This is an appeal from the decision of the United States Patent and Trademark Office (the "PTO") Board of Appeals (the "board") sustaining the examiner's rejection of claims 1-3, 8, 10-13, 18, 23, 25, 26, 28 and 29 in reissue application serial No. 163,300, filed June 27, 1980. Original U.S. Patent No. 4,161,449, entitled "Powdered Carpet Composition" was issued on July 17, 1979, from an application filed September 2, 1977. All of the claims at bar stand rejected as unpatentable under, inter alia, 35 U.S.C. § 102(b).1 We affirm the rejections under section 102(b).

Appellants' invention relates to a vacuumable, powdered carpet composition which is used as a carpet and room deodorizer and a method for treating carpets by using the powdered carpet composition. The patented product, CARPET FRESH, after application on the carpet in powdered form and removal by vacuuming, imparts deodorizing, antistatic and antisoil characteristics to the carpet. Appellants' invention is best described by their broadest independent claim:

1. A powdered carpet-treating composition comprising a blend of from about 40.0-98.99%, by weight, of an inorganic salt carrier selected from the group consisting of sulfates, chlorides, carbonates, bicarbonates, borates, tripolyphosphates, nitrates and blends thereof, substantially all of the particles of said carrier being between 0.06-0.25 millimeters; from about 1.0-25.0%, by weight, of an agglomerating agent selected from the group consisting of starch, silica powders, grain flour, wood flours, talc, pumice, clays and calcium phosphate; from about 0.01-20.0%, by weight, of a volatile odorous agent and an effective amount, up to about 15%, by weight, of an antistatic agent, said composition being adapted for application onto a carpet and for removal by vacuuming to impart a deodorizing effect.

All of the remaining claims at bar recite the ingredients of claim 1. Claim 26 additionally recites a dedusting agent, described as "an alkyl phthalate dedusting agent, the combined concentration of phthalate and said odorous agent ranging up to about 5.0% by weight." The method claims at bar, claims 8, 10, 18 and 25, recite the application of the claimed composition onto a carpet in powdered form and the removal of the composition, such as by vacuuming, in order to impart deodorizing, antistatic and antisoil characteristics to the carpet.

Appellants applied for reissue in connection with a patent infringement suit pending in a United States district court.2 That action was between appellants' assignee, Airwick Industries, Inc. ("Airwick"), and the intervenor of this case, Sterling Drug, Inc., which appeared as a protestor3 before the PTO throughout the reissue prosecution. The district court issued a stay pending the reissue decision.

In August, 1976, more than one year before appellants' filing date of September 2, 1977, Airwick conducted a consumer test in St. Louis, involving 76 consumers. During the first stage of the St. Louis test carried out on August 9 and 10, 86 consumers viewed a video tape presentation of the product concept. Similar tests, consisting of only video tape presentations, had previously been used in Philadelphia and Houston. After viewing the video presentation, the consumers were questioned about the pricing of the product, the believability of the claims made for the product, and their purchase intent.

In an internal memorandum dated August 19, 1976, James A. Smith, director of research and development at Airwick and one of the co-inventors, wrote to Wesley M. Buckner, the vice president for marketing at Airwick:

Results [of the first stage test] were good, but not as good as noted in the previous two cities. The respondents [consumers] overall reactions were rather dull and lacked the highly enthusiastic research seen earlier. Outlined below is the purchase intent scores.

Purchase intent           # of Respondents    %
------------------------  ----------------  -----
Definitely would buy             21          24.4
Probably would buy               37          43.0
Might or might not buy           19          22.1
Probably would not buy            8           9.3
Definitely would not buy          1           1.2
                          ----------------  -----
                                 86         100.0

Again, suggested pricing was reviewed with the respondents and there apparently is no problem with $1.89 for 14 oz. It did not affect the purchase intent trial of the women [consumers] questioned.

Smith concluded this memorandum by stating that "[t]he results will guide our future product development activities especially in fragrance type and level as well as product form."

In the second stage of the test, 40 of the consumers were then given a granular version of the composition and 36 given a powdered version. The composition had the following ingredients:

Ingredients                                                  Percent
------------------------------------------------  -----------------------------
Sodium sulfate                                                52.2
Sodium bicarbonate                                            25.0
Aluminum Oxide (Dispal)4                                 10.0
4. "Dispal" is a trademark for a product consisting essentially of alumina
   and was a known carpet antistatic agent.
Starch                                                        10.0
Perfume                                                        2.5

This composition is identical to that of formulation 3 of Example III in appellants' original patent and in the instant reissue application. In an internal memorandum dated August 6, 1976, Smith observed that the powdery product had, inter alia, the following characteristics:

[P]roduct should have better deodorization, anti-static and anti-soil properties.

Easier to vacuum up (even on shag carpeting--according to Jim McLaughlin).

Would be easier to manufacture, and less costly to manufacture--it's a complete dry blend. [Emphasis in original.]

Moreover, Smith stated, with regard to the upcoming St. Louis test, that "[h]opefully, we will have some guidance as to product form--large or small particle size."

These 76 consumers were allowed to use the products in their own homes for about two weeks and were instructed to return any unused portion. They did not enter into any agreement of confidentiality with Airwick and were allowed to use the products in their own homes without legal restriction.

On August 29 and 30, 68 of the consumers (33 of whom received the powdery product) were personally interviewed by the appellants regarding their experience with the product. The record, however, does not indicate whether the appellants attempted to contact the remaining eight consumers or to retrieve the product. Smith detailed the results of the St. Louis test in his memorandum dated September 20, 1976. Smith pointed out:

There was a difference in the ease of vacuuming up the two product forms. * * * [T]he more powdery one, vacuumed up the best. * * * [T]the granular version, was definitely more difficult to vacuum up. The product bounced under furniture or up against the baseboard resulting in greater effort to vacuum up. Neither of the products got trapped in carpeting at all.

* * *

* * *

The overwhelming majority of participants stated the product does deodorize. Some commented upon specific applications to rid the carpeting and room of pet odors especially. In a few instances, the odor was so overpowering that no opinion could be formed. Again, this will be further evaluated in any revised formulas.

* * *

* * *

Although the product usage period was relatively short, no mechanical problems with any of the vacuum cleaners used was evident. There was no mention of any bags breaking or machines clogging due to the product, either.

* * *

* * *

[W]e are pursuing alternate fragrances to remove the sweet note of the current fragrance * * *.

* * *

* * *

[W]e will pursue only the powdery form in our final product development efforts because of favorable product testing results and anticipated lower manufacturing costs.

After the St. Louis consumer test, work continued at Airwick in experimenting with different formulations. For instance, Example I5 of appellants' original patent and the instant reissue application, which is the CARPET FRESH formulation as first made, was first recorded in an Airwick worker's notebook on October 5, 1976. Moreover, testing for the antistatic and antisoil properties of the CARPET FRESH product was not planned until early October, 1976.

In his affidavit dated December 22, 1980, James H. McLaughlin, one of the co-inventors, stated:

Airwick * * * was seeking to technically develop the product while testing consumer reaction to the concept and to various prototype forms. In preparation for this Affidavit, I previewed the film of the video presentation given in Philadelphia and Houston in the summer of 1976. The video shows white beads being sprinkled onto a carpet and bouncing about on the carpet surface. This was not a "commercial" nor a test of any "product", as urged by Sterling. It was clearly a rough mock-up of the concept to test consumer reaction and, in fact, rapidly solidified in the container. * * * The remaining consumer test in the summer of 1976 in St. Louis, is shown in the test report to be a more advanced concept test with the experimental formulations. One was a powder, the other was a granular material. This test was during a period of intensive experimental work in the laboratory to define the metes and bounds of the Carpet Fresh product. I did not consider these tests to be a commercial exploitation but an integral part of our development and experimental work.

The examiner rejected all the claims at bar under, inter alia, section 102(b) on the ground that the December, 1975 composition6 and the RD-144 compositions7 were in public use or on sale more than one year prior to appellants' filing date. In addition, the examiner stated that the evidence

establishes that McLaughlin considered the December 1975 composition to have been sold to Airwick in accordance with their August 18, 1976 agreement.8 * * * It appears to the examiner that the inescapable conclusion is that the "product" sold to Airwick on August 18, 1976 was RD-144 and/or the December 1975 composition. The experimental use exception argument is not persuasive. In order for this situation to be within the experimental use exception the amount of money involved would have had to have been incidental to the experimental use. However, the August 18, 1976 letter shows that a considerable amount of money was involved and that the primary objective of CNPD was profit, not experimental evaluation of RD-144 or the December 1975 composition. [Emphasis supplied.]

The board, however, reversed the examiner's section 102(b) rejection because (1) the December, 1975 and RD-144 compositions were not sold to anyone more than one year prior to the filing date and (2) the August 18, 1976 contract between CNPD and Airwick did not constitute commercial exploitation per se.

Following protestor's request and proceeding under 37 C.F.R. § 1.196(b),9 the board, however, rejected all the claims at bar under section 102(b) on the ground that compositions within the scope of the claims were in public use in this country more than one year prior to the filing date of the original application. The board stated:

We believe that the factual situation in the appeal before us is well nigh on all fours with the factual situation before the courts in Omark Industries, Inc. v. Carlton Co. et al., 652 F.2d 783, 212 U.S.P.Q. (BNA) 413 (9th Cir.1980). * * * The product furnished the public in St. Louis was formulation 3 of Example III of Appellants' patent * * *. This product meets in haec verba each of Appellants' claims 1 through 5, 11 through 15, 23 and 28 through 30. The public thereupon employed it in the precise method of Appellants' claims 8 through 10, 18, 19 and 25. Bearing in mind the Durrant et al.¹⁰ and Moncrieff11 prior art the remaining claims in the application, requiring the presence of an alkyl phthalate dedusting agent, are also rendered unpatentable under 35 USC 102(b) coupled with 103. Applying the tests of the leading case of In re Yarn Processing Patent Validity Litigation, 498 F.2d 271, 183 U.S.P.Q. (BNA) 65 (5th Cir.1974), cert. denied 419 U.S. 1057 [95 S. Ct. 640, 42 L. Ed. 2d 654], 184 U.S.P.Q. (BNA) 65 (1974), we conclude that Appellants' St. Louis market testing was motivated primarily by the desire for competitive exploitation, not for experimentation. See also * * * Drackett Co. v. Davis, 176 U.S.P.Q. (BNA) 344 (E.D.Mich., S.Div.1972). [Emphasis in original.]

In addition, the board rejected all the claims at bar under (1) 35 U.S.C. § 102(f) on the ground that compositions within the scope of the claims were not the invention of Smith and McLaughlin, and (2) 35 U.S.C. § 102(g) on the ground that compositions within the scope of the claims were the invention of McLaughlin alone who did not abandon, suppress, or conceal the invention.

Further, the board rejected claims 1, 11-13, 23 and 26 as obvious under 35 U.S.C. § 103 in view of prior art references. Last, the board rejected claims 11-13, 18, 23 and 25 as unpatentable under 35 U.S.C. § 251 on the ground appellants are claiming subject matter which had been cancelled in prosecuting the application for the original patent in order to obtain allowance of the claims in that patent.

Appellants argue that their consumer test in St. Louis was not a public use. Rather, that test comes under the experimental use exception as enunciated by the Supreme Court in City of Elizabeth v. American Nicholson Pavement Co., 97 U.S. 126, 24 L. Ed. 1000 (1878). To illustrate this proposition, appellants cite Johnson & Johnson v. Kendall Co., 215 F. Supp. 124, 137 U.S.P.Q. (BNA) 505 (N.D.Ill.1963), rev'd on other grounds, 327 F.2d 391, 140 U.S.P.Q. (BNA) 289 (7th Cir.), cert. denied, 377 U.S. 934, 84 S. Ct. 1337, 12 L. Ed. 2d 297 (1964). In that case the patentee's advertising agency conducted a survey in which 100 housewives were asked to determine the efficacy of bandages. The court stated:

The Court is not impressed, irrespective of defendant's failure to give notice thereof, with the contention that the 1949 consumer test was a public use. As plaintiff [patentee] states, it was a test by its advertising agency on 100 housewives, to determine the efficacy of the crinoline and paper-faced bandages. The test was done in the presence of interviewers and all bandages were used up in the course of the test, and so was a permissible test and not an invalidating public use within the meaning of the patent statute.

Id., 215 F.Supp. at 139, 137 U.S.P.Q. (BNA) at 516. The appellants also cite International Silver Co. v. Pomerantz, 271 F.2d 69, 123 U.S.P.Q. (BNA) 108 (2d Cir.1959) to buttress the Johnson & Johnson proposition. In International Silver, the patentee permitted the claimed invention of a design patent to be viewed by 100 customers during a survey. The invention remained in the possession of the inventor's agents throughout the survey.

Appellants argue that they intended the testing activities to be an integral part of their research and development program. Their intent was not to exploit the invention and gain competitive advantage over others. They contend that the test was to obtain a fair and unbiased opinion of the product's operation and usefulness, especially in light of the fact that two versions of the product were presented to consumers which indicated that appellants' product was still in the experimental stage. Appellants contend that the fact that (1) the St. Louis test included only an evaluation of the fragrance level and the vacuumability of the product and (2) further formulation work continued after the test indicates that their product was still in the experimental stage. In support of their arguments, the appellants cite Poole v. Mossinghoff, 214 U.S.P.Q. (BNA) 506 (D.D.C.1982) for the proposition that their product was not commercially practical and therefore falls within the experimental use exception. In Poole, the inventor demonstrated a breadboard prototype to potential consumers. The demonstration was conducted under confidentiality and restriction, and the prototype was set up behind partitions. That court (Markey, C.J., sitting by designation) stated:

There is a difference between the requirements of a reduction to practice and industry requirements for a commercially practical device. There may be an experimental use following reduction to practice as long as the experiments are, as they were here, part of an attempt to further refine the device.

Id., 214 USPQ at 510.

Moreover, appellants argue that the board's reliance on Omark Industries, Inc. v. Carlton Co., 652 F.2d 783, 212 U.S.P.Q. (BNA) 413 (9th Cir.1980) was misplaced and clearly in error. In Omark, the patentee conducted two different tests regarding a brush cutting chain, one test by the engineering department and a second test by the sales department. As to the former, the chains were provided to workers at a dam site with Omark's engineers visiting the site to determine performance characteristics. The performance of the chains was recorded and the worn-out chains were taken back to the laboratory for analysis. This activity, which took place in 1957, was not considered by the court to be a public use. The second test, which took place in 1958, was conducted by the sales department to obtain customer reaction to the chains. Sample chains were given to distributors who in turn gave them to customers. The follow-up was informal--generally third-hand reports from customers to the distributors who relayed the information back to the sales department. This activity was held to be a public use. The appellants in the instant case argue that their St. Louis activities differ from the activities in Omark because their test was supervised, the information derived therefrom was reported to the Airwick scientists, and the test was conducted by the market research group.

Last, appellants argue that the board's reliance on Drackett Co. v. Davis, 176 U.S.P.Q. (BNA) 344 (E.D.Mich.1972) was similarly misplaced. In Drackett, an earlier inventor (i.e., not the patentee) permitted unrestricted, unmonitored use of his sponge mops by 24 consumers in a "market survey." The inventor subsequently interviewed some of the consumers, approximately two weeks and then six months after the survey began.

We, however, disagree with appellants' arguments.

Since we find the section 102(b) issue to be dispositive in this case, we therefore focus our analysis on this issue and do not reach the other issues.

* "Public use" of a claimed invention under section 102(b) has been defined as any use of that invention by a person other than the inventor who is under no limitation, restriction or obligation of secrecy to the inventor. Egbert v. Lippmann, 104 U.S. 333, 336, 26 L. Ed. 755 (1881). See also In re Blaisdell, 44 C.C.P.A. 846, 242 F.2d 779, 113 U.S.P.Q. (BNA) 289 (1957).12 Such use, however, has been held not to be a statutory bar to patentability if the use was primarily for bona fide experimental purposes. The Supreme Court stated in City of Elizabeth, supra, 97 U.S. at 134:

The use of an invention by the inventor himself, or of any other person under his direction, by way of experiment, and in order to bring the invention to perfection, has never been regarded as such a [public] use.

The experiment to improve and perfect the invention must be the real purpose in such public use and not merely incidental and subsidiary. Smith & Griggs Manufacturing Co. v. Sprague, 123 U.S. 249, 266, 8 S. Ct. 122, 130, 31 L. Ed. 141 (1887). See also Ushakoff v. United States, 164 Ct. Cl. 455, 327 F.2d 669, 140 U.S.P.Q. (BNA) 341 (1964).

The experimental use exception, however, does not include market testing where the inventor is attempting to gauge consumer demand for his claimed invention. The purpose of such activities is commercial exploitation and not experimentation. See Smith & Davis Manufacturing Co. v. Mellon, 58 F. 705, 707 (8th Cir.1893). See also Cataphote Corp. v. DeSoto Chemical Coatings, Inc., 235 F. Supp. 936, 143 U.S.P.Q. (BNA) 229 (N.D.Cal.1964), 356 F.2d 24, 148 U.S.P.Q. (BNA) 527, modified on other grounds, 358 F.2d 732, 149 U.S.P.Q. (BNA) 158 (9th Cir.), cert. denied, 385 U.S. 832, 87 S. Ct. 71, 17 L. Ed. 2d 67 (1966).

Where, as here, the inventor made the allegedly public use, he has the burden of going forward with convincing evidence that the public use activities fall within the experimental use exception. Smith & Griggs, supra, 123 U.S. at 264, 8 S. Ct. at 129. See also Blaisdell, supra, 242 F.2d at 784, 113 U.S.P.Q. (BNA) at 292-93. Cf. Strong v. General Electric Co., 434 F.2d 1042, 1043-44, 168 U.S.P.Q. (BNA) 8, 9 (5th Cir.1970), cert. denied, 403 U.S. 906, 91 S. Ct. 2207, 29 L. Ed. 2d 681 (1971).

Although the appellants argue that their St. Louis activities were used to obtain technical data for further technical development of the claimed invention, the evidence indicates that the activities did not fall within the experimental use exception. In determining the purpose of the alleged experimental use, objective evidence indicating a purpose for such testing and experiment is generally preferred. An inventor's subjective intent is generally of minimal value. See Robbins Co. v. Lawrence Manufacturing Co., 482 F.2d 426, 431, 178 U.S.P.Q. (BNA) 577, 581 (9th Cir.1973). Cf. In re Theis, 610 F.2d 786, 792, 204 U.S.P.Q. (BNA) 188, 193 (C.C.P.A.1979).

Objective evidence may include, inter alia, whether the inventor inspected the invention regularly, see Cataphote, supra; whether the inventor retained control over the invention, see Magnetics, Inc. v. Arnold Engineering Co., 438 F.2d 72, 168 U.S.P.Q. (BNA) 392 (7th Cir.1971); and whether the commercial exploitation was merely incidental to the primary purpose of experimentation, see Theis, supra.

Contrary to appellants' contention that the St. Louis test was needed to obtain scientific data on their invention's operation and usefulness, such data could have been easily obtained in their own facilities. The operability and other properties of the claimed invention could have been verified without the assistance of "typical housewives" (consumers). Instead, there was a more dominant purpose behind the St. Louis test, viz. to determine whether potential consumers would buy the product and how much they would pay for it--commercial exploitation. What the appellants wanted was information in order to gear their product to maximize potential sales before they would place their commercial product on the market. Appellants wished to obtain marketing data such as the purchase intent of the consumers. The St. Louis test was an experiment geared toward marketing and only incidentally toward technological improvement. See Theis, supra.

Contrary to appellants' contention, the use of two versions of the product does not in itself indicate experimentation. It was only a test to determine which version the typical consumer would prefer. Once that determination had been made, the appellants intended to gear the product to the form most acceptable to consumers. For example, Smith commented in his August 6, 1976 memorandum that the appellants hoped the test would provide some guidance as to the product form--large or small particle size. This evidence suggests that the test was a determination as to which particle size was more marketable and not a determination as to which particle size was technically superior.

Appellants' contention that the St. Louis test was an experiment because only the fragrance and vacuumability aspects of their invention were evaluated is inapposite. Since fragrance strength is an unclaimed element of the invention, it is therefore not a part of the invention. It is settled law that the experimental use exception is not applicable to experiments performed with respect to unclaimed features of an invention. In Minnesota Mining and Manufacturing Co. v. Kent Industries, Inc., 409 F.2d 99, 101, 161 U.S.P.Q. (BNA) 321, 322-23 (6th Cir.1969), the court stated:

The claims of the patent and not the specifications measure the invention * * *, and here the Cumar ingredient, although beneficial, was not included in the patent claims. The further experiments after May 5, 1942, were necessary to attain the added shelf-life, but the tests were for a non-patented ingredient or feature and therefore it did not prevent the statute from operating to invalidate the patent. * * * In Atlas v. Eastern Airlines, Inc., 311 F.2d at 160, [136 USPQ at 7,] supra, the Court stated: "[T]he rule is that an invention is deemed functional for public use purposes when it can perform the general function for which it has been developed, even though the device might later be refined." [Citations omitted.]

See also Theis, supra, 610 F.2d at 793, 204 U.S.P.Q. (BNA) at 193-94; Gould Inc. v. United States, 217 Ct. Cl. 167, 579 F.2d 571, 198 U.S.P.Q. (BNA) 156 (1978). There is no evidence that the test was related to the claimed aspects of the invention such as the particular carrier or its particle size, the particular agglomerating agent, the volatile odorous agent broadly, the antistatic agent broadly, or the proportions of the ingredients. Appellants' determination of consumer-acceptable fragrance strengths was a marketability experiment and not a technical experiment.

In addition, the result of the St. Louis test, as evidenced by Smith's September 20, 1976 memorandum, indicated that there was no problem in vacuuming with either form and that the powder version was more favorably received by the consumer. Such data could have been obtained by testing in appellants' facilities.

As to appellants' contention that further formulation work after the St. Louis test indicates that their product was still in the experimental stage, we again disagree. Although the appellants stated that antistatic and antisoil tests were not planned until October, 1976, the record indicates that these tests were not planned with technological reasons in mind. For example, Smith had already stated in his August 6, 1976 memorandum that the powdery product should have better antistatic and antisoil properties than the granular type. Since the appellants already knew the technical efficacy of these properties, tests directed toward these properties were unnecessary. These tests were planned primarily for marketing purposes, similar to the purposes of the St. Louis tests for fragrance level and vacuumability.

Moreover, the procedures used by the appellants suggest that the test was designed primarily to determine how well the product would sell, not to isolate systematically technical problems which remained in the product. Omark, supra, 652 F.2d at 787. Appellants did not control the actual testing of the composition. For example, the testing of the composition in the instant case was not conducted in the presence of appellants. Nor were restrictions placed on the consumers as to the use of the product. See Egbert, supra. Contrary to appellants' contention, this lack of restriction is similar to the situation in Omark where the cutting chains were used by consumers without restriction and in Drackett where the sponge mops were used without restriction. The instant case differs from the situation in Johnson & Johnson where the bandages were tested in front of patentee's interviewers and from International Silver where inventor's interviewers retained control over the invention at all times. Moreover, three of the 36 consumers who received the powder version were not interviewed and the product was not retrieved.

Appellants also contend that Omark is distinguishable because appellants' testing was conducted by the market research group and not a sales department. We disagree. In either instance, the investigating entity was testing the marketability of a product and not the operability or technical aspects of the product. In addition, as contrasted to the confidential evaluation of the prototype by consumers in Poole, the consumers in the instant case were not bound by any confidentiality agreement. See Egbert, supra. Moreover, the consumer interview and the subsequent evaluation in the instant case dealt with marketing aspects such as which version the consumers would purchase, the price range, etc. It was not an evaluation of the technological aspects of the product. See Omark, supra.

The totality of the evidence indicates that the St. Louis test was an experiment to determine whether consumers would buy a product.13 Such market testing does not fall within the experimental use exception. As stated in Cataphote, supra, 356 F.2d at 27, 148 U.S.P.Q. (BNA) at 529, citing Smith & Davis, supra:

[T]he entire course of appellant's conduct * * * was to promote product sales, and any testing that occurred had as its primary objective an ascertainment of the product's marketability. This activity * * * is typical of "a trader's, and not an inventor's, experiment," and "such a use does not carve an exception out of the statute." [Emphasis in original.]

Since the appellants have failed to prove by clear and convincing evidence that the St. Louis test falls within the experimental use exception, we hold the claims at bar are unpatentable and the decision of the board is affirmed.

AFFIRMED.

NICHOLS, Circuit Judge, with whom COWEN, Senior Circuit Judge, joins, dissenting.

Respectfully, I dissent.

In the majority opinion the whole case is made to turn on what appears to me to be the weakest ground of the several on which the decision below stands, public use of the invention more than one year before the date of the patent application. This "public use" boils down to tests performed in St. Louis August 1976. Some 76 female heads of households were given, without charge, granular and powdered versions of the invention to try in their homes. They were instructed how to use the samples. They were not required to sign any agreement of confidentiality and could use as much as they wanted without legal restriction, being asked to return any unused portion. After about two weeks they were interviewed. Questions largely focused on whether or not they would buy, but inquiry also included ease or difficulty of taking up the product in the home vacuum cleaner, with varying types of carpets, efficiency of the product in deodorizing, particularly in the 37 percent of homes that included pets, and with respect to smoke odors, objectionable odor from the product itself (found to be a major problem), quantity of product used in relation to carpet area cleaned, mechanical problems with the vacuum cleaners (none reported), and problems dispensing the product from its container. Why females exclusively were selected for the test is not stated, but may be surmised.

The majority assumes that all this could have been duplicated in the laboratory, except the consumer reaction testing not considered an acceptable part of the experimentation required to complete "reduction to practice." The idea that the other tests could have been duplicated in the laboratory seems highly improbable and one would like to know the record support for it. It seems to be supposed that the laboratory technician could anticipate all the varieties of carpets used in St. Louis homes, and all their degrees of age and wear; all the varieties of vacuum cleaners, with their variegated workings and degree of wear and tear; and all the ways the householder could produce malfunctions by dealing with the product in an unforeseen manner, with all the brands of engineering ineptitude to be anticipated in the average home. He would also have to know how strong the odor of the product would have to be to be offensive to the "female head of household" (or the family cat), and how activity and uncleanliness of pets might generate problems for the product to solve. It would seem this supposititious laboratory technician would have to be quite a paragon, and the inventor might be much impressed with his talents and still wish to learn by actual experience in the home, some of the ways that Murphy's law might operate to wreck his hopes.

What the court is really saying is that the inventor of a product for use by amateur engineers in the home cannot test the product by amateur engineers in the home without the occurrence of a "public use" starting the year of § 102(b) to run. It appears to me such a rule is well calculated to thwart and nullify, in the consumer product category, the experimental use exception itself. Yet this exception is well established. E.g., In Re Yarn Processing Patent Validity Litigation, 498 F.2d 271, 285, 183 U.S.P.Q. (BNA) 65 (5th Cir.), cert. denied, 419 U.S. 1057, 95 S. Ct. 640, 42 L. Ed. 2d 654 (1974). The board did conclude that the test was "motivated primarily by the desire for competitive exploitation." All this seems to mean is that the developers desired to compete with others and make money, and this they doubtless would be the last to deny. In order to make money they needed to know, without costly failure and recall after marketing, whether the product would work under home conditions. The test was in no way a start towards commercial exploitation itself even as a pilot or test marketing project. The further contrary deduction that all the tests could be performed in the laboratory seems to be the court's, not the board's. Assuming as the court does that our function is only partly appellate, because we are not sitting in review of any credentialed fact finder entitled to the "clearly erroneous" standard of review, and that we make our own fact findings from the raw record, it seems to me we should make fact findings that are more plausible than these.

The court lays stress on the absence of effort during the test to keep the composition of the product secret. The results of the test inured to the inventors alone, so a would be copyist who secured one of the samples would not know how successful the test was, though he could know by "reverse engineering" how the sample was made up. In view of the court's freedom in fact findings, I find that such a ritual of accounting for the samples and confidentiality agreements would have annoyed the housewives and made their reactions calculated and unspontaneous, and possibly they would even have refused to participate. Evidently some risk of disclosure was taken because it was not seen as a substantial risk.

Obviously the test was performed in part to obtain marketing information: how many housewives would want the product and what price they would pay. Evidently the test did nothing to discover the size of the potential consuming public or its ratio to the entire public. When asked if they would buy, housewives had already tested the product and found it satisfactory, thus separating them as a discrete body from the mass of consumers. The test users, how collected or how representative of the entire public unknown, were all "female heads of households" age 18 to 60, who "had carpeting in living room, dining room, or den." The test discovered that this sample, representative of a public of unknown size, would buy, if shown by a test in their own home, that they could put the product to a useful purpose. Thus the marketing testing and product testing aspects were inseparable and each was useless without the other. In those circumstances, how the court is able to find as a fact that the market testing aspect was primary, is beyond me.

When Mr. Fulton invented the steamboat, he had to test whether persons of ordinary education could feed his boiler and keep it producing steam, but not explode it, before he could test how many persons would want to ride up the Hudson to Albany as passengers. By my perception, the test at St. Louis paralleled the first, not the second, and was therefore a typical experimental use.

Inventors will feel compelled to run "female heads of houses in the age range of 18-60 years" through the laboratory instead of running the product through their homes serving as laboratories. The cost of such tests will be greater and the results less dependable. This decision, therefore, is one of the many disasters inflicted every year upon that unhappy target, the American consumer. It is not required by the cases cited and serves no discernible policy, either of the court or of the Congress that enacted the statute.

It is unnecessary to say much about the cases cited. The one most often referred to and a CCPA case binding as a precedent on us, is In Re Theis, 610 F.2d 786, 204 U.S.P.Q. (BNA) 188 (C.C.P.A.1979). The product there was actually sold to customers more than one year before the patent application. Some of the invoices bore the notation "6 months evaluation" but the court viewed the evaluation as possibly for the benefit of the customer, so he could return the product if it did not meet his needs. Presumably he would keep it if satisfied. The difference of this from the case at bar is obvious. The board relied very heavily on and calls an "all fours" precedent Omark Industries, Inc. v. Carlton Co., 652 F.2d 783, 212 U.S.P.Q. (BNA) 413 (9th Cir.1980). There, however, the product, a chain saw, was thoroughly tested by the company's engineering department in the hands of selected typical users, and this was held to be a valid experimental use. Thereafter the sales department took over and arranged for further trials by users. These were held to be sales-oriented as no further awareness of design problems was sought or obtained. This court understandably does not rely on the board's alleged "all fours" precedent. It is obvious the Omark court was willing to allow, as the board was not, an opportunity to try the product out in the hands of typical users, and the case should have embarrassed the board.

None of these cases involve the peculiar problems incident to perfecting a product to be used by housewives in the home in connection with their own variegated household appliances. It is the refusal to recognize the existence of these problems here that divorces the court from reality and produces a decision that will be quite harmful in its effect.

If the court must willy-nilly deny Smith and McLaughlin the benefit of their invention, it could find in the record grounds that are at once more plausible and less pernicious to the public weal.