In Fogel v. Chestnutt, 533 F.2d 731 (2 Cir. 1975)(Fogel I), cert. denied, 429 U.S. 824, 97 S.Ct. 77, 50 L.Ed.2d 86 (1976), we reversed a judgment of Judge Wyatt in the District Court, 383 F.Supp. 914, for the Southern District of New York which dismissed on *103the merits a derivative action in which two stockholders of American Investors Fund, Inc. (the Fund) sought to recover brokerage commissions and like amounts which allegedly could have been recaptured for the benefit of the Fund if it or an affiliate had become a member of the National Association of Security Dealers, Inc. (NASD) and the Philadelphia-Baltimore-W ashington Stock Exchange (PBW). The defendants were American Investors Corporation which had been investment adviser and manager of the Fund until July, 1966, and it successor Chestnutt Corporation, to which we shall refer collectively as the Adviser, and four of its directors. These were George A. Chestnutt, Jr., who had been the chief figure in both the Fund and the Adviser and owned 47% of the latter’s stock; Stanley L. Sabel, senior vice president, secretary and a director of the Fund, and vice president, secretary, general counsel and a director of the Adviser, and owner of 16% of its stock; Warren K. Greene, vice president and a director of the Fund, and vice president of the Adviser, and owner of .4% of its stock; and John Currier, a director of the Fund and owner of 2% of the Adviser’s stock.
In agreement with the result although not with all the reasoning in the well-known decision of the First Circuit in Moses v. Burgin, 445 F.2d 369, cert. denied, 404 U.S. 994, 92 S.Ct. 532; 30 L.Ed.2d 547 (1971), we held the defendants liable because of their failure adequately to investigate the possibilities of recapture and make a full report to and secure informed and deliberate consideration of this by the independent directors, see 533 F.2d at 749-50, in violation of the Investment Company Act (ICA), contrast Tannenbaum v. Zeller, 552 F.2d 402 (2 Cir.), cert. denied, 434 U.S. 934, 98 S.Ct. 421, 54 L.Ed.2d 293 (1977). We remanded the case to the district court for the determination of damages, with certain instructions on that subject, 533 F.2d at 755 — 57, hereafter discussed. Familiarity with the opinion in Fogel I will be assumed.
The district court referred the determination of damages to Magistrate Schreiber as a special master. He received testimony from an expert witness for plaintiffs and eleven witnesses, several of them experts, for the defendants. The Magistrate largely sustained plaintiff’s claims with respect to recapture of brokerage commissions on business actually transacted by the Adviser on the PBW, of tender offer fees that would have been payable if the Adviser or an affiliate had become a member of the NASD, and of underwriting discounts, commissions and allowances obtainable by having a brokerage affiliate become a member of the underwriting or selling group. However, he rejected plaintiffs’ claims with respect to the recapture of commissions for reciprocal brokerage, a subject with which, as was also the case with underwriting discounts, our opinion in Fogel I did not specifically deal. After deducting added expenses to which the Adviser would have been subject, the Magistrate arrived at a total of damages for the years 1967-74 of $349,013.04 to which there was to be added interest at the legal rates in effect at the end of each year.
Exceptions were filed by both sides and were disposed of by the district court in an opinion reported in 493 F.Supp. 1192 (1980). The district judge accepted, with some downward modifications, the recommendations of the Magistrate favorable to plaintiffs. However, he sustained plaintiffs’ exceptions to the large item relating to reciprocal commissions. The result was a judgment in the amount of $3,919,220, consisting of $2,370,357 of principal and $1,548,863 of interest.
Despite our recognition that defendant Currier was “a special case” and our suggestion that “equity would suggest the imposition of primary liability on the Adviser, which profited from the failure to recapture”, 533 F.2d at 750, the attorney, Clendon H. Lee, who had represented all the defendants in Fogel I, continued to do so in the district court after remand. After noting the relevant passage on this subject in Fogel I, the Magistrate said:
Because all defendants were represented by the same attorney during these proceedings, and no proof of individual re*104sponsibility was introduced, a finding of degrees of fault of the individual defendants is not feasible.
The district court upheld this and directed that judgment be entered holding all defendants jointly and severally liable in the full amount, 493 F.Supp. at 1204.
All defendants have appealed. The defendants other than Currier, while continuing to be represented by Mr. Lee, retained new attorneys who, in addition to challenging the amount of the award, argue that no award was permissible since in their view there is no private cause of action for damages for violations of the ICA other than that provided in § 30(f) which admittedly is not applicable here,1 and the one created by the addition of § 36(b) in 1970.2 We shall sometimes refer to the defendants so represented as “defendants” or “appellants”. Currier at long last has retained independent counsel who, while joining in the arguments made on behalf of the other defendants, place particular emphasis on the unfairness of subjecting Currier, who was not a part of the management group, to exposure to so large a judgment.3 In Part I of this opinion we shall discuss defendants’ attempt to raise at this stage of the case the issue of the existence of a private cause of action for damages under the ICA. Part II will deal with the issues raised in respect of damages. In Part III we shall discuss Currier’s arguments for treatment as a “special case”.
*105I. Liability
Our opinion in Fogel I necessarily assumed that an adviser’s and a director’s breach of fiduciary duty under the ICA gave rise to a private cause of action for damages. Congress had sought to deal with the problem that the relationship between a mutual fund and its investment adviser and underwriters was rife with opportunities for self-dealing, see Moses v. Burgin, supra, 445 F.2d at 376, by requiring a minimum proportion of disinterested directors, § 10(a), and by vesting them with special responsibilities with respect to contracts between the fund and advisers and underwriters, § 15(c). In Fogel I, 533 F.2d at 745, we agreed with Chief Judge Aldrich, writing for the First Circuit in Moses v. Burgin, supra, 455 F.2d at 377, that
If management does not keep these directors informed they will not be in a position to exercise the independent judgment that Congress clearly intended.
There was no occasion for us to discuss the existence of a private cause of action to enforce this duty since we had long since decided the point in Brown v. Bullock, 294 F.2d 415 (2d Cir. 1961), and, doubtless in view of this and the Supreme Court’s decision in J. I. Case Co. v. Borak, 377 U.S. 426, 84 S.Ct. 1555, 12 L.Ed.2d 423 (1964) (§ 14 of Securities Exchange Act), which seemingly confirmed it, defendants’ counsel did not raise the issue in brief or argument or in his petition for reconsideration.4
Defendants now argue that we — and they — were mistaken in the basic assumption that plaintiffs were entitled to recover if they could show violations of the ICA by the defendants, and that decisions of the Supreme Court, particularly Transamerica Mortgage Advisors, Inc. v. Lewis, 444 U.S. 11, 100 S.Ct. 242, 62 L.Ed.2d 146 (1979); see also Cort v. Ash, 422 U.S. 66, 95 S.Ct. 2080, 45 L.Ed.2d 26 (1975); Piper v. Chris-Craft Industries, Inc., 430 U.S. 1, 97 S.Ct. 926, 51 L.Ed.2d 124 (1977); Touche Ross & Co. v. Redington, 442 U.S. 560, 99 S.Ct. 2479, 61 L.Ed.2d 82 (1979); and Kissinger v. Reporters Committee for Freedom of the Press, 445 U.S. 136, 100 S.Ct. 960, 63 L.Ed.2d 267 (1980), although not dealing with the ICA, compel the conclusion that no implied private causes of action for damages exist under that statute. In addition to disputing defendants’ contention, plaintiffs assert that we need not and should not reach its merits. For this they rely both on the preclusive effect of the pretrial order and on the principle of the law of the case.
Defendants’ initial response is that the existence of an implied cause of action under the ICA is jurisdictional and hence may be raised at any time. This argument is simply one more instance of the fallacy, to which courts as well as counsel have not been immune, of confusing the question whether a court has jurisdiction with the question whether a complaint states a claim upon which relief can be granted.5
The complaint based jurisdiction on the ICA, as well as the Investment Advisers Act of 1940 and the Securities Exchange Act of 1934. It thus stated a claim arising under a law of the United States, of which the district court had jurisdiction under 28 U.S.C. § 1331(a), as well as § 44 of the ICA. If in fact the statutes relied upon by the plaintiffs did not authorize recovery of damages because of violations of the sort alleged, the complaint failed to state a claim upon which relief can be granted and was subject to dismissal upon that ground. *106However, as defendants would have been the first to assert if they had prevailed in that position and plaintiffs brought another action for the same relief, the dismissal would have been upon the merits and not for lack of jurisdiction.
This lesson has been taught as often in decision as it has been ignored in argument and dicta. Justice Holmes preached it long ago in The Fair v. Kohler Die and Specialty Co., 228 U.S. 22, 25, 33 S.Ct. 410, 411, 57 L.Ed. 716 (1913), when he wrote:
[W]hen the plaintiff bases his cause of action upon an act of Congress jurisdiction cannot be defeated by a plea denying the merits of this claim. It might be defeated, no doubt, in a case depending on diversity of citizenship by a plea to the citizenship of parties. Interior Construction and Improvement Co. v. Gibney, 160 U.S. 217, 219, 16 S.Ct. 272, 272-73, 40 L.Ed. 401. We are speaking of a case where jurisdiction is incident to a Federal statutory cause of action. Jurisdiction is authority to decide the case either way. Unsuccessful as well as successful suits may be brought upon the act, and a decision that a patent is bad, whether on the facts or the law, is as binding as one that it is good. See Fauntleroy v. Lum, 210 U.S. 230, 235, 28 S.Ct. 641, 642, 52 L.Ed. 1039. No doubt if it should appear that the plaintiff was not really relying upon the patent law for his alleged rights, or if the claim of right were frivolous, the case might be dismissed. In the former instance the suit would not really and substantially involve a controversy within the jurisdiction of the court, Excelsior Wooden Pipe Co. v. Pacific Bridge Co., 185 U.S. 282, 287, 288, 22 S.Ct. 681, 683, 46 L.Ed. 910, and in the latter the jurisdiction would not be denied, except possibly in form. Deming v. Carlisle Packing Co., 226 U.S. 102, 109, 33 S.Ct. 80, 83, 57 L.Ed. 140. But if the plaintiff really makes a substantial claim under an act of Congress there is jurisdiction whether the claim ultimately be held good or bad.
Justice Black reaffirmed the principle, in Bell v. Hood, 327 U.S. 678, 682-83, 66 S.Ct. 773, 776, 90 L.Ed. 939 (1946). Dealing with a case where the district court had dismissed for want of jurisdiction an action in which the plaintiff sought damages for violations of constitutional rights by FBI agents, his opinion stated:
Jurisdiction, therefore, is not defeated as respondents seem to contend, by the possibility that the averments might fail to state a cause of action on which petitioners could actually recover. For it is well settled that the failure to state a proper cause of action calls for a judgment on the merits and not for a dismissal for want of jurisdiction. Whether the complaint states a cause of action on which relief could be granted is a question of law and just as issues of fact it must be decided after and not before the court has assumed jurisdiction over the controversy. If the court does later exercise its jurisdiction to determine that the allegations in the complaint do not state a ground for relief, then dismissal of the case would be on the merits, not for want of jurisdiction. Swafford v. Templeton, 185 U.S. 487, 493, 494, 22 S.Ct. 783, 785-786, 46 L.Ed. 1005; Binderup v. Pathe Exchange, 263 U.S. 291, 305-308, 44 S.Ct. 96, 98-99, 68 L.Ed. 308. The previously carved out exceptions are that a suit may sometimes be dismissed for want of jurisdiction where the alleged claim under the Constitution or federal statutes clearly appears to be immaterial and made solely for the purpose of obtaining jurisdiction or where such a claim is wholly insubstantial and frivolous. (Footnote omitted.) 6
Justice Jackson illuminated the distinction, which the court of appeals had failed to note, in Montana-Dakota Utilities Co. v. Northwestern Public Service Co., 341 U.S. *107246, 249, 71 S.Ct. 692, 694, 95 L.Ed. 912 (1951). Chief Justice Burger did so again in Duke Power Co. v. Carolina Environmental Study Group, Inc., 438 U.S. 59, 71-72, 98 S.Ct. 2620, 2629-30, 57 L.Ed.2d 595 (1978).
Burks v. Lasker, 441 U.S. 471, 99 S.Ct. 1831, 60 L.Ed.2d 404 (1979), is even closer to our case. The Court was there dealing with the applicability of the business judgment rule to a private right of action asserted under the ICA and the Investment Advisers Act. After noting that the courts below assumed that an implied right of action existed under each act, the Court said that it would assume without deciding that respondents had such causes of action — a course which would have been improper if in fact the lower courts had no jurisdiction. The Court’s awareness of this was manifested in fn. 5:
The question whether a cause of action exists is not a question of jurisdiction, and therefore may be assumed without being decided.
It cited for this Mt. Healthy City Board of Education v. Doyle, 429 U.S. 274, 279, 97 S.Ct. 568, 572, 50 L.Ed.2d 471 (1977), which was to substantially the same effect.
Commendably recognizing that the footnote in Burks v. Lasker “is difficult to distinguish”, Reply Brief p. 2, defendants cite a sentence from Kissinger v. Reporters Committee for Freedom of the Press, 445 U.S. 136, 149-50, 100 S.Ct. 960, 968-69, 63 L.Ed.2d 267 (1980), where the Court said, in deciding that there is no private right of action under the Records and Records Disposal Acts, 44 U.S.C. §§ 2901 et seq., 3303 et seq.:
Congress has not vested federal courts with jurisdiction to adjudicate that question upon suit by a private party.
This shows only that the Supreme Court in using “jurisdiction” language can make the same mistake for which it has repeatedly chided lower courts. It is impossible to believe that the Court intended by a mere stroke of the pen to obliterate a distinction that it had consistently drawn for many decades. Rather this was another instance where, as Justice Jackson observed in Montana-Dakota Utilities Co. v. Northwestern Public Service Co., supra, 341 U.S. at 249, 71 S.Ct. at 694:
As frequently happens where jurisdiction depends on subject matter, the question whether jurisdiction exists has been confused with the question whether the complaint states a cause of action.
As Justice Frankfurter once said, “ ‘[¡jurisdiction’ competes with ‘right’ as one of the most deceptive of legal pitfalls.” City of Yonkers v. United States, 320 U.S. 685, 695, 64 S.Ct. 327, 333, 88 L.Ed. 400 (1944) (dissenting). Thus concluding that the question whether there is an implied cause of action for damages under the ICA goes to the merits rather than to jurisdiction, we pass on to plaintiffs’ contentions with respect to the effect of the pretrial order and the principle of the law of the case.
On September 4, 1973, the district judge entered a pretrial order pursuant to F.R. Civ.P. 16. The order began by reciting that the pleadings had been deemed amended and supplemented by plaintiffs’ supplemental complaint and that “[pjlaintiffs abandon no issues raised in this supplemental pleading.” (A-71) This was followed by several pages of stipulated facts. Next came a listing of the contentions of plaintiffs and defendants. The latter made no mention of any claim, such as may charitably be deemed to have been asserted in paragraphs 22 and 23 of the answer, see note 5, that the ICA did not authorize private recovery of damages for violations, although defendants did raise numerous issues of law, including an assertion (A-77e) that “[ajt all relevant times defendants and all directors have complied in all respects with the provisions of the Investment Company Act.... ” The order then proceeded to list materials on which the parties would rely and witnesses whom they expected to call.
Plaintiffs’ argument that the pretrial order constituted an abandonment of any defense that there is no private cause of action for violation of the ICA other than those provided in § 30(f) and § 36(b) is forceful. F.R.Civ.P. 16 directs that the pretrial order “when entered controls the sub*108sequent course of action, unless modified at the trial to prevent manifest injustice.” We find no merit in defendants’ argument that Rule 16 applies only to issues of fact. Clause (1) speaks broadly of “[t]he simplification of the issues”, and there can be no better way to do this than to abandon a legal defense which, if sound, would eliminate any occasion for trial. In United States v. Hougham, 864 U.S. 310, 315, 81 S.Ct. 13, 17, 5 L.Ed.2d 8 (1960), the Court spoke of the pretrial order there under consideration as having “conclusively established the issues of fact and law in the case”. In Fogel I we applied this very pretrial order to preclude an argument by defendants that we should make a “threshold” disposition without reaching the merits, on the ground of plaintiffs’ failure to allege with sufficient particularity their efforts to obtain action from the directors or their reasons for failing to do so, as required by F.R.Civ.P. 23.1. We held that point to have been abandoned since, although the answer had denied the paragraph of the complaint which sought to show compliance with F.R.Civ.P. 23.1, “no issue of non-compliance was included in defendants’ 17 contentions listed in the pretrial order.” Hence we regarded “the point as having been abandoned.” 533 F.2d at 738 n.7. We fail to see why the same logic should not apply to paragraphs 22 and 23. It would have been a simple matter to have included these in the list of contentions in the pretrial order or to reserve them as plaintiffs did with respect to allegations of their complaint. The cases cited by defendants, Laird v. Air Carrier Engine Service, Inc., 263 F.2d 948, 953-54 (5 Cir. 1959), and Padovani v. Bruchhausen, 293 F.2d 546, 548 (2 Cir. 1961), are not to the contrary.
If there is any answer to plaintiffs’ contentions with respect to the pretrial order, it must be rather that defendants’ abandonment of the no private cause of action defense, if we generously assume that paragraphs 22 and 23 of the answer adequately alleged this, was due to a justified belief that assertion of this would have been futile in view of our decision in Brown v. Bullock, supra, 294 F.2d 415, the uncontested assumption of an implied cause of action in Rosenfeld v. Black, 445 F.2d 1337 (2 Cir. 1971), cert. denied, 409 U.S. 802, 93 S.Ct. 24, 34 L.Ed.2d 62 (1972),7 the First Circuit’s decision in Moses v. Burgin, supra, 445 F.2d 369, and the view with respect to implied causes of action under the securities laws then generally prevailing in light of the Supreme Court’s decision in J. I. Case Co. v. Borak, supra, 377 U.S. 426, 84 S.Ct. 1555,12 L.Ed.2d 423. Rule 16 authorizes a district court to modify a pretrial order “at the trial to prevent manifest injustice.” We assume a court of appeals has similar power, although the party seeking relief must present a much stronger case since the adverse party has been subjected to the burden of a trial, and a still stronger one when an earlier appeal and extensive proceedings with respect to damages have occurred. Defendants scarcely meet this test — particularly since the 1970 amendment to § 36(b) which provides the basis for their principal argument against an implied cause of action under the ICA was available to them at the time of the pretrial conference in 1973 as a ground for seeking to distinguish Brown v. Bullock, which antedated 1970, and Rosenfeld v. Black, which did not involve the receipt of advisory fees, the subject matter of § 36(b).
Defendants are also precluded from now litigating the existence of an implied cause of action for damages for their violations of the ICA because of the principle of the law of the case. We begin by dismissing defendants’ contention that this principle is inapplicable because in Fogel I we did not expressly affirm the existence of a private cause of action. The principle applies as well to everything decided by necessary implication. See, e.g., Walston v. School Board, 566 F.2d 1201, 1205 (4 Cir. 1977); Terrell v. Household Goods Carriers’ Bureau, 494 F.2d 16, 19 (5 Cir.), cert. denied, *109419 U.S. 987, 95 S.Ct. 246, 42 L.Ed.2d 260 (1974). It would be absurd that a party who has chosen not to argue a point on a first appeal should stand better as regards the law of the case than one who had argued and lost.
However, in applying the principle of the law of the case, we must give serious consideration to defendants’ contention that Supreme Court decisions subsequent to Fogel I have demonstrated that the assumption on which we acted without contest from them was in error. We unhesitatingly accept that, as we stated in Zdanok v. Glidden Co., 327 F.2d 944, 951 (2 Cir. 1964), “if, before a case in a district court has proceeded to final judgment, a decision of the Supreme Court demonstrates that a ruling on which the judgment would depend was in error, no principle of ‘the law of the case’ would warrant a failure on our part to correct the ruling.” Defendants’ problem is with the requirement of demonstration. It is not enough, as explained by us in Zdanok, supra, 327 F.2d at 951-53, and by Judge Magruder in White v. Higgins, 116 F.2d 312, 317 (1 Cir. 1940), that defendants could now make a more persuasive argument against the existence of an implied cause of action than we would have thought likely when the case was last here. As Judge Magruder said, “mere doubt on our part is not enough to open the point for full reconsideration.” The law of the case will be disregarded only when the court has “a clear conviction of error” with respect to a point of law on which its previous decision was predicated, Zdanok v. Glidden, supra, 327 F.2d at 953, citing Johnson v. Cadillac Motor Car Co., 261 F. 878, 886 (2 Cir. 1919). We have no such clear conviction that a private action will not lie under the ICA for an omission of the Adviser and interested directors to communicate fully and frankly to disinterested directors with respect to the Adviser’s refusal, in furtherance of its own interests, to take action financially beneficial to the Fund which had been repeatedly called to the attention of the Adviser by the SEC and had been taken by other funds.
Appellants must concede that the Supreme Court has not yet passed on the question whether private causes of action for damages can be implied from the ICA. The closest approach that has been cited to us is Transamerica Mortgage Advisors, Inc. v. Lewis, 444 U.S. 11, 100 S.Ct. 242, 62 L.Ed.2d 146 (1979), which held that no private cause of action for damages can be implied from the Investment Advisers Act of 1940, although a cause of action for equitable relief could be implied. That distinction rested on a combination of two factors: (1) that § 215 of the Act provided only that contracts made in violation of its terms “shall be void” — a term ordinarily taken to mean merely that the person with power to avoid the contract “may resort to a court to have the contract rescinded and to obtain restitution of consideration paid”, 444 U.S. at 18, 100 S.Ct. at 246, and (2) that the jurisdictional section, § 214, earlier drafts of which had incorporated by reference a provision of the Public Utility Holding Company Act of 1935 giving the federal courts jurisdiction “of all suits in equity and actions at law brought to enforce any liability or duty created by” the statute, had been changed during the course of enactment to the final version under which the jurisdiction was limited to “suits in equity to enjoin any violation”, omitting any reference to actions at law or to liability. 444 U.S. at 21-22, 100 S.Ct. at 247-48.8
*110The question of implying a cause of action for damages for violation of the ICA differs in many ways. The two factors expressly relied on in Transamerica to negate a private cause of action for damages do not exist; the jurisdictional section, § 44, like similar sections in other securities laws, e.g., § 27 of the Securities Exchange Act, grants the district courts jurisdiction “of all suits in equity and actions at law brought to enforce any liability or duty created by, or to enjoin any violation of this title or the rules, regulations or orders thereunder.” Section 1(b) of the ICA declares “that the national public interest and the interest of investors are adversely affected — ”
(2) when investment companies are organized, operated, managed, or their portfolio securities are selected, in the interest of directors, officers, investment advisers, depositors, or other affiliated persons thereof, in the interest of underwriters, brokers, or dealers, in the interest of special classes of their security holders, or in the interest of other investment companies or persons engaged in other lines of business, rather than in the interest of all classes of such companies’ security holders ....
The section concluded by declaring that
the policy and purposes of this title, in accordance with which the provisions of this title shall be interpreted, are to mitigate and, so far as is feasible, to eliminate the conditions enumerated in this section which adversely affect the national public interest and the interest of investors.9
The Transamerica Court also noted, 444 U.S. at 20, 100 S.Ct. at 247,
Under each of the securities laws that preceded the Act here in question, and under the Investment Company Act of 1940 which was enacted as companion legislation, Congress expressly authorized private suits for damages in prescribed circumstances.
In contrast to the three court of appeals decisions, all of very recent date, see 444 U.S. at 14 n.4, 100 S.Ct. at 244 n.4, sustaining the availability of a private cause of action under the Investment Advisers Act, the cases in the courts of appeals upholding the private cause of action for damages for violation of the ICA go back to our decision in Brown v. Bullock, supra, in 1961, and include the Third Circuit’s 1963 decision in Taussig v. Wellington Fund, Inc., 313 F.2d 472, 476, cert. denied, 374 U.S. 806, 83 S.Ct. 1693, 10 L.Ed.2d 1031 (1963); the First Circuit’s 1964 decision in Levitt v. Johnson, 334 F.2d 815 (1964), cert. denied, 379 U.S. 961, 85 S.Ct. 649, 13 L.Ed.2d 556 (1965); the Tenth Circuit’s 1968 decision in Esplin v. Hirschi, 402 F.2d 94, 103 (1968), cert. denied, 394 U.S. 928, 89 S.Ct. 1194, 22 L.Ed.2d 459 (1969), and numerous other decisions of these courts of appeals, several subsequent to the 1970 amendment of § 36(b) on which defendants rely as negating an implied cause of action.10 The sole exception to the recognition of an implied cause of action for damages under the ICA was Brouk v. Managed Funds, Inc., 286 F.2d 901 (8 Cir.), cert. granted, 366 U.S. 958, 81 S.Ct. 1921, 6 L.Ed.2d 1252 (1961), judgment vacated and case remanded with instructions to dismiss, 369 U.S. 424, 82 S.Ct. 878, 8 L.Ed.2d 6 (1962), pursuant to a settlement, see 378 F.2d at 793 n.6.11 However, in Greater Iowa Corp. v. McLendon, 378 F.2d 783, 793 (8th Cir. 1967), although not finding it nec*111essary to decide the point, the Eighth Circuit, after quoting from J.I. Case Co. v. Borak, supra, 377 U.S. at 433, 84 S.Ct. at 1560, said that:
In coming down strongly in favor of courts implying private civil remedies where none are expressed, the strong indications are, that if given the opportunity, the Supreme Court would also find an implied civil liability in the Investment Company Act and thereby overrule our opinion in Brouk.
While, as defendants contend, this nose count is not quite so impressive as that with respect to Rule 10b-5 when the Supreme Court upheld an implied cause of action in Superintendent of Insurance v. Bankers Life & Casualty Co., 404 U.S. 6, 92 S.Ct. 165, 30 L.Ed.2d 128 (1971), it is impressive enough.12 Defendants argue that none of the cases upholding private causes of action under the securities laws existed in 1940 when the ICA was enacted, but neither were there such decisions when §§ 10(b) and 14 of the Securities Exchange Act of 1934 were adopted. Actually the case for an implied cause of action under the ICA is stronger than were those in Superintendent of Insurance as regards Rule 10b-5 and in J. I. Case as regards § 14 since the 1934 Act had not been the subject of extensive overhaul when those cases were decided, whereas the ICA was significantly amended in 1970 and again in 1975, long after the implied cause of action under it had been recognized. At least this is true unless defendants are right in asserting that the creation in 1970 of a new express remedy in § 36(b) shows a Congressional intention to negate all private causes of action which the courts of appeals had recognized as regards other sections of the ICA.
Although we do not find it necessary to decide the point, we do not think § 36(b) was intended to negate implied causes of action which the courts have found under sections of the Act other than § 36. Indeed, we are already on record to this effect. See Tannenbaum v. Zeller, supra, 552 F.2d at 417. The problem to which § 36(b) was addressed was that with which the SEC had dealt in pages 125-49,154 of its Report on Public Policy Implications of Investment Company Growth (PPI), H.R.Rep.No.2337, 89th Cong., 2d Sess. (1966). This was that advisers’ fees, generally stated as a percentage of the market value of the managed assets, which had been altogether reasonable when a fund was launched, may have become unreasonably high when the fund grew to enormous size. See Jennings & Marsh, Securities Regulation: Cases and Materials, 1394, et seq. (4th ed. 1977). Section 36 of the 1940 Act had proved to be an ineffective vehicle for dealing with this problem since it expressly authorized action only by the SEC and imposed a very high standard of proof, to wit, “gross misconduct or gross abuse of trust”. Although” our decision in Brown v. Bullock sustained a complaint with respect to advisers’ fees as pleading causes of action under other sections of the ICA, it was unlikely that a plaintiff could prove them. See Jennings & Marsh, supra, at 1395-96. Actions in state courts were likewise ineffective since those courts required objecting stockholders to show corporate “waste”. Id. at 1396, citing the well-known case of Saxe v. Brady, 40 Del.Ch. 474, 184 A.2d 602 (1962); PPI at 133-38; Sen.Rep.No.91-184, 91st Cong.2d Sess. 5, reprinted in [1970] U.S.Code Cong. & Admin.News 4897, 4901. In order to enable the courts to deal more effectively with this problem, the SEC proposed a bill amending the Act to provide that advisory fees must be “reasonable” irrespective of approval of the advisory contract by the directors or shareholders, and that shareholders as well as the SEC should be entitled to sue to enforce the new standard. Surely the last thing the SEC intended was the abolition of implied causes of action which the courts had recognized under other sections of the ICA, as the SEC had consistently favored, e.g., in its amicus brief to this court in Brown v. Bullock. That *112§ 36(b) was addressed only to cases where the advisory fee was attacked as such is demonstrated by the provision in § 36(b)(3) whereby any recovery is limited to the amount of compensation received.
As recounted in Jennings & Marsh, supra, at 1396-97, the mutual fund industry bitterly opposed the bill which “languished for over three years” until “largely at the prodding of Congress, the SEC and industry representatives reached a compromise which ultimately was embodied as an amendment to Section 36 of the Investment Company Act.” The authors state, at 1397, that “The language of subsection (b) is a lesson in the art of studied ambiguity in drafting of statutes.” We might agree with defendants that § 36(b), with the severe limitations of subsection (3), constitutes the exclusive remedy insofar as a private claim alleges solely that compensation of an adviser subsequent to June 14, 1972, the effective date of § 36(b), is so excessive in the sense of surpassing any reasonable relation to the services rendered that its payment is a breach of fiduciary duty. However, there is no reason to conclude that, by adopting a modified version of the SEC’s proposal to afford an express private remedy with respect to one problem as to which the 1940 Act had proved ineffective, the 1970 Congress meant to withdraw the implied private cause of action in other areas which had been recognized over the previous decade by four courts of appeals, with the lone dissenting court having indicated strong willingness to reconsider. Tannenbaum v. Zeller, supra, 552 F.2d at 417. The present action is not one “for breach of fiduciary duty in respect of . . . compensation or payments paid by such registered investment company or by the security holders thereof to such investment adviser or person” to which § 36(b) applies. It is rather for breach of the fiduciary duty to make a full disclosure to the independent directors of opportunities available to the Fund, even though a motivation for and the effect of the nondisclosure were to increase the gross fees by stimulating growth of the Fund and the net fees both by this and by obtaining from brokers “research” services which the Adviser would otherwise have had to supply at its own cost.
We are thus far from having “a clear conviction of error”, see Zdanok v. Glidden Co., supra, 327 F.2d at 953, in the assumption in Fogel I, unchallenged at the time, that a private cause of action lay in respect of plaintiffs’ complaints. In adopting a statute intended as a thorough and pervasive regulation of the investment company industry, in part because of the inadequacies of the 1933 and 1934 Acts “to cope with the grave abuses and evils that had developed in some quarters of the investment company business”, Brown v. Bullock, supra, 194 F.Supp. at 217, and in the face of the declaration in § 1, it seems to us highly unlikely that Congress intended that, save for suits to recover short-swing profits, § 30(f), enforcement should be solely the task of the SEC and of the criminal law, and that injured investors should have no recourse in a federal court. While we recognize that the question of the existence of a private cause of action under the ICA has become more debatable than we or the defendants thought in 1975, we thus perceive no justification for departing from the law of the case.13
*113II. Damages
As noted, we ended our opinion in Fogel I by saying, 533 F.2d at 755, that:
In remanding to the district court for the determination of damages, we are bound to give such guidance as we can.
We began by fixing the periods during which liability for various sorts of recapture should exist; appellants voice no complaints with respect to these. We then said, 533 F.2d at 755-57:
The amount for which defendants are to be held liable will depend on the attempt, difficult but ineluctable, of seeking to find what would have been. In deciding this we must mediate between the two principles, given expression in cases arising under the antitrust laws, that while “a defendant whose wrongful conduct has rendered difficult ascertainment of the precise damages suffered by the plaintiff, is not entitled to complain that they cannot be measured with the same exactness and precision as would otherwise be possible,” Eastman Kodak Co. v. Southern Photo Materials Co., 273 U.S. 359, 379, 47 S.Ct. 400, 405, 71 L.Ed. 684 (1927), yet recovery “cannot be had unless it is shown, that, as a result of defendants’ acts, damages in some amount susceptible of expression in figures resulted.” Keogh v. Chicago & Northwestern Ry., 260 U.S. 156, 165, 43 S.Ct. 47, 50, 67 L.Ed. 183 (1922). The Moses court held the defendants (other than unaffiliated directors) liable for Crosby’s failure to avail itself of NASD recapture (the only method there being pressed) on all transactions on the PBW and Pacific Coast Exchanges, since any lesser measure of recovery would mean that the business judgment defense, although foreclosed on the issue of liability because of lack of effective disclosure to unaffiliated directors, would be entering through the back- door. In a footnote, 445 F.2d at 385, n.25, the court added that the district court “may well find that at a reasonable interval after a final determination to bank give-ups should have been made,” during which interval the unfeasibility of achieving give-ups except on the PBW and Pacific Coast Exchanges might have been ascertained,” [sic] proper practice would have been, consistent with best execution, to make as many transactions as possible on the two favorable exchanges.”
The Moses result may appear somewhat harsh, particularly in a case like this involving a medium-sized no-load fund where there were stronger business reasons against seeking recapture, at least for the period when reciprocals or give-ups to brokers in return for sales efforts were in vogue. However, we think the considerations against allowing defendants to attempt to prove that, after independent investigation by the disinterested directors, the board might reasonably have concluded not to recapture, or at least not to go all out for recapture, see Part IV, supra, similarly foreclose defendants with respect to damages, as long as damages are limited to the business as actually conducted, namely, to tender fees and to those transactions carried out on the PBW and Pacific Coast Exchanges. On the other hand, if plaintiffs desire to pursue the path, noted in fn. 25 to Moses, of urging that the Adviser should have routed more of Fund’s business to the PBW'and Pacific Coast Exchanges, and of endeavoring to estimate what would have been recapturable if it had, compare Weiss v. Chalker, supra, 55 F.R.D. [168] at 171 [(S.D.N.Y.1972)], defendants should be allowed to develop the practical arguments against doing this, although not to relitigate the issues of *114possibility and legality of recapture. If this should seem a departure from strict logic, it would not be the first time this has occurred in the law of damages. (Footnotes omitted.)
Appellants’ first argument with respect to damages is in substance an objection to our precluding, to the extent that we did, the presentation of evidence that recapture would have been adverse rather than beneficial to the Fund. They argue that such preclusion eliminated an essential link in proof of damages, namely, causation, since thoroughly informed independent directors could well have decided against recapture, as the directors of the Chemical Fund did, see Tannenbaum v. Zeller, supra, 552 F.2d 402. We reject the argument. The record revealed an utterly perfunctory presentation of the question of recapture to the independent directors, 533 F.2d at 746-50, whom Congress had mandated in order “to supply an independent check on management and to provide a means for the representation of shareholder interests in investment company affairs”, S.Rep.No.91-184, 91st Cong., 1st Sess. 32 (1969), reprinted in [1970] U.S.Cong. & Admin.News 4897, 4927 (1970). This “caused” the Fund to forgo recapture, by removing the switch which, after proper disclosure, the independent directors might have set so as to send the Fund down the recapture road. To require a plaintiff to prove that the independent directors would not have opted for recapture no matter how full the disclosure would make the disclosure requirement a dead letter. A pertinent analogy exists in the law with respect to misrepresentation. There it is settled that in order to show causation a plaintiff need not prove that the recipient “would not have acted or refrained from acting as he did unless he had relied on the misrepresentation.” Restatement (Second) of Torts § 546, Comment b (1977). In this case failure of the defendants to make an adequate disclosure to independent directors is the legal equivalent of misrepresentation. Justice Blackmun’s observation in Affiliated Ute Citizens v. United States, 406 U.S. 128, 154, 92 S.Ct. 1456, 1472, 31 L.Ed.2d 741 (1972), that “This obligation to disclose and this withholding, of a material fact establish the requisite causation in fact”, is fully applicable here. We reiterate our agreement with Moses v. Burgin, supra, 445 F.2d at 385, that “Management having wrongfully prevented the matter from coming up, must bear the full consequences.”
Apart from the argument just rejected, appellants’ criticisms with respect to damages relate solely to the allowances for underwriting discounts and commissions, and commissions that could have been earned on the PBW Exchange as “reciprocals” for Fund business given to brokers on the NYSE and AMEX. Neither of these was mentioned in the quoted portion of our opinion relating to damages. Thus it became necessary for the Magistrate and the district judge to categorize these claims as either business actually conducted or business rerouted to a regional exchange since Fogei I permitted a business desirability defense only as to the latter.
Plaintiffs’ claim with respect to the recapture of underwriting discounts was that these could readily have been accomplished by establishing a brokerage affiliate with NASD membership and causing the affiliate to become a member of the underwriting or selling group which marketed the stocks in which the Fund invested. As a member the affiliate would have been entitled to commissions on the securities purchased by the Fund without having to do any actual marketing work, and could lawfully have paid such fees in the form of dividends to the Adviser, which would then have been obliged to credit them to the Fund. The Magistrate considered this to be an issue where full inquiry as to desirability, possibility and legality was required. He found possibility by combining our conclusion in Fogei I that NASD membership was available to an affiliate of the Adviser, 533 F.2d at 752, and a finding that the Adviser had sufficient leverage to force its way into the selling group. He found legality on the basis that although the rules of the NASD prohibited a broker acting on *115behalf of a mutual fund to transmit commissions to the fund either directly or by way of crediting an advisory fee, nothing prevented an adviser-owned affiliate from declaring dividends representing underwriting profits, with the Adviser then crediting such dividends to the Fund’s benefit. On the basis of these findings he concluded that at least 25% of the underwriting discounts, commissions and allowances on underwritten securities purchased by the Fund were recapturable. Apparently no evidence was offered that having such an affiliate designated as a member of the selling group would have been an undesirable business practice or that the Fund lacked the necessary leverage. Indeed there was evidence that the Fund had caused the lead underwriter to designate as members brokers who had assisted in the sale of Fund shares. All these conclusions were in line with those that had been reached by Judge Frankel in the case of another fund, Papilsky v. Berndt, [1976-1977] Fed.Sec.L.Rep. ¶ 95,627, at 90,128-29 (S.D.N.Y. June 24, 1976).
The district court considered the subject and in a well reasoned discussion, 493 F.Supp. at 1197-99, agreed with the Magistrate except for thinking that, in light of the expenses and risks of underwriting, an underwriting syndicate might resist according a 25% share to an affiliate bearing none of these, and thus adopted a 20% share as more conservative.
Two specific criticisms are made by appellants with respect to the district judge’s conclusion on this subject. First, they claim that “This category of damages should ... have been open to a full defense.” (Reply Brief, p. 19). The short answer is that the Magistrate so held and there is no indication that the district judge disagreed. Second, they assert that “the way damages were computed for this category was wholly conjectural.” Id. at 19 n.25. However, there was no doubt that some underwriting discounts and commissions could be recaptured, as conceded by defendant Greene and as found in Papilsky v. Berndt, and the district judge’s 20% figure was modest enough.
A more controversial subject is plaintiffs’ claim for reciprocal brokerage. Plaintiffs’ expert, Sherman O. Jones, who had headed a brokerage affiliate of Waddell & Reed, Inc., adviser to United Fund, which had been one of the leaders in recapture, described this practice as follows: Because of the attractiveness of mutual fund business, typically consisting of large amounts of securities, under the then fixed commission rate structure, brokers receiving orders from advisers for execution on the NYSE or AMEX were willing to pay a reward by sharing with affiliates of advisers who were members of regional stock exchanges a portion of the commissions on business which the brokers conducted on such exchanges. In Fogel I, we referred to the portion of the SEC’s PPI report dealing with that subject, 533 F.2d at 736, and also found on our own that at least until the abolition of reciprocals on July 15, 1973, an affiliate of the Adviser as a member of the PBW exchange could “obtain PBW Exchange business from NYSE brokers in return for execution on that exchange”, 533 F.2d at 752. Plaintiffs’ expert testified that only 60% of the commissions payable to brokers for business transacted on the NYSE and AMEX would have provided leverage for reciprocal dealing since the other 40% would be needed as a reward for research or other services. The established practice of reciprocation at the rate of $100 in shared commissions on the PBW for every $200 in commissions for fund transactions executed on NYSE and AMEX would result in recovery of 30% of the total commissions (60% subject to recapture times 50% recapture rate yields 30%).
The Magistrate considered this to be an area where the desirability defense was entertainable. He found that:
Demands to share commissions entailed considerable risks due to the delicately balanced financial and psychological factors inherent in the bargaining process of the securities market, and the possibilities for recapture appear to have been far fewer than plaintiffs contend.
The principal basis stated for this was that:
During the period at issue in this suit only about one in eight or ten mutual *116funds was recapturing although the potential for recapture was generally known (Tr. 245), and although the PBW was actively soliciting members and using the possibility of recapture as a selling point.
He thought that:
This evidence raises an inference that plaintiffs’ arguments in regard to the Fund’s potential for recapture are overly optimistic.
He also criticized plaintiffs’ witness for not analyzing “the Fund’s trading figures in detail to arrive at a reasonably precise proportion of recapturable trade commissions”, concluded that it was “plaintiffs’ burden to elicit from their witnesses the detailed basis of the analysis, if there was any”, and held they had failed to sustain this.
The district judge took a different view, 493 F.Supp. at 1199-203. He disagreed with the Magistrate’s assessment that plaintiffs’ estimates of the possibilities of recapture through reciprocals were overly optimistic. He thought that this conflicted with our opinion in Fogel I and with Judge Frankel’s opinion in Papilsky v. Berndt and with the testimony that in addition to the advisers mentioned in the PPI Report and in our opinion who had pursued this method of recapture, the same device had been employed by Dreyfus Corp., adviser for one of the largest aggregations of mutual funds. He did not think detailed analysis of the Fund’s trading figures was essential or even “meaningful”. He cited testimony of plaintiffs’ witness that beginning in 1969 Waddell & Reed’s affiliate had increased the rate of reciprocal business from 50% to 75% and that over the period from 1966 to 1973 had recaptured 42% of commissions paid on all exchanges. ' However, in the interest of conservatism and because of possibly distinguishing factors, he reduced the percentage of NYSE and AMEX commissions that would be recaptured through reciprocal brokerage from Jones’ 30% figure to 20%. On the basis of Jones’ 2 — 1 formula, this was equivalent to finding that only 40% rather than 60% of the NYSE and AMEX commissions were subject to reciprocity.
We think the Magistrate was mistaken in the legal test he applied to reciprocals. The second category outlined in the quoted passage of our opinion in Fogel I was addressed to possible claims by plaintiffs “that the Adviser should have routed more of Fund’s business to the PBW and Pacific Coast Exchanges, and of endeavoring to estimate what would have been recapturable if it had”. 533 F.2d at 756 (emphasis added). What we had in mind was that the Adviser might not have been able to obtain the best possible execution if large amounts of the Fund’s business had been rerouted to these regional exchanges and that the defendants should be allowed to establish this if they could. The plaintiffs’ claim here at issue did not involve any change in the Fund’s allocation of its own business among the various exchanges. The computations were based on the volume of transactions actually conducted for the Fund by brokers on the NYSE and AMEX and unrelated business done by those brokers on the PBW Exchange. Defendants’ evidence was not that recapture of reciprocals was impracticable but that it was undesirable for various reasons, primarily the obvious one that the brokers executing trades for the Fund on NYSE and AMEX would like to retain full commissions on all their business on the PBW Exchange and thus would give some preference on a variety of matters to funds that allowed them to do this. This was precisely the kind of information that could and should have been placed before the disinterested directors so that they could weigh the intangible advantages of receiving more enthusiasm from brokers both in the execution of orders and in pushing sales and furnishing research against the ascertainable amounts that were obtainable by insisting on reciprocity. If the independent directors had received opinions such as those adduced from defendants’ experts at the damages trial and had then decided not to seek reciprocals, we would have a different case, see Tannenbaum v. Zeller, supra, 552 F.2d at 416-29. Since the Magistrate applied a wrong legal standard, his findings are not, as urged by appellants, protected by the “unless clearly erroneous” provision *117of F.R.Civ.P. 53(e)(2). See, e.g., United Artists Television, Inc. v. Fortnightly Corp., 377 F.2d 872, 874 n.2 (2 Cir. 1967) (Rule 52(a)), rev’d on other grounds, 392 U.S. 390, 88 S.Ct. 2054, 20 L.Ed.2d 1176 (1968); NLRB v. Alterman Transport Lines, Inc., 587 F.2d 212, 220 (5 Cir. 1979); 5A Moore’s Federal Practice K1Í 52.03[2], 53.12[5] (2d ed. 1980) (Rules 52(a), 53(e)(2)).14
We therefore sustain the district court’s computation of damages.
III. Currier
Our opinion in Fogel I recognized special problems with respect to the defendant John Currier. When elected as a director of the Fund in 1962 as an outside business man, he was an “unaffiliated” director, see 533 F.2d at 746 n.15. He remained so after his acquisition of a 2% stock interest in the Adviser in 1964. However, he became an “interested” director as a result of § 5 of the Investment Company' Amendments Act of 1970, which amended § 10(a) to substitute the word “interested” for “unaffiliated” in the requirement for a minimum of 40% of independent directors and defined “interested person”, §§ 2(19)(A)(iii), (B)(iii), so that Currier’s stock interest in the Adviser, which had been reduced to 1.2% by that time, brought him within that term. While this made him ineligible to meet the 40% requirement, nothing in the Act explicitly imposes any special duties on directors who are “interested” but not a part of management.
All the defendants in Fogel I were represented by the same counsel, Clendon H. Lee. No argument was made to us that any difference existed as regards the liability of the Adviser, of the three management directors — Chestnutt, Sabel and Greene, and of Currier. The defense was that there was no liability on the part of anyone since Moses v. Burgin, supra, had been wrongfully decided or in any event was inapplicable to a no-load fund which sold its own shares without the intervention of an affiliated underwriter. However, we did take note of the problem with respect to Currier on our own initiative, saying, 533 F.2d at 750:
We therefore conclude that liability exists with respect to the Adviser and to defendants Chestnutt, Sabel and Greene. Under the principle laid down in Moses, *118supra, 445 F.2d at 384, we would see no basis for imposing liability on directors who were not interested. In this respect defendant Currier is a special case; although he could be counted as unaffiliated until December 14, 1970, because of his small stock ownership in the Adviser he was in fact not disinterested. We therefore include him with the defendants held to be liable. While we do not rule on the apportionment of liability, equity would suggest the imposition of primary responsibility on the Adviser, which profited from the failure to recapture.
Defendants’ petition for reconsideration was addressed solely to the broad issue we had decided; nothing was said as to Currier’s special position.
Although our opinion ought surely to have alerted Lee and Currier, if they had not already been so, of Currier’s need for separate counsel, Lee continued to act for all the defendants. Currier was not called as a witness at the hearings before Magistrate Schreiber. In the course of oral argument held after the submission of evidence, the Magistrate said to Lee:
I do want you to take up the issue that the Chief Judge had noted on page 750 as to the apportionment of the liability as to primary responsibility. If this Court is going to make a ruling on damages, it seems to me it falls upon the Court to accept the responsibility of apportioning the damages or are you of the contrary opinion?
Lee declined to take advantage of the opportunity, saying that “the Court is required to dismiss the case against the individuals.” The Magistrate concluded in his report:
The Second Circuit suggested that primary liability should be imposed on the Advisor, and that defendant Currier’s liability was of a lesser degree than that of the other individual defendants, but it did not rule on the apportionment of liability. Because all defendants were represented by the same attorney during these proceedings, and no proof of individual responsibility was introduced, a finding of degrees of fault of the individual defendants is not feasible.
While by this time it had become crystal clear that Currier needed separate counsel, nothing was done about it,15 and it does not appear that any exception was taken to this portion of the Magistrate’s Report. In any event the district judge concluded, 493 F.Supp. at 1204:
The Master correctly found (Report, p. 4) that no evidence would justify a distinction between the defendants as to the extent of their liability. The judgment will be against all of the defendants in the same amount.
Judgment was entered on June 27, 1980, finding all defendants jointly and severally liable.16
At this time Currier did what should have been done long ago, namely, engage separate counsel. They promptly filed a motion on July 9, 1980, for an order pursuant to F.R.Civ.P. 59 and 52 for a new trial or to alter and amend the judgment “so as to diminish his [Currier’s] liability and subordinate it to that of the other defendants.” The district court denied the motion without opinion.
In their briefs in this court, Currier’s counsel argue that our opinion in Fogel I was in error in holding Currier to be liable for a breach of fiduciary duty. They contend, inter alia, that he was never part of the Adviser’s management and that rather than owing a duty of disclosure to nonmanagement directors as representatives of the *119Fund’s shareholders, he was owed one; that we gave undue weight to the fact that by virtue of the Investment Company Amendments Act of 1970 he became an “interested” director; and that even if we were right in doing so, any liability because of this change should be limited to damages accruing after December 14, 1970. Citing Johnson v. Cadillac Motor Car Co., 261 F. 878, 886 (2 Cir. 1919), they urge that, for these and other reasons, we should reconsider our holding with respect to Currier in Fogel I despite the doctrine of the law of the case. Alternatively they ask us to cause the district court to do what we had expected, namely, to make an equitable apportionment of liability, either by modifying its judgment with respect to Currier or by reversing its denial of Currier’s post-trial motion.
Without passing on the soundness of Currier’s arguments, we at least agree that if these had been presented to us six years ago, there would have had to be much more extensive consideration whether Currier was liable at all than the short paragraph we have quoted. On the other hand, it would be a dangerous precedent to depart from the law of the case simply because new counsel have come up with arguments that could equally well have been presented when the case was first here. Currier does not fit the picture of the guileless innocent that counsel seek to draw. He is a graduate of Penn State University and the Harvard Business School and has had a successful and extensive business career. Our opinion in Fogel I, knowledge of which he does not deny, should have alerted him to the need of having his “special case” presented by petition for reconsideration or, failing that, by offering proof before the Magistrate. Instead he continued to allow himself to be represented by Lee, whom, as we learn from Currier’s affidavit supporting his post-trial motion, was doing this without additional compensation. Beyond this, in light of the assurances given by plaintiffs’ counsel, see note 3, and the absence of any suggestion that the other defendants are unable to pay the judgment, there seems to be little reason to fear that Currier will suffer actual financial loss.
However, we recognize that the very existence of the judgment constitutes a hardship. Although we fully understand the reluctance of the magistrate and the district judge to take action on our suggestion with respect to apportionment in light of Lee’s stubborn refusal to address the issue even after the Magistrate’s request, the record contained sufficient evidence for doing this and we see no reason why we should not do it ourselves rather than prolong this case by a second remand.17 We continue to believe that primary responsibility should rest on the Adviser which profited, or at least was intended to profit, from the failure to recapture, 533 F.2d at 750, see I Palmer, The Law of Restitution §§ 2.11, *1202.12 at 158 (1978), not indeed in the actual receipt of cash but in increased advisory fees due to the sale of shares by brokers who favored the Fund because of its no-recapture policy and by avoiding the expense of conducting research done by the brokers. Of any balance not recoverable from the Adviser, Currier should be liable only in the ratio of his stock ownership in the Adviser in any year to the total stock ownership of all the individual defendants. For reasons indicated in footnote 16, we draw no distinction between damages before and after December 14, 1970. These same principles shall apply to any further judgment awarding plaintiffs expenses and attorneys’ fees.
The judgment is affirmed as so modified. Plaintiffs may recover their costs against defendants other than Currier. No costs as between plaintiffs and Currier.
. This is a short-swing profits provision corresponding to § 16(b) of the Securities Exchange Act of 1934.
. This provides:
For the purposes of this subsection, the investment adviser of a registered investment company shall be deemed to have a fiduciary duty with respect to the receipt of compensation for services, or of payments of a material nature, paid by such registered investment company or by the security holders thereof, to such investment adviser or any affiliated person of such investment adviser. An action may be brought under this subsection by the Commission, or by a security holder of such registered investment company on behalf of such company, against such investment adviser, or any affiliated person of such investment adviser, or any other person enumerated in subsection (a) of this section who has a fiduciary duty concerning such compensation or payments, for breach of fiduciary duty in respect of such compensation or payments paid by such registered investment company or by the security holders thereof to such investment adviser or person. With respect to any such action the following provisions shall apply:
(1) It shall not be necessary to allege or prove that any defendant engaged in personal misconduct, and the plaintiff shall have the burden of proving a breach of fiduciary duty.
(2) In any such action approval by the board of directors of such investment company of such compensation or payments, or of contracts or other arrangements providing for such compensation or payments, and ratification or approval of such compensation or payments, or of contracts or other arrangements providing for such compensation or payments, by the shareholders of such investment company, shall be given such consideration by the court as is deemed appropriate under all the circumstances.
(3) No such action shall be brought or maintained against any person other than the recipient of such compensation or payments, and no damages or other relief shall be granted against any person other than the recipient of such compensation or payments. No award of damages shall be recoverable for any period prior to one year before the action was instituted. Any award of damages against such recipient shall be limited to the actual damages resulting from the breach of fiduciary duty and shall in no event exceed the amount of compensation or payment received from such investment company, or the security holders thereof, by such recipient.
(4) This subsection shall not apply to compensation or payments made in connection with transactions subject to section 80a-17 of this title, or rules, regulations, or orders thereunder, or to sales loans for the acquisition of any security issued by a registered investment company.
(5) Any action pursuant to this subsection may be brought only in an appropriate district court of the United States.
(6) No finding by a court with respect to a breach of fiduciary duty under this subsection shall be made a basis (A) for a finding of a violation of this subchapter for the purposes of sections 80a-9 and 80a-48 of this title, section 78o of this title, or section 80b-3 of this title, or (B) for an injunction to prohibit any person from serving in any of the capacities enumerated in subsection (a) of this section.
. Plaintiffs have assured Currier that execution will not be sought against him until efforts to collect the judgment from other defendants have been exhausted. Brief, pp. 13-14.
. A feeble reference to the point was made in the petition for certiorari, p. 30, where Cort v. Ash, 422 U.S. 66, 95 S.Ct. 2080, 45 L.Ed.2d 26 (1975), which had been handed down three months before Fogel I was argued and six months before it was decided, was cited.
. The confusion is well illustrated by paragraphs 22 and 23 of defendants’ answer:
22. The court lacks jurisdiction over the subject matter of this action in that there is no . .. claim of wrong or injury upon which relief can be granted.
23. The complaint fails to state a claim under the Constitution, Laws or Treaties of the United States, or upon which relief can be granted under the Investment Company Act of 1940 (15 U.S.C. § 80a-l et seq.), the Investment Advisers Act of 1940 (15 U.S.C. § 80b-1 et seq.), and the Securities Exchange Act of 1934 (15 U.S.C. § 78a et seq.) ____
. The Court rejected the position taken by Chief Justice Stone and Justice Burton in dissent, 327 U.S. at 685-86, 66 S.Ct. at 777-78, that where “neither the constitutional provision nor any act of Congress affords a remedy to any person, the mere assertion by a plaintiff that he is entitled to such a remedy cannot be said to satisfy jurisdictional requirements.” This is the precise position taken by the defendants here.
. No question as to the existence of an implied cause of action had been raised by the distinguished counsel representing the defendants in Rosenfeld and our opinion there as in Fogel I did not discuss the subject.
. The Court noted, 444 U.S. at 22-23 n.13, 100 S.Ct. at 248-249 n.13,
In 1975, the Commission submitted a proposal to Congress that would have amended § 214 to extend jurisdiction, without regard to the amount in controversy, to “actions at law” under the Act. See S. 2849, 94th Cong., 2d Sess., § 6 (1976). The Commission was of the view that the amendment also would confirm the existence of a private right of action to enforce the Act’s substantive provisions. See Hearings on S. 2849 before the Subcommittee on Securities of the Senate Committee on Banking, Housing, and Urban Affairs, 94th Cong., 2d Sess., 17 (1976); Hearings on H.R. 12981 and H.R. 13737 before the Subcommittee on Consumer Protection and Finance of the House Committee on Interstate and Foreign Commerce, 94th Cong., 2d Sess., 36-37 (1976). The Senate *110Committee reported favorably on the provision as proposed by the Commission, but the bill did not come to a vote in either House.
. In his elaborate opinion in Brown v. Bullock, 194 F.Supp. 207, 218 (S.D.N.Y.), aff’d, 294 F.2d 415 (2 Cir. 1961), Judge Herlands said of this statément “[t]his is more than a mere exordium. It is a directive to the courts.”
. Although defendants are correct in arguing that Herpich v. Wallace, 430 F.2d 792, 815-16 (5 Cir. 1970), did not explicitly decide the question of a private cause of action for damages under the ICA, the discussion leaves little doubt that the Fifth Circuit would have joined the ranks if decision had been needed.
. In fact, Brouk could have been and was in part rested on much narrower grounds. See 286 F.2d at 912, 918; Brown v. Bullock, 194 F.Supp. at 247.
. What is perhaps most significant about the Superintendent of Insurance case in this context is that the entire discussion of the existence of an implied cause of action was a five-line footnote, 404 U.S. at 13 n.9, 92 S.Ct. at 169 n.9.
. Citing Crane Co. v. American Standard, Inc., (Crane III), 603 F.2d 244 (2 Cir. 1979), appellants argue that departure by a court of appeals from the law of the case may be justified even in the absence of a Supreme Court decision directly demonstrating that the previous ruling of the court of appeals was in error. We have no quarrel with the argument but find it inapplicable. We held in Crane III that our ruling in Crane I, Crane Co. v. Westinghouse Air Brake Co., 419 F.2d 787 (2 Cir. 1969), cert. denied, 400 U.S. 822, 91 S.Ct. 41, 27 L.Ed.2d 50 (1970), that Crane, a tender offeror, possessed a cause of action for monetary damages under §§ 9(e) and 10(b) of the Securities Exchange Act and Rule 10b-5, could not be sustained in the aftermath of Piper v. Chris-Craft Industries, Inc., 430 U.S. 1, 97 S.Ct. 926, 51 L.Ed.2d 124 (1977), which actually dealt with § 14(e). However, as the opinion carefully pointed out, there was simply no rational basis for thinking that the Supreme Court would be more willing to imply a cause of action for a defeated offeror under the general provisions of Rule 10b-5 than under § 14(e), see 603 F.2d at 249. Hence the Piper case was effectively controlling. *113Here, as we have sought to show, it is far from certain that the Supreme Court would hold that plaintiffs had no implied cause of action for damages under the ICA as enacted or that, if they did, their rights were destroyed by the 1970 amendment creating § 36(b). Implied causes of action under the ICA can quite readily be assimilated to those still recognized under § 10-b and Rule 10b-5 and § 14 of the Securities Exchange Act rather than held to be governed by the Transamerica holding under the Investment Advisers Act. Whether they will be, only time will tell.
. Even if the “unless clearly erroneous” test were applicable, there would be sufficient ground for the district judge to have upset the Magistrate’s conclusions. The Magistrate never found that demanding reciprocity for 60%, much less 40%, of the Fund’s NYSE and AMEX business would have been disadvantageous. His conclusions were based rather on his belief that the “possibilities for recapture appear to have been far fewer than plaintiffs’ contend” and on the failure of plaintiffs’ expert to make a detailed analysis of all the Fund’s portfolio transactions over the six year period. The principal ground for his belief that the possibilities of recapture were fewer than plaintiffs contend was his inference from the fact that a relatively small percentage of mutual funds engaged in recapture through insisting on reciprocity. This fact had little meaning in the absence of evidence that the failure of most funds to follow the reciprocal route was due to a considered decision that on balance this would be disadvantageous to the fund rather than what Judge Wyatt characterized as “the selfish and conflicting interests of their advisers which, as in the case of the Adviser here, followed age-old patterns of human nature and preferred to use the ‘fat’ in the NYSE and AMEX commissions for their own benefit rather than for the benefit of the funds.” 493 F.Supp. at 1201. The criticism for failure to make a detailed analysis of the Fund’s transactions ignores the large deductions, 40% in the case of the plaintiff’s expert and 60% in the case of the district judge, for transactions where it would not have been feasible to demand reciprocity. We have found nothing to indicate that these deductions would not have been ample to cover block trades and crosses— the two types of transactions as to which defendants’ witnesses most articulately explained the undesirability of requiring reciprocity. In fact, plaintiffs’ expert indicated that "due to the smaller size of the Fund’s trades as compared to those of United Funds, on which his experience was based, it would have been easier for the Fund to negotiate reciprocal brokerage. We agree with the district judge that in the absence of contention and proof by defendants that there was something unusual about the Fund’s transactions, it would impose too great a burden to require plaintiffs’ expert to examine hundreds of transactions and come to a judgment whether it would have been possible to obtain reciprocity with respect to each. Jones’ extensive experience in doing exactly what plaintiffs asserted entitled him to express an informed judgment without getting into such detail.
. Currier averred in his affidavit in support of a motion for a new trial referred to below that after the Magistrate’s report was issued, Lee advised him that the Magistrate had not recommended any assessment of damages against the individual defendants and that Currier “was, in effect, out of the case.” We are, of course, unable to pass on the truth of this.
. Currier exaggerates the extent to which our holding in Fogel I was based on the 1970 change in the ICA. As the quoted passage shows, our statement was predicated on the point that because of his stock holding in the Adviser, Currier “was in fact not disinterested” —not on the basis that the 1970 amendments resulted in his being classified as “interested” ' rather than unaffiliated.
. None of the parties has argued that imposition of primary responsibility on the Adviser and apportionment of any liability undischarged by it among the individual defendants would go beyond our authority. The recent decisions of the Supreme Court in Northwest Airlines, Inc. v. Transport Workers, 451 U.S. 77, 101 S.Ct. 1571, 67 L.Ed.2d 750 (1981), and Texas Industries, Inc. v. Radcliff Materials, Inc., 451 U.S. 630, 101 S.Ct. 2061, 68 L.Ed.2d 500 (1981), against the imposition of contribution under the Equal Pay Act and Title VII of the Civil Rights Act of 1964, and § 4 of the Clayton Act rested on the basis that the statutes there under consideration expressly created the private damage actions and failed to provide for contribution, although § 11(0 of the Securities Act of 1933 and § 18(b) of the Securities Exchange Act demonstrated that Congress knew how to do this when it wished. The Court left open the question whether federal courts could properly provide for contribution when an implied right of action was found to exist Under the securities laws, Northwest Airlines, 101 S.Ct. at 1580 n.24; Texas Industries, 101 S.Ct. at 2066 n.11. It noted that some courts have implied such a right to contribution, see, e.g., Heizer Corp. v. Ross, 601 F.2d 330, 331-34 (7 Cir. 1979) (and cases cited at 332-33); Globus, Inc. v. Law Research Service, Inc., 318 F.Supp. 955 (S.D.N.Y.1970) (Frankel, J.), aff’d per curiam, 442 F.2d 1346 (2 Cir.), cert. denied, 404 U.S. 941, 92 S.Ct. 286, 30 L.Ed.2d 254 (1971). Here Currier assumes that he would be entitled to contribution if the judgment were left untouched (Reply Brief at 6 n.**) but argues that this remedy is insufficient. The present action is in essence a suit in equity against defendants for violation of their fiduciary duties, and equity courts traditionally have had wide discretion to mold their decrees in the interest of justice, Hecht Co. v. Bowles, 321 U.S. 321, 329-30, 64 S.Ct. 587, 591-92, 88 *120L.Ed. 754 (1944). Even in an action at law for damages under Rule 1 Ob-5, we have said that a court may limit the liability of a class of defendants. Shapiro v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 495 F.2d 228, 242 (2d Cir. 1974).