This is an appeal from an order of the District Court for the Southern District of New York dismissing, for want of subject-matter jurisdiction, a Rule 10b-5 action by a Luxembourg investment trust and its liquidators, 462 F.Supp. 209 (1978). It again raises vexing questions with respect to the reach of the anti-fraud provisions of our securities laws with respect to transactions having substantial foreign elements. Here, as in other cases on this subject,1 we are obliged to pick out boundaries as best we can although the statutory language gives little aid.2 The appeal illustrates the infi*913nite variety of situations that may arise; the ground rules we have endeavored to lay down, notably in Bersch, supra, 519 F.2d at 993, although alleged by both sides to be dispositive in their favor, do not lead ineluctably to one result or the other, at least as to one of the transactions here at issue. Decision is further complicated by the fact that, as in Schoenbaum, supra, the issue of subject-matter jurisdiction arises in the context of a situation in which the managers of the defrauded corporation are allegedly implicated in the fraud. Finally, if subject-matter jurisdiction is found to exist, our task is far from ended, since we are then faced with serious issues whether the complaint is sufficient with respect to the defendants who are before us and whether the action is time-barred.
I.
IIT’s Transactions in King-related Securities
Many members of the cast of characters in this case are not new to our courtroom. Plaintiff-appellant IIT, an International Investment Trust, was organized under the laws of the Grand Duchy of Luxembourg in 1961. Before it and its liquidators were forced to spend most of their time in court,3 IIT provided an investment vehicle by which fundholders could participate in a portfolio of securities chosen and managed by allegedly “[qualified professional investment counsel.”4 IIT was controlled and managed by IIT Management Company, S.A. (Management), a Luxembourg corporation, which was in turn controlled by its parent Investors Overseas Services, Ltd. (IOS), first a Panamanian and then a Canadian corporation whose “troubled existence”, see 519 F.2d at 1003, has spawned many actions besides the present one. Both Management and IOS were operated out of Geneva, Switzerland, although plaintiffs allege that “all the top persons” controlling the once vast financial empire were Americans, notably Bernard Cornfeld and Edward M. Cowett. The transactions which form the basis of IIT’s complaint occurred before Cornfeld lost control of IOS to Robert Ves-co.
IIT currently has 144,496 fundholders residing in 154 countries. Some 218 reside in the United States, although it is unclear how many of these are American citizens.5 At the height of its prosperity in the late 1960’s and early 1970’s, IIT held assets worth $375 million, about forty percent of which were in American securities. This prosperity, however, was short-lived. Late in 1972 the Securities and Exchange Commission charged that Vesco was looting the assets of the IOS funds and, in the wake of the resulting scandal, the Grand Duchy of Luxembourg placed all Luxembourg investment funds under supervision of the Bank Control Commissioner. One year later, upon petition of that Commissioner, the Luxembourg district court declared IIT an involuntary bankrupt. Georges Baden, Jacques Delvaux, and Ernest Lecuit, were appointed liquidators of the fund and are co-plaintiffs in this action.
*914The transactions giving rise to the present case, which occurred between January 16 and October 26, 1969, involved three series of acquisitions by IIT of securities related to a complex of companies controlled by one John M. King, an American oil and gas entrepreneur based in Denver. King allegedly controlled King Resources Company (KRC), a publicly traded Maine corporation, and The Colorado Corporation (TCC), a private company largely owned by him. Both the public side of the King complex (KRC) and the private side (TCC) bought and sold natural resource properties and offered a variety of investments in the nature of tax shelters. The two companies had numerous subsidiaries. One of these, King Resources Capital Corporation, N.V. (KRCC), a wholly-owned Netherlands Antilles subsidiary of KRC, figures prominently in this case. Like IIT, King and his companies have fallen on hard times, but were not named as defendants in this action because stays were issued by courts in bankruptcy proceedings involving them.
IIT’s first acquisition of King-related securities occurred between January 16 and October 26, 1969, during which period IIT bought about $8 million face value of KRCC subordinated convertible debentures. The debentures had been issued in Europe on November 27, 1968, to raise $15 million in the eurodollar market. This offering was closely coordinated with a domestic offering of an additional $25 million in debentures of KRC which occurred on November 26. The KRCC debentures were guaranteed by KRC and convertible into KRC common stock. The bulk of IIT’s purchases were made abroad, although IIT alleges it purchased $50,000 face value of the debentures through defendant Arthur Lipper Corporation (Lipper) in the United States. IIT sold its KRCC debentures between July 28, 1970 and February 5,1971 at a loss of $8,765,698.
IIT’s second acquisition of King-related securities was the purchase between January 16 and March 20,1969, of 200,000 shares of KRC common stock. IIT purchased its shares in the United States over-the-counter market for $16.8 million, availing itself of the brokerage services Lipper performed for IIT and the other members of the IOS complex. These shares were sold between October 6 and November 4, 1970, at a loss of approximately $14 million.
IIT’s final acquisition of King-related securities was a July, 1969 purchase of a $12 million 15 year convertible note from TCC. IIT alleges that the purpose of this loan was to make TCC a seemingly attractive merger partner for KRC, although no such merger took place. TCC defaulted on the note and has never paid any principal or interest to IIT. As' noted, TCC is now in bankruptcy.
II.
The Complaint
IIT commenced the instant action by filing a complaint in the District Court for the Southern District of New York on July 17, 1975. The complaint, correctly characterized by Judge Goettel as “flagrantly verbose,” 462 F.Supp. at 212 & n.7, alleged violations of sections 5, 11,12, 15, and 17 of the Securities Act of 1933; section 10(b) of the Securities Exchange Act of 1934. and Rule 10b-5 adopted thereunder; section 206 of the Investment Advisers Act; section 352-c of the General Business Law of New York; breach of fiduciary duty and contract; misappropriation; and failure to disclose these numerous violations of the law. As the briefs and arguments indicate, however, the focus of the complaint is Rule 10b-5. See 462 F.Supp. at 217, 221 & n.25.
According to the complaint and various affidavits and memoranda submitted by plaintiffs in the district court,6 the three *915transactions outlined above were the result of a conspiracy to defraud IIT between those in control of IOS and Management, together with Lipper and those in control of the King complex. The King empire allegedly required “continuous injections of vast sums of cash to survive,” some of which it obtained from IIT’s purchases. For their part, the IOS and Management defendants received personal kickbacks, opportunities to join in KRC tax avoidance schemes, and the ability to over-value King-related assets so as to increase their management fees and performance bonuses;7 Lipper, the United States broker for the IOS complex, was allegedly involved in all three transactions. Its recompense included not only the sizable commissions it gained from the IOS brokerage business but also a special right, allegedly given in connection with the TCC note transaction, to purchase 10,000 shares of TCC stock at what was thought to be a bargain price. Lipper and individuals at Lipper also allegedly partook of KRC investment and tax avoidance schemes involving projects as diverse as the development of Sinai oil properties and the leasing of jet aircraft.
In addition to alleging this general conspiracy to raid IIT for the benefit of the King complex, Lipper, and those in control of IOS and Management, the complaint also charged that the KRCC debenture prospectus was false and misleading in various respects, and that IIT “relied upon the prospectus and other false and misleading statements or nondisclosures in connection with the purchase of” the KRCC debentures and the KRC common stock as well, “or, in the alternative, the defendants are estopped from denying such reliance.” Chief among the misrepresentations and omissions charged were failure to disclose a KRC and TCC fraudulent investment scheme in near-worthless arctic properties; failure to disclose fully transactions with related companies like TCC; failure to disclose the conspiracy with those in control of IOS; failure to state revenues and net income properly and to account properly for expenses; and failure to disclose KRC’s constant need for a vast, steady supply of cash. The complaint does not charge any relation between the allegedly deficient prospectus and the TCC note, although it does allege that certain of the defendants involved with the King complex and TCC failed to disclose important facts in connection with this transaction.
The defendants with which we are here concerned fall into three groups: (1) the international accounting firm of Arthur Andersen & Co. (Andersen); (2) Bear, Stearns & Co., Adams & Peck and Burnham and Company (the underwriter defendants); and (3) Arthur Lipper and Arthur Lipper Corp. (referred to collectively as Lipper), securities brokers for IIT during the period embraced by this complaint. See Arthur Lipper Corp. v. SEC, 547 F.2d 171 (2 Cir. 1976), cert. denied, 434 U.S. 1009, 98 S.Ct. 719, 54 L.Ed.2d 752 (1978).
Andersen is alleged to have “aided and abetted” the aforementioned conspiracy, partly' because of its accounting work on the false and misleading prospectus. Andersen served as the independent certified public accountant for KRC and the IOS complex during the period in question and for IIT from its inception. Andersen terminated its relationship with the IOS companies on May 1, 1971, after the audit work for 1970 was completed. It became auditor for TCC early in 1970 but allegedly had done other work for TCC prior to that time. Andersen did its work for the King complex out of its Denver office, and there audited and certified the financial statements used in both the dollar and eurodollar debenture prospectuses. According to the complaint, Andersen “failed to prepare financial statements for King Resources Company reflecting the true financial condition of said company and failed to insure that full and correct financial information was provided in the Eurodollar prospectus.” IIT also stresses that Andersen’s role as auditors for *916both the King and the IOS companies put. it in a unique position to see the developing relationship between the two groups, yet it never informed either the IIT fundholders or the appropriate regulatory bodies.
The underwriter defendants Bear, Stearns & Co., Adams & Peck, and Burn-ham & Co. and its successors are also alleged to be liable as aiders and abettors of the conspiracy to defraud IIT.8 The three were lower-bracket underwriters with a total participation of $600,000 in the eurodol-lar debenture offering out of the $15 million face value total of the offering and of $750,000 in the dollar debenture offering out of the total of $25,000,000. The complaint charges that the underwriter defendants “knew, or should have known” that the prospectus which they circulated contained the material misrepresentations and omissions discussed above. The complaint further alleges that the underwriter defendants failed to take proper steps to learn the true condition of KRCC and KRC, failed to make a proper investigation of the facts, failed to disclose the involvement of IOS with the King complex, and failed to insure that the proceeds of the eurodollar offering were used for the purposes stated in the prospectus.
The Lipper defendants were charged both as principals and as aiders and abettors, as more fully discussed below.
IIT’s complaint concluded with a request for compensatory damages totalling $35 million and another $35 million in punitive damages, plus attorneys’ fees.
The case came before Judge Goettel on various, motions filed by Andersen, Lipper, and the underwriter defendants. They moved to dismiss for lack of subject-matter jurisdiction under Fed.R.Civ.P. 12(b)(1) and for failure to state a claim upon which relief can be granted under Fed.R.Civ.P. 12(b)(6). They also asserted that plaintiffs’ claims were barred by the statute of limitations, and Andersen moved to dismiss on the ground that plaintiffs lacked capacity to prosecute the action pursuant to Fed.R. Civ.P. 17(b). See 462 F.Supp. at 211 & n.l. In an opinion delivered on December 7, 1978, Judge Goettel denied Andersen’s motion under Rule 17(b),9 but granted the motions to dismiss the complaint as to all defendants for lack of subject-matter jurisdiction. He did not reach the issues of failure to state a claim on which relief can be granted or the statute of limitations. Plaintiffs filed a timely notice of appeal on December 26, 1978.
III.
The District Court’s reasons for finding lack of subject-matter jurisdiction
Judge Goettel began his discussion of subject-matter jurisdiction by rejecting *917the argument that Rule 10b-5 applied to the transactions here in question because of their effects within the United States. Distinguishing Schoenbaum, supra, because the victim in that case was a corporation whose shares were listed on the American Stock Exchange, with a substantial minority of American shareholders, he cited Bersch, supra, as holding that an unparticularized deleterious effect on the American economy from lessened ability to attract offshore investment funds did not provide the necessary effect, 519 F.2d at 989, and Vencap, supra, as holding that such effect was not provided “simply because half of one percent” of the shares of the allegedly defrauded fund were “held by Americans.” 519 F.2d at 1017. In this the judge was clearly right, and we need say no more about “effects” as a basis of subject-matter jurisdiction save in one respect noted in Part V below.
Turning to jurisdiction based on acts within the United States, the judge focused on the complicity of Management in all the fraudulent transactions. He thought that “[s]o long as the derivative action is one alleging total complicity on the part of foreign management, the ultimate focus of the theory remains a deception of foreign fundholders by foreign ‘directors’.” 462 F.Supp. at 224. “Since virtually all the fundholders were foreign nationals residing in foreign countries, the deception, if it could be proved, must have occurred outside of the United States.” Id. (emphasis in original). Furthermore, insofar as our decision in Goldberg v. Meridor, 567 F.2d 209 (2 Cir. 1977), cert. denied, 434 U.S. 1069, 98 S.Ct. 1249, 55 L.Ed.2d 771 (1978), relied, as regards causation, on the ability of the deceived shareholders to have sought injunctive relief if they had known the facts, the court thought this case to differ from “a domestic case”, 462 F.Supp. at 223, because, for reasons not clearly stated and somewhat contradicted by fn. 35 on p. 224, it assumed that any such suit would have had to be brought in Luxembourg. This would place “the plaintiffs in a curious position in that by establishing their right to an injunction under Luxembourg law, they could prove 10b-5 materiality; simultaneously, however, they would be offering a good reason not to apply Rule 10b-5 to the transactions, since the availability of relief under foreign law would then be at least partially evident.”10 Viewing the action as one that had “its genesis abroad . . . with a group of foreign managers of a foreign investment trust violating what would appear to be their fiduciary duties to their fundholders, and the foreign managers merely enlisting the aid of American aiders and abettors”, 462 F.Supp. at 225, the court found no basis for subject-matter jurisdiction, even as to transactions consummated within the United States, see id. at 224 n.34.
We see no sufficient ground for this characterization of the transactions here at issue. Our decision in Goldberg v. Meridor, supra, did not find the nub of the action to be the directors’ breach of fiduciary duty to the shareholders, see 567 F.2d at 221; indeed, that was the very ground that had been ruled out by Santa Fe Industries, Inc. v. Green, 430 U.S. 462, 97 S.Ct. 1292, 51 L.Ed.2d 480 (1977). The holding rather was that, as we said in speaking of Schoenbaum, an action under Rule 10b-5 can lie if “there is deception of the corporation (in effect, of its minority shareholders) when the corporation is influenced by its controlling shareholder to engage in a transaction adverse to the corporation’s interests (in effect, the minority shareholders’ interests) and there is nondisclosure or misleading disclosures as to the material facts of the transaction.” 567 F.2d at 217. The basic principle was that where the directors are parties to the fraud, deception, as stated by Chief Judge Seitz in Pappas v. Moss, 393 *918F.2d 865, 869 (3 Cir. 1968), “is fairly found by viewing this fraud as though the ‘independent’ stockholders were standing in the place of the defrauded corporate entity . .” The relevance of the wrongdoing of the directors and managers is in relieving the corporation of having their knowledge attributed to it. The judge was thus mistaken in viewing all the American participants, even including the King group, as mere aiders and abettors of Management in perpetrating a fraud on IIT. While that is a fair description of the asserted role of Andersen, the underwriter defendants and perhaps even Lipper, the members of the King complex and other defendants were claimed to have been perpetrators of a fraud upon the fundholders, and the three sets of defendants here before us could be held, on proper allegations, for aiding and abetting a deception originating in the United States. An actual participant in a fraud is no less a principal because someone else originated the plan. IIT and its liquidators are complaining of deception practices on IIT by both the King complex, whose acts were primarily in the United States, and Management, whose acts were mainly outside it, both allegedly aided and abetted by the defendants here before us, without any attribution to IIT of knowledge on the part of Management. The ability of such a victim to maintain such an action was decided in Goldberg, we see no reason to depart from that decision, and we shall discuss subject-matter jurisdiction in that light.
IV.
Subject-matter jurisdiction: the King Resources common and TCC convertible note transactions
So viewing the case, we have no difficulty in finding subject-matter jurisdiction with respect to IIT’s purchases of the KRC common stock and the TCC convertible note. Apart from the fact that these were securities of American corporations, the transactions were fully consummated within the United States. It should be evident by now that “the presence or absence of any single factor which was considered significant in other cases dealing with the question of federal jurisdiction in transnational securities cases is not necessarily dis-positive” in future cases, Continental Grain (Australia) Pty. Ltd. v. Pacific Oilseeds, Inc., 592 F.2d 409, 414 (8 Cir. 1979). Hence we do not mean to suggest that either the American nationality of the issuer or consummation of the transaction in the United States is either a necessary or a sufficient factor, see, e. g., Leasco, supra, 468 F.2d at 1336-37 (subject-matter jurisdiction sustained over a transaction in securities of a foreign issuer consummated abroad where many acts of deception had occurred in the United States), but rather that the presence of both these factors points strongly toward applying the anti-fraud provisions of our securities laws.
The complaint alleged that IIT purchased the 200,000 shares of KRC common stock. (600,000 shares after a 3-1 split) in the United States. Virtually all these shares were bought by Lipper for IIT in the over-the-counter market, and the certificates were kept in IIT’s sub-custodial account in New York City with The Bank of New York. We see nothing foreign in this transaction except that the purchaser was a foreigner and the orders were transmitted from abroad, by a devious method whereby Management advised the Montreal Trust Company in Toronto, the custodian of IIT’s securities, to receive the KRC common stock through its sub-custodian, with payment to Lipper’s London office to be made by IIT’s cash custodian, Credit Suisse. None of our cases or any others intimate that foreigners engaging in security purchases in the United States are not entitled to the protection of the anti-fraud provisions of the securities laws. See, e. g., Arthur Lipper Corp. v. SEC, supra, 547 F.2d at 179 (argument that injury was only at expense of off-shore funds “ignores that the fraud charged by the SEC was perpetrated in the United States by payments from one registered broker-dealer (Lipper Corp.) to another (IPC) in connection with the purchase and sale of securities in the *919United States over-the-counter market”); United States v. Cook, 573 F.2d 281, 283-84 (5 Cir.), cert. denied, 439 U.S. 836, 99 S.Ct. 119 (1978) (“It is an absurd notion that Congress intended activity in the United States involving American securities to be exempt from the fraud provisions of the securities acts simply because the victims are not American citizens.”); SEC v. Kasser, 548 F.2d 109, 116 (3 Cir.), cert. denied, 431 U.S. 938, 97 S.Ct. 2649, 53 L.Ed.2d 255 (1977) (anti-fraud provisions designed “to insure high standards of conduct in securities transactions within this country” as well as to protect domestic markets and investors).
A similar analysis applies to the TCC convertible note. Here the order to receive given the Montreal Trust Company indicated that payment for the note was to be made in Denver, Colorado, and that the note, like the KRC common stock shares, was also to be held by The Bank of New York, IIT’s sub-custodian for United States securities.
V.
Subject-matter jurisdiction: the KRCC convertible debentures
The defendants stress that IIT purchased its KRCC eurodollar convertible debentures, except perhaps for the $50,000 purchased from Lipper,11 in the European after-market. They argue that lack of jurisdiction over these purchases follows from our conclusion in Bersch, supra, 519 F.2d at 993, that the anti-fraud provisions of the federal securities laws “[d]o not apply to losses from sales of securities to foreigners outside the United States unless acts (or culpable failures to act) within the United States directly caused such losses”, since with the exception noted all purchases were made abroad. When the quoted statement is read in the context of the facts in Bersch, it does not have the effect contended.
The first difference is that in Bersch we were dealing wholly with foreign securities. All three of the offerings were of common stock of IOS, Ltd., a Canadian corporation having its center of activities in Geneva, Switzerland. The prospectuses for the primary offering of the shares underwritten by the Drexel group for the 5,600,000 share offering and for the secondary 3,950,000 share offering by IOB, of which the shares involved in the action were a part, stated that the shares “are not being offered in the United States of America or any of its territories or possessions or any area subject to its jurisdiction”; the secondary offering of 1,450,000 shares underwritten by J. H. Crang & Co. of Toronto was sold entirely in Canada. Here the primary offering was of $25,000,000 of KRC debentures, United States securities offered in the American market. The offering of $15,000,000 of KRCC eurodollar bonds was, in substance, an integral part of this financing. There is abundant evidence that the United States and foreign offerings were closely coordinated; while this was also true of the three offerings in Bersch, 519 F.2d at 980, there none of the offerings was of an American security or was made domestically to anywhere near the same extent as the coordinated debenture offerings, $25,000,000 in the United States and $15,000,000 abroad, in this case. Although the eurodollar debentures were nominally the obligations of a wholly owned Netherlands Antilles subsidiary of KRC, this corporation was inserted into the total offering simply because European investors were reluctant to purchase debentures issued directly by an American corporation, since interest payments would then be subject to United States withholding tax. The Netherlands Antilles corporation had no operating assets,12 the debentures issued by it were guaranteed by KRC, and they were con*920vertible into KRC common stock. We have previously refused to be deterred from considering the real facts by the interposition of a foreign subsidiary of this kind. Leasco, supra, 468 F.2d at 1337-38.
The fact that we are dealing here with debentures which in substance were American rather than foreign securities has bearings of several sorts. The first goes back to the effects test. We think Congress would have been considerably more interested in assuring against the fraudulent issuance of securities constituting obligations of American rather than purely foreign business. Our statement in Vencap, supra, 519 F.2d at 1017: “We do not think Congress intended to allow the United States to be used as a base for manufacturing fraudulent security devices for export, even when these are peddled only to foreigners” applies with even greater force when, as here, the securities are essentially American. Our very next sentence, id., “This country would surely look askance if one of our neighbors stood by silently and permitted misrepresented securities to be poured into the United States” reads with particular strength on a situation where the securities are essentially of the pourer’s own nationals. This concern is only partially diminished by the fact that the prospectus for the eurodollar offering stated that the securities were not registered under the Securities Act of 1933 and were not being offered within the United States or to Americans, and by the SEC’s grant of no-action treatment under the 1933 Act. None of this amounts to saying that if fraud had been committed in the United States in connection with the issuance of the debentures, American courts would look away. See Bersch, supra, 519 F.2d- at 986.
Here there was also greater relative American participation than in Bersch in other respects. While two of the six underwriters of the primary offering in Bersch were American banking houses, these had European offices from which they apparently did much of their work, and all the others and the underwriters of the two secondary offerings were foreigners. Here Dempsey-Tegeler & Company, Inc., an American firm, was the sole lead under-; writer of the dollar offering and co-lead underwriter of the eurodollar offering along with a Luxembourg bank.
Perhaps most important of all, a consequence of the KRCC debentures being essentially of an American security is that the activities occurring in the United States, which on their surface may appear similar to those held in Bersch to be “merely preparatory” and thus insufficient .to have “directly caused” loss to foreigners, assume a different aspect. The fact that the drafting of the final prospectus in Bersch was done in Europe was not just “a formal or ultimate act . staged in Europe”, as the Bersch district court found, 389 F.Supp. at 446, 457. The Bersch prospectus was mainly drafted in Europe because that was where IOS’ records and principals were. Here . the prospectus was wholly drafted in the United States because the offering, for largest part in form and for all in substance, was of securities of an American based corporation. Similarly while there was some domestic accounting work in Bersch, 389 F.Supp. at 456, most of the field work was and in the nature of things had to be done abroad. Here all the accounting work was and had to be done in the United States. Similarly the fact that the prospectuses in Bersch were printed in Europe while those in this case (including the prospectus for the eurodollar offering) were printed in the United States, while not of particular significance in and of itself, reflects the fact that in Bersch the work on the prospectuses had mainly been done in Europe, so that Europe was the natural place for printing, whereas here most of the work had been done in the United States and there was no reason to ship the prospectus elsewhere for printing. In sum while many of the acts in the United States in this case were similar to those in Bersch, the relativity is entirely different because of the lack here of the foreign activity so dominant in Bersch, 519 F.2d at 987 (“We see no reason to extend [jurisdiction] to cases where the United States activities . are relatively small in comparison to those abroad.”) Determination whether American activities “directly” caused losses to foreigners depends not only on how much *921was done in the United States but also on how much (here how little) was done abroad.13
We see little force in defendants’ argument that sustaining jurisdiction here will somehow affront Luxembourg. The problem of conflict between our laws and that of a foreign government is much less when the issue is the enforcement of the anti-fraud sections of the securities laws than with such provisions as those requiring registration of persons or securities. The primary interest of Luxembourg is in the righting of a wrong done to an entity created by it. If our anti-fraud laws are stricter than Luxembourg’s, that country will surely not be offended by their application. If they are weaker — which is not seriously suggested — the liquidators made their choice, doubtless at least in part because of difficulty in securing personal jurisdiction in Luxembourg, and after Andersen attacked their capacity, see note 9 supra, they obtained a second order from the Luxembourg district court reaffirming their right, so far as Luxembourg was concerned, to bring suit on behalf of the fundholders. The defendants with whom we are here concerned acted in the United States and cannot fairly object to having their conduct judged by its laws.
VI.
Failure of the complaint to state a claim upon which relief can be granted
Our holding that the district court erred in dismissing the action for want of subject-matter jurisdiction compels us to confront an issue which it was able to avoid — namely, whether the complaint set forth a claim upon which relief can be granted against the appellants. This issue must be discussed separately with respect to Lipper, the underwriter defendants and Andersen. While the underwriter defendants and Andersen are charged as aiders and abettors, Lipper seems to be charged both as a principal and as an aider and abettor.14
With respect to Lipper, it makes little difference whether the complaint is sufficient to make out a case that it was a party to the conspiracy to defraud, since paragraphs 64-76 of the complaint and paragraphs 61-85 of the affidavit of Neal J. Morse in opposition to Lipper’s motion to dismiss, which the judge treated as properly pleaded, amply state a claim for aiding and abetting. The peculiarly close relationship between Lipper, on the one hand, and IOS and the funds, including IIT, which IOS or one of its affiliates, here Management, served as investment adviser, on the other, are described in our opinion in Arthur Lipper Corp. v. SEC, supra, 547 F.2d 171. The papers here alleged that Lipper Corporation was no ordinary stockbroker but had the most intimate ties both with IOS and its affiliates and with the King group; that when it purchased the KRC common stock and the $50,000 eurodollar debentures for IIT, it knew that these were in violation of IIT Fund Regulations and investment poli*922cy; that Lipper was rewarded for its acts and silence by being given the right to purchase shares of TCC at a special price as well as participation in other King-related investment and tax avoidance schemes; that Lipper had made himself a part of the King empire and knew that information being disseminated by it was false and fraudulent. As our subsequent more detailed discussion of the elements of aiding and abetting will show, the allegations with respect to Lipper were plainly sufficient. See particularly Brennan v. Midwestern United Life Ins. Co., 417 F.2d 147 (7 Cir. 1969), cert. denied, 397 U.S. 989, 90 S.Ct. 1122, 25 L.Ed.2d 397 (1970).
The questions with respect to the underwriter defendants and Andersen are more difficult. The cases generally list three prerequisites to aiding and abetting liability:
(1) the existence of a securities law violation by the primary (as opposed to the aiding and abetting) party;
(2) “knowledge” of this violation on the part of the aider and abettor; and
(3) “substantial assistance” by the aider and abettor in the achievement of the primary violation.
See Rolf v. Blyth, Eastman Dillon & Co., Inc., 570 F.2d 38, 47-48 (2 Cir.), cert. denied, 439 U.S. 1039, 99 S.Ct. 642, 58 L.Ed.2d 698 (1978); SEC v. Coffey, 493 F.2d 1304, 1316 (6 Cir. 1974), cert. denied, 420 U.S. 908, 95 S.Ct. 826, 42 L.Ed.2d 837 (1975); Woodward v. Metro Bank of Dallas, 522 F.2d 84, 94-97 (5 Cir. 1975); Marbury Management, Inc. v. Kohn, 470 F.Supp. 509, 515 (S.D.N.Y.1979). Although this list of prerequisites has become commonplace, the exact content of the rather vague phrases, especially “knowledge” and “substantial assistance”, is still being delineated by the courts. Moreover, the three requirements cannot be considered in isolation from one another. Satisfaction of the scienter requirement will, for éxample, depend on the theory of primary liability and, as will be seen, there may be a nexus between the degree of scienter and the requirement that the alleged aider and abettor render “substantial assistance”. After studying the many cases we might be inclined to wonder whether the elaborate discussions have added anything except unnecessary detail to Judge L. Hand’s famous statement, made in a criminal context,15 that, in order to be held as an aider and abettor, a person must “in some sort associate himself with the venture, that he participate in it as something that he wishes to bring about, that he seek by his action to make it succeed.” However, we shall discuss the question in the terms that have become conventional.
(1) Primary Violation. While there is little doubt that plaintiffs have adequately alleged a number of primary violations, the allegedly false prospectus forms the sole link tying the underwriters and the principal link tying Andersen to the alleged fraud. These defendants claim that since the complaint alleges complicity in the fraud by' Management, the prospectus itself could have played no role in its decision to purchase the various King securities, and that therefore they cannot be held liable as aiders or abettors. Andersen thus contends that the suit is essentially one for breach of fiduciary duty and not for a securities law violation; the underwriter defendants argue that “causation” is lacking under plaintiffs’ theory since -the prospectus was not a causative factor in the decision to purchase. These arguments simply ignore the holdings in Goldberg that where the management and the directors are parties to the fraud, 567 F.2d at 219:
[T]he test must be whether the facts that were not disclosed or were misleadingly disclosed to the shareholders “would have assumed actual significance in the deliberations” of reasonable and disinterested directors or created “a substantial likelihood” that such directors would have considered the “total mix” of information *923available to have been “significantly altered.”16
and that causation is furnished by the possibility that informed stockholders could have obtained an appropriate remedy if the facts had not been concealed.17 See Note, Suits for Breach of Fiduciary Duty under Rule 10b-5 after Santa Fe Industries, Inc. v. Green, 91 Harv.L.Rev. 1874, 1893-98 (1978).
(2) Scienter. Recognizing that there can be no certainty on the subject until the Supreme Court fleshes out footnote 12 of Ernst & Ernst v. Hochfelder, 425 U.S. 185, 193-94, 96 S.Ct. 1375, 1380-81, 47 L.Ed.2d 668 (1976), we shall continue to follow our own decisions, see, e. g., Lanza v. Drexel & Co., 479 F.2d 1277, 1300-02 (1973) (en banc), as well as those of other courts of appeals with which we agree, that reckless conduct will generally satisfy the scienter requirement. However, there are special considerations in applying this general principle to aiders and abettors. In Rolf, supra, 570 F.2d at 44, a divided panel concluded “that at least where, as here, the alleged aider and abettor owes a fiduciary duty to the defrauded party, recklessness satisfies the scienter requirement.” Among the decisions of other circuits reaching a similar conclusion which were cited with approval, Woodward v. Metro Bank of Dallas, supra, 522 F.2d at 97 (pre-Ernst & Ernst) and Sundstrand Corp. v. Sun Chemical Corp., 553 F.2d 1033, 1044-45 (7 Cir.), cert. denied, 434 U.S. 875, 98 S.Ct. 225, 54 L.Ed.2d 155 (1977), are especially notable. In Edwards & Hanly v. Wells Fargo Securities Clearance Corp., . 602 F.2d 478, 484-85 (2 Cir. 1979), cert. denied,-U.S.-, 100 S.Ct. 734, 62 L.Ed.2d 731 (1980), dealing with the degree of scienter required where there was no such fiduciary duty, Judge Gurfein quoted with approval Judge Goldberg’s statement in Woodward v. Metro Bank of Dallas, supra, 522 F.2d at 95: “The scienter requirement scales upward when activity is more remote; therefore, the assistance rendered should be both substantial and knowing.”
The scienter allegations with respect to the underwriter defendants are primarily in paragraph 81:
The Underwriter Defendants aided and abetted and joined in the conspiracy by circulating a prospectus which they knew, or should have known, contained material misrepresentations of facts and material omissions.
If this stood alone, it might well run afoul of Fed.R.Civ.P. 9(b). However, paragraph 52 of the complaint alleged five major misrepresentations and omissions in the prospectus, which we quote in the margin;18 *924these gave the defendants sufficient notice of what they must meet, and the Rule itself says that knowledge “may be averred generally.”
The “or should have known” clause in paragraph 81 would not qualify under the definition of recklessness contained in the cases above cited, see Rolf, supra, 570 F.2d at 47 n.16, Edwards & Hanly, supra, 602 F.2d at 478-79. However, paragraph 81 alleges in the disjunctive that the underwriter defendants “knew” of the misrepresentations and omissions alleged in paragraph 52. As a matter of pleading, more is not required, see Fed.R.Civ.P. 8(e)(2). One may indeed doubt whether plaintiffs will be able to demonstrate actual knowledge on the part of these three minor underwriters at a trial, but in the absence of a sufficiently supported motion for summary judgment on this subject, a course still open to the underwriter defendants, they are entitled to an opportunity to do this.
The allegations of scienter with respect to Andersen and the financial statements in the prospectus are significantly weaker than those with respect to the underwriter defendants. Paragraph 62 alleged that “Arthur Andersen, among other things, failed to prepare financial statements for King Resources Company reflecting the true financial condition of said company and failed to insure that full and correct financial information was provided in the Eurodollar prospectus and elsewhere . .” and that it certified “false and misleading financial statements for King Resources Capital Corp., N.V.” This is a long way from the most liberal notion of an allegation of knowledge or recklessness.
Although the allegations with respect to Andersen’s role in preparing the financial statements are thus insufficient, the complaint also charges Andersen with other wrongs. Paragraph 62(c) accuses Andersen of:
Not disclosing that IIT’s “investments” in King Resources Company and King Resources Capital Corp., N.V. and IIT’s loan to The Colorado Corporation were contrary to the IIT Fund Regulations and contrary to IIT’s investment policy set forth in IIT’s prospectuses,
and paragraph 63 alleges that:
In addition, Arthur Andersen, as IIT’s independent certified public accountants, failed to disclose to IIT and its fundholders IIT’s improper loan to The Colorado Corporation, improper purchase of King Resources common stock and improper purchase of the Eurodollar Debentures.
There is no specific allegation of knowledge or recklessness concerning this failure to disclose in the section of the complaint la-belled “Role of Arthur Andersen”, which contains the above-quoted charges. Since by contrast, the sections of the complaint headed “Role of Lipper and the Lipper Corporation” and “Role of the Underwriter Defendants” both contain the “knew or should have known” charge, the omission as to Andersen could hardly have been inadvertent. Paragraph 87 of the complaint, however,- includes the general allegation that “[e]ach of the defendants knew, or should have known, that the sale, and purchase of the Eurodollar Debentures, the purchases of the King Resources common stock, and IIT’s loan to The Colorado Corporation were illegal” — just in what respects the complaint does not describe. -Reading this language in conjunction with the more specific allegations as to Andersen’s role in preparation of the financial statements, the distinction evidently attempted by the draftsman is that while he was unable to, charge Andersen with knowledge or recklessness for its contribution to the prospectuses which related to KRC and KRCC, he was charging some kind of conduct more culpable than negligence once Andersen became aware of IIT’s acquisition of King securities.19 This would explain plaintiff’s *925vague references to Andersen’s role as accountants on both sides of the IOS and King relationship. The same conclusion follows if we take account, as the district court did, 462 F.Supp. at 212 n.4, of the various affidavits and memoranda subr mitted by plaintiffs. We find it hard to see how these allegations that Andersen knew or should have known that something was “wrong”, not in the financial statements in the prospectuses but in the general relationship between IIT and King — some of Andersen’s alleged “concerns” relating to events subsequent to IIT’s purchases — met the requirement for aider and abettor liability under the circumstances. Here we find helpful Judge Goldberg’s statement in Woodward v. Metro Bank of Dallas, supra, 522 F.2d at 97, that:
When it is impossible to find any duty of disclosure, an alleged aider-abettor should be found liable only if scienter of the high “conscious intent” variety can be proved. Where some special duty of disclosure exists, then liability should be possible with a lesser degree of scienter.
We echoed this sentiment in Edwards & Hanly, supra, 602 F.2d at 484, as noted above.
The complaint does not allege any special duty of disclosure by Andersen. Although Andersen may have had a duty of disclosure as to errors which it found in the financial statements, see infra, the complaint does not allege any failure to disclose as to these statements, or even that Andersen ever knew that the statements were false or misleading. The failure to disclose concerned more general “wrongs” not dependent on or tied to Andersen’s previous role as an accountant. As to the financial statements themselves nothing more than negligence is alleged. As to the general failure to disclose, the complaint does not charge “scienter of the high ‘conscious intent’,” or, as we phrased it in Edwards & Hanly, “something closer to an actual intent to aid in the fraud.” 602 F.2d at 485. However, we can pretermit decision on the adequacy of the allegations of Andersen’s scieúter as to the general failure since the complaint against it fails on another related ground.
(3) Substantial assistance. There is much practical appeal in the suggestion of the underwriter defendants that their participation — $600,000 of the $15,000,000 eurodollar offering and $1,350,000 of the total $40,000,000 offering, was not “substantial assistance” in the ordinary meaning of those words. It is doubtless true that if these firms had not enlisted in the underwriting, it would have proceeded with others in their place. Still the underwriter defendants did, in Judge Hand’s language, associate themselves with the venture, participated in it as something they wished to bring about, and sought by their action to make it succeed. Moreover, we must consider the issue in the light of the allegation that these three underwriters actually knew of the specified falsities in the prospectus. If the underwriter defendants had such knowledge and had blown the whistle, the entire underwriting might have collapsed.
The substantial assistance claimed on the part of Andersen is twofold, its activity in connection with the preparation of the prospectuses and its failure to inform either the IIT fundholders or Luxembourg or United States authorities of what was afoot. As we have seen, the former ground fails because of the lack of any allegation of scienter. The question how far mere inaction, here Andersen’s failure to inform, can fulfill the requirement of substantial assistance is unsettled. See Fischer v. NYSE, 408 F.Supp. 745, 753 (S.D.N.Y.1976) (“No cases in the Second Circuit provide direct guidance in this uncharted area”); Hirsch v. du Pont, 553 F.2d 750, 759 (2 Cir. 1977) (specifically leaving open issue whether mere inaction, in the absence of a duty of disclosure, can give rise to aiding and abetting liability); Woodward, supra, 522 F.2d at 96 (“Most problematic in this area is the issue whether, or to what extent, silence and inaction can fulfill the requirement.”).
Several cases in other circuits refuse to impose aiding and abetting liability for in*926action except when there existed an independent duty to disclose. One of the leading cases to this effect is Wessel v. Buhler, 437 F.2d 279 (9 Cir. 1971). There plaintiff sought to hold an independent public accountant, Jordan, to aiding and abetting liability. Jordan had prepared three nonpublic financial statements, which may have formed the basis for figures in a fraudulent prospectus, but was not himself in any way responsible for the prospectuses or the figures appearing in them. Jordan did become aware of financial deficiencies, however, and plaintiffs theory was that “Jordan owed a duty to prospective investors to disclose his knowledge of RMC’s irregular financial conduct and of deficiencies in its financial records, and that his failure to perform that duty placed Buhler and his associates in a position to dupe the investors by launching stock with the third prospectus; therefore, Jordan aided and abetted Buhler and should be held as a principal.” Id. at 283. Judge Hufstedler gave short shrift to this argument:
There is not a scrap of authority supporting this extraordinary theory of Rule 10b-5 liability, and we will not supply any in this case.
We find nothing in Rule 10b-5 that purports to impose liability on anyone whose conduct consists solely of inaction.
. On the contrary, the exposure of independent accountants and others to such vistas of liability, limited only by the ingenuity of investors and their counsel, would lead to serious mischief. Id.
This language was quoted by us in a slightly different context, Lanza v. Drexel & Co., 479 F.2d 1277, 1300 (1973) (en banc), and Judge Frankel cited the case as persuasive in Gold v. DCL, Inc., 399 F.Supp. 1123,1128 (S.D.N.Y.1973). While here Andersen did prepare and certify the financial statements appearing in the fraudulent prospectus, there is, as we have noted, no adequate allegation of scienter by Andersen in that activity. Other courts have taken the view that mere inaction can constitute substantial assistance even in the absence of an independent duty to disclose if but only if there was a “conscious intention" to forward the violation of Rule 10b-5. See Woodward, supra, 522 F.2d at 96-97; SEC v. Coffey, supra, 493 F.2d at 1317; Rochez Brothers, Inc. v. Rhoades, 527 F.2d 880, 889 (3 Cir. 1975); Gould v. American-Hawaiian S. S. Co., 535 F.2d 761, 780 (3 Cir. 1976). We approached this position in Edwards & Hanly, supra, 602 F.2d at 484-85 where Judge Gurfein, in dealing with an alleged failure to discover, stated that “[fjinding a person liable for aiding and abetting a violation of 10b-5, as distinct from committing the violation as a principal, requires something closer to an actual intent to aid in a fraud, at least in the absence of some special relationship with the plaintiff that is fiduciary in nature.” Perhaps the leading example of “actual intent” aiding and abetting by what came close to mere inaction is Brennan v. Midwestern United Life Insurance Co., supra, 417 F.2d 147. Since this was a pioneering case in the general aiding and abetting field and few would disagree with its result, it is worth stating the facts.
Dobich, a dealer in Midwestern stock, was committing securities law violations of which Midwestern was aware and which had the effect of increasing the value of Midwestern’s over-the-counter stock, to the benefit of Midwestern’s efforts to accomplish a merger with another company. Some correspondence took place between Midwestern and Dobich, with Midwestern threatening to report Dobich to state authorities if he persisted in his violations. Later, however, when merger negotiations began, Midwestern took a much softer line and contrary to its threats, did not report Dobich. In a class action by persons who purchased Midwestern stock from Dobich but failed to receive delivery, the court held that Midwestern was liable as an aider and abettor. Although Professor Ruder stresses 20 the court’s statement, 417 F.2d at 154,
It is our view that the district court was correct in concluding that Midwestern’s *927acquiescence through silence in the fraudulent conduct of Dobich combined with its affirmative acts was a form of aiding and abetting cognizable under Section 10(b) and Rule 10b-5 (Emphasis supplied),
in contrast to the district court’s holding that Midwestern’s failure to report Dobich’s activity to state and federal regulatory authorities alone sufficed, 259 F.Supp. 673, 682 (N.D.Ind.1966) (on motion to dismiss); 286 F.Supp. 702, 704, 727 (N.D.Ind.1968) (alternative holding on merits), the “affirmative action” was slight indeed. It involved telling Dobich that Midwestern would do nothing and advising complaining purchasers to contact Dobich, thereby affording Dobich an opportunity to satisfy their complaints while maintaining the fraud. A number of courts have read Midwestern as a case of mere inaction. See, e. g., Rochez Brothers, Inc. v. Rhoades, supra, 527 F.2d at 889; Woodward, supra, 522 F.2d at 96; Coffey, supra, 493 F.2d at 1317. Even so viewed, Midwestern upholds aider and abettor liability in the absence of some independent duty to act only when there is clear evidence of the required degree of scienter, see Edwards & Hanly, supra, 602 F.2d at 484-85, and a conscious and specific motivation for not acting on the part of an entity with a direct involvement in the transaction. In Midwestern, not only did the evidence show that Midwestern knew of Do-bich’s fraudulent activities, but the company’s merger negotiations with Mid-Continent would have failed if the Dobich violations were made public. Midwestern, supra, 417 F.2d at 153. Here plaintiffs do not allege that Andersen intended by its silence to forward completion of the fraudulent transactions in the expectation of benefiting from the success of the fraud. Moreover, application of the tripartite test should not obscure the basic proposition that mere bystanders, even if aware of the fraud, cannot be held liable for inaction since they do not, in Judge Hand’s words, associate themselves with the venture or participate in it as something they wish to bring about. United States v. Peoni, supra, 100 F.2d at 402.21
Apart from a case like Midwestern, inaction can create aider and abettor liability only when there is a conscious or reckless violation of an independent duty to act. See Ruder, supra, 120 U.Pa.L.Rev. at 644. Accountants do have a duty to take reasonable steps to correct misstatements they have discovered in previous financial, statements on which they know the public is relying. See Fischer v. Kletz, 266 F.Supp. 180, 188 (S.D.N.Y.1967). Paragraph 52(C) of the complaint alleged failure of the KRC and KRCC prospectuses to disclose the conspiracy with IOS as a material omission, and arguably alleged that Andersen acquired knowledge of this after the prospectuses had been issued. However, the omission alleged in H 52(C) does not relate to that portion of the prospectus, the financial statements, over which Andersen had responsibility. Andersen had no independent duty to see to the correction of portions of the prospectus other than the financial statement it prepared. Even if we should assume adequate allegations of scienter on the part of Andersen with respect to other wrongs claimed by plaintiffs, the complaint fails because it neither alleges a conscious and specific motivation for not acting, as in the Midwestern case, nor alleges an independent duty to report.
A final word on this branch of the case must be said with respect to the claim in IIT’s reply brief that if there are defects in the pleadings, it should be permitted to amend. Applying the liberal standard enunciated in Fed.R.Civ.Pro. 15(a) and explicated in Foman v. Davis, 371 U.S. 178, 83 S.Ct. 227, 9 L.Ed.2d 222 (1962), we still see no justification for allowing this. IIT’s claims were set out four years ago by experienced counsel, in a complaint 38 pages in length; there has been extensive discovery; and the district judge in effect has already permitted amendment by treating the vari*928ous memoranda and affidavits submitted by plaintiffs as well-pleaded portions of the complaint, see note 4 supra. Although the complaint was filed shortly before Ernst & Ernst, this circuit had required allegations of scienter at least since Lanza v. Drexei, supra, 479 F.2d at 1305-06.22 We can only assume that IIT’s failure to allege scienter on the part of Andersen with respect to the financial statements in the prospectus was because it had no fair basis for doing so. We have been cited no new facts that would justify an amendment on this score. Plaintiffs having elected to stand on their complaint and the broad supplementation which the district court allowed, having compiled a record which occupies some four feet of shelf space there and having filed an appendix of 2711 pages on this appeal, their suggestion of further amendment now comes too late. Cf. Freeman v. Continental Gin Co., 381 F.2d 459, 469-70 (5 Cir. 1967).
VII.
The Statute of Limitations
Although all three of the moving parties sought to have IIT’s complaint dismissed as barred by the statute of limitations and briefed the point extensively below, the district court did not reach the issue, see 462 F.Supp. at 227, and only the underwriter defendants and the plaintiffs have addressed it in their briefs in this court. Since we have examined the briefs in the district court and believe the argument fails as a matter of law, we may as well dispose of it.
Federal law looks to analogous state law to supply the period of limitations for Rule 10b-5 actions, see Ernst & Ernst v. Hochfelder, supra, 425 U.S. at 210 n.29, 96 S.Ct. at 1389 n.29; Mittendorf v. J. R. Williston & Beane Incorporated, 372 F.Supp. 821, 830 n.4 (S.D.N.Y.1974). This circuit has uniformly found the analogy in the limitations period which the state had provided for an action based upon common law fraud. Stull v. Bayard, 561 F.2d 429, 431 (2 Cir. 1977), cert. denied, 434 U.S. 1035, 98 S.Ct. 769, 54 L.Ed.2d 783 (1978); Phillips v. Levie, 593 F.2d 459, 462 (2 Cir. 1979).
The New York Civil Practice Law and Rules has two separate relevant limitations periods:
The following actions must be commenced within six years:
8. an action based upon fraud; the time within which the action must be commenced shall be computed from the time the plaintiff or the person under whom he claims discovered the fraud, or could with reasonable diligence have discovered it. CPLR 213(8), former CPLR 213(9). Except as provided in article 2 of the uniform commercial code, where the time within which an action must be commenced is computed from the time when facts were discovered or from the time when facts could with reasonable diligence have been discovered, or from either of such times, the action must be commenced within two years after such actual or imputed discovery or within the period otherwise provided, computed from the time the cause of action accrued, whichever is longer. CPLR 203(f).
At first blush the coexistence of the two statutes is rather baffling. The explanation seems to be that an older statute, C.P.A. § 48(5), a predecessor of CPLR § 213(8), allowed six years after the fraud or the plaintiff’s discovery of the facts constituting the fraud. This was evidently thought to be too liberal and § 203(f) shortened the additional period allowed in cases of late discovery to two years. See Hoff Research & Development Laboratories, Inc. v. Philippine National Bank, 426 F.2d 1023, 1025-26 (2 Cir. 1970). The combined effect of CPLR 213(8) and 203(f) thus is “two separately-timed and alternative limitations periods in the case of a delayed discovery: *929six years from accrual or two years from discovery whichever is longer.” Hoff Research & Development Laboratories, Inc. v. Philippine National Bank, supra, 426 F.2d at 1026.
While federal law looks to state law for the relevant period of limitation when none has been provided in the federal statute, federal law controls on the question when that period begins to run. Moviecolor Limited v. Eastman Kodak Company, 288 F.2d 80, 83 (2 Cir.), cert. denied, 368 U.S. 821, 82 S.Ct. 39, 7 L.Ed.2d 26 (1961); Arneil v. Ramsey, 550 F.2d 774, 780 (2 Cir. 1977). The federal rule with respect to discovery of fraud is that “the statute commences to run when the plaintiff has actual knowledge of the alleged fraud or knowledge of facts which in the exercise of reasonable diligence should have led to actual knowledge.” Stull v. Bayard, supra, 561 F.2d at 432; Phillips v. Levie, supra, 593 F.2d at 462.
The underwriter defendants argue that between January 8,1971 and September 14, 1972, extensive publicity about the disastrous crumbling of the King complex appeared in the New York Times and the Wall Street Journal. Further, the King Resources securities which form the basis of IIT’s complaint had already plummeted in value by the second half of 1970, as the complaint itself makes clear. Finally, two actions containing allegations against King substantially similar to those involved in this case were brought by other plaintiffs as early as 1971. SEC v. King Resources Company, (D.Col.) (filed January 25, 1971); Deitrich Corporation v. King Resources Company, (D.Col.) (filed September 28, 1971). According to the underwriter defendants, IIT, “in the exercise of reasonable diligence,” should therefore have discovered the fraud it alleges well before two years prior to the time it commenced the action on July 17, 1975, and damages would have to be limited to those caused by frauds perpetrated after July 17, 1969. If, as held in Stull v. Bayard, supra, a cause of action for deception accrues when the deception is practiced, IIT would be time-barred with respect to the purchase of the KRC common stock and $5,300,000 of the KRCC debentures. Indeed, the underwriter defendants assert with excessive enthusiasm that any claims based on the remaining $2,715,000 debenture purchases are also time-barred even though they were executed after July 17, 1969, since, for purposes of applying the six-year period, IIT is charged with “assumed knowledge of the fraudulent wrong,” Stull v. Bayard, supra, 561 F.2d at 432, and, with such knowledge, should have refrained from making the additional purchases of King securities.
The New York statutes of limitations necessarily assume the existence of a plaintiff having capacity to sue. McCarthy v. Prudential Ins. Co., 252 N.Y. 459, 464, 169 N.E. 645 (1930). Hence, if Andersen is right in its contention,(see footnote 9, supra, that IIT had no existence as a legal entity but was simply an “indivisión organisee”, or organized co-ownership, which had no capacity to sue although the co-owners did,23 the statute would never have started to run against IIT, the running would have begun only upon the appointment of the liquidators, and suit was brought by them well within the two years allowed by CPLR 203(f), let' alone the six years allowed by CPLR 213(8). However, we need not rely on this.
Although IIT advances numerous other arguments to support its position that the statute had not run, we need consider only one which we find sufficient. This is that, quite apart from the point just mentioned, the two year period provided by CPLR 203(f) did not begin to run until the liquidators were appointed by the Luxembourg court. Since this occurred on December 18, 1973, and they commenced the instant action on July 17, 1975, the action was brought within two years from the time in which they themselves discovered or should have discovered the fraud. During the earlier period in which the underwriter defend*930ants claim the King fraud was easily discoverable, Management was controlled by Vesco, the successor to Cornfeld who lost control sometime in 1970. Vesco was allegedly involved in even more grandiose schemes to bilk the fundholders, and he certainly could not be expected to bring an action on their behalf when such an action would simply focus attention on his own wrongdoing. While the precise extent of the turnover in personnel between the Cornfeld and Vesco eras is not clear, it is plain that there was no significant change in terms of the likelihood of a suit similar to the present one being brought. Of course the fraud was discoverable by those in control of IIT at the time of the newspaper publicity; indeed, it was discoverable long before without the aid of the Wall Street Journal or the New York Times if, as the plaintiffs have alleged and the defendants have accepted, those people were actually involved in the fraud themselves or, as with Vesco, other schemes to injure IIT and the fundholders. It would seem anomalous and unfair, however, to hold the liquidators to the knowledge of those in charge of IIT when the liquidators are complaining precisely of wrongs perpetrated by them. When the liquidators assumed office, they did so free of any imputation of knowledge of the previous faithless managements.
Dudley v. Allen, CCH Fed.Sec.L.Rep. 193,273 (W.D.Kentucky, 1971), is almost precisely in point. In that case, plaintiff receiver sued defendants for allegedly conspiring with the directors of the company for which he was receiver to defraud the company and its shareholders. The applicable Kentucky statute of limitations for fraud provided that an action must be commenced within five years from the date it was or should have been discovered. The fraud occurred on January 23,1964, and the receiver filed the complaint on April 22, 1969. The court ruled that the statute of limitations was no bar since the receiver was not appointed until September, 1967. “[T]he alleged fraud was not ‘discoverable’ until Receiver Dudley was appointed . inasmuch as no one would have brought suit prior to that time" (emphasis in original). So here no one would have brought the present suit until those in control of IIT, either Cornfeld or his successor Vesco, were replaced by the disinterested liquidators. See also Developments — Statutes of Limitations, 63 Harv.L.Rev. 1177, 1200 (1950) (“The commencement of the statutory period has occasionally been delayed despite the existence of a theoretical right to recovery, until the occurrence of some later event the absence of which made suit impossible or improbable. . . .”).
Although the question of determining the date when the New York two-year period began to run in this case is one of federal law as shown by the cases cited above, precedents interpreting state law have tended to the same result reached by the district judge in Dudley v. Allen, supra. In Curtis v. Connly, 257 U.S. 260, 42 S.Ct. 100, 66 L.Ed. 222 (1921), Justice Holmes considered when the statute of limitations should begin to run on a claim by Curtis, a receiver, against defendant directors of a bank for paying dividends out of assets, making improper loans and carrying loans at inflated value. The question was whether the bank, on whose behalf the receiver sued, should be charged with knowledge of the fraud while it was managed by the defendant directors. Rhode Island law which was considered applicable contained a provision similar to the federal law of concealment. Id. at 262, 42 S.Ct. at 100. Justice Holmes held that the receiver was charged with the knowledge of the bank solely because as to most of the items the wrongdoing was recorded on the bank’s books which the stockholders had a right to inspect and, as to the carrying of loans at inflated values, three new and unimplicated directors had come on the board after the alleged wrongdoing and had not taken action themselves within the limitations period. Id. at 263, 42 S.Ct. at 101.24 Here *931there is nothing to indicate that inspection of IIT’s books would have revealed the fraud; we cannot assume that Justice Holmes included in “books” such things as correspondence files and other data which no fundholder would be likely to attempt to examine if, indeed, he had access to them.
In Michelsen v. Penney, 135 F.2d 409, 405-16 & n.2 (2 Cir. 1943) (Clark, J.), where a depositors’ committee sued the former chairman of the board of directors of a bank that had fallen into receivership, the court, having ruled that the bank was the plaintiff and the depositors were considered as bringing the suit on its behalf, concluded that “the statute is tolled while a corporate plaintiff continues under the domination of the wrongdoers ... at the earliest, the three-year period began to run from the date of appointment of Receiver Spurway . . Th'en defendants’ domination of the bank ceased and notice of defendants’ wrongs for the first time became imputable to the bank.” Id. at 415-16. See also Ikfov-iecolor Limited v. Eastman Kodak Co., supra, 288 F.2d at 88, 90; and Developments — Statutes of Limitations, supra, 63 Harv.L.Rev. at 1215.
These and other cases were reviewed in International Railways of Central America v. United Fruit Co., 373 F.2d 408, 412-17 (2 Cir.), cert. denied, 387 U.S. 921, 87 S.Ct. 2031, 2035, 18 L.Ed.2d 975 (1967), a suit by minority stockholders against a third party which allegedly controlled the corporation. The principle we announced there was that to toll the statute on the ground of adverse domination “at least once the facts giving rise to [partial] liability are known, plaintiff must effectively negate the possibility that an informed stockholder or director could have induced the corporation to sue,” id. at 414, and that on the special facts of that case the burden of negation had not been met. Here it has been. The situation in this case differs totally from the unique facts in the International Railways case, where there had been a prior successful derivative action and there was a substantial number of disinterested directors, see the detailed discussion at 373 F.2d at 414-16.25
Such New York authority as has been found is to the same effect. In Brown v. Brown, 93 N.Y.S.2d 63 (Sup.Ct., Oneida County, 1948), modified on other grounds, 275 A.D. 1068, 93 N.Y.S.2d 86, aff’d, 302 *932N.Y. 556, 96 N.E.2d 443 (1949), while plaintiff was incompetent in the state mental hospital, his committee made an allegedly improper sale of his property. When plaintiff sought to attack the sale, defendant asserted that any attack on the judicial proceedings which approved the sale was barred by the statute of limitations. The court rejected this defense, saying:
[I]t must be clear that the statute of limitations will not begin to run until there is someone capable of suing. Here again it was most unlikely that a committee sue itself and it is the general rule that where one person represents both sides of the conflicting claims, the statute does not run. ... A committee was in charge of his property and of course was a party to the proceedings which he has since attacked. It was not at all likely that his committee would bring any action to set aside the deed in question. Id. at 75-76.
It is true, of course, that if the alleged deception had not been practiced, individual IIT fundholders might have had various remedies. But the affidavits in support of the motions to dismiss on the basis of the statute of limitations make no adequate allegation that the circumstances alleged to have given notice of the fraud would reasonably have become known to the fundholders, only a few of whom lived in the United States.
The order dismissing the action for lack of subject-matter jurisdiction is reversed. The dismissal is nevertheless affirmed with respect to the underwriter defendants on the claims relating to the TCC note and with respect to Andersen on all claims, for failure to state a claim with respect to which relief can be granted. The order denying the motion to dismiss for lack of capacity to sue is affirmed. The court is directed to deny the motions to dismiss on the basis of the statute of limitations. The cause is remanded for further proceedings consistent with this opinion. No costs.
. The principal cases in this circuit have been Schoenbaum v. Firstbrook, 405 F.2d 200 (2 Cir.), rev’d with respect to holding on merits, 405 F.2d 215 (2 Cir. 1968) (en banc), cert. denied sub nom. Manley v. Schoenbaum, 395 U.S. 906, 89 S.Ct. 1747, 23 L.Ed.2d 219 (1969); Leasco Data Processing Equipment Corp, v, Maxwell, 468 F.2d 1326 (2 Cir. 1972); Bersch v. Drexel Firestone, Inc,, 519 F.2d 974 (2 Cir.), cert. denied, 423 U.S. 1018, 96 S.Ct. 453, 46 L.Ed.2d 389 (1975); and IIT v. Vencap, Ltd., 519 F.2d 1001 (2 Cir. 1975). Familiarity with these decisions will be assumed.
. See Bersch, supra, 519 F.2d at 993. The SEC has not attempted to define the transnational scope of the anti-fraud provisions by any exercise of its rulemaking powers. See Comment, *913The Transnational Reach of Rule 1 Ob-5, 121 U.Pa.L.Rev. 1363, 1364-65 (1973).
. See, e. g., IIT v. Vencap, Ltd., supra; IIT v. Lam, 531 F.2d 463 (10 Cir. 1976). The liquidators of IIT list no less than thirteen actions which they have litigated or are litigating on IIT’s behalf in federal district courts from Puer-to Rico to Colorado.
. The quotation is from the December 1, 1969 IIT prospectus. The prospectus describes IIT as “an open-ended mutual fund,” and the complaint in the present action also characterizes IIT as a “mutual fund”. Although there are indeed many similarities between IIT and the open-ended mutual funds common in this country, this action, unlike the previous one involving IIT, IIT v. Vencap, Ltd., supra, 519 F.2d at 1003 n.1, has also disclosed some differences. For example, unlike the typical open-ended American mutual fund, IIT itself has no directors.
. As of the IIT v. Vencap, Ltd., litigation, IIT had 178 American fundholders — 111 resident abroad and 67 resident in this country — and 159 fundholders who were not American citizens but resided here. See 519 F.2d at 1016 n.28. See also 462 F.Supp. at 223 n.33. Since 1967 “IOS and all of its affiliates” have operated under an SEC consent order which prohibits the sale of shares in IIT to American citizens wherever located. Securities Exchange Act Release No. 8083 (May 23, 1967).
. The district judge considered these papers to be well-pleaded for purposes of ruling on the motions which are the subject of this appeal, 462 F.Supp. at 212 n.4, and we will do so as • well. Although the judge, in doing this, may have been thinking of our decision in Goldberg v. Meridor, 567 F.2d 209, 213 (2 Cir. 1977), cert. denied, 434 U.S. 1069, 98 S.Ct. 1249, 55 L.Ed.2d 771 (1978), his action went far beyond what we there directed, namely, treating a proposed amended complaint as properly before the district judge. We think the generosity here ac*915corded was undue unless defendants’ motions were ' considered to be “speaking motions” within the last sentence of F.R.Civ.P. 12(b).
. The fee IIT Management received from IIT was a percentage of the fund’s total net assets.
. The district judge found it “unclear how these defendants allegedly participated in the purchases of King Resources common and the loan to the Colorado Corp.,” 462 F.Supp. at 213. Although plaintiffs claim that the eurodollar prospectus was the basis not only of IIT’s purchase of the debentures but also of the KRC common stock, no such claim is made with respect to the TCC note. No apparent connection between the underwriters and the TCC note transaction can be gleaned from the complaint or any of the affidavits and memoranda considered well-pleaded by the district judge. The underwriter defendants assert in their brief on appeal that IIT does not seek recovery from them for damages arising from the TCC note transaction, and IIT does not challenge this assertion in its reply brief. The complaint against the underwriter defendants with respect to the TCC note transaction thus should have been dismissed for failure to state a claim on which relief can be granted and our discussion of their liability should be read in that light.
. This issue was raised only by Andersen and if our decision, infra, that the complaint did not state a claim upon which relief can be granted against it should stand, Andersen presumably no longer has an interest in the point. If, for any reason, the point should retain vitality, we agree with Judge Goettel’s well reasoned conclusion, see 462 F.Supp. at 215-17, that, whatever the legal capacity of IIT may or may not have been, a United States court, as a matter of comity, should defer to the decrees of the Duchy of Luxembourg appointing the liquidators of IIT, authorizing them to institute proceedings on IIT’s behalf “before any court of law in the Grand Duchy of Luxembourg or abroad”, and confirming that the “liquidators act as the proxies and representatives of the investors considered as principals.” We are totally unimpressed with Andersen’s procedural challenges to these decrees.
. The court did not explain how a Luxembourg court could obtain personal jurisdiction over most of the defendants. We likewise do not understand the failure to consider the availability of relief in an American court; Goldberg is not distinguishable on this ground since it also dealt with a foreign corporation and the causal relation was found in the availability of injunctive relief in the New York courts and not in a foreign forum if the deception of the stockholders had not occurred.
. According to a memorandum in opposition to the motions to dismiss filed by IIT below, the $50,000 of eurodollar debentures purchased from Lipper were, like the KRC common stock, kept in IIT’s sub-custodial account at The Bank of New York. Memorandum in Opposition, 26.
. The prospectus stated that “[a]s of the date of this offering K.R. Capital has not engaged in any activities other than those incident to its formation and the offering described herein.”
. A comment has suggested that in refusing jurisdiction with respect to foreigners Bersch may not have taken adequate account of the fact that “Use of American names [as underwriters, accountants and lawyers] might be particularly attractive if foreign purchasers were thereby misled into believing that their investments were protected by the strict American securities laws”, and that “In such a situation, the United States clearly has an interest in adjudicating claims against the defendants, regardless of the nationality of the plaintiffs, in order to discourage other potential perpetrators of fraud from taking advantage of American resources and prestige.” Note, American Adjudication of Transnational Securities Fraud, 89 Harv.L.Rev. 553, 570 (1976). While we believe Bersch correctly denied subject-matter jurisdiction on its facts, we are not disposed to expand that holding to cases where conduct was preponderantly in the United States and the securities were American in practical effect.
. Paragraph 30 of the complaint charged that Lipper, together with others, “combined and conspired to do or cause to be done the acts alleged in this complaint which damaged IIT and its fundholders. Arthur Andersen and the Underwriter Defendants, jointly and severally, aided and abetted such conspiracy or conspiracies.” On the other hand, in paragraph 65 it is alleged that Lipper “aided and abetted the conspiracy by effectuating the purchases of the King Resources common stock.”
. United States v. Peoni, 100 F.2d 401, 402 (2 Cir. 1938), cited and approved in Nye & Nissen V. United States, 336 U.S. 613, 619, 69 S.Ct. 766, 770, 93 L.Ed. 919 (1949).
. The inner quotes are from TSC Industries v. Northway, Inc., 426 U.S. 438, 449, 96 S.Ct. 2126, 2132, 48 L.Ed.2d 757 (1976).
. IIT asserts that an IIT fundholder or Credit Suisse, the cash depository for IIT might have brought suit under Provision 806 of the Luxembourg Code of Civil Procedure to block the transfer of IIT funds. Actions in United States or Canadian courts might also have been mounted or requests made for investigation by the SEC.
Andersen argues that Goldberg should not apply to a mutual fund such as IIT, primarily because there is a larger turnover in fundholders than there was in the shareholders involved in Goldberg. We do not see that this makes any difference so far as the existence of a securities law violation is concerned. Andersen cited no cases supporting the distinction, and courts have generally treated mutual funds in the same manner as corporations in applying the antifraud provisions. See, e. g., Financial Industrial Fund, Inc. v. McDonnell Douglas Corp., 474 F.2d 514 (10 Cir.), cert. denied, 414 U.S. 874, 94 S.Ct. 155, 38 L.Ed.2d 114 (1973); Kauffman v. Dreyfus Fund, Inc., 434 F.2d 727 (3 Cir. 1970), cert. denied 401 U.S. 974, 91 S.Ct. 1190, 28 L.Ed.2d 323 (1971).
. 52. The November 1968 King Resources Capital Corp., N.V. Prospectus contained numerous misrepresentations and omissions. Chief among them were:
A. Failure to disclose the on-going fraudulent scheme whereby FOF and FOF Prop, were “investing” millions of dollars in worthless and/or extremely overvalued arctic properties;
B. Failure to fully disclose transactions with related companies such as The Colorado Corporation;
C. Failure to disclose the conspiracy with those in control of the IOS complex;
D. Failure to state revenues properly and to properly account for expenses; and
E. Failure to disclose that King Resources Company required a vast, steady supply of cash to repurchase outstanding limited partnership interests held by private investors.
. This interpretation is supported by the allegations in fl 88 of the complaint: “The IOS Defendants, the IIT Defendants, Arthur Andersen, Lipper and the Lipper Corporation knew, or should have known, that the transactions between the King empire and IIT violated the IIT Fund Regulations, were contrary to the investment policy set forth in IIT’s prospectuses and were not arm’s length transactions.” These allegations concern IIT’s acquisition of King securities and not the preparation of the KRC and KRCC prospectuses.
. Ruder, Multiple Defendants in Securities Law Fraud Cases: Aiding and Abetting, Conspiracy, In Pari Delicto, Indemnification and Contribution, 120 U.Pa.L.Rev. 597 (1972).
. In Peoni Judge Hand also said that the aider and-abettor must “by his action” seek to make the fraud succeed. Relaxation of the action requirement affords all the more reason to stress the association and participation aspects of the Peoni formulation.
. Ignorance of this requirement or a belief that it was mistaken would not authorize amendment at this stage. See Goss v. Revlon, Inc., 548 F,2d 405, 407 (2 Cir. 1976), cert. denied, 434 U.S. 968, 98 S.Ct. 514, 54 L.Ed.2d 456 (1977). In any event Ernst & Ernst was decided on March 30, 1976, and there was ample time to seek leave to amend before submission to the district court.
. Andersen cited the Luxembourg Code de la Legislation Commerciale, Societes Commer-ciales, § 1, art. 2, which lists six “judicial indi-vidualities” not including an indivisión.
. Justice Holmes stated that “knowledge of the facts by the new directors was knowledge by the bank, and none the less that according to the bill they in their turn were unfaithful. It is not alleged that they conspired with the defendants whose case we are considering. *931They came to the board as the eyes of the bank.” 257 U.S. at 264, 42 S.Ct. at 101. As noted above, we do not view the change from Cornfeld to Vesco control as a significant.one for present statute of limitations purposes. Although there is no allegation here that Cornfeld and Vesco conspired together, it is alleged that Vesco succeeded Cornfeld with the aim of defrauding IIT and the fundholders and that he subsequently did so. The “unfaithfulness” noted by Justice Holmes in Curtis apparently refers to the failure by the new directors to bring an action or recall the loans, and not to any subsequent wrongs similar to those of the original fraudulent directors. See id. at 264, 42 S.Ct. at 101. (“Any one of [the new directors] having notice was bound to do what he could to avert or diminish the loss.”). Curtis thus did not involve facts such as those before us, where the new management not only did not redress the wrong perpetrated by the old but actually embarked on more grievous wrongs of its own, also at the expense of plaintiffs.
. Another case upholding the defense of the statute, Costello v. Atlas Corporation, 297 F.Supp. 19 (N.D.Cal.1967), where plaintiffs, trustees in bankruptcy, sued defendant corporation in a diversity action for fraudulent dealings with the bankrupt company, illustrates the distinction. In ruling that the trustees were charged with the knowledge of the directors who knew of the wrong and were not charged with any wrongdoing themselves, the court went to considerable length to stress that the board was at all times independent and disinterested and, in particular, that the transactions were at arm’s length.
Here, the undisputed facts as taken from plaintiffs answers to defendants’ interrogatories establish that at all material times, TCC and TAL operated under independent boards of directors and officers, that Atlas was never even represented on either board, and that the boards acted properly and in good faith. Moreover, plaintiffs’ interrogatory answers affirm the fact that the Atlas-TCC arguments were negotiated at arm’s length, and were ratified or approved by TCC’s directors and shareholders.
This clearly implies that knowledge of the directors would not be imputed to trustees in a case such as ours, where the directors were implicated in the wrongdoing and the charge was precisely that the dealings were not at arm’s length.