delivered the opinion of the court:
This is an appeal from the decision of the Board of Appeals of the United States Patent Office, affirming that of the Examiner of Interferences in awarding to appellees Waite and Martin priority of invention defined by the two counts involved.
The interference, when before the board, also included a third party, Paul R. Pine, whose assignee was the Harshaw Chemical Company of Cleveland, Ohio, but who, after receiving an adverse decision from the board, took no appeal to this court. The interference before us involves the application of Waite and Martin (whose assignee is the McGean Chemical Company, also of Cleveland), filed April 27, 1935, and an application of the junior party Wirshing and Clifton, filed October 28,1935, the application being assigned to the General Motors Corporation of Detroit, Michigan, which company in turn assigned to the Udylite Corporation, also of Detroit.
*1053The invention involves the addition of a sulphate of lauryl alcohol (as expressed in count 1), or a sulphuric ester of a wider range of alcohols (as set forth in count 2), to a conventional acid plating bath. Nickel plating could be done in the acid bath without the addition •of such an agent, but the resulting product frequently was defective in that it was pitted. The invention here is concerned solely with the problem of preventing pitting. At the time appellants entered the field, hydrogen peroxide was used as an anti-pitting agent in dull finish nickel baths and was found to be satisfactory in that type of bath. If a brighter finish of the dull nickel plate was required, the plated product was buffed to a state of brightness. Hydrogen peroxide was not satisfactory as an anti-pitter in a bright nickel bath since it had an oxidyzing effect on the organic materials used as brightening agents.
The counts follow:
1. An electro-depositiong bath comprising an aqueous acicl solution of a compound 'of nickel, in which solution the solvent is preponderantly water, and a minor amount of a sulphate of lauryl alcohol.
2. In the art of electroplating, the process which comprises electrodepositing nickel from a plating solution containing a compound thereof in association with an addition agent comprising a sulphuric ester of a normal primary aliphatic alcohol having .from eight to eighteen carbon atoms- in the molecule.
It is obvious that the counts are broad enough to include the use of the additional chemical in both dull and bright plating.
Wirshing and Clifton, being the junior party, were required to show by a preponderance of the evidence that they were the first inventors and in an effort to do so took the testimony of a number of witnesses. The record consists of 775 pages, the greater portion of which is testimony. The substance of the evidence of appellants will be stated here as briefly as circumstances would seem to require.
Appellants base their claim of priority upon their alleged proof of actual reduction to practice in January and February 1934. It is conceded that if they did not reduce to practice at the time alleged, they have shown no diligence at any time prior to the filing of their application as aforesaid. In view of our conclusion, the only question necessary for us to consider is whether or not the January and February 1934 activities of the junior party amounted to a complete reduction to practice.
Appellants both testified at some length. They stated that they began experimenting with Gardinol — an addition material responding to the counts manufactured by E. I. Du Pont de Nemours & Company, Inc. — to see if it acted as a good cleaner or brightener, and that they discovered that it had certain anti-pitting properties. In conformity with the usual practice of General Motors, accurate reports were made by each of the appellants as to the work done, the materials used, and *1054the character of the resultant product. These reports show that some of the products which appellants made in conformity with the counts were free from pits. Appellants tested the same only by looking at the articles and determining that there were no pits present and that there was no scaling or separation of the plate from the material plated. No bending or flexing test was given the product to determine its brittleness or ductility.
The record shows that these reports came under the constant observation of one William M. Phillips, who was “an executive engineer, consultant for General Motors on matters of plating and finish,” and whose office was adjacent the laboratory in which appellants made their experimentations; that Phillips was in constant touch with the experiments and during the experimentation paid strict attention to the reports and had frequent conferences with the appellants. Phillips testified that some of the resulting products made in accordance with the reports were satisfactory. There is no showing, however, that Phillips at any time observed the mixing of the bath or the actual plating of the materials. Obviously, he depended upon the accuracy of the reports as to the materials used and the steps taken.
C. F. Nixon, assistant chief engineer of the Ternstedt Manufacturing Division of the General Motors Corporation in Detroit, Michigan, testified for the junior party that he had charge from the technical angle of the plating work of his company. The record shows that a plating bath was furnished by the Ternstedt company to appellants while working for General Motors with the view of experimenting thereon in an attempt to cure defects in the bath which were resulting in badly pitted articles. The record shows that they used Garclinol in this bath without success, and Nixon so testified.He, however, gave it as his opinion that, one of the plates said to be produced by appellants in one of the said experiments (not with the said Ternstedt bath) in the early part of 1934 was satisfactory from a pitting standpoint.
We have examined the testimony of each of appellant’s witnesses with great care and we find that there is no testimony in the record (except as to the Ternstedt bath) which corroborates appellants as to the material used in the bath and the actual plating of the exhibits in controversy in that bath.
The fact that Phillips or others may have testified to the quality of the plated exhibits which the reports show were made in accordance with the counts is not sufficient corroboration under the circumstances. Collins v. Olsen, 26 C. C. P. A. (Patents) 1017, 102 F. (2d) 828. The reports must necessarily, under the circumstances at bar, be self-serving in character.
*1055In deciding whether or not an actual reduction to practice was made and the invention fully completed and its utility determined, it is often proper to look to the subsequent conduct of the interested parties. Evidently believing that Gardinol was no better anti-pitter in dull plating than hydrogen peroxide which they were then using, and knowing of no other use for it or market for its sale (bright plating not being in mind), nothing further was done after the failure encountered in connection with the Ternstedt hath. No activity took place from that time on until Phillips learned from those connected with the assignee of Waite and Martin that Gardinol was one of the ingredients used in the electroplating bath which appellees’ assignee had installed and put in operation at the Chevrolet Flint plant of General Motors. The facts of record showing the conduct of appellants subsequent to> their January and February 1934 efforts would seem to emphatically suggest that they had not completed their invention and that their work amounted only to an inconclusive and abandoned experiment.
The Examiner of Interferences in part said (reference numerals omitted) :
The picture presented as to Wirsliing and Clifton broadly appears to be that in January 1934, after they, at the request of the Du Pont Company, tried fruitlessly to find a use for Gardinol WA as a metal cleaner, they discovered in the laboratory that under certain conditions Gardinol WA eliminated or reduced pitting in an acid nickel plating bath; that in February 1934 they added Gardinol WA to a 550-gallon tank of badly pitted solution at the Ternstedt Company but were unable to eliminate the pitting; that thereafter they dropped activity on the invention until early October 1935 when upon learning from the McGean Company that Gardinol WA was one of the ingredients of McGean baths in use at the General Motors plant they proceeded to prepare and file their involved application on the basis of their work in January and February of 1934.
These circumstances are considered to set forth either concealment of the invention or abandoned experiment depending on whether the early 1934 experiments constituted a reduction to practice or merely conception of the invention. The fact that work was dropped by Wirshing and Clifton on Gardinol WA after February 1934 until the date of filing, October 28, 1935, is an indication that Gardinol WA was not considered a sufficiently satisfactory addition agent. Phillips’ testimony lends support to this viewpoint. On the basis of these facts the conclusion is deemed warranted that the experiments of January and February 1934 were not sufficient to establish reduction to practice but did establish conception of the invention by Wirshing and Clifton. * *
The Board of Appeals said:
The examiner, in considering the record as to Wirshing and Clifton, held the early tests in February of 1934, not a satisfactory reduction to practice. He held, however, that, independently of the 'holding as to such tests, the long delay in filing the application brings the facts under the doctrine of Mason 1. Hepburn, as Wirshing and Clifton concede being stirred into activity *1056by their opponents. At the time of the tests in January and February, 1934,. the party Wirshing and Clifton were attempting to cure pitting in a metal plating bath and in order to be sure of the effect in the test in February, 1934, they added other elements to the bath which would accentuate pitting. Furthermore, they made use of samples from commercial baths in which pitting took place to an excessive degree. As a result of some tests, however, the test solution was thrown away and the record is not very convincing that these tests would be considered a satisfactory reduction to practice. Nor can we excuse the delay of twenty-one months in filing the application because of the size of the General Motors Corporation and the fact that the corporation was not using bright nickel baths during 1934. Such bright nickel baths were only used in 1935.
Appellees, in this court, present that which appellants state is a contention not previously made. The contention is to the effect that, irrespective of any view that may be taken of the testimony of the inventors or of the alleged corroborating witnesses, the record fails to show that appellants completed the invention by making-proper tests and that a proper and necessary test which must be applied to nickel plating in order to determine its utility is to give it the test of bending. In this manner the ductility of the nickel plating will be ascertained which will determine the probability of cracking or peeling.
We would be reluctant to consider this question if presented here for the first time, but, in view of our conclusion that irrespective of this fact there is no corroboration which sustains a finding of reduction to practice during the critical period, it becomes unnecessary to pass upon the question.
Nor is it necessary here to consider the question of the applicability of the principle akin to the doctrine of equitable estoppel laid down in Mason v. Hepburn, 13 App. D. C. 86, because the principle is never applied where there is no reduction to practice.
Reverting to the question of the sufficiency of the corroborative proof, appellants urge that since Phillips testified that some of the plates produced were satisfactory, and since he not only kept “in very close touch with the Wirshing and Clifton experiments but he made almost daily inspections of the product and checked his own opinion of the condition of the product with Clifton’s laboratory notes,” and since the products are available and are in court, it should not be held' that the testimony of the inventors has not been corroborated.
As before stated, there is nothing to show that Phillips’ knowledge is not entirely confined to what the appellants said to him verbally or in their reports. Both of these circumstances might be regarded as evidence of conception, but certainly cannot be regarded as corroborative evidence of reduction to practice. In Collins v. Olsen, supra, the facts relating to the question of corroboration of reduction to practice *1057were quite similar to those at bar. Reports of the work which Olsen had done ar the Battle Creek, Michigan, laboratories were sent to New York for approval. The reported work done by Olsen was approved and the value of the reported invention was determined by Olsen’s superior. It was also shown that during a portion of the time Olsen’s immediate superior was in the laboratory with him. The record, however, failed to disclose a corroboration of what Olsen had actually done, but was a corroboration only of the satisfactory results obtained upon Olsen’s statement (self-serving of course) of how he had performed the test. We held there, citing many authorities, as we hold here, that while the evidence would be good for conception, it is self-serving on the question of reduction to practice. We think the holding in the Collins v. Olsen case points unmistakably to the correctness of our conclusion herein reached.
Appellants having failed to prove that they at any time reduced said invention of the counts to practice prior to their filing date, which was subsequent to the filing date of the appellees, and having conceded that there was no diligence after February 1934 until they filed their application, it follows that priority of the invention was properly awarded by the concurring decisions of the tribunals below to the appellees, and the decision of the Board of Appeals is affirmed.