594 F.2d 30
100 L.R.R.M. (BNA) 2871, 85 Lab.Cas. P 11,161
AMERICAN MANUFACTURING ASSOCIATES, INC., Petitioner,
v.
NATIONAL LABOR RELATIONS BOARD, Respondent.
No. 78-1087.
United States Court of Appeals,
Fourth Circuit.
Argued Dec. 7, 1978.
Decided March 8, 1979.
John Haworth, High Point, N.C. (Haworth, Riggs, Kuhn, Haworth & Miller, High Point, N.C., on brief), for petitioner.
Arnold B. Podgorsky, N.L.R.B., Washington, D. C. (John S. Irving, Gen. Counsel, John E. Higgins, Jr., Deputy Gen. Counsel, Carl L. Taylor, Associate Gen. Counsel, Elliott Moore, Deputy Associate Gen. Counsel, John H. Ferguson, N.L.R.B., Washington, D. C., on brief), for respondent.
Before WINTER, RUSSELL and WIDENER, Circuit Judges.
DONALD RUSSELL, Circuit Judge:
The employer American Manufacturing Associates, Inc., a North Carolina corporation, is a small manufacturer engaged in fabricating welt cord and edge roll for use in the furniture trade. Its only plant is located in High Point, North Carolina, where it employs less than fifteen employees, working on a three-shift basis. The personnel on its first shift is greater than that on the other two shifts. Because of the larger complement of employees on the first shift, its employees normally perform some operations on that shift not duplicated on either the second or third shift.
The employer has been in business since 1907. Until 1971 Thomas B. McCormick owned forty-eight percent of the stock in the corporation. In 1971 McCormick acquired all of the stock and became the sole owner.1 Though he owned the corporation, McCormick, because of his other business to which he gave full time, installed his son-in-law Elbert Bowman as the plant manager, in charge of the employer's production. Beginning in 1973, however, the business began to experience problems in both quantity of production and quality of product. The low production resulted in delays in fulfilling orders on time, thereby causing considerable dissatisfaction on the part of customers. Increasing complaints of defective product, also, were being received from customers. As a result, the business became unprofitable. Its losses increased from 1973 on, due to this low production and poor quality. To off-set these losses, McCormick personally was forced on a number of occasions to contribute additional funds to the corporation. Concerned about the unprofitability of the business, he began to visit the plant from time to time, to observe the conditions, and to discuss with Bowman ways to increase production and quality. In particular, he brought to Bowman's attention his observations of dawdling away from their places of work by some of the employees when he had visited the plant and impressed on him the importance of disciplining employees responsible for defective work. Bowman said that he had no method of identifying the employees guilty of defective workmanship and he evidenced no disposition to take any steps to compel greater attention by the employees to their work or to overcome the problem of low production. As a result of the continued deterioration in production and quality and the accumulation of losses, McCormick decided on June 14, 1976, that he had no choice but to remove Bowman as production manager and to replace him with Bodenheimer, a new employee.
At the time McCormick employed Bodenheimer, the two of them prepared a set of working rules. One of the purposes of these rules was to assure that the employees worked a 40-hour week and did not take excessive breaks during working hours. Accordingly, the rules provided definite commencement hours, the time and extent of morning and afternoon breaks, the time and extent of lunch breaks, and leaving time for each shift. In introducing Bodenheimer to the employees as their new production manager, with full and exclusive authority to hire and fire, McCormick reviewed these new rules with the employees and, after adverting to the low production and poor quality at the plant, warned that both production and quality had to improve.
When initially employed, Bodenheimer was not familiar with the business. However, he began actively to acquaint himself with the plant's operations and the duties of the employees. His earliest efforts were directed at increasing production so that delays in meeting delivery dates for orders might be avoided. Bodenheimer also took certain positive steps to improve quality and to secure increased production. As we have already observed, Bowman had expressed an inability to do anything about these problems because he professed an inability to identify the delinquent employees. Bodenheimer realized, however, that in some way he had to secure greater production and improved quality. This was particularly so, in view of McCormick's visit to the plant on August 10, with a number of defective products returned by customers. McCormick was very upset about the situation. He called all the employees together in the break-room. He "told us that this was going to be his last trip that he made over there as far as to talk about the quality of the product, that he would not be over there again to talk about the quality of the product." Recognizing the urgency for some correction, Bodenheimer devised two tests to do what Bowman had claimed he was unable to do, I. e., identify the employees responsible for poor quality of product and low production. He, also, began an active effort to increase production so as to eliminate the back-log of orders and the delays in delivery, which had become an increasing complaint by customers. To do this he sought to recruit new employees in order to increase the work force on the second and third shifts.
To identify the employees who might be shirking on production, Bodenheimer sought to compare the production of the employees on the first shift with the production of those doing like work on the third shift. This represented a comparison of production between the most experienced and the least experienced. There was one shortcoming to this method: the third shift did not cover as many types of jobs as the first shift and thus the comparison had to be limited to those employees performing similar jobs on the two shifts. To meet the problems of quality control, on the other hand, Bodenheimer began to tag the product as it was completed so as to identify the employee responsible for the fabrication of the particular product. This enabled Bodenheimer to identify the employees responsible for a defective product when a customer returned such product. Both of these methods of quality and production analysis were inaugurated in August, 1976. By the end of the month, Bodenheimer had determined from the comparisons that, on the jobs compared, the new employees were substantially out-producing the older more experienced employees on the first shift. An analysis of the defective products returned had enabled him, also, to identify the employees responsible.
Bodenheimer reviewed with McCormick in August the results of his tests on production and quality. After considering these results, McCormick said that he "want(ed) to check further. I want to be sure of what is what." The next report showed the same results and in early September, McCormick instructed Bodenheimer to discharge Debbie Ellis, Lillie Marie Scearce, Linda Tucker, Barbara Gardner and Mary Hughes.2 Bodenheimer followed these instructions and discharged the indicated employees, giving to each a slip stating that they were discharged for "work unsatisfactory with regard to production and quality."
Some six weeks later, four of the discharged employees, Ellis, Scearce, Tucker and Gardner, filed charges of discrimination under the National Labor Relations Act, asserting that they were discharged because of their participation in protected activity in order to secure a wage increase. A complaint charging a violation of the Act was filed against the employer as a result of these charges. A trial on the complaint was held before an administrative law judge, who found that the discharge of the employees was based on the employees' poor work records, and not because of any protected activity under the Act. The Board, by a divided vote, and without disturbing expressly the administrative law judge's finding that the discharged employees were unsatisfactory employees, reversed the administrative law judge's finding that the protected activity of the employees had not been the motivating reason for their discharges. The employer has petitioned for review of the decision and order of the National Labor Relations Board3 and the Board has filed a cross-application for enforcement of its order. We deny enforcement.
The motivating reason for the discharges, in the opinion of the Board, was that the employees discharged had engaged some month or more before in a strike or work-stoppage in order to coerce a wage increase. Assuming for the moment that there was such a work-stoppage or sit-down, the first problem for the Board in establishing a violation of the Act and it is a crucial one was to establish that the employer knew of the work-stoppage or sit-down, which the Board found had motivated the discharges.4 Manifestly, if the individual responsible for the discharges was ignorant of the strike or sit-down, such strike could not have been a motivating factor. There is no dispute that the decision to discharge the employees was solely that of McCormick. That was the testimony of the employer; that was the express finding of the Board itself. It follows that the initial issue in this case then is whether McCormick knew of the work-stoppage or sit-down before he directed the discharges.
The burden of establishing that knowledge rested squarely on the Board. Stone & Webster Engineering Corp. v. N. L. R. B. (1st Cir. 1976) 536 F.2d 461, 464. It could have satisfied this burden, it is true, by either direct or circumstantial evidence, N. L. R. B. v. Columbia University (2d Cir. 1976) 541 F.2d 922, 929. But, as we said in an early case, "the evidence must be of circumstances which do more than give rise to a mere suspicion. They must be of such a character that they can reasonably be accepted as establishing as a fact the matter which is in issue." National Labor Rel. Bd. v. Shen-Valley Meat Packers (4th Cir. 1954) 211 F.2d 289, 293. This is but a reiteration of what the Supreme Court itself said in Universal Camera Corp. v. Labor Bd. (1951) 340 U.S. 474, 484, 71 S.Ct. 456, 462, 95 L.Ed. 456, that a finding of knowledge of union activity may not rest on "suspicion, surmise, implications, or plainly incredible evidence."
The record evidences no direct testimony that McCormick knew of the strike or work-stoppage.5 McCormick himself denied such knowledge; Bodenheimer testified that he never told or "notified" McCormick of the strike or work-stoppage. There is no contradiction by any acceptable evidence in the record of this direct testimony of McCormick and Bodenheimer. While the trier of facts is not required to accept uncontradicted testimony such as this, "he may not do so without good reason," N. L. R. B. v. Klaue (9th Cir. 1975) 523 F.2d 410, 414, and in the absence of " 'something in the evidence or in the tale, itself, which furnishes a basis for discrediting it because of its inherent improbabilities,' " Sabbagha v. Celebrezze (4th Cir. 1965) 345 F.2d 509, 512.6 The Board, in its opinion, makes no attempt at justifying a disregard of this uncontradicted evidence or of pointing to any circumstantial evidence which might support a refusal to accept it. Indeed, it blandly declares that, "(t)he record clearly reveals that McCormick, who made the decision to discharge the employees, was notified by Bodenheimer of the employees' participation in the work stoppage." It cites no evidence in the record to support this finding. This is understandable: there is none unless we are to engage in pure surmise. The Board identifies no witness who testified to such notification or to any circumstance or fact considered by the Board to lend credence to such finding of notification. It is true that the administrative law judge had found that the denial by Bodenheimer that he had told McCormick of the work-stoppage did "not ring true." The only basis for such finding, as given by the administrative law judge, was:
"Bodenheimer had been on the job not much more than 1 month when the incident occurred. It is extremely unlikely that this would not have been reported to McCormick particularly when, as they testified, McCormick would check at least twice a week with Bodenheimer as to how the operation was running."
This was, in essence, a mere surmise or speculation, without any circumstantial evidence whatsoever to indicate that McCormick did know of the work-stoppage or that Bodenheimer had told McCormick of it.
It is significant that neither the Board nor the administrative law judge adverted to Bodenheimer's explanation of why he did not report to McCormick the incident. First of all, Bodenheimer said he did not consider the incident to be a work-stoppage or "strike." He proceeded to explain the reasons for not considering it a work-stoppage or strike. As he detailed the circumstances surrounding the incident, it began when he found five or six employees in the break room on the occasion in question. This was not unusual and prompted no special concern on his part, according to his undisputed testimony. He observed nothing out of the ordinary in the fact that six employees were present: the employees, as he testified, generally took their break at the same time. Moreover, it was not a rare event for them to take such breaks from time to time. This was a practice which interferred with efficient production but one the employer could not control even with the new rules McCormick and Bodenheimer had issued on work-breaks. There was no outward indication that the employees were engaged in a strike when Bodenheimer encountered them in the break room. They made no demands at the time beyond asking Bodenheimer to convey to McCormick their request for a raise. Again, this was not unusual, since Bodenheimer said he was continuously being requested for a raise by the employees. When Bodenheimer told the employees that he could not reach McCormick, the employees did not say they were on strike or that they intended to refuse to work until they received a satisfactory reply to their demands. They demurely returned to their jobs without any further request.
That does not represent the normal scenario for a strike or work-stoppage. Under the circumstances, there was nothing unreasonable in Bodenheimer's testimony that he did not regard the incident as a work-stoppage or strike. He stated it was, in his view, merely another request of the employees for a wage increase, which he and McCormick had already discussed and agreed upon. It cannot be said that Bodenheimer's reaction was either unreasonable or unwarranted. If his reaction was reasonable, Bodenheimer's testimony that he did not report the incident to McCormick is not "plainly incredible" or "inherently improbable." Yet it was on the basis that this reaction was unreasonable and that basis alone that the administrative law judge reached his conclusion that the uncontradicted evidence of want of knowledge on McCormick's part did not "ring true." This, however, is more than the Board did, for it assigned no reason whatsoever for disregarding the uncontradicted testimony and for its incredible statement that Bodenheimer "notified" McCormick of the strike without the slightest explanation of why it denied credibility to the undisputed evidence or where in the record it found evidence to support its bald finding of notification.
The action of the Board in this case is similar to its decision in N. L. R. B. v. Whitfield Pickle Company (5th Cir. 1967) 374 F.2d 576, 581, where the Board disregarded the testimony of the employer that he lacked knowledge of the complaining employee's union activity. In denying enforcement to the Board's decision because of this, the Court said:
"The trial examiner and the Board find it 'incredible' that the company officers did not know of Mrs. Goodwin's union activities, but the barrenness of the record on this point indicates that their incredulity springs from nothing more than the assumption or suspicion that the officers' denials are false. The officers gave sworn testimony in an attempt to prove a negative: it can be well-nigh impossible to prove that one did not know something in the past by any other means than sworn denials. Such sworn testimony cannot be 'arbitrarily disregarded' because of the merest suspicion or assumption, without more, that the denials are lies."7
But, if it be assumed that McCormick knew of a work-stoppage, the Board still has a substantial burden to sustain before it can ask for the enforcement of its order. There can be no real dispute that the employer in this case had a valid ground for discharging these employees. Three of them had been identified as responsible for defective production; two of the others, by comparison, had production substantially lower than the new employees on the third shift. The complainant Scearce had long been observed by McCormick not working, loitering about the plant away from her job. McCormick had called this to Bowman's attention, while the latter was responsible for production, and had suggested that Scearce should be disciplined. When Bodenheimer took over the management, he noted the same conduct on Scearce's part. Scearce obviously was not an efficient worker. Defective workmanship and low production the two reasons for these discharges were not new problems for this plant and they were undoubtedly due in part to the kind of inattention to the job manifested by Scearce and demonstrated by the tests on the part of Gardner, Ellis and Tucker. These problems had plagued the plant's production for months. McCormick had demanded of Bowman that he take some appropriate action to correct them. The difficulty in correcting the problems, as Bowman argued, was that the employees responsible could not be identified. Bodenheimer's tests overcame this difficulty and pointed the finger of responsibility at the employees who were discharged. Considering the extended period over which these problems had persisted and his repeated demands that something be done to correct them, it is understandable that McCormick, chafing under several years of financial losses, would act when Bodenheimer brought the reconfirmed results of the tests to his attention.
The Board seeks to impugn the integrity of the tests which established the delinquencies of the employees discharged. It declares that the tests were not conducted "in a thorough and systematic manner." This apparently refers to the simplicity of the tests but the very simplicity of the tests argues for their reliability. Thus, in comparing the production of the complainants Tucker and Gardner on the first shift with the employees on the third shift doing similar work, Bodenheimer tagged the production by the complainants Tucker and Gardner, on the one hand, and the production of the new employees on the third shift on the other. It is undisputed that Tucker and Gardner, though more experienced, were substantially lower in production than the third shift employees. A similar method identified the employees responsible for poor quality. The method used by Bodenheimer in this instance conformed to that used by many manufacturers. Many products include in the carton or on the product an identification by number of the employee responsible for passing the product. This method is employed for the very same reason that Bodenheimer tagged the product, I. e., to identify the employee responsible for the defective product. It cannot thus be said that the tests were unreasonable. Nor did the administrative law judge find them to be unreasonable. On the contrary, he found that the tests were fairly devised and conducted, and that they provided a valid cause to discharge the employees in question.
The Board seeks to undermine the validity of the employer's claimed justification by pointing to the pay increases received by the discharged employees during their employment. It would, however, disregard the testimony of the complainants themselves that most, if not all, these increases were mandated by changes in the minimum wage under the Wage-Hour Law. Nor were the increases given these employees in tribute to their merit; they were plant-wide increases given all employees at the time. It hardly seems reasonable to read into legally mandated wage increases admissions of employer satisfaction with an employee's performance. And this is particularly so here, since it is undisputed that for an extended period the employer had been bedevilled by poor quality and low production and had been complaining of the poor work of the employees.
When an employer has, as did the employer in this case, a perfectly valid reason to discharge or discipline an employee, it is not sufficient in order to establish a violation of the Act "for the Board to declare that the * * * (discharges were) * * * 'pretextual,' " but " 'the burden which is on the Board is not simply to discover some evidence of improper motive, but to find an affirmative and persuasive reason why the employer rejected the good cause and chose a bad one' * * * and to 'present a substantial basis of believable evidence pointing toward the unlawful one.' "8 The Board sought to meet this burden by asserting that the claim of low production and defective work as the justification for the discharges was "pretextual" and that the real reason was the work-stoppage or strike. It is the thesis of the Board that the employer intended to fire all those who had taken part in the pay protest, to "clean house," as one of the complainants said, and that this was the real reason the complainants were discharged rather than the pretextual one of their poor work performance. But this claim loses any force or persuasion when it is noted that not all those who had protested were fired. Hilton, who had joined the six who talked with Bodenheimer in signing the written request for a raise (a request which was "merely putting into writing" what the six had said to Bodenheimer) was not fired. Meachum was an active member throughout of the group engaging in the so-called work-stoppage and in submitting the written protest. She was not fired. A third member of the group, Hughes, did not join in any complaint, even though she had been discharged along with the others. In short, out of seven who participated in the protest, only four have made any charge of discriminatory discharge.
The Board does not seek in any way to reconcile its finding that McCormick, in discharging the four complaining parties, was seeking to "clean out" all who had participated in what the Board denominates a work-stoppage with the significant failure of McCormick to order the firing of Meachum and Hilton. If the theory of the Board was correct, there had to be some explication of a reason for this apparent flaw in its reasoning. The Board undertakes none; it simply disregards the inconsistency. Nor can anything in support of the Board's theory be found in the opinion of the administrative law judge, who similarly did not comment on this circumstance. This, though, is understandable in the case of the administrative law judge, who had not embraced the Board's theory but had, on the contrary, found that McCormick had not ordered these four charging parties fired because of their work-stoppage but rather because they had proven themselves sloppy and inefficient workers.
But, while the Board appears to have treated the failure of McCormick to fire Meachum and Hilton, along with the other five participating in the critical incident, as an irrelevance, the charging parties themselves and the General Counsel were not blind to the weakness in their claim of discriminatory firing provided by this failure to fire at the same time Meachum and Hilton. They sought to shore up this weakness with some hearsay testimony of Bowman. In this hearsay testimony, Bowman was said to have explained the failure to fire Meachum by declaring that Bodenheimer did not know that she had been engaged in what the Board described as a work-stoppage. He offers no explanation for Hilton's retention. The administrative law judge refused to credit this hearsay evidence. He found incredible and unbelievable that Bodenheimer did not know Meachum had engaged in the so-called work-stoppage. Bodenheimer had met with the participants in this incident and Meachum was one of the group. Meachum was not unknown to Bodenheimer, as Bowman's alleged statement would suggest. She had worked for him for almost two months. He had less than fifteen employees working for him. It is inconceivable that he didn't know her as well as he knew the other employees in the break-room at the time. This entire hearsay testimony was further discredited, as the administrative law judge observed, by Bowman's admission that he had not even known, and was surprised when told, that any of the complainants had been fired. Such an admission, proved by the General Counsel, made unbelievable any attempt by Bowman to give a discriminatory color to the discharges and the administrative law judge properly dismissed it. Moreover, the Board recognized and accepted the administrative law judge's finding of incredibility for it expressly declared in its opinion that it did not rely on this alleged conversation with Bowman in making its findings.
In its brief on appeal, the Board offers, with some diffidence in a footnote, the suggestion that McCormick may not have fired Meachum and Hilton in order to conceal his illegal purpose in discharging the other five. It would seem sufficient answer to this suggestion that it was not incorporated in the reasons assigned by the Board for its critical finding of discriminatory motive in its decision in this case and cannot now be injected into the case as a belated explanation for the Board's finding. Beyond that, there is nothing in the record to lend credence to any such finding and the indisputable fact remains that the Board has made no effort to reconcile its finding that McCormick was intending to "clean out" all employees who had any part in the work-stoppage with the employer's retention of Meachum and Hilton. Without some reasonable reconciliation, the Board's finding that McCormick ordered the discharges to rid the business of all participants in the work-stoppage is manifestly wrong. Yet it is this finding which in turn is critical to the Board's ultimate finding of discriminatory purpose in the discharges; without it the finding of discriminatory purpose fails.
Actually, the Board offered in its opinion only four reasons for its finding that McCormick's justification for the discharges was pretextual. The first of these was that the employer had given to these employees from time to time pay raises. We have already seen why this reason is without merit and it is unnecessary to repeat those circumstances. It next seeks to challenge the adequacy of the tests by which the employer identified the offending employees. Again, this objection has been earlier canvassed and found unsupportable. The third reason is that, if the employer's justification for discharge was really low production and defective product, this justification had existed since 1973 and the fact that it was only asserted after the work-stoppage against a majority of those participating in such work-stoppage casts a suspicious cloud over the good faith of the asserted justification. The Board is correct that the employer had been troubled by poor product and low production in the plant since 1973. The record is undisputed that McCormick had raised this point often with Bowman, who was in charge of the plant, between 1973 and 1976. Again, it is undisputed that Bowman refused to take any action because he said he couldn't identify those employees who were responsible for defective product and who were not performing efficiently. All that, as we have observed elsewhere, was undisputed. These circumstances provide a complete answer to any finding that the employer knowingly winked at the low production and poor workmanship of the discharged employees from 1973 to 1976. When Bodenheimer, by his tests, identified the responsible employees did what Bowman professed he couldn't do the employer acted. There is no evidence in the record which supports any finding different from that of the administrative law judge, which was that the discharges were motivated solely by the low production and improper workmanship of the discharged employees.
The final basis for the Board's finding is an inference, based on its assumption that the employer began to employ persons for the third shift after the work-stoppage, and that these new employees were hired as replacements for the employees participating in the work-stoppage, whose discharges the employer had already determined upon. This reasoning, however, completely disregards the undisputed fact that the employer was behind on its orders and was under great pressure from its customers to make timely deliveries. In early August, McCormick had sternly demanded that the problems in the plant be promptly corrected. To alleviate this back-log and to comply with McCormick's demand for improvement, Bodenheimer attempted to add employees to the third shift. This, like the other findings of the Board in this connection, fails when tested by the undisputed record.
It is manifest that the Board has failed to meet its burden of providing any substantial evidence of some reason why the employer in this case chose the "bad" reason for discharging these employees and not the "good" one. In the absence of such a record, and of the failure by the Board to establish by any substantial and believable evidence that McCormick knew of the work-stoppage, the decision of the Board is reversed, the complaint is dismissed and enforcement of the Board's order is denied.
ENFORCEMENT DENIED.
Carolina Foam and Paper Company, the selling agent for American Manufacturing Associates, Inc., was also owned by Thomas B. McCormick
Tucker and Gardner were the employees on the first shift, who, on the comparison with similar workers on the third shift, had production substantially lower than those on the third shift. Ellis, Gardner and Tucker were the employees whose work had been returned as defective by the employer's customers. Scearce had been often observed and reprimanded for being absent from her place of work over a considerable period of time
234 NLRB No. 105
P.S.C. Resources, Inc. v. N. L. R. B. (1st Cir. 1978) 576 F.2d 380, 383; Amyx Industries, Inc. v. N. L. R. B. (8th Cir. 1972) 457 F.2d 904, 906; N. L. R. B. v. Gotham Industries, Inc. (1st Cir. 1969) 406 F.2d 1306, 1310; National Labor Rel. Bd. v. Chronicle Publishing Co. (7th Cir. 1956) 230 F.2d 543, 549
There was some hearsay testimony that Bowman had said McCormick ordered the discharges because he was "mad" over the work-stoppage. This testimony was found incredible and unworthy of belief by the administrative law judge, who gave compelling reasons for his conclusion. The Board apparently accepted the conclusions of the administrative law judge, for it expressly refused to rest its findings on this alleged statement of Bowman
See, also, Day v. Weinberger (9th Cir. 1975) 522 F.2d 1154, 1156; Hassler v. Weinberger (7th Cir. 1974) 502 F.2d 172, 178
See, also, NLRB v. Klaue, supra (523 F.2d at 414)
Firestone Tire & Rubber Co. v. N. L. R. B. (4th Cir. 1978) 583 F.2d 1268, 1273; See also, Firestone Tire & Rubber Co. v. N. L. R. B. (4th Cir. 1976) 539 F.2d 1335, 1337; N. L. R. B. v. Patrick Plaza Dodge, Inc. (4th Cir. 1975) 522 F.2d 804, 807; N. L. R. B. v. Consolidated D. Elec. Co., Div. of C. Corp. (4th Cir. 1972) 469 F.2d 1016, 1024-28; N. L. R. B. v. Covington Motor Company (4th Cir. 1965) 344 F.2d 136, 138; Maphis Chapman Corporation v. N. L. R. B. (4th Cir. 1966) 368 F.2d 298, 304; Hazel-Atlas Glass Co. v. National Labor Relations Bd. (4th Cir. 1942) 127 F.2d 109, 114 and 117; Famet, Inc. v. N. L. R. B. (9th Cir. 1974) 490 F.2d 293, 296