The State of New York v. The Nuclear Regulatory Commission

WATERMAN, Circuit Judge: This case arises under the National Environmental Policy Act of 1969 (“NEPA”). Plaintiff-appellant State of New York brings a consolidated appeal from three interlocutory orders of the United States District Court for the Southern District of New York, William C. Conner, District Judge. These three orders refused two requests by appellant for preliminary injunc-tive relief, denied appellant’s motion for summary judgment, and granted a motion to dismiss made by two of the defendants-appellees. Appellant, a sovereign state, commenced this civil action on May 5, 1975 on behalf of itself and, as parens patriae, on behalf of all residents and citizens of the State of New York. Named as defendants were seven federal agencies and the chief executive officer of each agency. 1 Appellant, seeking both declaratory and injunctive relief, asserted in its complaint that each of these agencies exercised a power of regulation or of approval over the transportation by air and related connecting transportation 2 of *747 plutonium and other special nuclear material (“SNM”) 3 or, in the case of the Energy Research and Development Administration (“ERDA”), that the agency itself made shipments by air of SNM without having to procure licenses or submit the reports which would be required of a private commercial carrier of such material. All the defendants were alleged to have violated Section 102(2)(C) of NEPA, 42 U.S.C. § 4332(2)(C), 4 by licensing, approving, allowing or executing the transportation by air of plutonium and other SNM without having compiled an environmental impact statement (“EIS”) relating to the environmental consequences of air shipment of SNM into, out of, within or over the City and State of New York and the United States and its territories. In addition to declaratory relief, appellant also sought the issuance of an order annulling any existing licenses, approvals or other actions of the defendants which permitted the transportation by air of plutonium and other SNM into, out of, within or over the City and State of New York and the United States and its territories. The appellant further prayed that the defendants be enjoined from issuing any such licenses or taking any other actions which would permit or would cause such air shipments in the future to be executed. Appellees, while never conceding that an EIS is required in the situation presented here, are in the advanced stages of preparing such a document. After some delay from the originally scheduled date of completion, the projected completion date is now set for early February, 1977. This consolidated appeal consists of appeals from four portions of three interlocutory orders which disposed of two motions for preliminary injunctions, a motion by two defendants to dismiss the complaint as to them and a motion by plaintiff-appellant for summary judgment. The first such order, dated September 9, 1975, denied plaintiff-appellant State of New York’s motion for a preliminary injunction. By this motion plaintiff had sought, pending a disposition on the merits, an injunction annulling all existing, and restraining the issuance of all future, licenses and approvals, and restraining all other actions of defendants-ap-pellees which permit, or would permit, the transportation by air of plutonium and other SNM into, out of, within or over the City and State of New York and the United States and its territories. Appellant had also requested by way of this motion that the court enjoin any actions of defendants-appellees which would cause the transportation by air of SNM into, out of, within or over New York and the United States and its territories. Following the district court’s denial of this motion, appellant filed a timely notice of appeal on November 7, *748 1975. The second order from which appellant has timely appealed, dated December 23, 1975, is one granting the motion of defendants-appellees Civil Aeronautics Board (“CAB”) and United States Customs Service (“Customs”) to dismiss so much of the complaint as is directed against them on the ground that the complaint fails to state a claim upon which relief can be granted against them. Finally, appellant also timely appeals from a third order of the district court, dated May 7,1976. This order denied two motions which had been made by appellant on December 12, 1975, one being a motion for summary judgment and the other a purportedly new motion for a preliminary injunction. The motion for summary judgment had sought both a declaration that appellees were in violation of NEPA and also the establishment of a mandatory timetable for the development and completion of the EIS appellant claims is required in this case. The new motion for a preliminary injunction differed from the first only in that, with respect to one SNM, enriched uranium, the second motion sought to enjoin only the commercial air transportation of this material, leaving the military-assisted air transportation of such material beyond the scope of the requested relief. For the reasons stated in this opinion, we affirm the district court’s refusal to grant the two motions for a preliminary injunction; and we dismiss the appeal from the district court’s grant of Customs’ and CAB’s motions to dismiss the complaint as to them; and we also dismiss the appeal from the district court’s denial of appellant’s motion for summary judgment. Before considering the consolidated appeal, we think it would be helpful to describe briefly the general role some of the defendants play in the regulation and transportation of SNM, the beneficial uses to which SNM is routinely put, and the potential for abuses or accidents involving these materials. The federal agencies named as defendants in this lawsuit are all alleged to be, to some extent or other, involved in the transport of SNM or in the regulation of its transportation. The two agencies whose involvement is most direct are the Nuclear Regulatory Commission (“NRC”) and ERDA which, pursuant to the Energy Reorganization Act of 1974, 42 U.S.C. § 5801 et seq., are the joint successors to the Atomic Energy Commission, the federal agency formerly charged with primary responsibility in the area. The NRC’s functions are regulatory in character and the most important of these is the licensing of importers, exporters and domestic carriers of SNM. These licenses do not, however, require that any particular mode of transportation be utilized. In contrast to the regulatory role played by the NRC, ERDA produces SNM at its own facilities and transports or arranges for the transportation of SNM. In making such shipments, which as a rule are connected with national defense programs, ERDA is not required to be licensed by the NRC. For our present purposes we are concerned only with two varieties of SNM, plutonium and uranium enriched in the isotope U-235. Plutonium is manufactured by subjecting natural uranium to neutron bombardment within the confines of a nuclear reactor. Uranium enriched in the isotope U-235, on the other hand, is produced by increasing in natural uranium the percentage of the isotope U-235 normally found in that element. As to the other SNM specifically mentioned in 42 U.S.C. § 2014(aa), uranium enriched in the isotope U-233, we need not elaborate upon the nature of that material because it is not currently being shipped by air and appellant acknowledges that no relief is being sought with respect to it. Any references in this opinion to enriched uranium should therefore be understood as pertaining only to uranium enriched in the isotope U-235. Plutonium has a number of uses in the fields of military technology, electrical power generation, medical technology, and in research in the areas of nuclear physics and nuclear materials. As a fuel for nuclear reactors, plutonium is being used only experimentally at the present time, but it is used extensively as a power source for *749 heart pacemakers, and as a power source for electric generators installed in spacecraft and deep-sea diving devices. More importantly for purposes of this lawsuit, however, plutonium is a potential primary ngredient for a nuclear weapon and, when ísed in such a device, can produce damage >f frightening proportions. This damage vould not be limited solely to the instantaneous physical devastation resulting from the explosion itself. Plutonium is a highly radiotoxic substance which, when inhaled, can cause lung cancer. The extensive aerial dispersion of plutonium particles caused by the detonation of a nuclear explosive device would therefore present an independent source of injury. Moreover, of serious concern to appellant is the possibility that a deadly dispersion of plutonium particles could conceivably result from some source other than a nuclear explosion. For instance, much of the evidence before Judge Conner concerned the potential consequences, under various hypothetical conditions, which could result from the crash of an aircraft transporting plutonium. Subsequent to the inception of this lawsuit, and undoubtedly with knowledge of the suit and in response 5 to some of the specific concerns expressed by appellant, Congress imposed strict limitations on the air shipment of plutonium. Specifically, § 201 of the NRC Appropriations Act, 89 Stat. 413 (1975) (“NRC Act”) prohibits the NRC from licensing the air transportation of plutonium until such time as the NRC certifies to a particular congressional committee that “a safe container has been developed and tested which will not rupture under crash and blast-testing equivalent to the crash and explosion of a high-flying aircraft.” Exempted from these otherwise absolute prohibitions are air shipments of plutonium “contained in a medical device designed for individual human application.” By way of a similar appropriations act for ERDA, see 89 Stat. 1063 (1975) (“ERDA Act”), Congress imposed similar restrictions on the air shipments of plutonium ERDA is itself authorized to make. The exceptions in this act are somewhat broader than those in the NRC Act, but, nonetheless, the overall impact of the act is clearly to effectuate some reduction in the number of air shipments of plutonium being made. Besides these significant restrictions on the air shipment of plutonium, federal regulations, see 10 C.F.R. §§ 73.30-73.36, require that special security arrangements be undertaken whenever amounts of plutonium exceeding two kilograms are being transported. It should be mentioned that substantially more than two kilograms of plutonium are needed to manufacture even a primitive nuclear bomb. Uranium enriched in the isotope U-235 is the other SNM with which we must concern ourselves on this appeal. If natural uranium is enriched so that it consists of at least 20% U-235 by weight, it is classified as “high enriched”; otherwise, it is designated as “low enriched.” Uranium with a low enrichment of from two to four per cent U-235 is principally used as nuclear reactor fuel to generate electrical energy. Low enriched uranium cannot be used to manufacture nuclear bombs and therefore no special security precautions are required when it is transported. Special security arrangements are required, however, for shipments in excess of five kilograms of uranium with a high enrichment of U-235. There seems to be general agreement between the parties to this lawsuit that the air shipments of high enriched uranium do not present as severe a possible hazard as the air transportation of plutonium. While both could be used to produce bombs, the possible release of high enriched uranium, regardless of the degree of enrichment, which might follow an aircraft accident would not create the hazard of dispersal that release of plutonium might create. This difference in risk seems to be reflected in the congressional reaction to the poten *750 tial dangers involved in the air transportation of the two types of SNM. Although the NRC and ERDA Acts definitely restricted the air carriage of plutonium, they left untouched similar transportation of enriched uranium. Unless permitted pursuant to the exceptions contained in 49 U.S.C. § 1807, however, air shipments of enriched uranium, as well as those of plutonium, are now restricted to all-cargo flights. That statute requires that the United States Department of Transportation promulgate regulations which prohibit the transportation of any radioactive materials (and this includes enriched uranium and plutonium) on any passenger aircraft unless the materials do not pose an unreasonable hazard to health and safety during transportation and are intended for use in research or in medical diagnosis or treatment. The regulations which the statute mandated are now in effect. I With this background in mind, we now proceed to a consideration of the appeal from the district court’s September 9, 1975 denial of appellant’s initial motion for a preliminary injunction. We have frequently recited the standards which are to be applied by a district court in this circuit in determining when such a motion should be granted. To obtain the preliminary relief he seeks the movant must make ‘“a clear showing of either (1) probable success on the merits and possible irreparable injury, or (2) sufficiently serious questions going to the merits to make them a fair ground for litigation and a balance of hardships tipping decidedly toward the party requesting the preliminary relief.’ ” Triebwasser & Katz v. American Telephone & Telegraph Co., 535 F.2d 1356, 1358 (2d Cir. 1976) (emphasis in original), quoting Sonesta International Hotels Corp. v. Wellington Associates, 483 F.2d 247, 250 (2d Cir. 1973); see Pride v. Community School Board, 482 F.2d 257, 264 (2d Cir. 1973). Further, it should be emphasized that such preliminary injunctive relief can be awarded only upon a clear showing that the movant is entitled to the relief, Triebwasser & Katz v. American Telephone & Telegraph Co., supra at 1358; Sonesta International Hotels Corp. v. Wellington Associates, supra at 250, and that in making such a showing the movant bears a heavy burden, Pride v. Community School Board, supra at 264. Moreover, though not evident on its face, the “language of the second prong of the Sonesta test does not eliminate the basic obligation of the plaintiff to make a clear showing of the threat of irreparable harm. That is a fundamental and traditional requirement of all preliminary in-junctive relief . . ..In sum, the balancing of hardships test of Sonesta necessarily includes the showing of irreparable harm.” Triebwasser & Katz v. American Telephone & Telegraph Co., supra at 1359 (emphasis supplied). In the case at bar, Judge Conner assumed for purposes of argument that appellant had made a sufficient showing of likelihood of success on the merits. The judge then determined that the appellant had not established either that there was a threat of irreparable harm or that the balance of hardships tipped in appellant’s favor. However, in view of Triebwasser’s categorical reaffirmation that the threat of irreparable harm is a necessary precondition to the granting of any preliminary injunctive relief, it was not necessary, though it was certainly not ill-advised, for the judge, once he had determined that appellant had not established the threat of irreparable injury, to balance the hardships. Inasmuch as we agree with Judge Conner that appellant has failed to demonstrate by the required showing the threat of irreparable injury, we affirm the district court’s denial of the first motion for a preliminary injunction. Accordingly, we need not discuss appellant’s argument that the balance of hardships tips in its favor so as to have required the issuance of a preliminary injunction. As a preliminary matter, it is not altogether clear what standards of review should govern our consideration of this appeal from the denial of preliminary injunc-tive relief. It is, of course, beyond cavil that the usual rule is that the grant or denial of such relief “will not be disturbed [on appeal] unless there is an abuse of [the *751 trial court’s] discretion ... or unless there is a clear mistake of law”. Triebwas-ser & Katz v. American Telephone & Telegraph Co., supra at 1358. A number of cases in this circuit, however, have taken the position that the appellate court has “broader discretion” and can apply “a stricter standard of review” on an appeal from the grant or denial of a preliminary injunction when the district court’s decision was based not on any testimonial evidence but rather solely on a written record consisting of pleadings, depositions, affidavits and other documents. See, e. g., Munters Corp. v. Burgess Industries, Inc., 535 F.2d 210, 211 n. 4 (2d Cir. 1976); San Filippo v. United Brotherhood of Carpenters & Joiners, 525 F.2d 508, 511 (2d Cir. 1975); Fur Information & Fashion Council, Inc. v. E. F. Timme & Son, Inc., 501 F.2d 1048, 1050-1051 (2d Cir.), cert. denied, 419 U.S. 1022, 95 S.Ct. 498, 42 L.Ed.2d 296 (1974); Dopp v. Franklin National Bank, 461 F.2d 873, 878-879 (2d Cir. 1972). While there may be legitimate criticism concerning the validity of this concept, 6 we find that, in any event, *752 the concept is inapposite in the present case, for what is highly significant about the line of cases in this circuit following Orvis v. Higgins, 180 F.2d 537 (2d Cir.), cert. denied, 340 U.S. 810, 71 S.Ct. 37, 95 L.Ed. 595 (1950) is that the appellate court’s enlarged right of review is a discretionary right and not a duty to burrow into the documentary record in exactly the same way the district court should. As was stated in Pan American World Airways, Inc. v. Aetna Casualty & Surety Co., 505 F.2d 989, 1004 (2d Cir. 1974) (emphasis supplied), “Orvis stands for the proposition that a record consisting only of pleadings, depositions, and affidavits may, at the reviewing court’s discretion, be reviewed de novo.” See Dopp v. Franklin National Bank, supra at 879; San Filippo v. United Brotherhood of Carpenters & Joiners, supra at 511; Munters Corp. v. Burgess Industries, Inc., supra at 211 n. 4. Obviously, the important question then is when should that discretion be exercised. A careful study of Dopp and San Filippo provides some guidance. In Dopp, where the “drastic remedy” of preliminary relief had been granted by the trial court, the question being reviewed by this court was whether the successful movant had made a clear showing of probable success on the merits. Dispositive to the resolution of this broad question was the subquestion of the existence or nonexistence of an alleged oral agreement. The documentary evidence on this point raised a classic question of credibility — one of the alleged participants to the alleged oral agreement denied its existence; the other swore to it. The district judge, without a hearing and by comparing “one piece of paper to another”, Dopp v. Franklin National Bank, supra at 879, quoting Sims v. Greene, 161 F.2d 87, 88 (3d Cir. 1947), “resolve[d] the bitterly disputed facts in favor of the party who [had] the burden of establishing his right to preliminary relief.” 461 F.2d at 879. Unfortunately, in so doing, the judge “did not articulate any reasons for concluding that the victory belonged to” the movant. Id. Thus, the expanded review undertaken in Dopp was triggered, it can be surmised, by the trial court’s grant of the “drastic remedy” of preliminary relief without an articulation of reasons for the court’s acceptance of and reliance upon one of the two conflicting versions of the crucial incident in the case. Moreover, this failure to articulate reasons occurred in the context of a record from which it was impossible to divine any such reasons. In San Filippo also, the district court, which this court noted had applied the wrong legal standard, see 525 F.2d at 512, failed to set forth specific findings of fact to support its generalized finding that there was no threat of irreparable harm there. Id. at 511. Thus, the appellate court, in exercising a broad review, was merely attempting to supply detailed findings which preferably should have been made by the district court below. These cases arguably suggest several very general guidelines to assist us in deciding how broad a review we should employ in the case at bar. First, there is perhaps some hint in Dopp that there is more justification for expanded review when the “drastic remedy” of a preliminary injunction has been granted, rather than denied, on the basis of a documentary record. Second, the breadth of appellate review is to some extent dependent upon the specificity of the findings made by the trial court. As the findings and the reasons supporting them become more specific, detailed and clearly articulated, the scope of review is likely to be narrower. Finally, in both Dopp and San Filippo there were defects, apart from the findings themselves, in the district court proceedings, which this court *753 apparently believed justified a broader review of the lower court’s proceedings than the review the law sets forth shall be given findings of fact examined in connection with an appellate review of the grant or denial of preliminary injunctive relief. None of the factors which we ■ have gleaned from these cases — factors which might influence us to expand our review of Judge Conner’s decision — are present in the appeal before us. The district judge here assiduously and specifically set forth reasons to support his finding that appellant did not establish by the required showing the threat of irreparable harm. In other respects his decision contains no glaring infirmities which might cast suspicion over the entire opinion. Also, of some importance perhaps is the fact that the district court here refused rather than granted the “drastic remedy” of a preliminary injunction. Having decided that under the circumstances there is no reason for us to review the district court’s findings any more comprehensively than we ordinarily would if they were derived from testimonial as well as documentary evidence, we now turn to a consideration of Judge Conner’s finding that appellant failed to establish by the required showing a threat of irreparable harm. We cannot find that the district court abused its discretion in making that finding and in denying the motion for a preliminary injunction; to the contrary, Judge Conner’s memorandum decision denying the motion is thorough and persuasive and, substantially on the basis of the reasoning in that opinion, we affirm the district court’s denial of the first motion for a preliminary injunction. On this appeal, as it did in the district court below, appellant relies on two distinct theories in an attempt to meet its heavy burden of clearly establishing the threat of irreparable harm. Appellant first contends that any violation of NEPA, in and of itself and as a matter of law, constitutes irreparable harm of sufficient magnitude to compel the issuance of a preliminary injunction. In the alternative, appellant asserts that, even if the violation of NEPA which appellant alleges occurred here does not as a matter of law prove the threat of irreparable harm, nonetheless the record developed in this case establishes a substantial enough possibility of irreparable harm to require that a preliminary injunction be awarded. We address these arguments seriatim. Appellant’s first argument must be summarily rejected. It is true, as Judge Conner stated, that appellant has pointed out cases which do appear to support appellant’s position that any NEPA violation constitutes, per se, irreparable harm so as to require the issuance of preliminary injunctive relief, see, e. g., Environmental Defense Fund v. Tennessee Valley Authority, 468 F.2d 1164, 1184 (6th Cir. 1972); Scherr v. Volpe, 466 F.2d 1027,1034 (7th Cir. 1972); Izaak Walton League v. Schlesinger, 337 F.Supp. 287, 295 (D.D.C.1971), but, as appellees correctly assert, the law in this circuit is clear on this issue and directly contrary to the position appellant would have us adopt. In Conservation Society of Southern Vermont, Inc. v. Secretary of Transportation, 508 F.2d 927 (2d Cir. 1974), vacated on other grounds and remanded, 423 U.S. 809, 96 S.Ct. 19, 46 L.Ed.2d 29 (1975), we were confronted with the precise argument appellant advances here and we responded as follows: Although the procedural requirements of NEPA must be followed scrupulously and cost or delay will not alone justify noncompliance with the Act, where the equities require, it remains within the sound discretion of a district court to decline an injunction, even where deviations from prescribed NEPA procedures have occurred. 508 F.2d at 933-934 (footnotes omitted). This stance has been reaffirmed in The East 63rd Street Association v. Coleman, Docket No. 76-6083, at 3 (2d Cir., May 18, 1976) (order), ** in which we explained: “We also note that even if there were violations of the National Environmental Policy *754 Act, which we by no means find, the district court has substantial discretion to determine whether an injunction should issue.” See Greene County Planning Board v. Federal Power Commission, 455 F.2d 412, 424-425 (2d Cir.), cert. denied, 409 U.S. 849, 93 S.Ct. 56, 34 L.Ed.2d 90 (1972). Moreover, the cases offered by appellant in support of its first argument are, as Judge Conner found, distinguishable from the case at bar. One way of distinguishing appellant’s cases is that in each of them the action sought to be enjoined would have produced an irreversible and irretrievable commitment of resources that would have made it virtually certain that the missing EIS, even if eventually completed, would never serve the purpose it was intended to serve — that of insuring that federal projects having a significant effect on the environment might proceed only on the basis of a “careful and informed decision-making process”. Scherr v. Volpe, supra at 1034. This denial of the right of the citizenry to have federal projects which affect the environment proceed only on the basis of “a careful and informed decision-making process” was indeed found to provide sufficient irreparable injury to support the issuance of a preliminary injunction in Scherr v. Volpe, supra at 1034; Environmental Defense Fund v. Tennessee Valley Authority, supra at 1184; and Izaak Walton League v. Schlesinger, supra at 295. Of course, when massive resources are committed to and are expended on a project without any forethought having been given to its environmental impact, modification of that project is unlikely despite the fact that undesirable environmental consequences are disclosed by the EIS when it is ultimately drafted. Scherr v. Volpe, supra at 1034; Environmental Defense Fund v. Tennessee Valley Authority, supra at 1183-1184; Izaak Walton League v. Schlesinger, supra at 294. It is clear that in the case before us there is no imminent irretrievable and irreversible commitment of resources similar to those in the three cases cited, and that here, where the challenged shipments have been made without accident over a period of 25 years, there has been no new extraordinary commitment of resources by the defendants while the preparation of the EIS has been delayed. A second basis upon which at least Scherr and Tennessee Valley Authority are distinguishable is that in each the action sought to be enjoined necessarily would have caused immediate, demonstrable and irreparable damage to the environment. Specifically, in Environmental Defense Fund v. Tennessee Valley Authority, supra at 1183, the “District Court [h]ad found that the activities relating to irreparable defacement of the environment were continuing”. This irreparable defacement consisted of “the cutting and burning of timber, the movement of massive amounts of earth, the construction of large earthworks, and the relocation of roads and bridges.” Id. And in Scherr v. Volpe, supra at 1029, the “district court [had] specifically found that the immediate effects [of the highway conversion project] would include an alteration of natural wildlife habitats, stripping of forest land with attendant erosion, increased levels of noise, air and water pollution, and interference with the natural beauty and recreational value of the area.” There is an obvious difference between the situations in Scherr and Tennessee Valley Authority, on the one hand, and, on the other hand, the situation in the present case. Unlike the situation in either Scherr or Tennessee Valley Authority, the action sought to be enjoined here, the air shipment of SNM, is by no means certain to produce the cataclysmic consequences appellant fears will ensue; indeed, the chance that such consequences will ever eventuate is infinitesimally small. The circumstances under which the “drastic remedy” of a preliminary injunction should be awarded become more apparent if we recall that the purpose of such relief is not to award the movant the ultimate relief sought in the suit but is only to preserve the status quo by preventing during the pendency of the suit the occurrence of that irreparable sort of harm which the movant fears will occur. When trees are felled or massive excavations commenced without the requisite environmental study having been performed and the resulting state *755 ment compiled, it is beyond question that irreparable ecological damage has occurred. In such circumstances, therefore, the preliminary injunction is properly issued because it is necessary to preserve the status quo. The present case is in stark contrast. While it is obvious that the status quo would have been altered in Scherr and Tennessee Valley Authority, here the chances are indeed remote that the feared harm, an accidental or intentional release of radioactive materials, would occur and thereby result in the destruction of the status quo while the case is pending in court. The status quo is destroyed when trees are cut; forests are stripped, soil eroded, and natural wildlife habitats destroyed. The status quo is not destroyed when an aircraft transporting SNM rushes down the runway and becomes airborne. Appellant alternatively argues that not only is there here the threat of irreparable harm per se but the record in this case affirmatively discloses such a threat. The threat is claimed to exist in two forms. First, an aircraft transporting plutonium, the highly radiotoxic character of which is undisputed, might crash accidentally, and a catastrophic dispersal of this deadly radioactive element might occur. The second category of potential injury which appellant fears could occur involves terrorist operations. The terrorists might intentionally shoot down an aircraft carrying plutonium so as to cause a lethal dispersal of the radioactive particles, or an airplane transporting either plutonium or enriched uranium might be hijacked, or the SNM stolen, so that the terrorists could use the materials to make a bomb or some other weapon. Judge Conner found that the threat that any of these or similar terroristic or unsocial events might occur was far too remote to justify him in issuing a preliminary injunction. We agree. Appellant, relying upon Chelsea Neighborhood Associations v. United States Postal Service, 389 F.Supp. 1171 (S.D.N.Y.), aff’d, 516 F.2d 378 (2d Cir. 1975) and Sones-ta International Hotels Corp. v. Wellington Associates, supra, contends that the threat of irreparable harm need only be possible, rather than of some greater certainty, to justify the issuance of a preliminary injunction. To be sure, these two cases do set forth the standard as being one of “possible” irreparable harm. But, in Chelsea, the district court was not concerned about harm which could be characterized, as we believe the harm in the present case can be, as extremely remote. By the court’s own words, the harm in Chelsea, which consisted of the defendant, in effect, having irretrievably committed itself to a construction project by letting a contract, was “imminent.” 389 F.Supp. at 1185-1186. In Sonesta, the harm, the pending consummation of an illegal tender offer, was viewed as irreparable because “once the tender offer has been consummated it becomes difficult, and sometimes virtually impossible, for a court to ‘unscramble the eggs.’ ” 483 F.2d at 250, quoting in part Electronic Specialty Co. v. International Controls Corp., 409 F.2d 937, 947 (2d Cir. 1969). The threat of this harm materializing was more than possible; it was certain and imminent. The case law informs us that the award of preliminary injunctive relief can and should be predicated only on the basis of a showing that the alleged threats of irreparable harm are not remote or speculative but are actual and imminent. Crowther v. Seaborg, 415 F.2d 437, 439 (10th Cir. 1969); Holiday Inns of America, Inc. v. B & B Corp., 409 F.2d 614, 618 (3d Cir. 1969); Capital City Gas Co. v. Phillips Petroleum Co., 373 F.2d 128, 131 (2d Cir. 1967). Relying upon this theory, the Third Circuit has said, in Holiday Inns of America, Inc. v. B & B Corp., supra at 618 (emphasis supplied): We must protect that which is protect-able, but, in so doing, we must limit the use of injunctive relief to situations where it is necessary to prevent immediate and irreparable injury. The dramatic and drastic power of injunctive force may be unleashed only against conditions generating a presently existing actual threat; it may not be used simply to eliminate a possibility of a remote future injury, or a future invasion of rights, be *756 those rights protected by statute or by the common law. The Tenth Circuit expressed a similar philosophy in Crowther v. Seaborg, supra at 439: [The movant’s case] is totally lacking in support of the necessary ingredient of irreparable damages. Such damages, under the undisputed evidence are very remote and not within the realm of probability. As they have in connection with the many other underground nuclear explosions detonated in the past, the Atomic Energy Commission and the other cooperating governmental agencies are exercising the highest degree of care, caution and expertise to prevent any possible damage to life, property and natural resources. This court, also, has commented upon how certain the threat of injury must be to justify the issuance of a preliminary injunction: “The most significant condition which must be present to support the granting of a temporary injunction is a showing that if the relief is not granted irreparable injury during the pendency of the trial will result.” Capital City Gas Co. v. Phillips Petroleum Co., supra at 131 (emphasis added). Judge Conner concluded that on the record before him the alleged threats of harm from either an accidental crash or terrorist activities provided no sufficient basis to justify the issuance of the preliminary injunctive relief sought by appellant. We believe the judge was correct. It is clear that appellant has failed to establish that there is any but the most remote of possibilities that an accidental crash of an airplane transporting SNM will occur or that, even if it did occur, the crash would result in the various catastrophic consequences appellant prophesies. It is agreed by all that the only risk connected with the transportation of uranium enriched in the isotope U-235 is the risk of theft by terrorists because this particular radioactive substance apparently does not display the dispersive characteristics of plutonium. With respect to an accidental or intentionally caused air crash then, only accidents involving aircraft carrying plutonium present any risk of lethal dispersal of radioactivity following impact. Appellant has clearly failed to establish that this risk constitutes a “presently existing actual threat” rather than “a possibility of a remote future injury.” See Holiday Inns of America, Inc. v. B & B Corp., supra at 618. The remoteness of this threat can best be appreciated by examining various factors which when considered sequentially illustrate the patent improbability of the risk. It is uncontested that commercial air shipments of plutonium began about 25 years ago and there has not been a single accident involving any release of plutonium in that entire quarter century. This is not at all surprising because the probability that any particular commercial aircraft will crash is extremely small. For example, although between 1970 and 1973 American commercial carriers logged over I2V2 billion miles, only 32 of these aircraft were destroyed and only 80 were substantially damaged in aircraft accidents. Moreover, all-cargo flights, which by virtue of 49 U.S.C. § 1807 must be used to transport shipments of SNM unless the SNM does not pose an unreasonable hazard to health and safety and is to be used for medical or research purposes, suffered only 3 accidents in 1970 despite traveling 61 million miles, had no accidents in 1971 or in 1972, one accident in 1973 and only two in 1974, but three accidents in four years and only six in five. In addition to the virtual absence of commercial carrier accidents generally, the threat of the irreparable harm resulting from plutonium dispersion following a crash is further diminished by the fact that there is only the most remote probability that one of the relatively few aircraft that do crash will be transporting plutonium at that very time. Even assuming that an aircraft did happen to be transporting plutonium at the time of the accident, one cannot, of course, perforce presume that the SNM cargo will be at all affected. As Judge Conner noted, many crashes in which aircraft are designated as “substantially damaged” or even “destroyed” may not in any way involve *757 injury to the cargo. Damage to the cargo is a product of many factors, including the placement of the cargo in the aircraft, the magnitude and location of stresses in the aircraft following the crash, and the stresses to which the cargo is subjected during the crash. Also tending to reduce the threat of plutonium dispersal upon crash is the fact that, pursuant to United States Department of Transportation regulations, the plutonium can only be transported in containers able to withstand certain specified levels of applied stress. Of further significance is the fact that over 46% of all aircraft accidents occur over water or soft soil. Crashes in these environments are not likely to result in destruction of the containers used to transport and protect the plutonium. Moreover, it cannot seriously be questioned that the recently enacted NRC and ERDA Acts have resulted in an overall reduction in the number of air shipments of plutonium being made. There are factors in addition to those already discussed which further reduce the risk of the catastrophic plutonium dispersal appellant would have us believe might occur. Let us assume for purposes of argument that the following extremely remote possibilities occur simultaneously: 1) an aircraft crashes; 2) the aircraft is one of the miniscule number carrying plutonium; 3) the aircraft is damaged so severely that the cargo is subjected to stresses; and 4) the container protecting the plutonium is unable to withstand the applied forces and it cracks. Even assuming the unlikely occurrence of all these conditions, the disaster appellant foresees happening would be further dependent on where the plane crashed, whether the plutonium was released beyond the fusilage of the aircraft, in what form the plutonium was being transported at the time of the accident, and whether the meteorological conditions were such that a high level of dispersal could be achieved. There may be other conditions which would have to coincide before a catastrophe would occur, but we think it is already eminently obvious that the threat of harm from an accidental air crash is indeed extremely remote. We turn now to Judge Conner’s treatment of the suggested problem of potential terrorist activities, and we do not disturb his findings. The district judge realized that if, as requested by appellant, the court enjoined the air transportation of plutonium or enriched uranium the shipments of these vital materials would then have to be made by surface modes of transportation. It was the court’s opinion that, as to terrorist activities, the shipment of the special nuclear material by air rather than by surface conveyances actually reduces the possibility of a successful terrorist operation directed at possessing or dispersion of the material. This conclusion was predicated on the rather obvious facts that using aircraft to transport the SNM reduces the time during which the material is in transit and also prevents terrorists from gaining access at times other than before the flight has begun and after it has terminated. While the district judge did not mention the possibility that an aircraft could be brought down by a terrorist missile, we cannot find his basic conclusions to be unsound. Having found that an injunction against the air transportation of special nuclear materials would increase rather than decrease the threat of harm, the district court quite properly rejected this alleged basis as a justification for the issuance of any preliminary injunctive relief. It should also be emphasized, as it was before, that the absolute number of air shipments of plutonium has been reduced by virtue of the passage of the NRC and ERDA Acts discussed earlier in this opinion. Moreover, as to both plutonium and enriched uranium, the regulations adopted pursuant to 49 U.S.C. § 1807 prohibit the transportation of any radioactive material (and this includes either enriched uranium or plutonium) on any passenger aircraft unless the materials are intended for use in research or in medical diagnosis or treatment and do not pose a hazard to health or safety during flight. Shipments of SNM which do not fall within the exemptions provided for by the statute and the regulations must be made on all-cargo aircraft where the threat of hijacking would obviously be much less significant. *758 II Appellant next appeals from that portion of the district court’s order of May 7, 1976 which held that the court lacked jurisdiction over appellant’s second motion for a preliminary injunction because that motion was an attempt to reargue the district court’s order denying the first motion for a preliminary injunction, an order from which an appeal had been taken to this court and was pending here. Judge Conner ruled that, although the evidence appellant submitted in support of the renewed request for injunctive relief did not differ substantially from that previously submitted, the four new affidavits which were offered were sufficient to raise a number of essentially factual questions which could be resolved only by holding an evidentiary hearing. Under the circumstances such a hearing, the district court concluded, could be held only with the permission of the court of appeals. We affirm Judge Conner’s refusal to grant the relief sought in appellant’s second motion for a preliminary injunction. At the threshold, we do not agree with the assertion of appellees that this appeal from the aforementioned portion of the district court’s order of May 7, 1976 must be dismissed for lack of appellate jurisdiction. Directing our attention to the most likely source of appellate jurisdiction here, that part of 28 U.S.C. § 1292(a)(1) which provides for jurisdiction over appeals from “interlocutory orders ... refusing . . . injunctions,” appellees argue that the district court’s order “does not contain any order refusing a preliminary injunction.” It is true that the order does not use the word “refuse.” Yet, it is abundantly clear that the practical, and immediate, effect of the order is the refusal of the injunction sought. See Allico National Corp. v. Amalgamated Meat Cutters & Butcher Workmen, 397 F.2d 727, 729 (7th Cir. 1968). We therefore hold that we have jurisdiction to consider the appeal from the district court’s refusal to award injunctive relief. With regard to that refusal, we are of the opinion that Judge Conner acted correctly and, accordingly, we affirm his May 7, 1976 order insofar as that order refused the issuance of a preliminary injunction. The district judge believed that the purported new motion for a preliminary injunction was little more than a poorly disguised effort to reopen and reargue the earlier unsuccessful motion. The judge was clearly correct in this belief. Each of the motions sought the same basic relief and relied upon virtually identical arguments and substantially similar evidence. The fact that some new evidence was offered in support of the second motion does not undermine the thesis that the second motion was, in essence, a request for reconsideration of the court’s denial of the first motion for a preliminary injunction. The four new affidavits were aimed squarely at some of the conclusions Judge Conner had reached in denying the earlier motion. In fact, in the words of two of the affiants, their purpose in submitting their affidavits was to augment and refine prior affidavits filed in connection with the first motion for a preliminary injunction. Moreover, in an affidavit submitted in support of the “new” motion, appellant’s attorney conceded that “prosecution of the appeal from the earlier order may not be necessary” if appellant were victorious on its new motion. Thus construing the later motion to be one seeking to reargue the earlier motion on the basis of the new affidavits, the district court believed that Rule 62(c) of the Federal Rules of Civil Procedure controlled the situation and that under our decision in Ideal Toy Corp. v. Sayco Doll Corp., 302 F.2d 623 (2d Cir. 1962), the district court lacked jurisdiction to tamper in any way with the order then on interlocutory appeal other than to issue orders designed to preserve the status of the case as it sat before the court of appeals. 7 We agree. It is *759 obvious, of course, that if on the basis of the new evidence Judge Conner had reconsidered his denial of the original motion and granted the preliminary relief appellant was again seeking, the judge would have acted in contravention of our holding in Sayco and would have accomplished precisely what Sayco intended to prevent — destruction, rather than preservation of the posture of the case as it sat before the court of appeals. Ill Appellant also appeals from that portion of the district court’s May 7, 1976 order which denied appellant’s motion for summary judgment. Appellees argue that, inasmuch as at this juncture in the lawsuit such an order is not a “final decision” within the meaning of 28 U.S.C. § 1291, this court presently lacks jurisdiction to review the district court order denying the motion for summary judgment. Appellant insists, however, that we do have appellate jurisdiction, claiming that we have the power to review the merits of the entire case whenever an interlocutory appeal under 28 U.S.C. § 1292(a) has properly been brought before us. It is, as appellees contend, well settled in this circuit that an order denying a motion for summary judgment is not a “final decision” within the meaning of 28 U.S.C. § 1291 and such an order is therefore ordinarily not reviewable until a final judgment has been rendered in the case. E.g., Chappell & Co. v. Frankel, 367 F.2d 197 (2d Cir. 1966); John Hancock Mut. Life Ins. Co. v. Kraft, 200 F.2d 952, 953 (2d Cir. 1953); Marcus Breier Sons, Inc. v. Marvlo Fabrics, Inc., 173 F.2d 29 (2d Cir. 1949); Drittel v. Friedman, 154 F.2d 653, 654 (2d Cir. 1946). It is equally as clear, however, that there is authority in this circuit indicating that under some circumstances an interlocutory order which is ordinarily nonappealable may be reviewable if the order is sought to be reviewed in connection with an appellate court’s review of an appealable interlocutory order. For instance, although “[a]s a general rule, when an appeal is taken from the grant or denial of a preliminary injunction, the reviewing court will go no further into the merits than is necessary to decide the interlocutory appeal . . ., this rule is subject to a general exception [that] if its examination of the record upon [the] interlocutory appeal reveals that [either the plaintiff’s or the defendant’s] case is entirely void of merit”, the appellate court may award judgment to the appropriate party. Hurwitz v. Directors Guild of America, Inc., 364 F.2d 67, 70 (2d Cir.), cert. denied, 385 U.S. 971, 87 S.Ct. 508, 17 L.Ed.2d 435 (1966). Hurwitz would thus, in effect, permit, in certain narrow instances, an interlocutory review of denials of motions for summary judgment which are generally not reviewable on an interlocutory basis. Another case exhibiting expansive review on an interlocutory appeal is Sanders v. Levy, Docket No. 75-7608 (2d Cir., June 30, 1976). There, this Court, again in the context of review of an interlocutory order which is by statute or court decision made appealable, recognized a slightly different basis for reviewing an order, such as one denying a motion for summary judgment, which is ordinarily not appealable prior to judgment. The approach taken in Sanders was that if an appealable interlocutory order is before the appellate court, then the court may also review an ordinarily nonappealable ruling made by the district court if “[t]here is sufficient overlap in the factors relevant to both [the issues involved in the appealable and those involved in the nonappealable district court rulings] to warrant [the appellate court’s] exercising plenary authority over” both. Id. at 4589-4590. 8 Somewhat *760 the same notion is expressed by Professor Moore: the appellate court acquires a type of ancillary jurisdiction when an order which is normally nonappealable prior to judgment involves a question basic to or underlying the issue raised by the concededly appealable interlocutory order. 9 Moore’s Federal Practice, 1110.25[1], at 272-273. Obviously Hurwitz and Sanders reflect distinct departures from rigid rules that an appellate court has jurisdiction to review final decisions only, see 28 U.S.C. § 1291, or those interlocutory decisions which are treated as final by case law or are expressly made appealable by statute, see, e.g., 28 U.S.C. § 1292(a). After carefully examining both Hurwitz and Sanders, however, we are convinced that neither case justifies our enlarging the scope of review on the interlocutory appeal properly before us so as to include jurisdiction over the attempted appeal from the order denying appellant’s motion for summary judgment. First of all, there is direct authority in this circuit holding that, despite the presence before us of a proper interlocutory appeal from a denial of a motion for a preliminary injunction, we nonetheless lack jurisdiction to review on an interlocutory basis an order denying a motion for summary judgment. John Hancock Mut. Life Ins. Co. v. Kraft, supra at 953; cf. Drittel v. Friedman, supra at 654. While it may well be that Kraft does not foreclose the possibility of review of a denial of a motion for summary judgment in connection with an interlocutory appeal from the denial of a motion for a preliminary injunction, we would be reluctant to stray from the path followed in Kraft absent subsequent categorical decisions allowing or compelling us to do so. Secondly, Hurwitz and Sanders each represent an exception to the well-established principles restricting the scope of interlocutory review; this is evident in Hurwitz in which it was carefully noted that the “cautious exercise of [the] power [exercised there is not] undesirable”. 364 F.2d at 70 (emphasis supplied). Thirdly, both Hurwitz and Sanders are phrased in discretionary and not mandatory language. In neither did the court believe itself compelled or required to assume an expanded jurisdiction. Rather, the circumstances in each were such that the court concluded that it might, see 364 F.2d at 70, assume jurisdiction, or that plenary jurisdiction was warranted, see Sanders v. Levy, Docket No. 75-7608, at 4590. Finally, in both Hurwitz and Sanders there were significant connections, absent in the case at bar, between the matters which were undeniably reviewable on an interlocutory basis and those which were not. What Hurwitz and Sanders have in common, and what distinguishes them from the present case, is that in each of them the expanded review undertaken by the appellate court required no greater expenditure of effort by that court than if it had strictly confined its review to the interlocutory order which was independently appealable. Specifically, in Sanders there was such an overlap of factors relevant to review of the appealable and the nonappealable district court actions that the appellate court was able to decide the ordinarily nonappealable aspect of the case with minimal, if any, expenditure of judicial effort beyond that required to decide the appealable portion of the case. In Hurwitz the trial court had denied a motion for a preliminary injunction on the dual grounds that the movant had established neither a likelihood of success on the merits nor the threat of any irreparable harm. Thus, if the disappointed movant was to succeed on the interlocutory appeal it took from the denial, the movant had to persuade this court that both of the grounds upon which the trial court based its ruling were erroneous. On the interlocutory appeal this court apparently chose to address first the issue involving the merits of the case, that is, whether there was a likelihood of success on the merits. In the course of its analysis of that issue, this court became convinced that the plaintiffs *761 had proven their case on the merits and the court then awarded judgment to the plaintiffs. Because the disposition made by this court, relief on the merits, was clearly beyond the usual ambit of appellate action on an interlocutory appeal from a denial of a motion for a preliminary injunction, Hur-witz might at first blush appear to be an invitation to wide-ranging appellate review on every interlocutory appeal. There is, however, no justification for attaching such significance to Hurwitz because, although the relief awarded on appeal there may have been unusual, that relief was awarded only as a result of this court’s routine and necessary review of an issue — the likelihood of success on the merits — which was an integral part of the interlocutory appeal there. In disposing of the interlocutory appeal from Judge Conner’s order denying the first motion for preliminary injunction, we have had no occasion to delve into questions related to the merits of the controversy. We agree with the district court’s finding that the appellant has failed to establish the threat of irreparable injury and have concluded that, on that basis alone, the first motion for a preliminary injunction was properly denied. Consideration of the appeal from the denial of the motion for summary judgment, however, would obviously require us to investigate matters to which we have devoted no attention in resolving the interlocutory appeal from the denial of preliminary injunctive relief. See General Motors Corp. v. City of New York, 501 F.2d at 648. Thus, while we certainly cannot disagree with the salutary proposition espoused by Professor Moore that “once a case is lawfully before a court of appeals, it does not lack the power to do what plainly ought to be done,” 9 Moore’s Federal Practice U 110.25[1], at 273, we think it apparent that the analysis upon which we have affirmed Judge Conner’s denial of the first motion for a preliminary injunction does not by any means make it plain what we ought to do if we were to review the denial of the motion for summary judgment. Accordingly, the appeal from that denial is dismissed. IV Finally, we consider the appeal taken from the district court’s grant of the motions to dismiss made by appellees CAB and Customs. As with the appeal from the denial of the motion for summary judgment, appellees assert here also that we lack jurisdiction and that this portion of the consolidated appeal must also be dismissed. We agree. Rule 54(b) of the Federal Rules of Civil Procedure provides that in an action involving multiple parties any order “which adjudicates . . . the rights and liabilities of fewer than all the parties shall not terminate the action as to any of the parties” unless the district court has expressly determined “that there is no just reason for delay” and directed the entry of judgment. As appellees point out, the district court was neither requested to make, nor did it make, the express determination and direction described in Rule 54(b). It is thus manifest that in the present posture of the lawsuit the dismissals of defendants CAB and Customs do not constitute final decisions within the meaning of 28 U.S.C. § 1291. Appellant maintains, as it does in the appeal from the district court’s denial on the motion for summary judgment, that the presence before us of the concededly proper appeal from the denial of the first motion for a preliminary injunction empowers us to review the entire case, including the order granting the motions to dismiss. We believe, however, that the same considerations which militated against our reviewing the denial of the motion for summary judgment apply here as well. As already explained, in reviewing the district court’s denial of appellant’s first motion for a preliminary injunction it has not been necessary for us to consider any factor other than the existence or absence of the threat of irreparable harm. Obviously, as was true in connection with any possible review of the denial of the motion for summary judgment, there is here not only “insufficient overlap” but rather no overlap at all between the factors *762 necessary to resolve the clearly appealable denial of the motion for a preliminary injunction and the ordinarily nonappealable dismissals of fewer that all of the defendants. Under these circumstances there is no justification for our arrogating to ourselves any broader jurisdiction than is specifically conferred upon us by 28 U.S.C. § 1292(a)(1). Accordingly, the appeal from the district court’s grant of CAB’s and Customs’ motions to dismiss the complaint as to them is dismissed. To summarize, we affirm the district court’s order of September 9, 1975 denying appellant’s first motion for a preliminary injunction. We dismiss the appeal taken from the district court’s order of December 23, 1975 granting the motions of appellees Civil Aeronautics Board and United States Customs Service to dismiss the complaint against them. We affirm that portion of the district court’s order of May 7, 1976 which refuses the issuance of a preliminary injunction; we dismiss the appeal from so much of that order as denied appellant’s motion for summary judgment. Footnotes [1] A regional official of Customs was also named as a defendant. [2] For the sake of simplicity, we shall not repeat references to the fact that the term “air transportation” or similar terms used in this opinion includes various forms of related connecting transportation used to bring special nuclear materials to the airport from their place of *747 origin and, after the flight, from the airport to the place of destination. [3] SNM is defined in § 11 (aa) of the Atomic Energy Act of 1954, 42 U.S.C. § 2014(aa) as: “(1) plutonium, uranium enriched in the isotope 233 or in the isotope 235, and any other material which the Commission, pursuant to the provisions of section 2071 of this title, determines to be special nuclear material, but does not include source material; or (2) any material artificially enriched by any of the foregoing, but does not include source material.” [4] 42 U.S.C. § 4332 provides, in pertinent part: “The Congress authorizes and directs that, to the fullest extent possible: (1) the policies, regulations, and public laws of the United States shall be interpreted and administered in accordance with the policies set forth in this chapter, and (2) all agencies of the Federal Government shall— Hi ;)c :fs (C) include in every recommendation or report on proposals for legislation and other major Federal actions significantly affecting the quality of the human environment, a detailed statement by the responsible official on— (i) the environmental impact of the proposed action, (ii) any adverse environmental effects which cannot be avoided should the proposal be implemented, (iii) alternatives to the proposed action, (iv) the relationship between local short-term uses of man’s environment and the maintenance and enhancement of long-term productivity, and (v) any irreversible and irretrievable commitments of resources which would be involved in the proposed action should it be implemented.” [5] See the remarks made by Senator Ribicoff during the Senate debate over the ERDA Act, 121 Cong.Rec. S14603 (daily ed. July 31, 1975), and the remarks made by Congressman Scheuer during the House debates over the NRC Act, 121 Cong.Rec. H5895-5896 (daily ed. June 20, 1975). [6] The rationale for this exception to the general rule is that the “clearly erroneous” test of Rule 52(a) of the Federal Rules of Civil Procedure, which otherwise applies to the findings required by Rule 52 to be made when injunctive relief is granted or denied, is not controlling when the district court’s findings were based exclusively on documentary evidence. This is so, the reasoning goes, because where the credibility of witnesses is not at stake the appellate court is in as good a position as the district court was to draw inferences and to make findings from the written record. See, e. g., San Filippo v. United Brotherhood of Carpenters & Joiners, supra at 511. It must be acknowledged, however, that the idea that the appellate court is not bound by the clearly erroneous standard when reviewing a documentary record has received far less than universal support. First of all, the idea has been categorically rejected by some other circuits, see, e. g., Merrill Trust Co. v. Bradford, 507 F.2d 467, 468 (1st Cir. 1974); Lundgren v. Freeman, 307 F.2d 104, 113-115 (9th Cir. 1962); and has been expressly rejected in cases too numerous to enumerate. Secondly, while it is true that there are many Second Circuit cases which recognize the concept, see, e. g., San Filippo v. United Brotherhood of Carpenters & Joiners, supra at 511; Munters Corp. v. Burgess Industries, Inc., supra at 211 n. 4; Fur Information & Fashion Council, Inc. v. E. F. Timme & Son, Inc., supra at 1051, one of the more recent of these contained a vehement dissent by a distinguished jurist, Justice Tom Clark, sitting by designation in the Second Circuit, see Dopp v. Franklin National Bank, supra at 886-889 (dissenting opinion); the concept has been severely criticized by a renowned former Chief Judge of this circuit, the late Judge Charles E. Clark, who also drafted the Federal Rules of Civil Procedure, see C. Clark, Special Problems in Drafting and Interpreting Procedural Codes and Rules, 3 Vanderbilt L.Rev. 493, 506 (1950) (“Hence we have the [clearly erroneous] rule now so overturned that when the appellate court wishes to apply the policy of nonreviewability of the original rule, it finds it necessary to utter an apology for seeming to violate the rule of the case law.”); and see Heim v. Universal Pictures Co., 154 F.2d 480, 491 (2d Cir. 1946), (Clark, J. concurring opinion); and there are a number of other cases in the Second Circuit which, at the very least, simply ignore the principle, see, e. g., Lamont v. Commissioner of Internal Revenue, 339 F.2d 377, 380-381 (2d Cir. 1964); Weddle v. Commissioner of Internal Revenue, 325 F.2d 849, 851 (2d Cir. 1963) (Friendly, J.); Ideal Toy Corp. v. Sayco Doll Corp., 302 F.2d 623, 624 (2d Cir. 1962); Austin v. Commissioner of Internal Revenue, 298 F.2d 583, 584-586 (2d Cir. 1962). Third, we are not aware of any United States Supreme Court decision which ratifies the idea that the clearly erroneous rule does not apply to review of a documentary record; indeed, it has been stated that “[t]he Supreme Court has never departed from [the] view” that the clearly erroneous test applies to all the district court’s findings, “regardless of the nature of the evidence.” 9 C. Wright & A. Miller, Federal Practice and Procedure: Civil § 2587, at 746, 745 (1971). See United States v. Singer Mfg. Co., 374 U.S. 174, 194 n. 9 (1963); Commissioner of Internal Revenue v. Duberstein, 363 U.S. 278, 291, 83 S.Ct. 1773, 10 L.Ed.2d 823 (1960); United States v. United States Gypsum Co., 333 U.S. 364, 394, 68 S.Ct. 525, 92 L.Ed. 746 (1948). Fourth, at least one of the most prominent authorities in the area of federal practice is highly critical of the notion that findings based on a documentary record need not be clearly erroneous to be overturned, see 9 C. Wright & A. Miller, Federal Practice and Procedure: Civil § 2587, at 740-749 (1971); but see 5A Moore’s Federal Practice Í! 52.04, at 2687-2688 (1975); and the Original Committee Note of 1937 to Rule 52 belies the concept (the clearly erroneous rule “is applicable to all classes of findings in cases tried without a jury whether the finding is of a fact concerning which there was conflict of testimony, or of a fact deduced or inferred from uncontradicted testimony.”) Finally, it can be argued that appellate reliance on such a concept presents an obvious invitation to abuse and miscon *752 ceives the respective functions that should be performed by trial and appellate courts. The traditional and proper role of the appellate court is to correct specific deficiencies in trial court proceedings. Yet, excessively intrusive review by the appellate courts can only encourage unsuccessful litigants to seek to retry their cases in the appellate forum. See Lundgren v. Freeman, supra at 114. Appellate courts were not intended to serve such a function and the mere suggestion that they might so serve cannot but exacerbate the already unwieldly appellate caseload and lessen the confidence that litigants should rightfully have in the federal district courts. [**] [Table reference, 538 F.2d 309]. [7] Judge Conner also recognized that, as interpreted by Sayco, Rule 62(c) also allows a district judge to modify an injunctive order after an appeal has been taken from that order if “the judge is satisfied that his order was erroneous”. Ideal Toy Corp. v. Sayco Doll Corp., *759 supra at 625. Judge Conner was also correct, however, in stating at page 13 of his memorandum and order of May 7, 1976 that modification for this reason may be made only on the basis of “evidence that was before the trial judge at the time he rendered the ruling under appeal.” [8] The importance of such overlap is emphasized in General Motors Corp. v. City of New York, 501 F.2d 639, 648 (2d Cir. 1974) in which *760 we stated: “The guiding principle to inform the discretionary application of pendent jurisdiction is whether review of the appealable order will involve consideration of factors relevant to the otherwise nonappealable order.”