COFFIN, Circuit Judge.
In 1966 CBS brought suit in the district court for the District of Massachusetts against Sylvania claiming the infringement of three patents relating to color television picture tubes. The three [722]*722patents are the Fyler and Rowe patent, No. 2,690,518, and the two Giuffrida patents, Nos. 3,179,836 and 3,222,172. After a lengthy trial the district court held that all three patents were valid and had been infringed. Sylvania brings this appeal.
The color television industry in the United States has been a game of high stakes. The race for a high quality, compatible color television system capable of being mass-produced, has seen the expenditure of millions of dollars and the dedication of talent on a large scale by every major company in the radio and television field. The history of both legal and technological developments is extensive and is set forth in detail in the district court opinion. 294 F.Supp. 468 (D.Mass.1968). The abbreviated discussion which follows recounts only so much of that history as is necessary for resolution of the issues before us.
In 1946 the Federal Communications Commission (FCC) authorized the broadcasting of monochrome (black and white) television. Although at that time research had been and was being conducted in color television, an acceptable system had not been developed and many believed that development of an acceptable system was at least ten years away.
However, in 1949 the FCC called for proposals on color television standards. In response to that request, both CBS and RCA demonstrated color systems to the FCC. The first CBS system was known as the “field sequential” system which consisted of a monochrome picture tube with a rotating color disc in front.1 While the CBS system produced a color picture, it was not compatible, i. e., it could not be received without special parts and attachments on both black and white and color receivers.
RCA’s system — the “dot sequential” system — was a compatible system. It consisted of three picture tubes (one for each color) — which utilized phosphor dots to produce color — and a mirror which combined the three images into one picture seen by the viewer. Despite the compatibility feature, the RCA system was not successful — primarily because of inadequate color quality and high production costs.
In the fall of 1949 RCA began an intensive program to develop a compatible all-electric system which would require only a single picture tube. By March 29, 1950, RCA was able to demonstrate such a system — the planar tricolor screen system. The basic components of the system were an electron “gun”, a tube, a mask, and a screen. The tube was constructed of glass and served as a receptacle for the three other parts. The screen was placed just behind the face of the tube and consisted of a flat panel (hence the name planar) upon which thousands of trios of phosphor dots (red, blue, and green) were deposited by a silk screen process.
At the other end of the tube was placed the “gun” which was simply a transmitter of electron beams. Color was produced by the electron beams striking the appropriate phosphor dot at the proper angle. In order to achieve the required precision, it was necessary to control the direction and velocity of the beam. RCA used a “shadow” mask for this purpose. The mask was placed between the “gun” and the screen and was a thin metal sheet containing a multitude of holes. Each hole was aligned with a trio of phosphor elements and was not only smaller than the trio but smaller than an individual dot. This construction permitted a beam to strike only one dot at a time, the other two remaining shadowed by the mask.
Obviously, proper alignment of the mask holes with the phosphor elements was critical. If alignment was incorrect, “misregistry” occurred, producing blurred color. A cause of misregistry [723]*723was heat-induced expansion of the mask because approximately eighty per cent of the electron beam energy was absorbed by the mask as the beams were channeled toward the screen. The heat generated by this absorption of energy resulted in a buckling of the mask as lateral expansion took place. The final result was improper alignment.
RCA’s approach to this problem was to pre-stress the mask by “hot blocking”, 1. e., the mask is heated and stretched across a metal frame and then cooled. The theory was that, as installed, a pre-stressed mask would have no capacity for heat expansion. But RCA was on the horns of a dilemma, for pre-stress-ing gave rise to serious production problems due to the fact that television picture tubes are required to be evacuated.2
In addition to the alignment problem, it was also thought desirable to have the face of the tube itself serve as the screen. In October of 1953 CBS announced a development directed at both problems. The announcement concerned a patent (the Fyler and Rowe patent) which had been applied for on June 1, 1953. The subject of the patent was a color television picture tube with a curved face and a matching curved mask. The face of the tube itself served as the screen with the phosphor elements being directly disposed thereon.
The curved screen and matching curved mask arrangement of the Fyler and Rowe patent purported to confer a number of advantages. Its most noteworthy attraction was that it affords permanent registry because the spherical mask responds to heat by expanding radially with no significant lateral movement.3 The mask holes therefore are not expanded as would be the case in lateral expansion. The permanent registry feature also provides significant production gains. Since the mask need not be prestressed, problems of tube evacuation — namely, the lengthy cooling cycles and delicate handling required— are obviated. Finally, by placing the screen on the face of the tube, maximum utilization of the tube face is obtained. The CBS disclosure met with an almost instantaneous favorable response in the industry and ultimately resulted in the concluding of licensing agreements with RCA and several foreign manufacturers. There can be little doubt that the curved screen and matching curved mask arrangement was a major development in color television.
Sylvania does not dispute the advantages of the Fyler and Rowe patent. Sylvania does, however, contend that the patent is invalid for three reasons: (1) obviousness; (2) anticipation; and (3) non-compliance with 35 U.S.C. § 112. Before discussing these challenges in detail, we note that Sylvania accepts, as it must, its burden of showing that the district court’s determination was “clearly erroneous”. Rule 52(a) Fed.R. Civ.Proc.
A. Section 112
The relevant portion of 35 U.S.C. § 112 reads as follows:
“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 [724]*724enable any person skilled in the art to which it pertains * * * to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention * * * ”
Sylvania argues that the patent discloses neither a useful way of making the screen nor the “best mode contemplated by the inventor * * CBS’s response is that the patent clearly and accurately assumed that screen-forming techniques were well-known in the art.4
The district court found no problem with respect to § 112 because it held that “it was generally known in the trade prior to October of 1953 how to utilize photo-resist techniques for forming tri-color phosphor screens on a curved inner surface of a tube.”
Sylvania’s first argument is that the district court focused on the wrong date — October 1, 1953, when CBS publicized the patent application rather than June 1, 1953, the date of application. We assume for the purposes of discussion that June 1, 1953 is the date as of which disclosure is to be tested for adequacy.5 Even, however, as of June 1, 1953, there is ample evidence that photographic techniques were well-known in the art. Both the pre-trial stipulation and the testimony of employees of RCA (notably, Moody and 'Grimm) indicate that photo-resist techniques were known perhaps as early as 1948 but in any event clearly before 1953.6 Perhaps most importantly, a paper on photo-resist techniques entitled “The Preparation of Phosphor Screens for Color Television Tubes” was presented by its authors, Levy and Levine, to the American Electrochemical Society in April of 1953.7 Finally, we note that the patent examiner must have agreed in issuing the patent that disclosure was unnecessary. While this fact is not conclusive, it is entitled to some weight. See Western States Machine Co. v. SS Hepworth Co., 147 F.2d 345 (2d Cir. 1945).
Sylvania’s second § 112 attack is directed at the “best mode” requirement. In October of 1953 CBS applied for a patent on photo-resist technique (Perry and Rowe patent No. 3,080,231) and in the application it was stated that “* * * no practical photographic methods have been devised to form phosphor screens directly upon a curbed surface.” Sylvania argues that CBS knew this as of June 1, 1953, and therefore, did not disclose the “best mode” of carrying out the Fyler and Rowe invention. Sylvania’s argument is susceptible of two interpretations: first, that CBS knew as of June 1, 1953, that there were no direct deposit techniques known in the art, or second, that as of June 1, 1953, CBS knew of Perry and Rowe’s invention but did not disclose it. As our prior discussion demonstrates the first contention is without merit.
Leaving aside the question of whether the statement quoted from the Perry and Rowe application could fairly be considered “puffing”, we fail to see [725]*725how Sylvania’s second contention is supported by it. In drafting the specifications for a patent application an applicant is required to disclose the “best mode” known to him; he is not obligated to develop a perfect process. Thus, the only way in which Sylvania’s “best mode” attack could be upheld would be if it were shown that as of June 1, 1953, CBS possessed knowledge of a better method of photo-resist technique but failed to disclose it. There is, however, no evidence to support such a finding.8
We hold that the district court correctly concluded that the Fyler and Rowe patent complied with the requirements of 35 U.S.C. § 112.
B. Anticipation
Sylvania contends that the Fyler patent was fully anticipated by three prior patents and a technical paper published in the Journal of the Society of Motion Picture and Television Engineers in July of 1952.
1. The Kaplan Paper
The first claim of anticipation relates to an article by Sam H. Kaplan entitled “Theory of Parallax Barriers” which discussed the geometric principles underlying the use of masking devices in color television. The district court relied on expert testimony in holding that the Kaplan paper did not anticipate the
Fyler patent. In our view the district court cannot be said to have been “clearly erroneous” in so holding.
While the Kaplan paper recognized that a color television mask and screen might be non-planar, and therefore spherical, no attempt was made to teach how such a mask and screen might be constructed. Indeed, Kaplan’s entire discussion is east in two-dimensional terms, and a spherical construction would obviously involve three dimensions.9 Moreover, Kaplan did not recognize the advantages that would be conferred by a spherical construction and he did not advocate the use of a spherical mask and screen.
The test for anticipation is whether the prior printed publication (or patent) describes the invention with sufficient clarity and specificity so that one skilled in the art may practice the invention without assistance from the patent claimed to have been anticipated. Ballantyne Instruments & Electronics, Inc. v. Wagner, 345 F.2d 671 (6th Cir. 1965), on remand, 260 F.Supp. 540 (S. D.Ohio 1966), aff’d, 386 F.2d 789 (6th Cir. 1967); Bros. Inc. v. W. E. Grace Mfg. Co., 351 F.2d 208 (5th Cir. 1965), cert, denied, 383 U.S. 936, 86 S.Ct. 1065, 15 L.Ed.2d 852 (1966). The district court was entitled to find that Kaplan’s failure to teach construction of a spherical tube did not satisfy the test.10
[726]*7262. The Avins Patent, No. 2,712,568
The Avins patent relates to “signalling circuits” and particularly to “col- or synchronization” for color television receivers — a process wholly apart from and precedent to the subject of the Fyler and Rowe patent. The patent specifications do not mention picture tubes; in fact, the application states that “It is not the intention here to limit the practice of the present invention to anjr single type of color image reproducing device.” The drawings, which accompanied the specifications, do however clearly represent a spherical mask and screen. But equally clearly, nothing is taught about construction. In our opinion, Judge Woodbury’s remark in Des Rosiers v. Ford Motor Co., 143 F.2d 907 (1st Cir. 1944), is dispositive of Sylvania’s contention:
“Drawings can anticipate as well as words * * * if they teach to the art what the patentee claims as his invention.” (Emphasis added.) 143 F.2d at 911-912.
Unlike the case of simple mechanical devices where drawings may by themselves teach all that is necessary for construction, the unexplained representation of a curbed screen and mask in a color television picture tube does not deal with factors such as stability or how the mask is to be attached, and therefore does not aid those skilled in the art in the production of such a model.
3. The Lawrence Patent, No. 2,692,532
The Lawrence patent dealt with color picture tubes and was specifically concerned with a lens system for “secondary deflection” of electrons which would result in a greatly reduced electrical power requirement.11 As in the case of the Avins patent, supra, anticipation is alleged on the basis of drawings in the application which depict a curbed screen and mask. But again, as in the Avins patent, there is no teaching as to practical use and we hold that the district court was clearly correct in finding that the Lawrence patent did not anticipate Fyler.12
4. The Hansen Patent No. 2,568,448
The Hansen patent, like the Lawrence patent, dealt with the problem of divergence of the electron beams. Hansen’s approach was to impose alternate color strips on the screen. In addition, Hansen called for a pitch in both the mask and the screen pursuant to a formula disclosed in the application. The claimed anticipation rests, not upon drawings, but upon the following statement contained in the application:
“* * * a slightly different equation will apply if the screen and the grid wire configuration are slightly curved as may be the case in many practical tube constructions.”
Hansen teaches nothing about spherical construction, and, in fact, we think that the statement quoted above may be described as merely speculation that a screen might be curved. It clearly does not constitute anticipation.
C. Obviousness
Sylvania’s final challenge to the validity of the Fyler patent is that its subject matter was, at the time of invention, obvious to one skilled in the art. 35 U.S.C. § 103.
The most authoritative pronouncement on the doctrine of obviousness is Graham v. John Deere Co., 383 U.S. 1, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966), where the Supreme Court laid down the following guide:
“* * * the scope and content of the prior art are to be determined; differences between the prior art and [727]*727the claims at issue are to be ascertained ; and the level of ordinary skill in the pertinent art resolved * * *383 U.S. at 17, 86 S.Ct. at 694.
In pursuing these- three lines of inquiry, the Court observed that:
“* * * Such secondary considerations as commercial success, long felt but unsolved needs, failure of others, etc. might be utilized to give light to the circumstances surrounding the origin of the subject matter sought to be patented * * 383 U.S. at 18, 86 S.Ct. at 694.
Sylvania charges that the district court relied on secondary considerations and failed to confront the fundamental issues posed in John Deere, supra. The district court’s opinion is not, however, so easily dismissed — as an analysis consistent with the approach mandated by the Supreme Court demonstrates.
1. The Prior Art
Our earlier discussion of the development of color television disclosed that by June 1, 1953, (the date of the Fyler and Rowe application), the art had progressed from a non-compatible “field sequential” system to a compatible “dot sequential” planar screen system. This latter system was the RCA system and as of 1953 RCA was the undisputed industry leader. As of 1953 a color picture tube having a curved screen and matching curved mask did not exist. It is true, however, that the possibility that such a configuration might be developed had been recognized — notably by Kaplan and perhaps arguably by Avins, Lawrence, and Hansen.
2. Differences Between the Fyler and Rowe Patent and the Prior Art
In contrast to the pre-stressed mask-planar screen combination employed in the prior art, the Fyler and Rowe patent called for a curved screen and a matching curved mask. The major contribution of the Fyler and Rowe patent was the mask configuration.13 The primary advantage of the curved mask was that it afforded permanent registry. It also had other attractive features in that “* * * it reduces the weight of the tube, it allows exhaust time to be reduced and it is simpler mechanically.” (Testimony of Grimm, RCA Senior Engineer.)
3. The Level of Ordinary Skill in the Art
As we noted earlier, color television development involved almost every major company in the communications industry. In view of the fact that millions of dollars were expended and in view of the scale of the organized application of technical intelligence to color television, it would be a classic understatement to say that the level of skill in the art was high.
We think it significant that, given this high level of skill, those trained in the art immediately recognized the Fyler and Rowe patent as a breakthrough development. See, e.g., United States v. Adams, 383 U.S. 39, 52, 86 S.Ct. 708, 15 L.Ed.2d 572 (1966); Otto v. Koppers Co., 246 F.2d 789 (4th Cir. 1957), cert. denied, 355 U.S. 939, 78 S.Ct. 427, 2 L.Ed.2d 420 (1958); Ibis Enterprises, Ltd. v. Spray-Bilt, Inc., 220 F.Supp. 65 (S.D.Fla.1963).14 Indeed, Sylvania’s own color tube manager, Mr. Lamb, stated in November of 1953 that the CBS tube “* * * is a very desirable advance in the state of the art, and in the long run will probably be the [728]*728method of making color tubes as long as an aperture mask is used.”
To the laudatory comments of those skilled in the art must be added the tangible action of RCA in licensing the Fyler patent. To be sure licensing has its limits as an indicator of obviousness, cf. Lear, Inc. v. Adkins, 395 U.S. 653, 89 S.Ct. 1902, 23 L.Ed.2d 610 (June 16, 1969), but absent a showing that the licensing arrangement was motivated by considerations other than those relating to evaluations of validity and desires to avoid infringement,15 the fact that the leader in color television development chose to license the Fyler patent is entitled to some weight.
Finally, Sylvania concedes that there was a technology gap in the mask structure process and it concedes that CBS was the first to fill that gap. But, in Sylvania’s view, the gap lay not in the Fyler and Rowe concept of curved configuration of mask and screen — not in itself patentable — but in the technology required to form a screen and mount a mask in order to utilize the concept.
The fundamental reason assigned for this argument is that because the principle of a curved mask and screen had been recognized in the art, Fyler did not “invent” anything but merely reduced that known principle to application. Were Sylvania’s premise true, its conclusion would inevitably follow. But therein lies the defect. At best the prior art can be said to disclose that non-planar surfaces were possible as a matter of geometric theory. There had been no experimentation concerning application and no procedures taught. In this respect the case before us is clearly distinguishable from Koppers Co., Inc. v. Foster Grant Co., Inc., 396 F.2d 370 (1st Cir. 1968). Of course, had Fyler not made his discovery, the art might well have eventually arrived at the same result. But arguably inevitable technological progress is not the test. United Chromium, Inc. v. International Silver Co., 60 F.2d 913, 916 (2d Cir. 1932), cert. denied, 288 U.S. 600, 53 S.Ct. 319, 77 L.Ed. 976 (1933). It is well settled that in retrospect the most creative developments often appear commonplace. The critical question is whether at the time • of invention the subject would have been obvious to one skilled in the art. There is ample evidence to support the district court’s finding that the Fyler patent “was far from obvious”.
In view of our disposition of the above argument, we need not consider Sylvania’s other contentions which are primarily directed at the district court’s interpretation of secondary evidence.16 We therefore affirm the district court’s holding that the Fyler and Rowe patent is valid.17
Finally, we consider the two remaining CBS patents involved in this case — the so-called Giuffrida patents. The two Giuffrida patents are closely related and deal with a second problem of proper registry — the effect of the earth’s magnetic field.18 At the outset [729]*729we wish to make it clear that we consider neither the issues of validity nor infringement to be foreclosed.19
As we observed earlier, color is produced by the electron beams striking the phosphor dots on the screen. The direction of the beam is critical, and, because of the force of the earth’s magnetic field, the trajectory of the beams is not a straight line. The effect of the earth’s magnetic field is to deflect a horizontal beam of electrons downward. Unless this effect is compensated for, misregistry will occur. The earliest attempt at compensation was magnetic shielding of the beam through use of a metal case. A second attempt at compensation was planned distortion which took two primary forms: color equalizer magnets and purity coils. Neither method was entirely satisfactory.20
Giuffrida’s approach to compensation was to compute the degree of deflection caused by the earth’s magnetic field in the northern hemisphere, and having obtained this data, to compute the degree of physical misalignment for the location of the light source sufficient to cancel out the magnetic field effect. Simply stated, Giuffrida’s idea was to form the screen with the phosphor dots located where the electron beam, when deflected by the magnetic field, would hit them.21
At trial Sylvania challenged the Giuf-frida patents on grounds of anticipation and obviousness. Contentions as to anticipation have not been pressed on appeal and are not before us. Although the issue of obviousness remains open, it is inextricably linked to questions concerning the scope of coverage of the patents. Essentially, the problem before us is to define the subject of the patents with some precision. Because the district court found infringement to be conceded, it did not confront the issue of coverage.
If the scope of the patents appeared to us clear, we would not hesitate to reach our own determination. Unfortunately, the matter is far more complicated. Although the patents themselves were in evidence at the trial, the file wrappers were not. We are unwilling to decide an important question to which the patent alone arguably gives an ambiguous answer.22
The relevance of the file wrapper as evidence of the patentee’s intentions be[730]*730fore the Patent Office, and as evidence of the inventor’s own evaluation of his work, has long been recognized. See, e. g., Graham v. John Deere Co., supra 383 U.S. at 33, 86 S.Ct. 684, 15 L.Ed.2d 545, citing Hogg v. Emerson, 11 How. 587, 13 L.Ed. 824 (1850); Progressive Engineering, Inc. v. Machinecraft, Inc., 273 F.2d 593, 598 (1st Cir. 1959). The file wrapper may prove to be particularly helpful in cases like the one before us where a broadly worded claim is based on more narrowly framed specifications.
That part of the judgment which held valid and infringed the Fyler and Rowe patent, No. 2,690,518 is affirmed. That part of the judgment which held valid and infringed the Giuffrida patents, Nos. 3,179,836 and 3,222,172 is vacated and the case remanded for further proceedings consistent with this opinion. Appellee recovers costs on appeal.