfoil sections. For the most part, the airfoil development work was unclassified, and the NACA published hundreds of pages of data in the form of technical reports and technical notes which were distributed thi’oughout the world prior to July 1945. CIO. The NACA’s objective in developing its airfoil sections and in publishing data on those sections was to provide the aircraft industry and wing designer with basic data which would enable wing designers to make knowledgeable choices of airfoil sections for wing design. As evidence of the acceptance of the quality and quantity of the NACA’s work, it is to be noted that Zien, writing his paper in Germany in 1938, used NACA airfoil sections in his example wing (Exhibit AP). Lachmann, writing for the Journal of the Royal Aeronautical Society in 1937, used NACA airfoil sections in his example (Exhibit AO). Zacher, in his article (Exhibit AJ) which describes the development of the D-30 Cirrus Glider notes that NACA mean lines were designated for the root and tip sections of the D-30 Cirrus. Virtually all of Dr. Garbell’s glider designs built in Italy in the 30’s and all of Dr. Garbell’s proposals and designs made while he was employed at CVAC featured NACA airfoil designations. It is undisputed, therefore, that the concept of the airfoil sections, the use and utility of airfoil sections and airfoil section data in wing design were commonly known and understood in the art well before the date of the application which matured into the patent in suit. D. COVERAGE OF THE CLAIMS. Dl. Claim 1 of the patent in suit relates to a lifting surface defined solely by specifying the relative mean line camber values of the three or more controlled airfoil sections in the wing. The least mean line camber must be in the controlled section at the root, the greatest mean line camber must be in the controlled section at the tip, and the mean line camber values at one or more interjacent sections must be greater than that which would be obtained at the same respective spanwise location of the interjacent sections by straight-line fairing between the root and tip airfoil sections. D2. Claim 2 of the patent in suit also defines a wing solely in terms of the relative mean line camber values of the controlled fluid-foil sections making up the wing. However, Claim 2 contains further limitation that “ . . . said three or more controlled fluid-foil sections having values of the mean line camber selected in such manner that the resulting spanwise distribution of maximum obtainable section lift coefficients of the three or more controlled sections forms a curvilinear polygon enveloping a curve representing the spanwise distribution of section lift coefficients for a given planform actually prevailing at the maximum attainable lift coefficient of the lifting surface”. There are three elements of the foregoing limitation; the spanwise distribution of maximum attainable section lift coefficients or Cimax distribution, the spanwise distribution of actually prevailing section lift coefficients, the Cj distribution, and the envelopment feature. ^ D3. Claim 3 is essentially the same as Claim 2 with the addition of a further limitation, “ . . . that the said resulting spanwise distribution of maximum attainable section lift coefficients for a given planform be so shaped that the first intersection with the spanwise distribution of actually prevailing section lift coefficients occurs in that interval of spanwise station in which stall inception is to be obtained”. This “tangency” limitation, describes where stall will occur on the wing based on the relationship between the Cjmax and Cj distributions. D4. Claim 7 relates to a lifting surface defined according to the same mean line camber relationship set forth in Claim 1 i. e., smallest, greater, greatest going from root to tip with the addition of a limitation as to the location of at least one interjacent section, i. e., where two lines drawn tangent to the lift distribution, Cj, intersect. D5. Other claims in the patent cover variants where the mean line camber of the interjacent section is less than that obtained by straight line fairing. D6. The patent issued on the continuation-in-part application 2,498,262, extended the patentee’s coverage to wings in which the camber of the interjacent section exceeded that at the tip. E. REDUCTION TO PRACTICE. El. Neither the patentee, Dr. Gar-bell, nor the plaintiffs ever actually reduced to practice the alleged invention covered by Claims 1, 2, 3 and 7 of the patent in suit. E2. Neither of the defendants, McDonnell Douglas and Boeing, designed, developed, constructed or performed work of any kind on wings for or on behalf of Dr. Garbell, and none of the wings employed in anywise by the defendants are or were reductions to practice of the invention of the patent in suit. E3. Dr. Garbell was never an employee of, nor performed any work for, either of the defendants, McDonnell Douglas and Boeing. E4. No one acting for or on behalf of Dr. Garbell ever actually reduced to practice the alleged invention covered by the claims of the patent in suit. E5. The tangible embodiments of the wings, which Dr. Garbell testified he designed for the CVAC Two-Engine Tailless, the XB-46, the Model 107, the Model 110 and the other CVAC airplanes, Models 240, 340, 440 and 880 having wings allegedly covered by the patent claims, cannot be relied upon by Dr. Garbell as actual reductions to practice for or on behalf of Dr. Garbell. E6. Dr. Garbell was paid by CVAC, as its employee, for such design and other work as he may have performed in connection with the wings above named. E7. CVAC manufactured all the above wings either for its own purposes or for purposes of the Government. E8. CVAC was not the agent of Dr. Garbell as to any of its operations, and, more specifically, was not Dr. Garbell’s agent as respects any wings it may have constructed which embodied a camber distribution covered by any of the claims of the patent in suit. E9. Dr. Garbell “constructively” reduced his invention to practice by the filing of the application of the patent in suit on July 16, 1946. The earliest date plaintiffs can rely upon as the date of invention is thus July 16, 1946 for purposes of applying Section 102. , F. ANTICIPATION — THE CURTISSWRIGHT DEVELOPMENT. FI. In a period from approximately 1935 to 1940, the Curtiss-Wright Company, St. Louis Division, built a series of single engine, low-wing monoplanes. The first of these airplanes was known as the Curtiss-Wright Model 19L. As originally built, the Model 19L had a constant camber wing, that is the camber at the root was the same as the camber at the tip with no variation between root and tip. During flight tests, this wing proved to have very poor stall qualities, with a tendency to stall near the wing tips causing dangerous roll behavior. Dr. Albert E. Lombard, Jr. of the Curtiss-Wright Company modified the 19L wing in order to improve its stall characteristics. The modifications were carried out by adding a “glove” to envelope the leading edge area of the 19L wing. The glove had the effect of increasing the camber of the wing near the tip with the increase in camber diminishing proportionately in moving inboard from the wing tip to the Rib 4 station. In addition to increasing the camber, the glove had the effect of increasing the leading edge radius of the wing. F2. The modified version of the Curtiss-Wright Model 19L became known as the Model 19R. Flight tests showed that the Model 19R wing was a substantial improvement over the 19L wing in terms of stall performance. Dr. Lombard disclosed his work to the public in an article entitled “Technological Improvements in the Curtiss-Wright Coupe”, published in the Journal of the Aeronautical Sciences, June, 1936 (Exhibit BC). F3. Dr. Albert E. Lombard, Jr.’s article illustrates the fact that it was well-known in the art prior to the filing date of the application which matured into the patent in suit that an increase in camber of a given airfoil section could bring about an increase in the maximum lift coefficient of the airfoil section and that the use of a highly cambered section in the tip region of an aircraft wing would help prevent tip stall; that is, that the airfoil sections near the tip would not reach their maximum lift coefficient (stall point) at the same time that sections further inboard would begin to stall. This fundamental principle, that an increase in camber tends to increase the maximum lift coefficient of an airfoil section, has also been taught in the earlier publications of the NACA, particularly in Technical Report 460 (Exhibit A). Dr. Lombard’s paper received wide attention as evidenced by the fact that it was subsequently cited by Zien in 1938 (Exhibit AP), written and published in Germany. Dr. Lombard’s paper is also cited by Lachmann (Exhibit AO), published in the British Journal of the Royal Aeronautical Society in 1937. F4. In 1938, the Curtiss-Wright Company undertook to design a fighter aircraft to be entered in a Government competition. The engineering decision was made to adopt the outer wing panel of the old Model 19R configuration for what would become the outer panel of the wing of the new Model 23 airplane. Up until the design of the Model 23 airplane, Curtiss-Wright had been building its single-engine monoplanes with fixed landing gear or partially retracted landing gear. In order to give the Model 23 greater performance, it was determined that the Model 23 would have flush retracting landing gear. F5. The Model 23 wing was designed to give improved aerodynamic performance as to both lift and drag. F6. The wings of the Curtiss-Wright Model 21B and the Model 23 have the same geometry and are formed of three controlled airfoil sections. F7. The Court has adopted the findings of the Special Master with respect to the Curtiss-Wright Model 21B and Model 23 as set forth on Page 41 of Exhibit 1 as follows; Airfoil Mean line shape Semi-span position Max. rise Ohordwise location of max. rise Rib 1 Ex; IH-10A, 246 and R. 970 0.0% .00 0.0% Rib 4 Ex. KE-1 26.5% .27 1.9% 30% Rib 11 Ex. KE 92.9% .48 to .54 3.5% 30-40% Faired airfoil Ex. KE — 2 26.5% .079 to .085 0.6% 30% F8. Rib 1 of the Curtiss-Wright Model 21B and Model 23 airplane wings is the controlled fluid foil section located at the root. F9. Rib 4 of the Curtiss-Wright Model 21B and Model 23 airplane wings is the interjacent controlled fluid foil section. F10. Rib 11 of the Curtiss-Wright Model 21B and Model 23 airplane wings is the controlled fluid foil section located at the fluid-dynamically effective tip. Fll. The Curtiss-Wright Model 21B and Model 23 wings were thus characterized by a camber distribution as called for by Claims 1, 2, 3 and 7 of the patent in suit in which the controlled airfoil section located at the root had the least mean line camber, the controlled airfoil section located at the tip had the greatest mean line camber, and in which the interjacent controlled airfoil section had a mean line camber at variance with, and greater than, the mean line camber of the airfoil section obtainable at the same interjacent span-wise station by means of straight-line fairing between the airfoil section at the root and the airfoil section at the tip of the wing. F12. Plaintiffs’ own evidence supports precisely that wing geometry which defendants sought to prove and which the Special Master found for the Model 21B and Model 23 wings, and shows the improvements in lift obtained by this wing over the previous wing used on the Model 19R. F13. Stall analysis of the CurtissWright Model 21B and Model 23 wing design (Exhibit IP) • shows that the spanwise distribution of maximum attainable section lift coefficients of the three controlled sections forms the equivalent of a “curvilinear polygon” which envelopes the curve of the span-wise distribution of actually prevailing section lift coefficients near the maximum attainable lift coefficient of the lifting surface, as covered by Claims 2 and 3 of the patent in suit. F14. The stall analysis for the Curtiss-Wright Model 21B and Model 23 wing design (Exhibit IP) shows that the first intersection of the curves of the maximum attainable and actually prevailing lift coefficients occurs in the vicinity of mid-semispan in accordance with the objectives of the patent in suit. F15. The stall analysis for the Curtiss-Wright Model 21B and Model 23 wing design (Exhibit IP) shows that the wing design substantially achieves the object of the patent in suit in avoiding tip stall and in creating mid-semi-span stall which spreads more rapidly inboard than outboard. F16. Analysis of the design of the Curtiss-Wright Model 21B and Model 23 wing design (Exhibit IP-1) shows that the interjacent fluid foil section is located near a spanwise point corresponding to the intersection of a tangent to the inboard portion of the curve of actually prevailing section lift coefficient and a substantially horizontal tangent to the highest point of the same curve as covered by Claim 7 of the patent in suit. F17. The Curtiss-Wright Model 21B and Model 23 wings embodied the same combination of elements as claimed in Claims 1, 2, 3 and 7. The Curtissr Wright Model 21B and Model 23 wings performed the same functions and achieved the same objectives sought by the patent in suit. F18. Flight tests of the CurtissWright Model 21B and Model 23 airplanes showed that each of the respective airplanes had highly satisfactory stall characteristics. F19. The Curtiss-Wright Model 23 airplane was built and offered for sale in the United States to the United States Government in 1939. F20. The Curtiss-Wright Model 21B was manufactured and placed on sale in the United States in 1940. F21. Aircraft of the Fighting Powers (Exhibit PK) states the following with reference to the Curtiss-Wright Model 21B: “The 21B was built solely for export and had never seen service on any battlefront when accepted by the Dutch. Nevertheless, when action was joined with the Japanese in the spring of 194