Emerson v. National Cylinder Gas Company

146 F. Supp. 581 (1956)

John H. EMERSON and J. H. Emerson Co.
v.
NATIONAL CYLINDER GAS COMPANY and Stanton Scientific Equipment Co.

Civ. A. Nos. 54-900, 55-905.

United States District Court D. Massachusetts.

November 19, 1956.

Robert L. Thompson, Boston, Mass., for plaintiff.

M. Hudson Rathburn, Mason, Kolehmainen, Rathburn & Wyss, Chicago, Ill., Herbert P. Kenway, Boston, Mass., for defendant.

ALDRICH, District Judge.

These are two actions for infringements of two patents for intra-pulmonary respirators which were consolidated for trial. The parties having submitted some 1,500 pages of oral testimony and depositions, 200 exhibits, and 250 pages of briefs, exclusive of briefs considered in connection with the motions for default, D.C., 131 F. Supp. 299, and for summary judgment, D.C., 135 F. Supp. 268, I cannot review the play without setting forth a more detailed account than that usually supplied by critics. (Nor could I be fairly described as an over-night critic.) The cast of principal characters, not necessarily in the order of appearance, is as follows:

Emerson — Inventor of patent No. 2,364,626, filed Nov. 9, 1942, issued Dec. 12, 1944, known as the '626 Patent, and of patent No. 2,468,741, filed Dec. 12, 1944, issued May 3, 1949, known as the '741 Patent, basis of action 54-900; plaintiff herein.

J. H. Emerson Co. — Emerson's company, manufacturer and seller of allegedly infringed resuscitators, and of other machines, notably the iron lung; co-plaintiff.

Sinnett — Inventor of Patent No. 2,268,172, filed June 12, 1941, issued Dec. 30, 1941, known as the Sinnett Patent, exclusively licensed to J. H. Emerson Co., the basis of action 55-905; employee of J. H. Emerson Co.

Stanton Scientific Equipment Co. — Manufacturer of accused resuscitator; named defendant in both cases, but not served; held to have appeared in 54-900; refused to appear in 55-905, although as between defendants the indemnitor and the one ultimately liable.

National Cylinder Gas Co. — Distributor of accused resuscitator; defendant herein; plaintiff in counterclaim for declaration of invalidity.

Erickson — Designer and seller of resuscitators and other equipment. Connected with E. & J. until about 1945.

E. & J. — E. & J. Manufacturing Co., manufacturer and seller of resuscitation and other equipment.

Goodner — Designer of resuscitators, associated with E. & J. 1927 to 1933; in business for himself 1937-1956.

N. L. Curtis, L. N. Curtis — Resuscitator salesmen, who did business with each other, but not in association.

Drager — German inventor of original resuscitator, known as "pulmotor."

Stolle, Raymond, Fox — Extra designers.

Act I (1908-1921)

In 1908 Drager invented a respirator for artificial breathing which operated solely by the pressure inhalation gas, permitting the gas to pass from the container cylinder into the lungs during the positive (inhalation) cycle, and drawing it out during the negative (exhalation) cycle, the built-up positive and negative pressure in the lungs operating automatically the mechanism to reverse the cycles. The suction "motor" was a Venturi, a simple, well-known device, see Gillman v. Stern, 2 Cir., 114 F.2d 28, certiorari denied 311 U.S. 718, 61 S. Ct. 441, 85 L. Ed. 468, having no moving parts, and consisting of a nozzle and chamber, which creates a partial vacuum in the chamber when gas under pressure passes through the nozzle. The positive gas was fed to a face mask through a tube, and, alternatively, the thus manufactured negative pressure was applied to the mask through a second tube. The change-over valves from positive to negative were flipped by a mechanism consisting essentially of a bellows, expanding or contracting responsively to the varying lung pressure, operating on a spring-controlled lever or toggle. The Drager respirator took several forms, resulting in a number of United States and British patents, some of which may not have been too practical; but none of which were inoperable. The essential device, Nos. 1,044,031 and 1,049,346, was entirely practicable, although somewhat cumbersome and elaborate. It operated at pressures, especially the negative, in amounts now considered excessive. All Venturi machines "waste" a certain amount of gas on exhalation because the gas utilized to create the negative pressure in the Venturi chamber discharges to the atmosphere instead of to the patient. Drager possessed what is called today an economy feature, meaning that the Venturi during inhalation drew in and mixed outside air with the pressure gas — hence less gas was required to fill the patient's lungs. By 1921 Drager "pulmotors" were condemned by the American Medical Association, hereafter termed the AMA. The record does not adequately satisfy me as to the reasons for this, but whatever they were they were not eliminated and the device ceased to be a factor on the American market.

In 1913 another German, Stolle, filed an application which in December, 1920 ripened into Patent No. 1,358,893. This device required manual adjustment to individual lung capacity, and was never produced. The plaintiffs seek to rebut the presumption of operability arising from the issuance of the patent by testimony that it would not work. Having in mind the low pressure presumably available, the necessary speed of operation, and the problem of metal to metal friction before the development of present-day materials, I am not unsympathetic with that contention, and could wish the testimony, either way, had been more specific. However, pistons can function, as subsequent machines proved, and I suspect the principal reason Stolle was never produced was because it lacked certain features possessed by Drager which made effective competition impossible. I find that plaintiffs have failed to overcome the presumption of operability,^[1] and I will not disregard certain teachings of Stolle, if material.

Act II (1927-1938)

In 1927 E. & J. put on the market its resuscitator, Goodner Patent 1,893,670, followed later by Erickson Patent 2,138,845. The E. & J. machine differed from Drager in two substantial respects. It substituted for the Venturi a piston and cylinder device which utilized the pressure of the inhalation gas itself to create the required negative (exhalation) pressure. No gas was wasted by discharge to the atmosphere without passing to the patient. In spite of this, I am satisfied that the piston and cylinder was a poorer device than the Venturi motor. It was cumbersome, expensive to manufacture, had many moving parts, and required maintenance. Trouble could come from the piston leathers. I believe Goodner designed it to escape the Drager patent, and possibly the effects of its AMA condemnation. The Erickson patent, which was the real E. & J. machine, was filed in 1935, after all Drager patents had expired, but retained the piston device. I suspect, however, that once having started with a piston machine, there may have been no particular incentive to change over to a Venturi until competition threatened.

A second major E. & J. change from Drager was that for the bellows valve-activator E. & J. substituted a diaphragm and toggle, with different type valves. This was an improvement. It was a more compact and efficient snap-acting mechanism, but it was not new, except in the resuscitator field. Incidental changes were in the oxygen flow circuit, which had the effect of eliminating most, if not all, possibility of internal contamination of the in-flow channel by the patient.

The E. & J. machine in the Erickson form was successful commercially. I am unable to tell to what extent it suffered indirectly from the AMA condemnation of the Drager pulmotor, but it managed, as have other machines since, without AMA approval. However, Stanton and Stephenson carried on a vigorous campaign to obtain approval of Erickson, and in 1939 were successful.^[2] In advance of anticipated approval various persons began designing Venturi machines.

Act III (1938-1941)

Although return to the Venturi was eventually accomplished, all attempts were not successful. In 1938 or 1939 Goodner produced a rotating one, which would not work in cold temperatures. An application filed April 26, 1939 on another device, resulted in Raymond Patent No. 2,273,790, but I am satisfied that this was entirely inoperable. Defendants argue that it could have been made to function by mere mechanical changes, but they have failed to demonstrate how. While it is true that an inoperable device may sometimes constitute a prior reference, and the new flow circuit, in itself, was an improvement, I cannot feel that merely showing a circuit which might be an improvement, had it worked, is sufficient disclosure. In other words, I do not find the facts to fall within defendants' cases, and I put Raymond aside.

Plaintiffs' Sinnett machine was designed in April, 1940, and the first sale made July 6, 1940 (to L. N. Curtis), which is the date of invention. Defendants claim that in 1938 Erickson almost exactly anticipated it. Erickson did so testify, with some corroboration. He stated that he made but one model of this alleged machine, and then set it aside. The model has been lost, and there are no contemporary identifying records. Plaintiffs say that defendants' proof, being based on oral testimony only, was insufficiently strong in view of the special burden applicable to this situation. Cf. United Kingdom Optical Co. v. American Optical Co., 1 Cir., 68 F.2d 637, affirming, D.C., 2 F. Supp. 174. I do not reach this question. Defendants rely on 35 U.S.C.A. § 102(g), which forbids a patent if "before the applicant's invention thereof the invention was made in this country by another who had not abandoned, suppressed, or concealed it."^[3] In view of Erickson's testimony that he put the device aside as soon as it was fully tested and found to work to his satisfaction, and showed it to no one, instructing his employees to secrecy, I find that the invention, if made, was suppressed or concealed. Cf. Gillman v. Stern, 2 Cir., 114 F.2d 28, certiorari denied 311 U.S. 718, 61 S. Ct. 441, 85 L. Ed. 468. Defendants seek to rebut this finding by arguing that part of the device was "carried forward" into the first Fox model, which came two years later, after Sinnett. It is true that when an invention is not disclosed because it is not perfected, or is being worked upon, and disclosure would be premature and injurious to the patent, a case of concealment is not made out. Cf. Kendall v. Winsor, 21 How. 322, 16 L. Ed. 165. However, I find, consistently with an argument made in defendants' brief,^[4] (in another connection, defendants not recognizing that a dilemma may have two horns), that Erickson's reason for failure to bring out his 1938 invention immediately (assuming he then had it), was simply because it would interfere with the sale of the piston model, of which his company was still well stocked. This was not an excusable reason. Hambuechen v. Schorger, 56 App.D.C. 141, 10 F.2d 1006.

In 1938 or 1939 Goodner designed a Venturi device with an economy feature. This feature always requires a Venturi operation for inhalation as well as for exhalation. Goodner first used a two-purpose, movable Venturi, alternating between the two flow circuits by means of a rotating drum. As previously stated, in cold temperatures this stuck. The machine was eventually redesigned with a fixed, double-acting Venturi, through which the gas passed alternatively in one direction or the other.^[5] In its second form Goodner might be cited in anticipation of Sinnett if it were timely. The date of its perfection was vigorously disputed. The burden of proof is upon the defendants, and I find they have not met it.

Sinnett — Validity

In this state of the art did the Sinnett machine amount to invention? Admittedly it was a combination of old elements — the Erickson diaphragm toggle-valve assembly, which it took over bodily, and the Venturi motor. The question, therefore, is, what did it produce that was novel or unexpected? Plaintiffs claim that Sinnett (1) passed no inhalation gas through the Venturi, eliminating its contamination by contagious patients; (2) did away with the economy feature which mixed air with the inhalation gas, and (3) eliminated or greatly reduced lubrication and maintenance. They caution the court that the fact that in retrospect this may all appear simple and easy may be hindsight only, and is to be considered with caution and in the light of the evident contemporary difficulty experienced by others. I believe the degree of such caution must depend upon the circumstances. I cannot agree with plaintiffs that any prior unsatisfied demand had been as long or as strong as they suggest. I believe they underestimate the indirect effects of AMA disapproval of Drager, and exaggerate the call for improvements over Erickson. I also disagree with their legal contention that the test of what would appear to be mere mechanical changes to persons skilled in the art must be down-graded if the only persons who happen to be skilled in the art lack ordinary mechanical or engineering ability. This is mirror-writing the principle. 35 U.S.C.A. § 103 is a negative limitation. The test of mechanical skill vs. invention is an objective one, "not restricted to the skill of any particular mechanic," Johnson Co. v. Pennsylvania Steel Co., C.C.Pa., 67 F. 940, 942, affirmed, 3 Cir., 70 F. 244, or we would have the absurd result that inconsequential inventions could be supported by proof of individual subjective incapacity.

I cannot find that the improvements effected by Sinnett were as valuable or striking as contended. With regard to reduction in maintenance, the troublesome, or at least cumbersome part of Drager appeared to be the bellows, etc., activating the valves. This had been solved by Erickson. Combining Erickson's diaphragm and toggle with Drager's Venturi, reduced maintenance over Erickson, but in no new or surprising manner. I do not find here the greatly increased efficiency or other unexpected physical improvements that helped carry the day in Ellis, Inc., v. Denis, 1 Cir., 224 F.2d 311.

The alleged advantage of eliminating the economy feature was not testified to by any witness. Counsel argued that it was desirable because the admission of outside air made administration unsafe in catastrophe atmospheres contaminated by smoke, escaping gas, etc. Plaintiffs still sell today in substantial numbers devices with the economy feature (the '626 Patent), as well as without it (Sinnett, and the '741 Patent), and, strangely enough, advertise that the economy type "is ordinarily supplied for non-medical rescue work." No witness testified that any catastrophe atmosphere in which a worker would be likely to apply the device is too unwholesome to give the patient in diluted form. I cannot regard advocacy as a satisfactory substitute for testimony, and I find that if Sinnett invented anything of substantial value it was the flow circuit apart from any question of the economy feature. Ordinarily, elimination of an element and of its function is not invention. Richards v. Chase Elevator Co., 159 U.S. 477, 16 S. Ct. 53, 40 L. Ed. 225; Grayson Heat Control, Ltd. v. Los Angeles Gas Appliance Co., 9 Cir., 134 F.2d 478. The economy feature was not an essential mechanical part, and it seems to me that there is no reason for not applying the general rule.

The final question, then, is whether the flow circuit itself, in passing the inhalation gas direct to the patient without going through the Venturi, constituted invention. From the standpoint of utility this was doubtless something of an improvement, although there was no testimony as to how hard it is to clean a Venturi after contamination by a contagious patient, or that it requires peculiar care or skill not possessed by persons normally using resuscitators.^[6] It is to be borne in mind that plaintiffs do not, and cannot, if they are to establish infringement, claim an invention whereby no part of the inhalation circuit serves as the exhalation circuit — in other words, a device in which the inhalation circuit is totally uncontaminable by a previous patient, so that none of it has to be cleaned. Therefore the improvement was less valuable than it would have been otherwise.^[7]

In spite of the warning properly given by counsel as to the ease of hindsight, I can see nothing in this picture but ordinary mechanical ingenuity, even a high degree of which does not constitute invention. Allied Wheel Products v. Rude, 6 Cir., 206 F.2d 752. As previously stated, I do not believe it constituted invention to eliminate Drager's economy feature. Mechanically, of course, it was easier to build a structure without it than with it. The difficulties experienced by Goodner and others, which plaintiffs point to as indicating invention on Sinnett's part, were due in large measure to the attempt to preserve that feature. Without it there was no need to run the inhalation gas through the Venturi, and the Venturi could be located anywhere in the exhalation circuit. What would be more natural than to place it, as was done, essentially where the piston motor was on Erickson? Nor, mechanically, does such location appear difficult. In spite of the presumption of validity arising from Patent Office action I cannot feel, in the light of the then state of the art, that Sinnett, in the form here claimed, did anything appreciably more than assort a good hand out of cards lying face up on the table.

Act IV (1942 —)

On November 9, 1942, Emerson filed an application for a patent on a hand-held portable resuscitator, which issued as the '626 Patent Dec. 12, 1944. This device had an economy feature, and for that reason is not claimed to be infringed. It is important only because claim 9 of the '741 Patent, filed Dec. 12, 1944, the basis of action 54-900, must be related back to it, or that action fails. After suit was brought on this later patent plaintiffs' counsel discovered that Emerson had made a sale to the British Air Ministry during the war which preceded the filing date by more than a year. Although it was a limited sale it was destructive of the patent, except that plaintiffs seek to save claim 9 under 35 U.S.C.A. § 120.

Between November, 1942, and December, 1944 there was activity in the field, which defendants discuss in their brief. Indeed, in some places they suggest that even the '626 Patent anticipates '741. All of this, of course, in view of plaintiffs' opening concession, is quite beside the point, and I will not consider it further.

The '741 Patent, which differs from '626 in that it has no economy feature, calls for a hand-held structure which contains, inside the casing, the Venturi and the valve mechanism. The face mask is directly affixed to the outside of the casing. In portability and flexibility this device marked a great improvement over prior resuscitators which were attached to the oxygen tanks, and whose mask, though separate therefrom, could be only a few feet away. However, on examination, there was very little mechanical change, except the physical fact of separation. So little change, in fact, that I do not believe it constituted invention.

The process of the '741 application through the Patent Office is interesting. At the outset the Examiner repeatedly took the position that there was no invention because applicant was merely using the fully disclosed Sinnett mechanism to actuate the Drager portable respirator, a hand-held device without an automatic valve, not previously considered in this opinion.^[8] Applicant finally convinced him he was wrong. He was, but only in his approach. Applicant was not bringing Sinnett to Drager, but Draer to Sinnett. He took the Drager idea of joining the casing of the motor chamber and the face mask together, and brought it to the Sinnett mechanism. He made no change in that mechanism itself, except, as defendants' expert testified, of a low grade mechanical order. During the prolonged discussion the Examiner seems to have picked up another track — the question of whether applicant's claim was generic, and whether it constituted double patenting over applicant's earlier '626 Patent. The more he mentioned that patent, the more applicant retreated to the Sinnett structure, and in that form claim 9 was eventually allowed.^[9]

Essentially the only difference between claim 9 and Sinnett is that bringing the motor and mask together eliminated the connecting tubes, two of which were needed in Sinnett to obviate the rebreathing problem, and substituted a single short passageway.^[10] In the light of the prior art I do not consider this elimination invention, and believe if the Examiner had not started off on the wrong foot, and been kept there by applicant, he would have, correctly, adhered to his original conclusion. Applicant does throw in as a make-weight that eliminating the tubes makes the motor additionally sensitive to the pressure in the patient's lungs. This may be so as a theoretical matter. I do not find it of any practical utility.

There is no question but that a portable resuscitator is a very handy thing. Indeed, defendants' accused device is named the Handy, and is advertised as an "ingenious * * * great new contribution to the first aid field." In spite of this seller's talk, defendants have nothing that plaintiffs do not have, and had long before them. From the standpoint of infringement their device quite apparently infringes.^[11] However, I cannot find that making Sinnett to a degree smaller and portable constituted invention.

In case I should be mistaken as to this I will deal somewhat briefly with the question of reading claim 9 back into the '626 application. Under § 120 this may be done, where the application was filed before '626 issued, and recited that it was a continuation in part of the earlier application, if the invention was "disclosed," but not claimed, in that application. This matter was considered in my opinion on defendants' motion for summary judgment, 135 F. Supp. 268. I think some of the reasoning may have been oversimplified. The question of what constitutes disclosure within the purview of this statute has been infrequently considered. I doubt the complete applicability of the interference cases, relied on by defendants, denying any equivalency, for the reasons pointed out in some of the opinions. See, e. g., In re Key, 76 F.2d 398, 400, 22 C.C.P.A., Patents, 1098; cf. Smith v. Carter Carburetor Corp., 3 Cir., 130 F.2d 555, 559. It seems to me that whatever is equivalent should be regarded as part of the disclosure. The purpose of that doctrine is to reduce the otherwise inordinate length to which language might have to be carried. Of course, what the language implies in any particular case depends upon the total circumstances, but I think it too narrow to say that "disclosure" must necessarily be limited to a literal reading. Applicant's difficulties in the present case arose in part from the presence or absence of the economy feature. In some respects he seems to have been caught between the upper millstone that omission of an element and its function does not constitute invention, (which I have already suggested with relation to this particular feature), and the nether stone that in a combination patent the claims are presumed to include and require every matter specified. It is apparent that in principle the respirator device will function, not only as well, but in the same way, with or without the economy feature. I do not think one can arbitrarily say that a disclosure with the economy feature does not disclose one without it. Whether it claims it is something else. It may be that it was proper for applicant in '741 to seek to cover, as a continuation of the '626 application which did not claim it, the device in this alternate form. Nor do I think the answer of "double patenting" which defendants seek to make need be sound — it may be that applicant did exactly what was permitted. Cf. B. F. Sturtevant Co. v. Massachusetts Hair & Felt Co., 1 Cir., 122 F.2d 900, 913, rehearing denied, 1 Cir., 124 F.2d 95, certiorari denied 315 U.S. 823, 62 S. Ct. 917, 86 L. Ed. 219. It is no answer to say this would be "unfairly" extending the patent. There is no ethical test of fairness when one is dealing with an arbitrary structure of statutory law. Chapman v. Wintroath, 252 U.S. 126, 40 S. Ct. 234, 64 L. Ed. 491. However, I will leave fuller consideration of these points, if material, to a court above me, as they are unnecessary for my decision, and are, with one exception, all based upon matters of record not involving disputed testimony.

The one factual dispute arises if defendants are correct in their legal contention that the word "disclosed" in § 120 means literally, with no scope for equivalency. In '741 the tube by which the exhalation gas was to be drawn from the motor chamber to the suction side of the Venturi was described as opening into the "wall of the motor chamber." A witness for plaintiffs testified that the comparable tube in '626 enters the wall of the motor chamber. I find it does not. Between the end of that tube and the motor chamber proper is a valve chamber, which connects with the motor chamber proper during exhalation, and with the outside atmosphere during inhalation, providing the economy feature. At all times it opens into the tube. Plaintiffs contend that the tube opens into the walls of the motor chamber because the valve chamber is part of the motor chamber, or, at least, its walls are part of the walls of the motor chamber. I do not agree with defendants that this is logically impossible. This is an intellectual, not a structural problem. The valve chamber is shut off from the motor chamber proper half of the time. Pressure in the valve chamber cannot affect the diaphragm "motor" at all during one cycle, and does so only very indirectly during the other. When the valve chamber was added it was to serve a tube function, to admit air therein during inhalation, rather than a motor function. I find this to be determinative. Under any doctrine of equivalents the valve chamber is part of the tube, and hence it may be said that in effect the tube opens into the motor chamber, but as a matter of literal reading, if this be the test, it does not.

Epilogue

I will hear counsel as to costs, if any, before final disposition.

NOTES

[1] In particular I reject as unconvincing an argument based upon pressures made for the first time in plaintiffs' brief.

[2] Although defendants made much of it, I do not consider it important who did or did not help to get approval, or for what machine. But I note that defendants repeatedly intimate that all resuscitators had been disapproved by the AMA, and flatly state that resuscitators were approved in 1939. In fact, only one specific device was involved in each instance.

[3] The absence of a comma before "who" would suggest that the burden of proving lack of abandonment or concealment is on the defendants. Cf. Gorski v. New York Life Ins. Co., 315 Mass. 17, 51 N.E.2d 761. But cf. Nystrom v. Mancuso, 64 F.2d 698, 20 C.C.P.A.,Patents, 934. However, in the case at bar defendants proved the reverse.

[4] "E. & J. was content to continue with the piston type machine. Obviously it would have been poor business judgment to scrap all of the tools and inventory required for production of the piston machine and to rush onto the market with a Venturi-type device * * * when there was no competitive reason to do so."

[5] Neither form of the Goodner machine need have been so complicated had Goodner been willing to sacrifice the economy feature, but to do so would have doubled the gas consumption.

[6] During the trial something was said about vomiting by patients. No evidence was introduced as to frequency of such, and I assume in any event that this would occur in catastrophe work, for which plaintiffs, paradoxically, recommend the other machine.

[7] The first two claims of Sinnett clearly indicate a device in which the inhalation and exhalation circuits are totally independent. Plaintiffs do not allege infringement of these claims. Defendants contend that the remaining claims are to be similarly construed. I do not find it necessary to decide this point. These claims were added by applicant late in the day. It may be that the Examiner in allowing them did not fully appreciate their possible breadth.

[8] Actually there was more than one of these Drager devices. Patent No. 1,150,508, Drager, showed a device where the motor and valve was in effect built into the mask, with a line leading off to the gas supply. British Patent No. 146,862, Dragerwerk, showed a mask separated by a single tube from the motor. The drawing showed this to be a short tube, although its length was not mentioned in the specifications.

[9] Claim 9 was originally claim 11. Claim 11 was rejected (along with a great many others) as generic, and as lacking invention over Drager and Sinnett. Applicant was instructed to elect only one species. Claim 11 was changed and refiled as claim 37. Claim 37 was rejected as "not readable on the elected species." On the same date certain other claims were rejected as constituting double patenting over '626. It seems apparent that this was because it was not clear that the inhalation gas did not pass through the Venturi, as it did in '626. To anticipate that point, in case it was going to be raised, applicant corrected claim 37. He made a number of other changes. In his "Remarks" he stated that the revised claim "clearly distinguishes from the arrangement disclosed and claimed in the prior patent." (Italics supplied.) He also changed this claim to a "tube, one end of which * * is fixed in an opening in the wall of the motor chamber." The effect of this gives rise to much discussion.

[10] If the motor was separated any distance from the mask and only one connecting tube were used, all the resuscitator would do would be to draw the air from the patient into the tube, and on the next cycle force it right back again. This problem was adverted to by Drager, and is so obvious that it occurred to the Court at a tender age in connection with elementary submarine apparatus. Sinnett, and all others before him when the motor and mask were separate, used two tubes running to the mask and open to it at all times. The gas was forced down one tube and drawn back through the other. It is apparent that as one brings the mask and the motor closer together the rebreathing problem and the need for two tubes lessens, and eventually disappears.

[11] The question of infringement may involve a matter of file wrapper estoppel. This, in turn, involves some of the same questions raised by the issue of relating the patent back to the '626 application, discussed infra.