OAKES, Circuit Judge:
On cross appeals from a judgment of the United States District Court for the Southern District of New York, Lawrence W. Pierce, Judge,1 plaintiff David E. Rolf (Rolf) endorses the district court’s holding on questions of liability, but challenges the district court’s measure of damages. Defendants in turn attack the district court’s theories of liability. In the court below, Judge Pierce imposed aiding and abetting liability on defendant Michael Stott (Stott) and derivative liability on his employer, Blyth, Eastman Dillon & Co., Inc. (BEDCO), for Stott’s substantial assistance to and participation in a web of securities fraud perpetrated by defendant Akiyoshi Yamada (Yamada),2 contrary to § 10(b) of the Securities Exchange Act of 1934 (SEA), 15 U.S.C. § 78j(b) and Rule 10b-5 thereunder, 17 C.F.R. § 240.10(b)-5. Judge Pierce also implied a private cause of action under New York Stock Exchange (NYSE) Rule 405 and under Article III, Section 2 of the constitution of the National Association of Securities Dealers (NASD). The Judge then awarded damages and interest totaling $55,790. We agree with Judge Pierce on the aiding and abetting liability of Stott. We therefore affirm as to him and in view of BEDCO’s acknowledgment for purposes of this appeal that it is vicariously liable by reason of Section 20 of the 1934 Act, 15 U.S.C. § 78t(a), affirm also as to it. Brief for Defendants-Appellees-Cross-Appellants at 28n.*. Accordingly, we do not reach the question whether there is an implied cause of action under the NYSE rule or the NASD constitution. We disagree with Judge Pierce on the measure of damages, however, and accordingly remand for reconsideration thereof.
I
Facts
Rolf is an ophthalmologist from Shaker Heights, Ohio. Long an investor and an aggressive trader in the stock market, he began his association with Eastman Dillon Union Securities & Co., BEDCO’s predecessor firm, in 1963 when he entrusted a discretionary account to S. Logan Stirling, a partner of the firm. The value of Rolf’s portfolio at that time was approximately $400,000. In March of 1969 Stirling was forced to retire owing to ill health. At the end of April, a BEDCO partner assigned the Rolf account to Stott, a registered representative with the firm for 11 years during 4 of which he was a manager of BED-CO branch offices. Stott then telephoned Rolf and offered his services. Rolf, however, wanted an investment advisor to manage his account, not simply a broker. Stott, therefore, at Rolf’s request, supplied the names of two investment advisors. Rolf ultimately interviewed and selected Yama-da, one of the “new breed” of young mon[42]*42ey-managers with supposed expertise in research and “special situations.”
The district court found that Rolfs investment intent was to combine Yamada’s and Stott’s strengths into a “Stirling-type” operation, 424 F.Supp. 1021, 1028 (1977). By combining Yamada’s youth and zeal with Stott’s reliability and supervision, Rolf hoped to realize, as he had with Stirling’s advice, substantial capital gain in an investment program emphasizing preservation and augmentation of capital. In furtherance of these investment objectives, Rolf executed a broad authorization giving his investment advisor, Yamada, full trading discretion. Rolf left the account and its accompanying trading commissions with Stott and BEDCO in return for Stott’s supervision of Yamada.
On May 9, 1969, the date of the trading authorization, Rolf’s equity in his portfolio stood at $1,423,000. The portfolio consisted of 21 good quality, listed securities and the warrants of two companies.3 By January of 1970 Yamada had liquidated the entire portfolio, selling 14 issues at a loss. The net value of Rolf’s portfolio had declined to approximately $712,000 of which $338,000 was invested in the restricted stock of Dela-nair, Inc.4 During this period of portfolio liquidation. Yamada and Stott were in daily contact. The district court found that out of 41 issues purchased for Rolf in the complaint period Stott “either recommended or was somehow involved with the decision to purchase” 12 securities, some of which were highly speculative. Id. at 1030.5
With the rash of new, unfamiliar securities which found their way into Rolf’s portfolio, Rolf became concerned and sought assurances from Stott as early as July, 1969. Specifically Rolf wished to ascertain that Yamada’s purchases were consistent with the former’s investment goals and strategy. To assuage Rolf’s fears, the district court found, Stott undertook a hand-holding operation whereby Stott would reassure Rolf of Yamada’s competence whenever Rolf questioned it. Id. at 1031. For example, when in August, 1969, Yamada decided to purchase nearly $400,000 in Dela-nair stock, Rolf checked with Stott who assured the doctor that if Yamada recommended the stock, then it was safe to proceed.
By March 29, 1970, the value of Rolf’s portfolio had dropped to $446,000 of which nearly one-half was tied up in Delanair. In early April, Rolf complained to Stott who began to assume the posture that he was a mere “order taker.” Rolf disagreed with this self-description, asking Stott to “work closely with Aki,” and reminded Stott that Stott was his “man in N.Y.” Id. at 1032— 33. Later, on December 14, 1970, Rolf again asked Stott to “keep [his] pulse on the situation.”
The district court discredited Stott’s testimony that he was not involved in the management of Rolf’s portfolio specifically finding that Stott was in fact so enmeshed. Id. at 1028, 1030, 1031. Stott and Yamada were in daily contact. Stott made numerous recommendations to Yamada for Rolf accounts using BEDCO research analysis, id. at 1030, but never counseled against a Yamada purchase. Id. at 1033. And most [43]*43of the trades were executed through Stott at BEDCO. For those stocks purchased through other brokerage houses, because of BEDCO internal rules, Stott received confirmation slips. Additionally, such securities were delivered to and held by BEDCO.
The district court’s finding on the question of Stott’s attitude toward the quality of the purchases is not altogether clear. The lower court states on the one hand that it gives “some weight to Yamada’s statement that Stott referred to the stocks in the Rolf account as ‘junk,’ ” i.e., of very low quality. Id. at 1033 (emphasis added). But the Judge goes on to conclude “that Stott did indeed consider many of the securities to be ‘junk’ and that he told this to Yama-da.” Id.
The district court unequivocally found, however, that Yamada was engaged in fraudulent stock manipulations, of which Stott was ignorant. The district court also concluded unambiguously that Yamada’s overall management of the account was fraudulent in nature, over and above the specific manipulations of which Stott was unaware. Id. at 1043. Stott was of course knowledgeable that many of the securities purchased for Rolf were highly speculative, “high-fliers.” Id. at 1035. Nevertheless, neither Stott nor BEDCO ever identified any security as unsuitable for Rolf. Id. at 1036. Stott’s services to Rolf consisted solely of certain “buy” recommendations, executing transactions and performing the accompanying paperwork. The court concluded that
Stott’s practice of continually voicing his confidence in Yamada and in Yamada’s investment decisions constituted a fraud upon Dr. Rolf, who sincerely believed that Stott had some basis for his statements. The statements of support and the assurances which were repeatedly made were made with willful and reckless disregard for whether they were true or false.
Id. at 1042.
II
District Court Holding
The district court based liability on alternative legal theories. The first was that Stott owed a fiduciary duty to Rolf which he breached; that by virtue of that breach Stott aided and abetted Yamada’s fraud and was therefore liable under § 10(b) of the SEA, 15 U.S.C. § 78j(b), and Rule 10b-5 thereunder, 17 C.F.R. § 240.10(b)-5; and that BEDCO’s liability for Stott’s participation in Yamada’s fraud derives alternatively from the common law doctrine of respon-deat superior or the securities law doctrine of controlling persons liability, § 20(a), SEA, 15 U.S.C. § 78t(a). The district court also rested liability on an implied private cause of action from violations of NYSE Rule 405 and the NASD constitution, Article III, Section 2.6 The Judge then awarded damages of $55,790, on a “churning” theory, even though he dismissed plaintiff’s churning claim, on the ground that any other measure of damages would be “speculative.” 424 F.Supp. at 1045.
III
Standard of Review
At the outset, we note that the evidence in this ease against Stott, while not overwhelming, is substantial. We are bound by the district court’s findings, as we discern them, on the basis of the clearly erroneous rule. Fed.R.Civ.P. 52(a); see United States Steel and Carnegie Pension [44]*44Fund, Inc. v. Orenstein, 557 F.2d 343 (2d Cir. 1977) (sub silentio). We have held on countless occasions that on review of district court findings, “we may not substitute our judgment on facts for that of the trial judge, who was in a superior position to appraise the evidence, and we may not reverse his findings unless, on the entire record, we are “ ‘ “left with the definite and firm conviction that a mistake has been committed.’ ” Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 123 [89 S.Ct. 1562, 23 L.Ed.2d 129] (1969), quoting United States v. United States Gypsum Co., 333 U.S. 364, 395 [68 S.Ct. 525, 92 L.Ed. 746] (1948).” Van Alen v. Dominick & Dominick, Inc., 560 F.2d 547, 550 (2d Cir. 1977). This is true because the trial judge is particularly able to assess the demeanor and credibility of witnesses. Id. at 551; see Newburger, Loeb & Co. v. Gross, 563 F.2d 1057, at 1070 (2d Cir. 1977).
IV
Liability
A. Stott
There is no doubt that Yamada perpetrated a gross fraud upon Rolf in violation of § 10(b) and Rule 10b-5. We conclude that Stott, by virtue of assurances of confidence in Yamada and in Yamada’s investment decisions and by virtue of his reckless disregard of whether those assurances were true or false and of substantial evidence that Yamada was improperly and fraudulently managing Rolf’s account, participated in and lent assistance to the fraud upon Dr. Rolf. The reasoning which leads us to this conclusion follows.
1. Reckless disregard of truth or falsity as constituting scienter. The starting point is Ernst & Ernst v. Hochfelder, 425 U.S. 185, 96 S.Ct. 1375, 47 L.Ed.2d 668 (1976). In Hochfelder, the Supreme Court explicitly failed to decide whether § 10(b) and Rule 10b-5 may, in appropriate circumstances, give rise to aiding and abetting liability and, if so, the elements of such a cause of action. Id. at 192 n. 7, 96 S.Ct. 1375. Although the Supreme Court has, therefore, not passed on the issue, our court has adopted the position that § 10(b) and Rule 10b-5 do permit the imposition of aiding and abetting liability. Hirsch v. DuPont, 553 F.2d 750, 759 (2d Cir. 1977); see Brennan v. Midwestern United Life Insurance Co., 417 F.2d 147 (7th Cir. 1969), cert. denied, 397 U.S. 989, 90 S.Ct. 1122, 25 L.Ed.2d 397 (1970); Note, Accountants’ Liabilities for False and Misleading Financial Statements, 67 Colum.L.Rev. 1437, 1448 (1967). Of course, the basic holding of Hochfeider, that scienter is an element of the § 10(b)/Rule 10b-5 cause of action,7 also establishes the standard for aiding and abetting liability.8
The question then becomes precisely what level of scienter is required in this type of 10b-5 case and whether the district court’s findings indicate that plaintiff’s proof satisfies that standard. We conclude on one of the questions left open by Hochfelder, 428 U.S. at 194 n. 12, 96 S.Ct. 1375, that at least where, as here, the alleged aider and abettor owes a fiduciary duty to the defrauded party,9 recklessness satisfies the scienter requirement. We arrive at this conclusion for several reasons.
First, by leaving open the possibility that recklessness might satisfy the scienter requirement, the Supreme Court recognized that in certain instances a recklessness standard might be appropriate.
[45]*45See Sundstrand Corp. v. Sun Chemical Corp., 553 F.2d 1033, 1044 (7th Cir. 1977) (“no hint in Hochfelder that the Court intended a radical departure from” common law analogue of fraud which imposes liability for reckless behavior), cert. denied, - U.S. -, 98 S.Ct. 225, 54 L.Ed.2d 155 (1977). The relationship most logically subjected to a recklessness standard, rather than some stricter standard involving proof of intent to defraud, is where the aider and abettor owes a direct fiduciary duty to the defrauded party. See Woodward v. Metro Bank of Dallas, 522 F.2d 84, 97 (5th Cir. 1975). Liability premised on the recklessness of one’s fiduciary in failing to perform his duty to disclose is a far cry from awarding damages for simple negligence. See Ernst & Ernst v. Hochfelder, supra. Clearly, Stott, as Rolf’s broker, owed Rolf a fiduciary duty. See Hanly v. SEC, 415 F.2d 589, 596, 597 (2d Cir. 1969).10
Second, on a linguistic level, the term scienter11 is used by the Supreme Court to mean, in the disjunctive, “knowing or intentional misconduct.” 425 U.S. at 197, 96 S.Ct. 1375.12 Use of the word knowing” implies conduct which is somewhat less directed and focused than “intentional” activity which commonly is characterized by a specific mental state whose animus is to bring about a particular result, see W. Prosser, Law of Torts § 107, at 700 (4th ed. 1971). “Knowing” is a word laden with common law connotations: at common law, reckless conduct is viewed as a form of knowing conduct. Id. at 701. For example, the common law requirement of scienter as an element of the tort of deceit or misrepresentation may be proved in a number of ways: ii
There is of course no difficulty in finding the required intent to mislead where it appears that the speaker believes his statement to be false. Likewise, there is general agreement that it is present when the representation is made without any belief as to its truth, or with reckless disregard whether it be true or false.
Id. (footnotes omitted). It is unquestionable that the common law has served as an interpretive source of securities law concepts. See Holdsworth v. Strong, 545 F.2d 687, 693-94 (10th Cir. 1976), cert. denied, [46]*46430 U.S. 955, 97 S.Ct. 1600, 51 L.Ed.2d 805 (1977). The common law tort of fraud has adopted a recklessness standard as one means of satisfying the requisite intent element of that cause of action.13 Similarly, securities law cases have recognized that recklessness may serve as a surrogate concept for willful fraud. Concurring in SEC v. Texas Gulf Sulphur Co., 401 F.2d 833, 868 (2d Cir. 1968) (en banc), cert. denied sub nom. Coates v. SEC, 394 U.S. 976, 89 S.Ct. 1454, 22 L.Ed.2d 756 (1969), Judge Friendly noted the distinction between a “merely negligent misstatement” and “the kind of recklessness that is equivalent to wilful fraud.” See Haimoff, Holmes Looks at Hochfelder and 10b-5, 32 Bus.Law 147, 162 (1976).
Third, it is consistent with, if not demanded by, precedent in this circuit14 to hold that reckless conduct satisfies the scienter requirement. See Lanza v. Drexel & Co., 479 F.2d 1277, 1306 (2d Cir. 1973) (en banc); Shemtob v. Shearson Hammill & Co., 448 F.2d 442, 445 (2d Cir. 1971); Bucklo, The Supreme Court Attempts to Define Scienter Under Rule 10b-5: Ernst & Ernst v. Hochfelder, 29 Stan.L.Rev. 213, 214 (1977). Lanza v. Drexel, supra, of course required scienter as an element of a 10b-5 cause of action prior to the Supreme Court’s decision in Hochfelder. The Lanza test was stated in terms of “willful or reckless disregard for the truth,” id. at 1306 (emphasis added), thereby recognizing that either intentional or reckless behavior is the predicate mental state for 10b-5 liability. The Hochfelder Court noted that a number of courts of appeals “have held that some type of scienter — i. e., intent to defraud, reckless disregard for the truth, or knowing use of some practice to defraud — is necessary in such an action” and then cited inter alia our decision in Lanza, 425 U.S. at 194 n. 12, 96 S.Ct. at 1381. Thus, on balance, we consider that Hochfelder left intact our rule that recklessness is a form of scienter in appropriate circumstances.15
[47]*47A final basis for applying a recklessness standard in certain instances rests perhaps on the practical problem of proof in private enforcement under the securities laws. Proof of a defendant’s knowledge or intent will often be inferential, see Ruder, Multiple Defendants in Securities Law Fraud Cases : Aiding and Abetting, Conspiracy in Pari Delicto, Indemnification, and Contribution, 120 U.Pa.L.Rev. 597, 635 (1972), and cases thus of necessity cast in terms of recklessness. To require in all types of 10b-5 cases that a factfinder must find a specific intent to deceive or defraud would for all intents and purposes disembowel the private cause of action under § 10(b).
2. Other elements of aiding and abetting liability. Given, then, that we find aider and abettor liability appropriate under § 10(b) and given that we believe at the very least that fiduciaries have acted with scienter when they have been reckless, we pass to the other well-established elements of the aiding and abetting cause of action. The first element that a plaintiff must prove is that the primary party, here Yamada, as distinguished from the secondary aiding and abetting party, committed a securities law violation. Woodward v. Metro Bank of Dallas, supra, 522 F.2d at 95; see SEC v. Coffey, 493 F.2d 1304, 1316 (6th Cir. 1974), cert. denied, 420 U.S. 908, 95 S.Ct. 826, 42 L.Ed.2d 837 (1975); cf. Landy v. Federal Deposit Insurance Corp., 486 F.2d 139, 162 (3d Cir. 1973) (requiring “independent wrong” instead of independent securities law violation), cert. denied, 416 U.S. 960, 94 S.Ct. 1979, 40 L.Ed.2d 312 (1974). The district court found, and we see no basis for concluding otherwise, that Yamada’s overall management of the account in violation of his fiduciary duties owed by reason of the investment advisory agreement was fraudulent. 424 F.Supp. at 1043. Indeed the district court called Yamada’s handling of the account a “gross fraud.” Id. Moreover, Stott and BEDCO in their brief acknowledge that Yamada committed securities fraud. Their very purpose is to point the finger at Yamada in order to exonerate Stott. Brief for Defendants-Appellees-Cross-Appellants at 28, 34, 42, 44, 47. There is therefore no reason to analyze Rolf’s portfolio on a stock by stock basis to determine which purchases and sales constituted frauds upon Rolf, a more specific and particularized investigation which we might have to undertake if Yamada had merely manipulated two or three stocks without engaging in a more exhaustive and all-encompassing web of fraud.
The second requirement for establishing the aiding and abetting violation is Stott’s knowledge of Yamada’s fraud. We have indicated above that the scienter element may be satisfied by proof of reckless conduct. While the evidence in this case is not overwhelming, we believe it is sufficient to sustain Judge Pierce’s findings on the basis of the clearly erroneous rule.
Reckless conduct is, at the least, conduct which is “highly unreasonable” and which represents “an extreme departure from the standards of ordinary care . to the extent that the danger was either known to the defendant or so obvious that the defendant must have been aware of it.” Sanders v. John Nuveen & Co., 554 F.2d 790, 793 (7th Cir. 1977).16 Stott was aware that the quality of the securities being purchased by Yamada was very low. The daily contact between Stott and Yamada gave Stott ample opportunity to supervise the investment advisor, an obligation which the lower court found Rolf had sought and Stott had undertaken to perform. Stott’s most egregious breach of his duty to Rolf was his constant reassurance that Yamada knew what he was doing and that if Yamada were purchasing stocks they must be satisfactory. These representations were conclusorily made to Rolf without investigation and with utter disregard for whether [48]*48there was a basis for the assertions. “A representation certified as true when knowledge there is none, a reckless misstatement, or an opinion based on grounds so flimsy as to lead to the conclusion that there was no genuine belief in its truth, are all sufficient upon which to base liability.” State Street Co. v. Ernst, 278 N.Y. 104, 112, 15 N.E.2d 416, 418-19 (1938). Stott’s representations and opinions were given without basis and in reckless disregard of their truth or falsity.
The third and final requirement to establish the aiding and abetting violation is that Stott rendered substantial assistance to Yamada in the fraudulent mismanagement of Rolf’s portfolio. One commentator has suggested that substantial assistance might include “repeating . . . misrepresentations (or aiding in their preparation), by acting as conduits to accumulate or distribute securities, by executing transactions or investing proceeds, or perhaps by financing transactions.” 2 A. Bromberg, Securities Law § 8.5 (515) (1974). In this case, Stott’s assistance was both active and passive. He processed many of the relevant securities orders; he reassured Rolf of Ya-mada’s competence; and he either recklessly failed to learn of or failed to disclose Yamada’s web of fraud. See Woodward v. Metro Bank of Dallas, supra, 522 F.2d at 96-97. The effect of Stott’s “hand-holding operation” was to prevent Rolf from discovering Yamada’s fraud. Stott’s acts, therefore, were a substantial causal factor in the perpetuation of Yamada’s fraud and in the cumulation of Rolf’s losses. See Lan-dy v. Federal Deposit Insurance Corp., supra, 486 F.2d at 163. We therefore affirm the trial court’s judgment to the extent that it found Stott an aider and abettor of Yamada’s fraud.
B. BEDCO
While there is disagreement among the circuits17 and, perhaps, even disharmony within the Second Circuit18 on the relationship between § 20(a) of the 1934 Act, 15 U.S.C. § 78t(a), and the common law doctrine of respondeat superior, BEDCO has conceded “for the purposes of this appeal” that if Stott is found liable it is vicariously liable under Section 20, even though the decision below stated that “BEDCO did not have actual notice of the fraud.”19
V
Damages
We are not capable of precisely measuring Rolf’s damages on this appeal, although we do not think that they were so speculative as to compel resort solely to damages as in a churning case, for commissions paid the broker (and interest thereon). Accordingly, we remand to the district court to determine damages in accordance with the guidelines set forth below.
[49]*49First, the district court should determine as near as possible the time when Stott began to aid and abet Yamada’s fraud20 and compute the market value of Rolf’s portfolio on that date. Second, the district court should subtract the value of the portfolio on the date when Stott’s participation in and assistance to the fraudulent scheme ceased from the value on the date when Stott became an aider and abettor. This amount is Rolf’s gross economic loss.21 See Note, Churning by Securities Dealers, 80 Harv.L.Rev. 869, 884 (1967). The district court should then reduce Rolf’s gross economic loss by the average percentage decline in value of the Dow Jones Industrials, the Standard & Poor’s Index, or any other well recognized index of value, or combination of indices, of the national securities markets during the period commencing with Stott’s aiding and abetting and terminating with its cessation.22 Thus if during the relevant period the stock market declined in value by 25%, then Rolf’s gross economic loss should be reduced by 25%. Because plaintiff’s theory of liability is to the effect that Stott’s aiding and abetting prevented Rolf’s discovery of Yamada’s fraud, it is not fair to reduce damages any further by the amount of loss on a discrete transaction to which Stott had no connection because he did not execute the transaction, recommend the security, or reassure Rolf with respect to that security.23 However, should the loss on the Delanair stock [50]*50be determined to have occurred during the aiding and abetting period, Rolfs damages must be reduced by $175,000, the amount for which he settled his claims against several Delanair-related defendants. We also hold that Rolf is entitled to a return of commissions paid to Stott and BEDCO, but only as to transactions falling within the aiding and abetting period, with interest thereon as determined by the district judge.
Finally, we request the district judge to reconsider his decision on the question of prejudgment interest in view of our obvious conclusion that Rolf was deprived of a principal sum. See Nelson v. Hench, 428 F.Supp. 411 (D.Minn.1977). While such an award is a matter of judicial discretion, Blau v. Lehman, 368 U.S. 403, 414, 82 S.Ct. 451, 7 L.Ed.2d 403 (1962); Norte & Co. v. Huffines, 416 F.2d 1189, 1191-92 (2d Cir. 1969), cert. denied sub nom. Muscat v. Norte & Co., 397 U.S. 989, 90 S.Ct. 1121, 25 L.Ed.2d 396 (1970), it is not unreasonable to request the district judge to set forth his reasons should he again deny prejudgment interest. See Wessel v. Buhler, 437 F.2d 279, 284 (9th Cir. 1971).
Judgment affirmed in part and remanded.