Opinion PER CURIAM.
PER CURIAM:In November 1978, the Mine Safety and Health Administration (MSHA) of the Department of Labor issued final regulations requiring coal operators to equip all underground miners with self-contained self-rescuers (SCSRs) that will provide oxygen in the event of a mine explosion or fire.1 The regulations gave coal operators two years, until December 21, 1980, to order and supply the devices. Today we do not review either the propriety of the 1978 regulations or the wisdom of the initial phase-in period.2 Instead we judge a more narrow question: whether on December 5, 1980, the Secretary of Labor acted reasonably when he deferred implementation of the regulations until June 21, 1981, and whether he had good cause to take that action without providing prior notice or an opportunity to comment.3 While we find the notice-and-comment question very close, we conclude that, in the special circumstances presented, the Secretary had good cause to dispense with usual rulemaking procedures and that his action was otherwise reasonable.
I. BACKGROUND
Each year, underground fires and explosions claim the lives of some coal miners.4 Miners who are not killed by the initial explosion or spreading flames may die from inhalation of noxious gases or simple lack of oxygen. In 1969 the National Academy of Engineering concluded that asphyxiation from smoke, carbon monoxide, or carbon dioxide caused twenty-two percent of the 566 mine deaths that occurred between 1950 and 1969.5 Responding to this type of information Congress in 1969 required coal mine operators to give all underground miners self-rescue devices that would protect them from poisonous gases for at least one hour.6 Regulations promulgated shortly thereafter directed mine operators to supply filter-type self-rescuers to all underground miners.7
*321Filter-type self-rescuers, however, are inadequate to the life-saving task for several reasons. They protect miners only against carbon monoxide, not against smoke, carbon dioxide, or other toxic gases. Even protection from carbon monoxide ceases if carbon monoxide levels exceed one percent of the atmosphere. At higher concentrations of carbon monoxide, the filter-type self-rescuer produces air that is too hot to inhale.8 Miners react by spitting out the filter-type device and thereupon die from carbon monoxide poisoning. In addition to these problems, filter-type self-rescuers do not supply an independent source of oxygen. Therefore, even if the devices successfully protect miners from carbon monoxide, a low oxygen level may cause miners to lose consciousness and eventually perish in the contaminated mine atmosphere.9
Aware of the filter-type device’s limitations, the Bureau of Mines began an extensive testing program to identify a better self-rescuer. In 1977 these efforts culminated in proposed regulations requiring coal operators to equip underground miners with SCSRs rather than filter-type self-rescuers.10 SCSRs contain an independent oxygen supply that lasts for an hour or more. Thus, they protect miners from all poisonous gases and give miners sufficient oxygen to attempt escape from a mine struck by a disaster. While SCSRs are much more effective than filter-type self-rescuers, the proposed rules did not contemplate an immediate switch from one device to the other. Instead, the proposed rules indicated that coal operators would have two years from the date of any final regulations to equip their miners with SCSRs.11
On November 21, 1978, the Department of Labor issued final regulations requiring the use of SCSRs in coal mines. As promised in the 1977 proposed rules, coal operators were given two years, until December 21, 1980, to select, order, and supply SCSRs. 1978 Regulations, supra note 1, at 54246. During this phase-in period, MSHA pledged to field test the SCSRs. Id. at 54244. While MSHA was confident that SCSRs were “reliable and safe to use and store in underground mines,” id. at 54243, the agency wanted to conduct field testing to determine how the use of SCSRs would “affect miners” in actual mining situations. Id. at 54244.
After promulgation of the final rules, MSHA promptly developed a program for field testing SCSRs. On February 28, 1979, Joseph O. Cook, MSHA Administrator for Coal Mine Safety and Health, sent Robert B. Lagather, Assistant Secretary of Labor for Mine Safety and Health, an action plan for evaluating SCSRs. The plan envisioned completion of testing well before the December 1980 deadline for implementing the regulations.12
Two circumstances, however, delayed the field testing program. First, the National Institute for Occupational Safety and Health (NIOSH), which had to approve SCSR models before they could be used in the tests, did not approve production model SCSRs until the summer of 1980.13 Second, *322the Joint Industry Health and Safety Committee, representing both a major coal operators’ association and the United Mine Workers of America (UMWA), would not recommend that their members participate in field testing until MSHA answered safety concerns about SCSRs that had surfaced early in 1980.14 The Bituminous Coal Operators’ Association (BCOA), the major industry association represented on the Joint Committee, did not finally approve field testing in its mines until November 19, 1980.15
MSHA worked diligently to overcome these roadblocks. When NIOSH approval for production model SCSRs was not forthcoming, the agency used hand-manufactured models that had been approved in 1979 for a first round of field tests.16 To *323calm lingering fears about the safety of SCSRs, MSHA asked the Bureau of Mines to conduct an extensive study of the devices. This study was completed during the summer of 1980 and presented to the Joint Industry Health and Safety Committee the following fall.17 And when BCOA continued to withhold approval for field testing throughout the fall of 1980, MSHA initiated tests in mines owned by members of the National Independent Coal Operators Association (NICOA).18
Despite delays in the field testing program, MSHA appeared determined throughout the fall of 1980 to complete field testing by December and implement the regulations on schedule.19 On November 5, 1980, Assistant Secretary Lagather wrote to the Joint Industry Committee, noting that the Committee had already received the Bureau of Mines study, and expressed his confidence that field tests could “be started at . . . affiliated mines [of Committee members] in the very near future.” 20 On November 12, 1980, Robert G. Peluso, MSHA Chief of Special Projects, wrote to representatives of the two manufacturers that had received approval for their SCSRs and asked how many SCSRs they would have on hand by December 21, how many devices they could manufacture during each month after that date, and how they planned to transport finished SCSRs to customers.21 Finally, on November 28, *3241980, Assistant Secretary Lagather reported to BCOA that seven mines had been selected for field testing, that testing would begin December 5, 1980, and that testing at three mines would be completed by December 15. While testing at the other four mines would not be completed until January of 1981, Lagather hoped that information from the first three tests would “provide some guidance in implementating the regulations.”22
Although Lagather’s November 28 letter implied that he planned to implement the regulations on schedule, MSHA announced a different position shortly thereafter. On December 5, 1980, MSHA published the order that gave rise to this petition for review.23 In a few brief paragraphs, MSHA stated that since it had not completed field testing SCSRs, and since only a small number of the devices were available,24 it had decided to defer implementation of the regulations until June 21, 1981. This order was preceded by neither notice nor a comment period; MSHA declared that “[i]n view of the imminence of the deadline and the circumstances stated above” it was “impracticable and contrary to the public interest to provide advance notice and opportunity for public comment on this amendment.” December 5 Order, supra note 3, at 80502.
The Council of the Southern Mountains (CSM), a representative of miners in Martin County, Kentucky, together with District 12 of UMWA and the Coal Employment Project, petitioned this court for review of the December 5 order. We permitted the American Mining Congress (AMC), an industry association, to intervene. Because the order deferring implementation of the regulations is now in effect,25 we have treated this case on an expedited basis. We turn now to the two issues that CSM most urgently presses on appeal.26
*325II. NOTICE AND COMMENT
Petitioners’ most substantial claim is that the Secretary improperly issued the December 5 order without prior notice or an opportunity to comment. The Administrative Procedure Act (APA) requires agencies to give “[gjeneral notice of proposed rule making” and to provide “interested persons an opportunity to participate in the rule making through submission of written data, views, or arguments with or without opportunity for oral presentation.” 5 U.S.C. § 553(b), (c) (1976).27 This notice-and-comment requirement permits interested parties to criticize projected agency action before that action is embedded in a final rule and allows the agency to benefit from the parties’ suggestions. See National Tour Brokers Association v. United States, 591 F.2d 896, 902 (D.C. Cir. 1978).
At oral argument the Government did not dispute that the December 5 order was the type of agency action ordinarily subject to the notice-and-comment procedure.28 The Government contends, however, that 5 U.S.C. § 553(b)(B) excuses its failure to follow the notice-and-comment route. That subsection creates a narrow exception, applicable only when an “agency for good cause finds . . . that notice and public procedure thereon are impracticable, unnecessary, or contrary to the public interest.” 29
We recently admonished agencies that circumstances justifying reliance on this exception are “indeed rare” and will be accepted only after the court has “examine[d] closely proffered rationales justifying the elimination of public procedures.” American Federation of Government Employees v. Block, 655 F.2d 1153, 1157 n.6 (D.C. Cir. 1981). We proceed, therefore, to “examine closely” both the circumstances surrounding the December 5 order and the Secretary’s stated rationale for failing to follow notice- and-comment procedures before issuing that order.
The Secretary’s justification for dispensing with notice and an opportunity to comment was “the imminence of the dead*326line” for implementing the SCSR regulations. This rationale is one that permits avoidance of APA procedures only in exceptional circumstances. Otherwise, an agency unwilling to provide notice or an opportunity to comment could simply wait until the eve of a statutory, judicial, or administrative deadline, then raise up the “good cause” banner and promulgate rules without following APA procedures. Because of this possibility for abuse, “the mere existence of deadlines for agency action . . . [can] not in itself constitute good cause for a § 553(b)(B) exception.” United States Steel Corp. v. United States Environmental Protection Agency, 595 F.2d 207, 213 (5th Cir. 1979), quoted in American Federation of Government Employees, supra, 655 F.2d at 1158.
In this case, however, we do not believe MSHA tarried through the fall of 1980, knowing that the SCSR regulations could not be implemented on time, and then postponed the implementation date at the eleventh hour. Instead, we believe that MSHA, recognizing the life-saving importance of the SCSR rules, was determined to implement the rules on time; that the agency believed through the end of November 1980 that on-schedule implementation was possible; and that only at the beginning of December, when it truly was too late to follow notice-and-comment procedures, did the agency concede that effective implementation of the regulations would be advanced by a relatively short-term postponement. Our conclusion is based on five factors that, in combination, render this a special, possibly unique, case.
First, the circumstances that ultimately forced MSHA to postpone implementation of the regulations were beyond the agency’s control. Field tests were not completed on schedule because of delays in obtaining approval for production model SCSRs and because both mine operators and miners refused to participate in the tests until MSHA answered last-minute safety concerns about SCSRs.30 Sufficient supplies of SCSRs were unavailable in December 1980 because of delays in obtaining approval for production models and because of industry’s failure to order the devices.31 MSHA’s actions caused neither of these problems.
Second, MSHA acted diligently both to initiate field testing and to overcome the hurdles created by other parties. The agency developed an action plan for field tests early in 1979.32 It did not dawdle at the outset and then attempt a rush in the final months. When roadblocks threatened to hold up the field testing program, MSHA reacted resourcefully. It completed some field tests with hand-manufactured SCSRs,33 requested a comprehensive Bureau of Mines study on the safety of SCSRs,34 and performed some field tests in mines that were not owned by members of the resistant BCOA.35
*327Third, the record strongly indicates that, even in November 1980, MSHA intended to implement the regulations on schedule. During the first half of that month a MSHA official wrote to the two manufacturers of approved SCSRs and asked how they intended to transport SCSRs to customers, how many of the devices they would have on hand by December 21, and how many units could be produced each month after that date.36 Moreover, once it finally received industry’s tardy approval for field testing on November 19, MSHA made immediate plans to complete as many of the tests as possible by December 15.37 And Assistant Secretary Lagather, writing to BCOA about these planned tests, indicated that the regulations indeed would be implemented on December 21.38 There is nothing in the record that shows MSHA wavering over the implementation date until the December 5 order itself issued.
Fourth, when the Secretary did postpone the implementation date, he deferred implementation for a relatively short time. If a rule ranks as a substantive regulation,39 the limited nature of the rule cannot in itself justify a failure to follow notice and comment procedures. We have recognized, however, that “[t]he more expansive the regulatory reach of [agency] rules, . . . the greater the necessity for public comment.” American Federation of Government Employees, supra, 655 F.2d at 1156. Conversely, the limited scope of the December 5 order influences our finding that the Secretary possessed good cause to dispense with prior notice and comment.
Finally, Government counsel assured this court at oral argument that field testing has been completed, that an evaluation of the tests should be available shortly, and that implementation of the regulations on June 21 of this year is fully anticipated. While these assurances were made after the December 5 order, they contribute substantially to our impression that MSHA has followed a persistent course toward implementation of the regulations without unnecessary delay. Absent these assurances, we might have entertained some doubt about MSHA’s good faith intention, throughout the fall of 1980, to implement the regulations on schedule. And were we not convinced of that determination, we could not excuse MSHA’s eleventh hour decision to defer implementation without affording interested parties prior notice or an opportunity to comment.
These five factors, taken together, persuade us that the Secretary had good cause to dispense with notice and comment before issuing his December 5 order. We emphasize again, however, that this is an extremely close case. Nothing in this decision diminishes the force- of our repeated admonition that “it should be clear beyond contradiction or cavil that Congress expected, and the courts have held, that the various exceptions to the notice-and-comment provisions of section 553 will be narrowly construed and only reluctantly countenanced.” New Jersey v. United States Environmental Protection Agency, 626 F.2d 1038, 1045 (D.C. Cir. 1980). We recognize an exception to that rule in this case, but we do so guardedly, based on the totality of the special circumstances presented.40
*328III. ARBITRARY AND CAPRICIOUS
Petitioners also contest the reasonableness of the Secretary’s decision to postpone implementation of the regulations. In contrast to the borderline question the notice-and-comment issue presents, we have scant difficulty concluding that the Secretary’s action was reasonable. During the phase-in period numerous parties, including both coal operators and miners, questioned the safety of SCSRs. Against this background the Secretary reasonably determined that field testing should be finished before the regulations were implemented. Completion of the field tests would help the Secretary satisfy any genuine safety concerns raised by parties governed by the regulations; in addition, the test results would enhance his effort to assure the cooperation of all affected parties when SCSRs were finally introduced into the mines.41
The unavailability of SCSRs42 moreover, is a factor relevant to our consideration of the Secretary’s decision to delay implementation. While the Government clarified at oral argument that unavailability of the devices was not an “independent justification” for the deferral, the fact that very few miners would have received SCSRs by December 21 fortified the Secretary’s decision to delay implementation of the regulations.43 As the Government now points out, the unavailability of SCSRs rendered less severe the practical consequences of the Secretary’s decision to defer implementation until field testing was complete.
Finally, we note a point stressed by Government counsel in briefing and arguing this case. As the December deadline approached, MSHA found that guidelines on the storage of SCSRs and the training of miners had not been completed.44 While the Secretary did not rely on this problem in his published notice of deferral, the lack of finished guidelines on these important matters was a factor that could reasonably have influenced the Secretary’s decision to defer implementation.
We regret that the long effort to equip miners with safe, oxygen-generating rescue devices has been prolonged an additional six months. We do not believe, however, that MSHA has been insensitive to the health and safety of underground miners. Satisfactory deployment of SCSRs will demand the cooperation of both coal operators and miners. To assure that cooperation, and to reassure the concerned parties that SCSRs would not introduce additional dangers into the hazardous mine environment, MSHA reasonably decided to postpone implementation of the SCSR regulations from December 21, 1980, to June 21, 1981.
CONCLUSION
Since we find the Secretary’s December 5 order neither procedurally infirm nor arbitrary and capricious, that order is
Affirmed.
. Use of Filter-Type and Self-Contained Self-Rescuers in Underground Coal Mines, 43 Fed. Reg. 54241 (Nov. 21, 1978) [hereinafter cited as 1978 Regulations]. The regulations are codified at 30 C.F.R. §§ 75.1714 to .1714-3 (1980).
. No party ever petitioned for review of the 1978 SCSR regulations.
. The order deferring implementation was published at 45 Fed.Reg. 80501 (Dec. 5, 1980) [hereinafter cited as December 5 Order].
. According to petitioners, 89 miners died in coal disasters between 1970 and 1978. Brief of Petitioners at 6 n.*. Before 1970, the number killed each year was higher. In 1968, a year that saw an unusually high number of mine deaths, 309 miners lost their lives. During that year each miner had a one in five hundred chance of perishing in the mines. S.Rep.No. 411, 91st Cong., 1st Sess. 8 (1969).
. Committee on Mine Rescue and Survival Techniques, National Academy of Engineering, Mine Rescue and Survival at 5-6 (Mar. 1970) (final report). These figures do not include 78 miners who died in a 1968 mine explosion in Farmington, West Virginia. The cause of death for 76 of those miners is not known. Id. at 6.
More recent surveys indicate that, between 1970 and 1978, 12 miners died from asphyxiation. See Brief of Petitioners at 6 n.*.
. Federal Coal Mine Health & Safety Act of 1969, Pub.L.No.91-173, tit. III, sec. 317(n), 83 Stat. 742, 787 (codified at 30 U.S.C. § 877(n) (1976)).
. 35 Fed.Reg. 17890, 17928 (Nov. 20, 1970); Brief for the Secretary of Labor at 3.
. When carbon monoxide constitutes 2% of the mine atmosphere, air inhaled through a filter-type self-rescuer reaches a temperature of 200 degrees Fahrenheit. See 1978 Regulations, supra note 1, at 54241.
. For a general discussion of the problems associated with filter-type self-rescuers, see 1978 Regulations, supra note 1, at 54241 — 42.
. 42 Fed.Reg. 59300 (Nov. 16, 1977). The Department of Interior published the proposed rules. As of March 9, 1978, Congress transferred responsibility for administering the coal mine safety and health laws to the Department of Labor. The final rules, as well as the December 5 order discussed in this opinion, were published by the Department of Labor. See 1978 Regulations, supra note 1, at 54241.
. 42 Fed.Reg. 59300, 59302 (Nov. 16, 1977).
. See Action Plan for an Evaluation Program for the Self-Contained Self-Rescuers at 2-3 (Record item B(43)), Joint Appendix (J.A.) at 313-14; id. attachment (“Milestones”) at 1-2, J.A. at 316-17; Brief for the Secretary of Labor at 5-6.
. See Letter from Anthony Robbins (NIOSH Director) to Robert B. Lagather (June 26, 1980), J.A. at 41; Letter from Robert B. Lagather to Congressman John P. Murtha at 1-2 (July 7, 1980), J.A. at 55-56; Letter from Frank R. Lee *322(Supervisory Electrical Engineer, MSHA Approval and Certification Center) and Samuel L. Terry (Supervisory Chemist, Respirator Section, Testing and Certification Branch, NIOSH Division of Safety Research) to Dr. Pasternack (Draegerwerk AG Lubeck) (June 9, 1980) (Record item E(49)); Letter from Ralph J. Touch, Jr. (Acting Chief, Testing and Certification Branch, NIOSH Division of Safety Research) to T.D. McConnell (Mine Safety Appliances Co.) (July 24, 1980) (Record item E(51)); Letter from Anthony Robbins to Davitt McAteer (Counsel for Petitioners) at 1 (Dec. 23, 1980), reprinted in Memorandum in Reply to Respondents’ Response to Petitioners’ Motion for Stay, as Attachment D.
At oral argument, counsel for CSM claimed that “commercially produced” SCSRs were approved as early as 1977, and that subsequent extensions of approval were sought only for minor design changes. The record items cited above, however, indicate that ready-to-manufacture SCSRs were not finally approved until June and July of 1980. Moreover, one of the “minor insignificant changes” referred to by CSM’s counsel was a change in the size of the SCSR’s straps. As subsequent tests proved; the straps are an important part of the device and may cause numerous difficulties if not properly designed. See Memorandum from Jerry W. Stengel (Supervisory Physical Scientist, Bureau of Mines) to Robert L. Marovelli (Director, Division of Minerals Health and Safety Technology, Department of Interior) at 5 (Nov. 3, 1980) (reporting that at October 1980 smoke test at Bruceton mine, participants stepped on “excessive straps” and found straps difficult to loosen), J.A. at 219. Alteration of the SCSR’s straps, therefore, may appreciably improve the utility of the device.
.Letter from Robert L. Vines (for the Bituminous Coal Operators’ Association) and Martin Connors (for the UMWA) to Robert B. Lagather at 1 (Feb. 15, 1980), J.A. at 25.
The American Mining Congress (AMC), another industry association and an intervenor in this proceeding, also suggested to MSHA that field tests should not begin until the agency had sponsored an “independent evaluation” of SCSRs. Letter from J. Allen Overton, Jr. (AMC President) to Robert B. Lagather at 1 (Feb. 22, 1980), J.A. at 27.
The safety concerns of both the Joint Industry Committee and the American Mining Congress evidently stemmed from an in-house status report on SCSRs prepared by the Research and Development Department of Conoco Inc. The report concluded that “potential hazards of K02 [the chemical agent used in some SCSRs] in a mine environment have been identified and need to be carefully tested and evaluated.” Mining Research Division, Research and Development Department, Conoco Inc., Technical Memorandum No. 34-1334-6, at 1 (Oct. 22, 1979), J.A. at 15. Joseph O. Cook, MSHA Administrator for Coal Mine Safety and Health, transmitted a copy of the Conoco report to E. Muessig, Deputy Assistant Secretary of Labor for Mine Safety and Health, on February 5, 1980. J.A. at 13.
. Letter from Joseph P. Brennan (BCOA President) to Robert B. Lagather (Nov. 19, 1980), J.A. at 301. The record does not show when UMWA approved participation of its members in field testing. On June 18, 1980, a UMWA representative signed a Joint Industry Committee letter to leaders of BCOA and UMWA. This letter recommended that more safety tests should be performed on SCSRs and, more startling, that filter-type self-rescuers should be used until the government could develop SCSRs small enough to wear on miners’ belts. Letter from Martin Connors (for the UMWA members) and Robert L. Vines (for the BCOA members) to Sam Church, Jr. (UMWA President) and Joseph P. Brennan (BCOA President) at 1-2 (June 18, 1980), J.A. at 36-37. These recommendations were sent to Lagather and other government officials on July 11, 1980. Letter from Sam Church (UMWA President) and Joseph P. Brennan (BCOA President) to Lindsay Norman (Bureau of Mines Acting Director), Robert B. Lagather, and Anthony Robbins (NIOSH Director) (July 11, 1980), J.A. at 58. By November 21, 1980, however, UMWA was protesting strenuously against any delay in implementing the SCSR regulations. Letter from Everett Acord (UMWA Safety Director) to Robert B. Lagather (Nov. 21, 1980), J.A. at 303-05.
. See Memorandum from Robert G. Peluso (MSHA Chief of Special Projects) to Edward Clair (Counsel for Coal Mine Safety and Health *323Standards and Regulations, Office of the Solicitor, Department of Labor) at 2-3 (June 22, 1979), J.A. at 7-8; Brief for the Secretary of Labor at 6 n.5. According to petitioners, MSHA finished this first phase of field testing, which called for MSHA inspectors to wear SCSRs during routine mine inspections, late in 1979. Inspectors wore SCSRs for a total of 683 hours and answered questionnaires about the devices. Brief of Petitioners at 36 n.*.
. See Letter from Robert B. Lagather to Robert L. Vines (BCOA Vice President for Health and Safety) and Everett S. Acord (UMWA Safety Director) at 1 (Nov. 5, 1980), J.A. at 220; R.W. Watson, W.J. Doyak, & A.L. Furno, Evaluation of the Safety of One-Hour Chemical Self Rescuers (PRC Report No. 4294 July 1980), excerpted in J.A. at 42-54. Interim findings by the Bureau of Mines were presented to interested parties on several occasions prior to preparation of the final report and those parties suggested additional tests that could be performed. See Letter from Robert B. Lagather to Congressman John P. Murtha at 2 (July 7, 1980), J.A. at 56; Letter from Robert B. Lagather to Martin Connors (UMWA) and Robert L. Vines (BCOA) (May 21, 1980), J.A. at 35.
The Bureau of Mines tests showed that SCSRs might start one fire in about eight years. Since statistics suggested that mines would experience about 129 fires from other sources during that same period, as well as 500 frictional ignitions of methane at the face, the Bureau of Mines concluded that SCSRs would have “an insignificant impact on the frequency of ignitions and fires already occurring” in mines. R.W. Watson, W.J. Doyak, & A.L. Furno, supra, at 67, J.A. at 53.
. Brief for the Secretary of Labor at 10. These tests, conducted in six mines, were completed by early October 1980. See Letter from Wayne E. Veneman (Chief, MSHA Office of Information) to J. Davitt McAteer (Counsel for Petitioners) at 1 (Dec. 16, 1980), reprinted in Memorandum in Reply to Respondents’ Response to Petitioners’ Motion for Stay, as Attachment G.
. All parties appear to share the understanding that implementation does not mean every miner will be supplied with an SCSR on the critical date. Rather, once the regulations are implemented, mine operators will be impelled to supply whatever devices are available and order additional devices on pain of exposure to statutory penalty. 30 U.S.C. § 820 (Supp. III 1979).
. Letter from Robert B. Lagather to Robert L. Vines (BCOA Vice President for Health and Safety) and Everett S. Acord (UMWA Safety Director) at 1 (Nov. 5, 1980), J.A. at 220.
. Letters from Robert G. Peluso to Thomas McConnell (Mine Safety Appliances Co.) (Nov. 12, 1980), J.A. at 293, 294; Letters from Robert G. Peluso to Wesley Kenneweg (National Mine Service Co. (United States representative for the German Draeger company)) (Nov. 12, 1980), J.A. at 295, 296.
One company, the National Mine Service Co., promptly informed Peluso that it would be able to transport finished SCSRs in compliance with Department of Transportation regulations and that 9000 units would be available for immediate delivery by December 31, 1980. Thereafter, the company expected to produce 1350 to' 1650 units per month, for a total of 16,000 to 20,000 units during 1981. Stárting in 1982, the company expected production to rise to 3000 to 5000 units per month. Letter from Wesley J. Kenneweg (National Mine Service Co.) to R. G. Peluso (Nov. 14, 1980), J.A. at 297; Letter from Wesley J. Kenneweg to Robert G. Peluso (Nov. *32417, 1980), J.A. at 298-99. Responses from the other company, the Mine Safety Appliances Co., are not in the record. According to the Government, Mine Safety Appliances had no units available for delivery in December 1980. Brief for the Secretary of Labor at 13 n.9.
. Letter from Robert B. Lagather to Joseph P. Brennan (BCOA President) at 1 (Nov. 28, 1980), J.A. at 306.
. December 5 Order, supra note 3.
. Only 9000 SCSRs were available for delivery in December 1980. See note 21 supra. Since there are approximately 133,000 underground coal miners in the United States, Brief for the Secretary of Labor at 13 n.9, the number of SCSRs available in December 1980 was woefully inadequate. In their written and oral presentations to this court, petitioners emphasized that manufacturers will remain without incentives to produce SCSRs in quantity until orders are placed for the devices.
Reluctance to place early orders for SCSRs is understandable. At oral argument, counsel for CSM estimated that SCSRs now cost approximately $550 apiece. If coal operators purchase one SCSR for each of the nation’s 133,000 coal miners, the cost will exceed 73 million dollars.
. On January 27, 1981, a motions panel of this court stayed the December 5 order. A few weeks later, however, the court vacated the January 27 stay and left the December 5 order in effect pending expedited review.
.In addition to the issues discussed below, CSM claims that the Secretary impermissibly engaged in ex parte contacts with industry representatives during the fall of 1980. We decline either to vacate the December 5 order because of these contacts or to require MSHA to disclose the substance of the communications in the record presented to this court for review. At the time these contacts occurred, MSHA was not engaged in rulemaking. Rather, it was endeavoring to implement a final rule. Successful implementation of a rule may depend on uninhibited exchanges between the agency and the parties affected by the rule. Moreover, given the limited nature of the Secretary’s December 5 order, the stated reasons for that order, and the circumstances under which the order was made, we do not believe disclosure of the ex parte conversations pointed to by petitioners would aid our review.
We find similarly unpersuasive intervenor’s argument that we lack jurisdiction to review the Secretary’s December 5 action. The Government, we note, does not share intervenor’s view. Brief for the Secretary of Labor at 1 n.1. In any case, we believe that 30 U.S.C. § 811(d) (Supp. Ill 1979) clearly gives this court jurisdiction to review the December 5 order. That section authorizes this court to review “mandatory health or safety standard[s] promulgated” by the Secretary of Labor. While the Secretary based his December 5 order on the general grant of rulemaking authority contained in 30 U.S.C. § 957 (1976), the order was in effect an amendment to a mandatory safety standard. Authority to review such an amend*325ment is implicit in the jurisdictional grant of § 811(d). Buttressing our position, “[t]he structure of the [Federal Coal Mine Health & Safety] Act ... makes it quite clear that Congress intended that all legal challenges to the Act, to its enforcement and to any regulations promulgated thereunder be heard by the Federal Courts of Appeals, not by the Federal District Courts.” Bituminous Coal Operators’ Association, Inc. v. Marshall, 82 F.R.D. 350, 352 (D.D.C.1979).
.The Act also requires agencies to publish a final rule “not less than 30 days before its effective date.” 5 U.S.C. § 553(d) (1976). Since the December 5 order was effective immediately, the Secretary may have violated this provision of the APA. The Federal Coal Mine Health & Safety Act, however, provides that “[a]ny mandatory health or safety standard ... shall be effective upon publication in the Federal Register unless the Secretary specifies a later date.” 30 U.S.C. § 811(a)(5) (Supp. III 1979). We need not determine whether this provision excused the Secretary’s failure to abide by § 553(d), since petitioners focus their fire on the Secretary’s violation of subsections (b) and (c) of that section, not on any violation of subsection (d). Cf. American Federation of Government Employees v. Block, 655 F.2d 1153, 1155-1159 (D.C. Cir.1981) (discussing the two notice requirements of § 553 and their exceptions).
. Section 553(b)(A) of the APA excepts “interpretative rules, general statements of policy, or rules of agency organization, procedure, or practice” from the notice-and-comment requirement. Intervenor asserts that the December 5 order fell within this exception since it was an “interpretative rule.” The Secretary’s published notice, however, did not invoke § 553(b)(A) and Government counsel told this court at oral argument that the order deferring implementation of the SCSR regulations “would most likely be considered a substantive rule” since it “did affect the rights of the parties involved.” We agree with the Government that the December 5 order was a substantive rule since, by deferring the requirement that coal operators supply life-saving equipment to miners, it had “ ‘palpable effects’ upon the regulated industry and the public in general.” National Helium Corp. v. Federal Energy Administration, 569 F.2d 1137, 1146 (Em.App.1977).
. Section 553(b)(B) mandates that if an agency invokes this good cause exception, it must “incorporate[ ] the finding [of good cause] and a brief statement of reasons therefor in the rules issued.” There is no dispute that MSHA complied with this requirement. See p. 579 supra.
. See notes 13-15 and accompanying text supra.
. See Letter from Anthony Robbins (NIOSH Director) to Davitt McAteer (Counsel for Petitioners) (Dec. 23, 1980), reprinted in Memorandum in Reply to Respondents’ Response to Petitioners’ Motion for Stay, as Attachment D.
. See note 12 and accompanying text supra. See also Memorandum from Robert G. Peluso (MSHA Chief of Special Projects) to Edward Clair (Counsel for Coal Mine Safety and Health Standards and Regulations, Office of the Solicitor, Department of Labor) (June 22, 1979) (describing field test program and inquiring whether it would be necessary to comply with the then Department of Health, Education and Welfare’s “Guidelines for Human Subjects” during the program), J.A. at 5-12.
. See note 16 and accompanying text supra.
. See note 17 supra.
. See note 18 and accompanying text supra. These first two factors suffice to distinguish National Association of Farmworkers Organizations v. Marshall, 628 F.2d 604 (D.C. Cir. 1980), cited by petitioners. There, the Secretary of Labor claimed that the imminent approach of a harvest season gave him good cause to dispense with notice and comment before promulgating regulations. This court, however, rejected the good cause plea in part because it found that “the time pressure posed by the impending harvest season was due in large part to the Secretary’s own delays.” 628 F.2d at 622.
. See note 21 supra.
. See Letter from Robert B. Lagather to Joseph P. Brennan (BCOA President) (Nov. 28, 1980), J.A. at 306.
. Id. See text accompanying note 22 supra.
. The APA excepts “interpretative rules,” statements of policy, and certain procedural rules from the notice-and-comment requirement. See note 28 supra.
. Petitioners argue that even if the Secretary’s failure to allow notice and comment did not violate § 553 of the APA, it contravened § 811 of the Federal Mine Safety and Health Act. Subsection (a) of the latter section prescribes notice-and-comment procedures MSHA must follow when promulgating or revising a mandatory health or safety standard. 30 U.S.C. § 811(a) (Supp. III 1979). We agree that the December 5 order was an amendment to a mandatory safety standard. See note 26 supra. Section 811(a), however, appears to incorporate the provisions of 5 U.S.C. § 553(b) (1976). The good cause exception contained in the latter section, therefore, is operative under 30 U.S.C. § 811(a) as well.
. We note that at oral argument, intervenor AMC stated emphatically that “industry intends to comply with the regulations on June 21.”
. See note 24 supra.
. Petitioners claim that in numerous other cases the Secretary has permitted safety regulations to go into effect, despite the unavailability of devices needed to comply with the regulations, and has then assessed token fines against parties who could show that they had ordered the devices. While it might have been reasonable for the Secretary to follow that course in this case, he was not compelled to do so. Instead, he could reasonably conclude that his energies were better devoted to completion of field testing to put to rest any tenable objections to ordering SCSRs.
.Responsibility for preparing these guidelines had been given to a private engineering firm. By the fall of 1980, these guidelines were still in draft form. See Foster-Miller Associates, Inc., Third Draft of Guidelines for Self-Contained Self-Rescuer Device Storage and Use in Underground Coal Mines (Oct. 23, 1980), J.A. at 78-207b.