Pursuant to the Occupational Safety and Health Act,1 the Secretary of Labor has promulgated a regulation, 29 C.F.R. § 1910.-95(b)(1), requiring employers to implement “feasible administrative or engineering controls” in order to protect their employees from exposure to harmful levels of occupational noise. In 1974, the Secretary cited West Point Pepperell, Inc., a textile manufacturer, for violations of that noise regulation, and an administrative law judge affirmed the citation in pertinent part. The Occupational Safety and Health Review Commission, however, vacated the citation, and the Secretary now petitions for review of the Review Commission’s order.
*981When all is said and done, there is less to this case than meets the eye. Resolution of this petition does not require us to decide delicate and important questions of policy and congressional intent in providing, “so far as possible, every working man and woman in the Nation safe and healthful working conditions.”2 Rather, it merely requires us to read the Review Commission’s order and then decide what the Commission said in that order. The Secretary vigorously contends that the order announces an erroneous legal principle— namely, that engineering controls are not technologically “feasible,” and therefore employers are not required to implement them under 29 C.P.R. § 1910.95(b)(1), unless such controls have been fully developed and marketed by disinterested third parties. West Point Pepperell, on the other hand, contends that the Commission simply made a factual determination, supported by substantial evidence, that the state of technology in the textile industry in 1974 was such that there were no “feasible” engineering controls for significantly reducing the level of occupational noise in its mill. We agree with West Point Pepperell’s reading of the order and therefore affirm.3
The Regulatory Background
The Occupational Safety and Health Act (the Act) empowers the Secretary of Labor to promulgate safety and health standards “reasonably necessary or appropriate to provide safe or healthful employment and places of employment,” see 29 U.S.C. §§ 655, 652(8), and obligates employers to comply with those standards and regulations validly promulgated by the Secretary, 29 U.S.C. § 654(a)(2). Numerous medical and psychological studies document the serious effects of excessive noise exposure.4 Pursuant to his statutory authority the Secretary of Labor, through the Occupational Safety and Health Administration (OSHA), has responded to the dangers of occupational noise and has promulgated 29 C.F.R. § 1910.95(b)(1). That regulation provides that
[w]hen employees are subjected to sound [levels] exceeding those listed in Table G-16, feasible administrative or engineering controls shall be utilized. If such controls fail to reduce sound levels within the levels of Table G-16, personal protective equipment shall be provided and used to reduce sound levels within the levels of the Table.
This standard reflects the determination that the preferable method for abating hazardous noise exposure is through technological or administrative controls such as the installation of noise-damping materials to machine surfaces and work-room walls, the substitution of quieter machines, processes, or parts for noisy ones, and the rotation of workers to minimize their exposure to high sound intensities. But if such engineering or administrative controls are not “feasible,” then workers shall be provided with and required to wear personal protective *982devices — i. e., earplugs and earmuffs — that successfully attenuate their exposure to excessive noise levels.
Table G-16 defines excessive noise exposure in terms of decibels (dB) as measured on the “A scale/slow response,” and prescribes 90 dBA as the maximum permissible exposure for an 8-hour workday.5 An important feature of decibel measuring is that decibel levels increase logarithmically rather than arithmetically. “Every increase of 10 dB represents an increase of approximately 300 percent in sound pressure. A 100 dB noise is, therefore, 3 times as intense as a 90 dB noise, rather than about 10 percent more intense, as might be expected.”6
Employer noncompliance with 29 C.F.R. § 1910.95(b)(1) or any other standard or regulation promulgated by the Secretary under the Act may result in the issuance of citations and the imposition of fiscal penalties. See 29 U.S.C. §§ 658(a), 659(a), 666. If the employer observes certain procedural requirements, it is entitled to an administrative hearing and review by the Occupational Safety and Health Review Commission of any such penalty or citation to abate a violation. 29 U.S.C. § 659(c). The Act provides that the administrative law judge (ALJ) who conducts the hearing shall make a “report of any * * * determination which constitutes his final disposition of the proceedings.” 29 U.S.C. § 661(i). That report becomes the final order of the Review Commission unless one of the three Commissioners exercises his discretion to have the report reviewed by the Commission itself, in which case the decision of the Commissioners constitutes the final administrative determination. See id.; see generally Accu-Namics, Inc. v. Occupational Safety and Health Review Commission, 5 Cir., 1975, 515 F.2d 828, 834-35, cert. denied, 1976, 425 U.S. 903, 96 S.Ct. 1492, 47 L.Ed.2d 752.
Pointing A Finger At West Point
Respondent West Point Peppered, Inc. (West Point), is a large textile manufacturer. In 1974, on OSHA compliance officer inspected West Point’s mill in Lindale, Georgia, and found steady noise levels of 90-97 dBA in Weave Room No. 2, which contained 112 Draper Shuttleless Looms (DSLs), and 98-103 dBA in Weave Room No. 3, which contained 1,178 Draper Fly-Shuttle Looms (DFSLs). All of these looms were manufactured by the Draper Division of the North American Rockwell Corporation, which is the nation’s largest producer of looms. The noise created by these looms was due to the vibration of various machine parts, the grinding of pinion gears, and the impact of various loom parts during normal operation of the looms. Weave room employees worked virtually their entire 8-hour shifts in the midst of this deafening noise, although some of the noise was muted by the personal ear protective devices West Point required them to wear. The Secretary of Labor cited West Point for failing to implement feasible engineering controls to reduce or eliminate the harmful loom noise in either of the two weave rooms, proposed a civil penalty of $300 for the two violations, and ordered West Point to submit an engineering compliance program and periodic progress reports.
West Point contested this citation in a timely fashion, and, after formal pleadings had been filed before the Review Commission, the case was tried before a Commission ALJ. Fourteen volumes of testimony *983was heard from a number of employees, scientists, engineers, and doctors.
A portion of the testimony concerned the effectiveness of earplugs and earmuffs in reducing the actual exposure of workers to loom noise. Evidence was introduced to the effect that all weave room employees at the Lindale Mill had been tested and fitted with personal ear protective devices, that employees were required to wear these devices at all times in high noise areas, and that if properly fitted and continuously worn, the devices reduced actual noise exposure 20-25 dBA, well within the 90 dBA maximum.
West Point employees also testified about the efforts made by the Company to reduce loom noise at its source. Some of these failed, such as attempts to substitute machine parts made of plastic and other soft materials (which were unable to withstand the wear and tear of normal operations) and the construction of shields or baffles (which interfered with observation of and access to the looms). Other efforts had limited success, such as mounting the looms on floor pads designed to isolate loom vibration and wider spacing of the looms. These efforts notwithstanding, however, it was undisputed that the noise levels in Weave Room Nos. 2 and 3 exceeded the 90 dBA level that 29 C.F.R. § 1910.95(b)(1) permits where feasible administrative or engineering controls are available.
The testimony of four witnesses related to the crucial question of whether in fact there existed in 1974 feasible administrative or technological methods by which significant noise reduction could be achieved in the two weave rooms.7 On the basis of this testimony, the ALJ concluded that while there was no persuasive evidence concerning administrative controls, with regard to engineering controls “the preponderance of the credible evidence establishes that it was feasible to reduce the noise level in Weaving Room No. 2 by at least 5 dBA.” He therefore affirmed the citation with respect to the shuttleless looms in Weave Room No. 2. Although he also concluded that materials and techniques were available to lower somewhat the sound levels of the fly shuttle looms in Weave Room No. 3, he vacated the citation with respect to that weave room because he found that the level in that room, even with the controls, would still exceed 90 dBA.8
The Review Commission’s Decision
The Review Commission exercised its prerogative to review the AU’s report9 and vacated the citation in its entirety. The Commission reviewed the testimony of the witnesses in some detail and concluded, in explicit terms, that “the evidence does not establish that the technology exists to achieve significant noise attenuation in either of the two weave rooms.”10 Contrary to the assertion of the Secretary on this appeal, and to the dissent of Commissioner (now Chairman) Cleary, the Commission did not predicate its decision on the principle that engineering noise controls are not technologically feasible unless they have been developed, tested, and marketed by a third-party supplier.11
*984As explained in Accu-Namics, supra, the Review Commission itself is the fact-finder under the Act, and the Court of Appeals reviews only the final order of the Commission. 515 F.2d at 834. Upon such judicial review, the Commission’s findings of fact are conclusive if supported by substantial evidence in the record considered as a whole. 29 U.S.C. § 660(a); Accu-Namics, 515 F.2d at 834-35. Having determined that the Commission’s decision was the rather straightforward factual one that in 1974 there were no feasible technological methods for significantly reducing noise levels in the weave rooms at West Point’s Lindale mill, the only issue that remains for us to determine is whether that factual finding was supported by substantial evidence.
Substantial Evidence?
The sole witness for the Secretary of Labor on the issue of technological feasibility was Dr. Ronald Bailey, who was an assistant professor of mechanical and aerospace engineering with a doctorate and extensive research experience in noise and vibration control. Dr. Bailey testified at length concerning the sources of DSL and DFSL noise and the application of basic engineering principles by which the looms could, at least in theory, be modified to reduce noise during their operation. He also described several loom modification kits that Draper was developing but had not yet marketed, as well as his own laboratory tests on a single loom using sound-absorbing and damping materials similar to those being tested by Draper. On the basis of the various engineering principles and these laboratory tests, Dr. Bailey stated that in his opinion the noise levels in both Lindale weave rooms could be reduced by at least 5 dBA. Upon questioning, however, he conceded that he would not make a “blanket recommendation” that textile manufacturers immediately install his proposed loom modifications because he had not conducted a “feasibility study.”12
Three of West Point’s witnesses also addressed the technological feasibility of reducing loom noise. The vice-president of another textile manufacturer agreed with the soundness of the engineering principles advanced by Dr. Bailey, but stated categorically that the “application of those principles * * * in a weave room is not today available [to] [p]rovide a substantial reduction in the noise level created by the loom.” He also testified that Draper’s developmental noise reduction kits, to which Dr. Bailey had referred, had been installed on a trial basis by one textile manufacturer and been found unsatisfactory. West Point’s director of engineering provided testimony about the practical difficulties West Point had encountered in its attempts to implement various noise attenuation devices.13 Al-
*985though he voiced his general skepticism that any noise reduction techniques currently known or proposed would prove to be practicable, he testified that West Point had ordered 24 of Draper’s latest developmental loom modification kits. The third witness offered by West Point possessed a degree in electrical engineering, had worked primarily in the field of acoustics, and was accepted by the ALJ as an expert witness on the design of noise reduction modifications for space and machinery. This witness testified at length on the doubtful prospects for significantly reducing loom noise at West Point’s Lindale mill, based on his personal investigation of the plant. He also stated that to the best of his knowledge there was no feasible engineering method available to reduce loom noises to 90 dBA or below.
The standard by which “substantial evidence” is measured — “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion 14 —is clearly satisfied in this case. We accordingly affirm the Review Commission’s decision that no feasible technological methods for significantly reducing loom noise in West Point’s weave rooms were in existence in 1974.15
If, and when, through various technological developments, such methods do become feasible, West Point is obligated by § 1910.-95(b)(1) to implement them. West Point is of course, by virtue of its extensive operations and prominent position in the textile industry, a part of the technology and must therefore exploit all reasonable opportunities, whether suggested by its own experience or that of others knowledgeable in the field, for solving this problem in a feasible manner. In the meantime, a feasible technological solution apparently, and unfortunately, exists only in the realm of theory.16
AFFIRMED.
. 29 U.S.C. §§ 651-678.
. See 29 U.S.C. § 651(b).
. The Seventh Circuit has recently held that the term “feasible” in the § 1910.95(b)(1) noise regulation standard includes a consideration of both technical and economic feasibility.
[Wle conclude that the standard should be interpreted to require those engineering and administrative controls which are economically as well as technically feasible. Controls may be economically feasible even though they are expensive and increase production costs. * * * But they will not be required without regard to the costs which must be incurred and the benefits they will achieve. In determining whether controls are economically feasible, all the relevant cost and benefit factors must be weighed.
Turner Co. v. Secretary of Labor and OSHRC, 7 Cir., 1977, 561 F.2d 82, 85 (quoting from the Review Commission’s decision in the same case, Turner Co., OSHRC Docket No. 3635, BNA 4 O.S.H.C.-(1976)).
This decision certainly comports with the plain import of the term “feasible,” but we need not endorse or reject the Seventh Circuit’s holding as both parties have restricted the issue in this case to one of technological feasibility.
. See, e. g., Anticaglia & Cohen, “Extra-Auditory Effects of Noise as a Health Hazard,” 31 Am. Indus. Hygiene Ass’n J. 278 (May-June 1970); National Institute for Occupational Safety and Health, Criteria for a Recommended Standard, Occupational Exposure to Noise (1972), at IV-1 to IV-16.
. Decibels, the basic unit of measurement of sound levels, are recorded on sound level meters according to several scales. On the A scale, the meter is more sensitive to higher pitched tones than those of a lower pitch, just as the human ear is. The “slow” response is another setting of the instrument by which it averages out high level noises of brief duration (such as hammering), rather than responding to the individual impact noises. See U. S. Dep’t of Labor, Guidelines to the Department of Labor’s Occupational and Noise Standards p. 3 (1971).
. U. S. Dep’t of Labor, Guidelines, supra, note 5, at 3. “Illustrated another way, if one machine produces a sound level of 90 dB, a second machine of the same kind placed next to it will result in a combined noise level of 93 dB, rather than 180 dB, which might be expected.” Id.
. We summarize this testimony later in the course of this opinion.
. Several months after the ruling of the ALJ, the Review Commission held that if feasible engineering controls will achieve a “significant reduction” in noise levels, they must be implemented even though the resulting noise levels still would exceed the limits specified in Table G-16 of § 1910.95(b)(1). Continental Can Co., OSHRC Docket No. 3973, BNA 4 O.S.H.C. 1541 (1976). However, the ALJ’s misinterpretation of § 1910.95(b)(1) as applied to Weave Room No. 3 is of no consequence in light of the Review Commission’s subsequent finding that the evidence did not establish the technological feasibility of reducing noise levels significantly in either of the two weave rooms.
. See 29 U.S.C. § 661(i); Accu-Namics, supra, 515 F.2d at 834.
. The Commission also concluded, after consideration of the costs that would be entailed in rotating weave room employees, that “the feasibility of administrative controls has not been established.”
. The Commission majority explained its explicit holding, quoted in text, supra, as follows:
The burden of establishing technological feasibility rests with complainant. * * * *984An analysis of the testimony of both respondent’s and complainant’s expert witnesses yields the conclusion that complainant has not satisfied this burden. The recommenda- . tions of complainant’s expert, which had never been tested in an operating weave mill, present the mere possibility that, if they were implemented, noise levels would be attenuated. In light of the experience of respondent’s noise experts revealing that the application of acoustical and resilient components does not substantially reduce the noise emissions in a functioning weave room and the unsatisfactory results of Draper’s field tests on its noise reduction kits, we are not convinced that it was feasible to achieve significant noise reduction in either weave room # 2 or # 3 by implementing engineering controls on DSL and fly shuttle looms.
. By a “feasibility study,” Dr. Bailey explained that he meant
Definition of a problem, development of alternative approaches to solving that problem, evaluation of these approaches, determination of the effectiveness of technique on prototype parts, the evaluation of the economics of this situation, all of these aspects would be entailed in what I mean by feasibility study.
. An amusing interchange, indicative of the divergent orientations with which the parties approached “technological feasibility,” occurred when the Secretary of Labor’s attorney questioned this witness on the criteria by which West Point evaluated noise reduction devices:
Q. And would you [test them] on all of the machines in that room?
A. No, sir. We could not. You might say we could not afford to try them on that basis. * * * [Depending on the cost of the part, we would try two or three parts, maybe a half dozen parts and evaluate those * * * *985mostly * * * to confirm that they would indeed wear well and stay in adjustment on the machine. This was the thing that usually rejected parts. It was not so much a matter of whether or not they quieted the machine, but if they seemed to have a tendency to quiet the machine and would not run on the machine, we had to discount them on that fact. We evaluated them primarily that they would operate.
Q. Not if they were quieter?
A. Well, they had to operate on the machine, first they had to do that, and then we hoped that they would indeed be [as] quiet as the supplier had suggested that they might be.
Q. So, but your testimony now is that your primary goal at least in the first instance was first to see if they ran?
A. Satisfactorily, yes, sir. It had to produce cloth on the loom. This was our primary goal.
. Universal Camera Corp. v. National Labor Relations Board, 1951, 340 U.S. 474, 477, 71 S.Ct. 456, 95 L.Ed. 456.
. Cf. American Petroleum Institute v. Occupational Safety and Health Administration, 5 Cir., 1978, 581 F.2d 493 (the Secretary may promulgate a standard requiring employers to implement technological methods not yet developed but looming on the horizon only if the Secretary can show by substantial evidence that the employment conditions imposed by the proposed standard are reasonably necessary to provide safe or healthful places of employment).
. This conclusion is shared by more than just the three witnesses who testified for West Point and two members of the Review Commission. In March 1975 (subsequent to the proceedings before the ALJ in this case), the Assistant Secretary of Labor for OSHA expressed the opinion that while in general the technology exists for compliance with either an 85 dBA or a 90 dBA limit through engineering controls, “there are some significant exceptions, such as textile weaving.” 40 Fed.Reg. 12336, 12338 (1975). Later, the Department of Labor (OSHA) held public hearings concerning a proposed reduction of the maximum permissible level of noise exposure for an 8-hour workday to 85 dBA. At these hearings (in October 1976), a vice-president of the corporation that manufactures Draper looms testified that when the field trials of Draper’s developmental loom noise reduction kits were finally evaluated, the company had had to revise its estimate of likely noise reduction to between zero and one-and-one-half dBA. From its repeated failures, the company had learned that “translation from the laboratory to the weave room did not produce the predicted results even under tightly controlled conditions,” and had concluded that with engineering concepts as then known, it was unlikely that noise levels in a typical operating weave room could be reduced for either shuttleless looms or fly shuttle looms to 90 dBA in the foreseeable future.