The parties in these two cases, involving interrelated factual situations, are, together with their associates and licensees, owners of two separate patents issued on tobacco harvesters. Their conflicting interests led to the institution of these two actions in which each patentee maintains that his patent is valid and infringed by the other patentee and his licensee.

The two cases were consolidated for trial after which the District Court held that the device of the Holliday patent No. 2,672,248 issued March 16, 1954, was a generic or pioneer invention entitled to a wide range of equivalents while the device of the Long patent No. 2,704,158 issued March 15, 1955, embodied certain patentable improvements. Since the machine actually manufactured and sold by each is substantially that disclosed in the Long patent, it was held that Long and his associates had infringed the valid Holliday patent and that Holliday and his licensee had infringed Long's valid improvement patent. It was the judgment of the District Court that each of the patentees should have injunctive relief and compensatory damages from the other.

The defendants in each case have appealed to this Court.

The bright leaf tobacco grown in the Carolinas and Virginia is a plant consisting of a single stalk, which at maturity may be from five to seven feet in height. At intervals, broad tobacco leaves, from six to twenty-four inches in length, grow directly from the stalk.

The leaves do not ripen uniformly. Those nearest the ground are the first to ripen while those at the top of the stalk are the last. It is necessary, therefore, to have successive harvestings of the same field. The lower, ripened leaves are selected and picked by hand, while the higher unripened leaves are left to ripen for later harvestings.

The commonly accepted method of harvesting and stacking such tobacco, prior to 1953, was entirely a hand operation. Pickers, or "primers," would walk through a field, selecting and picking the ripe leaves which they placed upon a cart or sled. When filled with leaves, this conveyance was taken out of the field where "handers" arranged the leaves in convenient bundles and passed the bundles to "loopers" who tied the bundles to a stick. Many such bundles were attached to each stick until a filled stick was placed in a rack from whence it was removed to the curing barn where it was again placed on racks.

Throughout the broad field of agriculture, the rising cost of farm labor, coupled with acreage restrictions, has made such manual operations more and more uneconomic and has resulted in a rapidly accelerating mechanization of agriculture.

Tobacco farmers were not unconscious of these pressures and Holliday and Long, apparently entirely independently, each conceived the idea of a mechanical tobacco harvester.

In 1952 Holliday constructed a crude high clearance machine having a platform above the height of mature tobacco plants and equipped with hand operated conveyor chains by which tobacco leaves might be passed from pickers on seats below the platform to "loopers" working on the platform. This machine as constituted in 1952 was not practical for harvesting, however, and its principal use during that year was for spraying and dusting.

Holliday continued his efforts to produce a machine which would be practical as a harvester and on July 3, 1953, he tested his machine which was then equipped with power operated conveyor chains.

Attached to the frame of the Holliday machine were vertical members. There was a sprocket at each end of each vertical member around each of which the conveyor chain made a 180° turn. Attached at intervals to the conveyor chain were fixed, spring-pressed clips into which tobacco leaves might be inserted by a picker below the platform and from which the leaves might be removed by loopers working above the platform. The pickers were carried in seats attached to the lower portion of the vertical members carrying the chains, the seats being adjustable in height so that throughout the harvesting season, the picker would be within convenient reach of the ripened leaves.

This was the machine of the Holliday Patent No. 2,672,248. It is shown in the margin in a diagrammatic drawing,1 and is fairly described in claim No. 7, the only claim of that patent involved in this controversy:

"A tobacco harvester comprising a frame, wheels supporting said frame, vertical members secured to said frame in spaced relationship to each other, vertically extending conveyor chains carried by said vertical members, drive means for actuating said chains, a platform secured on top of said frame, said chains extending upwardly above said platform from therebelow, said chains having spring pressed clips attached thereto for resiliently engaging and carrying tobacco leaves, a plurality of seats, and means adjustably suspending said seats from said vertical members in back of said chains."

The Holliday machine had certain practical deficiencies. The comparatively short vertical run of the conveyor chain above the platform did not permit loopers on the platform to keep pace with the pickers below. Leaves which were not removed on the rising run of the chain, being held in fixed clips, were inverted when they passed over the upper sprocket with resultant damage to the leaves.

Meanwhile, quite independently, Long had been working to produce a machine which would accomplish the same result sought by Holliday. Long was a manufacturer of farm implements, and he and his chief engineer, one Davis, worked concurrently upon two different approaches to the problem.

Long's approach, and the one ultimately adopted, was, as Holliday's, a high clearance, three wheeled, motorized machine, shown in a diagrammatic drawing in the margin.2 As in Holliday, loopers worked from the high level platform where they removed from the conveyor, bundled and stacked tobacco leaves which had been picked by pickers who were carried in seats, adjustable in height, below the platform. Each conveyor chain, unlike Holliday, however, ran over seven sprockets, affixed to a variety of members. These sprockets carried each chain in front of each of two pickers below the platform and provided it with a substantially horizontal run above the platform in the work space of the loopers. Attached to the chain were spring-pressed clips, pivotally mounted so that the tobacco leaves would hang downward at every point in the run of the conveyor. The adjustable seats for the pickers were affixed to members of the frame having no direct connection with other members carrying the sprockets of the conveyor system.

An essential element of each claim of the Long patent is the combination of a conveyor chain having "a substantially horizontal run of substantial extent" above the platform and pivotally mounted clips to carry the leaves of tobacco.

Long's machine was first tested in the field on July 13, 1953, ten days after Holliday's first test of his machine.

After a public demonstration of an improved version of the Long machine on Labor Day, 1953, and a showing of three of the machines at the North Carolina State Fair later that autumn, Long undertook to promote and manufacture his harvester. This was done at substantial expense, but it proved to be a commercial success. His company sold 1,429 of the machines in 1954 and 2,120 in 1955.

Holliday learned of the Long machine prior to its public demonstration in September 1953. He had talked with Davis, Long's chief engineer, and had obtained from Davis, pictures of the Long machine. Thereafter, on October 9, 1953, Holliday filed his patent application, but he achieved no other significant step until March 5, 1954, when he entered into a licensing agreement with Harrington Manufacturing Company, Inc. Holliday's patent was issued on March 16, 1954.

Instead of following the drawings and specifications of the Holliday patent, however, Harrington, with Holliday's assistance, proceeded to manufacture machines which, substantively, were literal copies of Long's machine. Subsequently, Harrington employed Long's former chief engineer, Davis, in connection with that and other work.

In 1954 Harrington manufactured approximately 60 harvesters and in 1955 approximately 1,050, all being substantially identical to Long's machine.

The parties, in argument in this Court, do not seriously question that each of the patented devices involved some novelty amounting to invention. Long's principal contention is that his machine does not infringe the Holliday patent while Holliday contends that the Long patent is invalid because of claimed public use of the Long machine more than one year prior to the date of Long's patent application.

Claim 7 of the Holliday patent provides for vertical members, secured to a frame, vertically extending conveyor chains carried by the vertical members and with means for adjustably suspending seats from the vertical members.

We are asked to construe the word "vertical" as used in Holliday's claim as a "random adjective" implying no more than the functional purpose of the conveyor to elevate material from one level to another, and the term "vertical member" as synonymous with all of the side members of the Long machine which, in the aggregate, happen to be in a substantially vertical plane. But Holliday's contribution, in the light of the prior art, does not warrant such liberal construction of the language of his claim.

An inventor has a right to define specific terms as used in a patent, and, while the drawings and specifications may not enlarge the scope of the claims, they may be considered in determining the inventor's definition of the terms he employed. Black & Decker Mfg. Co. v. Baltimore Truck Tire Service Corporation, 4 Cir., 40 F.2d 910; Hutzler Bros. Co. v. Sales Affiliates, Inc., 4 Cir., 164 F.2d 260; City of Grafton, W. Va. v. Otis Elevator Co., 4 Cir., 166 F.2d 816.

Should there be any doubt from the language of claim 7 that Holliday used the terms "frame" and "vertical member" in contradistinction to each other, it is clear from the drawings and specifications that he did. He showed in his drawings and described in his specifications "vertical members," carrying the conveyor chains and adjustable seats which are not an integral part of the frame but are merely "welded or otherwise secured to the transverse members" of the frame.

The Long machine, as any box-like structure, has combinations of members through which, in a geometric sense, vertical planes may be passed. It does not have "vertical members" in the sense that term is employed in the Holliday patent. The literal language of claim 7 of the Holliday patent cannot be read directly on the Long machine.

An examination of the prior art convinces us that Holliday's machine was not a pioneer invention and that the restrictive language of the claim was employed advisedly in order to avoid the prior art.

As the District Court found, "high clearance machines, in the farm implement art" designed to pass over the tops of growing plants "have long been used for such purposes as detasseling corn, spraying and dusting crops." They were also known and used for harvesting.

In 1918 patent No. 1,275,781 was issued to Stark et al., for a high clearance cotton harvester. Seats were provided for the pickers and there was a conveyor to carry the cotton to the upper level of the machine. The seats were not adjustable in height and the conveyor was not truly vertical.

In 1950 patent No. 2,518,965 was issued to Whitley for a high clearance tobacco harvester. Seats for the pickers were adjustable in height. There was, however, no mechanical conveyor system. The tobacco leaves were passed by hand from the pickers below to the loopers above the platform.

In 1955 patent No. 2,702,134 was issued to Alphin, upon an application filed in 1952, for a tobacco harvester having a conveyor belt to carry the leaves from the pickers in seats near the ground to loopers riding on a high clearance platform. The seats for the pickers were not adjustable in height and the machine was subject to the practical disadvantage of being propelled by a conventional tractor rather than a single motorized wheel. Holliday had seen a picture of the Alphin machine and knew of it when he was working upon his own.

What Holliday did, therefore, was to combine the well-known high clearance self-propelled platform with adjustable seats for the pickers as shown in the Whitley patent and to substitute a vertical chain conveyor with fixed, spring-pressed clips for the inclined conveyor belts of the Alphin and Stark patents.

While Holliday's machine may not have been wanting in invention, it was not the first mechanical harvester with mechanical means for elevating material from a lower level to workers on a high clearance platform, nor even the first specifically designed for harvesting tobacco. It was a combination of known elements and it was subject to practical deficiencies, mentioned above, of such consequence that it was never commercially produced.

The limited contribution of the Holliday machines does not warrant the broad construction urged upon us. To adopt it would extend the language of the claim far beyond the apparent intention of the inventor and raise grave doubt of the validity of this claim of the patent in the light of the prior art. Against the background of earlier use of conveyor systems in mechanical, high clearance harvesters, Holliday could not have claimed the use of conveyors for the function of elevating tobacco leaves. Quite properly he restricted his claim to the particular conveyor system disclosed by him. The restricted language of his claim, describing his conveyor system, should not now be expanded to preclude the use of other conveyor systems of substantially different design and arrangement.

Assuming that the Holliday machine involved patentable invention, his patent is entitled to a reasonable range of equivalents. In determining the extent of the range of equivalents, however, we must consider the state of the prior art, the novelty and contribution of the claimed invention, the nature and extent of the differences between the patented and the accused devices, the scope of the claim of the patent and the limitations inherent in it and other surrounding circumstances. Graver Tank & Mfg. Co. v. Linde Air Products Co., 339 U.S. 605, 70 S. Ct. 854, 94 L. Ed. 1097. Protection of a patent is not to be denied because of inconsequential differences, nor should one be allowed to work a fraud upon a patent by the substitution of obvious equivalents which serve no useful purpose. Graver Tank & Mfg. Co. v. Linde Air Products Co., supra; Black & Decker Mfg. Co. v. Baltimore Truck Tire Service Corp., 4 Cir., 40 F.2d 910; Saco-Lowell Shops v. Reynolds, 4 Cir., 141 F.2d 587; Wine Ry. Appliance Co. v. Baltimore & O. R. Co., 4 Cir., 78 F.2d 312. On the other hand, restricted claims and narrow combination or paper patents should not be so expanded as to preempt an entire generic field. Harford Agr. & Breeders Ass'n v. Puett Electric Starting Gate Corp., 4 Cir., 182 F.2d 608; Wheeling Stamping Co. v. Standard Cap & Molding Co., 4 Cir., 155 F.2d 6; Scott & Williams, Inc., v. Whisnant, 4 Cir., 126 F.2d 19.

Construing Holliday's claim as we do, its range of equivalents, in the light of the prior art, is, at most, comparatively narrow.

The "vertical members" specified in Holliday's claim are not found in Long's machine. The frames of the two machines are clearly equivalent, but portions of the frame of Long's are not the equivalent of the conveyor system of the Holliday device or of Holliday's "vertical members," which are the heart of his conveyor system.

We conclude that the Long machine does not infringe claim 7 of the Holliday patent.

In reaching this conclusion, we are mindful that Holliday and Long were working independently and contemporaneously. The record does not disclose that Long even knew of Holliday's work, or that he pirated anything from Holliday. The record is clear that Long's machine was not subject to the deficiencies of the device of the Holliday patent, that it met with immediate commercial success and that it was promptly copied by Holliday and his licensee.

It is true, of course, that one who merely improves the device of a basic patent may not appropriate to himself all of the rights of the basic patentee, but we should be reluctant to so extend a narrow and specific claim as to protect the copyist from the effective innovator.

We agree with the District Court that the machine of Long's patent No. 2,704,158 involved patentable invention. As indicated earlier, it combined, in a different way, the same known elements used by Holliday. The differences, however, were significant. The provision for a "substantially horizontal run of substantial extent" of the conveyor above the platform and the use of pivotally mounted clips overcame the deficiencies of the Holliday machine and resulted in the first really practicable tobacco harvester. It was the first to achieve commercial success, and it was promptly copied by Holliday and his licensee, Harrington.

These are all indicia of invention in a field in which others were conscious of the problem and had sought, without success, to solve it.

While many of the machines manufactured by Harrington were literal copies of the Long machine, others incorporated a variant which was introduced to avoid the claims of the Long patent. This variant was the introduction of three additional sprockets during the course of the extended horizontal run of the conveyor above the platform. These sprockets converted the substantially horizontal run of the conveyor above the platform from a straight line to a series of zigzags. The terminal sprockets above the platform were at substantially the same elevation, however, and the overall course of the zigzags was horizontal.

We agree with the finding of the District Court that the zigzags added nothing to the performance or function of the machine and were ineffectual to avoid Long's claim to a "substantially horizontal run."

Holliday and Harrington, in Long's suit, while setting up the usual defenses of invalidity for lack of invention and non-infringement, rely principally upon the claim that Long's machine was in public use more than one year prior to the filing of his patent application.

Long's machine was first tested on July 13, 1953. It was apparent in this test that the pickers could not keep the loopers busy so he modified the conveyor system to add two more pickers. This modification was accomplished that evening and the machine was again tested in the field on July 14, 1953. Long testified the machine performed very satisfactorily on the 14th.

During the following weeks it was used in the harvesting of tobacco in fields owned by Long's father and once in the field of a neighbor.

Long characterized these uses as experimental or tests and there is nothing in the record to indicate that Long derived any financial gain from them or that they approached a start of commercial exploitation.

Holliday, whose licensee now questions Long's use, had testified that he had felt it necessary to test his machine throughout the harvesting season to be certain it would operate in the same manner when pulling leaves near the top of the stalk as it did early in the season when the ripened leaves were close to the ground.

Long testified that his "tests," during the summer of 1953, resulted in a number of changes. He changed the brakes from the front to the rear wheels to give the machine greater stability. He redesigned the auxiliary power system to avoid its sucking in tobacco leaves. Finally, he replaced his pivotally mounted weighted clips with L-shaped clips, similarly mounted, because the weights interfered with the leaves passing over them.

The record indicates that perhaps all of Long's use of his machine prior to his advertised "public demonstration" on September 7, 1953 was experimental. Certainly, we find no reason to disagree with the finding of the District Court that the use prior to July 28, 1953, the critical date under the provisions of the statute, 35 U.S.C. § 102(b), was anything other than experimental. Long's expression of satisfaction with the operation of the machine on July 14th is not inconsistent with a recognition of the necessity of further testing and experimentation, particularly when the record affirmatively shows that further testing and experimentation was conducted.

Ever since Elizabeth v. American Nicholson Pavement Co., 97 U.S. 126, 24 L. Ed. 1000, it has been clear that necessary testing and experimentation by the inventor for a not unreasonable period of time is not a public use within the meaning of Section 102(b) of the Act.

On abundant evidence it has been found here that the inventor's use during the critical period was experimental only and did not constitute a public use. See: Smith & Griggs Mfg. Co. v. Sprague, 123 U.S. 249, 8 S. Ct. 112, 31 L. Ed. 141; Cline Electric Mfg. Co. v. Kohler, 7 Cir., 27 F. 638; General Electric Co. v. Minneapolis-Honeywell Regulator Co., 2 Cir., 118 F.2d 278; Picard v. United Aircraft Co., 2 Cir., 128 F.2d 632; Metallizing Engineering Co. v. Kenyon Bearing & Auto Parts Co., 2 Cir., 153 F.2d 516; Merrill v. Builders Ornamental Iron Co., 10 Cir., 197 F.2d 16; Lyon v. Bausch & Lomb Optical Co., 2 Cir., 224 F.2d 530, affirming, D.C., 119 F. Supp. 42.

A much longer period than the two weeks that were involved could hardly be regarded as excessive for testing the function and utility of a rather complicated machine, newly conceived and newly constructed and for experimentation looking toward essential changes and adjustments.

The judgment of the District Court in case No. 7383 is reversed and the cause remanded for the entry of final judgment for the appellants, Long Manufacturing Co., et al.

The judgment of the District Court in case No. 7384 is affirmed and the cause remanded for the assessment of compensatory damages and further proceedings in accordance with that judgment.

Case No. 7383, reversed and remanded.

Case No. 7384, affirmed.