The Solitron defendants appeal from the court's denial of their motion to decertify the class, made on the ground that the named representatives had purchased their shares after the date stipulated as the closing date for their subclass; from the court's decision not to set aside the verdict against them with respect to fiscal year-end financial statements for 1967, 1968, and 1970; from the amount of damages; and from the court's exclusion of certain proffered testimony. The plaintiffs cross-appeal from the court's decision to set aside the verdict against the Solitron defendants with respect to fiscal year-end financial statements for 1972, 1973, and 1974; and from the court's decision to set aside the verdict against Sternbach with respect to the 1967, 1968, and 1970 reports. We reverse the damage award and the decision to grant judgment in Sternbach's favor notwithstanding the verdict, and remand for redetermination of damages and determination of the amount of contribution owed Sternbach by the Solitron defendants.

Solitron, a manufacturer of electronic semi-conductors, was subject by virtue of its contracts with the United States government to the Renegotiation Act of 1951, as amended, 50 U.S.C.App. §§ 1211-1233. In 1972 the Eastern Regional branch of the United States Renegotiation Board, which enforces the Act's limits on profits from government contracts, determined that Solitron had realized renegotiable profits of.$3.2 million in fiscal year 1967 and $4.4 million in fiscal year 1968.

Solitron, seeking administrative review of the Board's assessment, retained Price Waterhouse & Co. to reexamine its financial statements for 1967 through 1970. Price Waterhouse concluded that those statements, prepared and certified by Sternbach and signed by Benjamin Friedman, Solitron's chief executive officer and largest shareholder, had substantially overstated inventories and sales. In 1973 and 1974 Solitron, seeking to avoid liability for excess profits, disclosed these overstatements to the Renegotiation Board in letters prepared by James S. Trager, a former Sternbach accountant who was then Solitron's assistant treasurer, and James P. Barry, then Solitron's treasurer. In January 1975 the Renegotiation Board issued a final determination that Solitron owed $3.9 million in excess profits from 1967 to 1970, which Solitron is contesting before the Court of Claims. In March 1975 the Securities and Exchange Commission (SEC) brought an action against Solitron for violations of the securities laws, but withdrew the charges the next month pending an investigation by special SEC counsel, who concluded in June 1978 that Solitron had overstated income, but had not done so fraudulently.

Named plaintiffs Howard Sirota (who purchased Solitron shares on June 1, 1971) and Family Restorations (who purchased Solitron shares on April 7, 1971) filed the instant action in March 1975. They contended that the misrepresentations of fact in the company's 1967-70 financial statements that were revealed in the proceedings before the Renegotiation Board demonstrated that defendants Solitron, Friedman, Trager, Barry, and Sternbach had violated section 10(b) of the Securities Exchange Act of 1934, 15 U.S.C. § 78j, and Rule 10b-5 promulgated thereunder, 17 C.F.R. § 240.10b-5. They complained that inventories had been overstated while consignments had been improperly treated as sales, and that Solitron had failed to provide any reserve for probable refunds to the government of renegotiated profits. Their complaint was consolidated with others against Solitron in February 1976.

In February 1976 the named plaintiffs sought certification as representatives of a single class of Solitron investors who purchased common stock during the period beginning February 28, 1967 and ending March 20, 1975, and who thereafter sold at a loss. Subsequently they moved for the designation of three subclass periods: (1) purchasers between May 5, 1967 and June 15, 1971 (which included Howard Sirota and Family Restorations); (2) purchasers between June 15, 1971 and June 22, 1972; (3) purchasers between June 22, 1972 and March 20, 1975. The second of these subclasses was dismissed at trial for want of proof of its claims.

The first subclass was originally certified by the court to include purchasers between May 5, 1967 and June 2, 1971, notwithstanding Solitron's argument that its December 14, 1970 press release announcing losses and inventory write-downs of $7 million, which plaintiffs contended was itself fraudulent, had nullified any effect of the alleged misrepresentations on post-December purchasers of Solitron stock. The parties stipulated at the end of trial that the subclass closed on December 16, 1970. The claims of this 1967-70 subclass prevailed before the jury, and the court, refusing to decertify, granted judgment notwithstanding the verdict to Sternbach but not to Solitron, against whom the court upheld the jury's damage award. The third subclass, certified by the court to include purchasers between June 8, 1972 and January 27, 1975, claimed that Solitron's annual reports for 1972-74, which stated that the Renegotiation Board had assessed Solitron's excess profits but failed to disclose the amounts, were materially false and misleading. Although this subclass also prevailed before the jury, the court granted judgment to the Solitron defendants notwithstanding the verdict.

The Solitron defendants appeal first from the court's decisions to certify and not to decertify the class represented by plaintiffs Sirota and Family Restorations. The court certified this subclass on February 13, 1979 to include purchasers of Solitron stock between May 5, 1967 and June 2, 1971 (two days after release of the 1970 annual report). The parties stipulated at the close of trial, however, that December 16, 1970 (two days after the press release reflecting $7 million dollar inventory write-downs) was the closing date for the class. Family Restorations bought its shares on April 7, 1971; Sirota on June 1, 1971. Thus the named plaintiffs were in the subclass as originally certified but out of it as stipulated. The Solitron defendants argue on appeal that the named plaintiffs were not qualified to represent the class. They argue that the court erred, first by not ruling before certification that the December 1970 press release had cut off any injury to the market or to the named plaintiffs from the 1967, 1968, and 1970 financial reports, and second by not decertifying the class when the plaintiffs rested their case without having shown that the December 1970 press release was fraudulent.

Judge Brieant certified the subclass as extending until June 2, 1971 on the ground that precertification "inquiry into the merits of a suit" was barred by Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 177, 94 S. Ct. 2140, 2152, 40 L. Ed. 2d 732 (1974). Eisen held that Fed.R.Civ.P. 23 gave a district court no authority to hold a precertification hearing on the merits in order to determine whether a suit may be maintained as a class action, and therefore that the lower court's allocation of the costs of class notice to defendants upon a preliminary determination that plaintiffs were "more than likely" to prevail was improper. Class certification motions are not subject to the same standards as motions for dismissal for failure to state a claim or for summary judgment. See Miller v. Mackey International, Inc., 452 F.2d 424, 428 (5th Cir. 1971), cited with approval in Eisen, 417 U.S. at 178, 94 S.Ct. at 2153.

On the other hand, there can be no doubt that it is proper for a district court, prior to certification of a class, to allow discovery and to conduct hearings to determine whether the prerequisites of Rule 23 are satisfied. "(A) preliminary hearing, addressed not to the merits of plaintiff's individual claim, but to whether he is asserting a claim, which, assuming its merit, will satisfy the requirements of Rule 23, has never been regarded as violative of the rule stated in Eisen ...." Doctor v. Seaboard Coast Line Railroad Co., 540 F.2d 699, 707 (4th Cir. 1976) (emphasis in original) (footnote omitted). Indeed a district court may be reversed for premature certification if it has failed to develop a sufficient evidentiary record from which to conclude that the requirements of numerosity, typicality, commonality of question, and adequacy of representation have been met. See, e.g., Chateau de Ville Productions, Inc. v. Tams-Witmark Music Library, Inc., 586 F.2d 962, 966 (2d Cir. 1978).

The issue here falls between the prohibition in Eisen and the obligation to rest certification under Rule 23 on something more than the pleadings, see, e.g., Professional Adjusting Systems of America, Inc. v. General Adjustment Bureau, Inc., 64 F.R.D. 35, 38 (S.D.N.Y.1974) (Gurfein, J.). If the Solitron defendants were arguing that a district court must determine whether the named plaintiffs have a meritorious claim before they can be certified as class representatives, they would plainly be wrong. See, e.g., Huff v. N.D. Cass Co., 485 F.2d 710, 714 (5th Cir. 1973) (en banc) (vacating dismissal of class action on ground that "a class plaintiff who otherwise meets the demands of 23(a) and (b) should not be found to be disqualified solely by an advance determination that his claim is predictably not a winning claim and that, therefore, he cannot adequately represent the class as mandated by 23(a) (4)"). Solitron's argument may be, rather, that some kinds of merits determinations are crucial to determining whether a class action is proper and that in such circumstances Eisen poses no bar.

The dictum in Huff did suggest that some very basic merits determinations-e.g., whether a named plaintiff suing his employer was ever employed by the defendant, 485 F.2d at 714-may be made prior to certification because they affect whether the named representative has the nexus with the class required by Rule 23. But the determination involved here is not so basic. No case cited by Solitron supports the proposition that a district court abuses its discretion by certifying when it later appears that plaintiffs lacked a meritorious cause of action. Doctor v. Seaboard Coast Line Railroad Co. cites the Huff dictum, but only to support the more obvious proposition that the court may properly "identify the character or type (but not the merits) of each plaintiff's claim and then ... determine whether there was a class to which such claim ... was common and of which it was typical...." 540 F.2d at 708-09 (emphasis in original). In Doctor the district court denied class representative status to plaintiffs whose claims were plainly unfit for a class action because they were not shared by any other employees; the court never reached nor thought it proper to reach the merits of the individual claims.

As Judge Gurfein wrote, in making a certification decision, a judge must look somewhere "between the pleading and the fruits of discovery .... (E)nough must be laid bare to let the judge survey the factual scene on a kind of sketchy relief map, leaving for later view the myriad of details that cover the terrain." Professional Adjusting Systems of America, Inc. v. General Adjustment Bureau, Inc., 64 F.R.D. at 38. On this view, it would be improper for a district court to resolve substantial questions of fact going to the merits when deciding the scope or time limits of the class. Even a case on which the Solitron defendants rely heavily, In re LTV Securities Litigation, 88 F.R.D. 134, 147-48 (N.D.Tex.1980), supports this view. In that case, Judge Higginbotham decided to close the class period on the date of a press release he regarded as cutting off further claims. He also chose the earlier of two suggested dates for opening the class period, however, thus including in the class those who purchased before the first announcement of restated earnings, stating that while "the court has reservations as to whether plaintiff can make out a claim with regard to the 1975 period, (it) believes that a sufficiently substantial question has been presented such that the class period must commence at the earlier time period proffered." Id. at 147. Thus where it was disputed whether part of a proposed class had a cause of action, Judge Higginbotham certified the class using the more inclusive time period.

Here too, the district court may properly have believed at the time of the decision to certify that there was a substantial question of fact for the jury whether the December 1970 press release had cured the market, barring post-December claims, or was itself fraudulent. Solitron's arguments about that release would then have appeared to be defenses on the merits. The parties' post-trial stipulation to a December 16, 1970 closing date for the subclass, even though it mooted the merits of this issue, cannot be applied retroactively to the pretrial decision to certify.1 We therefore find that it was proper for Judge Brieant to certify the class as he did, including post-December claimants within the first subclass and allowing Sirota and Family Restorations to act as class representatives.

Although a district court may decertify a class if it appears that the requirements of Rule 23 are not in fact met, it need not decertify whenever it later appears that the named plaintiffs were not class members or were otherwise inappropriate class representatives. Rather, the Supreme Court has stated, "provided the initial certification was proper ... the claims of the class members would not need to be mooted or destroyed because subsequent events or the proof at trial had undermined the named plaintiffs' individual claims." East Texas Motor Freight System, Inc. v. Rodriguez, 431 U.S. 395, 406 n.12, 97 S. Ct. 1891, 1898, 52 L. Ed. 2d 453 (1977) (citing Franks v. Bowman Transportation Co., 424 U.S. 747, 752-57, 96 S. Ct. 1251, 1258-61, 47 L. Ed. 2d 444 (1976)). Here certification was proper, and there was no need for decertification. But for the parties' stipulation to December 16, 1970 as the closing date, the question of the effect of the press release (and thus the question of the merit of the named plaintiffs' claims) would have been resolved by the jury verdict. The defendants have not shown that the named plaintiffs failed to represent the class adequately, and the class's victory before the jury might be thought to rebut such an argument in any case. Thus the subsequent definition of the class to exclude the individual named plaintiffs did not require decertification of the class.

1. Fraud by the Solitron Defendants, 1967-70

The Solitron defendants also appeal from the district court's decision not to set aside the jury's verdict that in the company's financial statements for 1967, 1968, and 1970 they knowingly and materially misrepresented inventories and characterized certain consignments as sales. As the court properly noted, judgment notwithstanding the verdict may be entered only if the evidence, viewed in the light most favorable to the non-movants without considering credibility or weight, reasonably permits only a conclusion in the movants' favor. Mattivi v. South American Marine Corp., "Huguenot", 618 F.2d 163, 167 (2d Cir. 1980); Simblest v. Maynard, 427 F.2d 1, 4 (2d Cir. 1970).

a. Inventory Overstatements

It is undisputed that the financial statements for the fiscal years ending 1967, 1968, and 1970 materially overstated inventory. The Solitron defendants argue collectively, however, as they did below to the jury and the court, that the inventory recosting that demonstrated the substantial overstatements was not even possible until Solitron had acquired for the first time a computer capability and a sophisticated inventory costing system based upon newly developed time-and-motion studies recommended by Price Waterhouse in 1972, some two years after the misstatements. They therefore argue that the evidence was insufficient to show scienter, i.e., that there was either actual knowledge of the misrepresentations or a reckless disregard for the representations' truth or falsity. See Ernst & Ernst v. Hochfelder, 425 U.S. 185, 96 S. Ct. 1375, 47 L. Ed. 2d 668 (1976); IIT, an International Investment Trust v. Cornfeld, 619 F.2d 909, 923 (2d Cir. 1980). See also Rolf v. Blyth, Eastman Dillon & Co., 570 F.2d 38, 47-48 (2d Cir.), cert. denied, 439 U.S. 1039, 99 S. Ct. 642, 58 L. Ed. 698 (1978), on appeal from remand, 637 F.2d 77 (2d Cir. 1980).

We agree with Judge Brieant's post-verdict assessment that there was sufficient evidence of scienter on the part of the company and the individual defendants to sustain the jury verdict against the Solitron defendants for 1967, 1968, and 1970. The jury could properly infer intent from subsequent admissions of misrepresentations, coupled with the defendants' continuous intimate knowledge of company affairs. Solitron benefited from overstating its inventory. From 1966 through 1970 its reported earnings per share surged upward 100%, 50%, 331/3, and 20% respectively. The price of the stock followed the reported earnings, enabling Solitron to make numerous acquisitions2 after a five-for-one split in April 1968. There was direct evidence that Friedman as well as Trager knew that inventory had increased from $1.5 million in 1966 to $2.94 million in 1967, while the inventory-turnover rate declined from 6 times a year to 4.5 times a year; that inventory had increased nearly $5 million in 1968, reducing inventory turnover to less than 2.5 times per year; and that in 1970 inventory had increased by nearly $4 million to $12.5 million, with a turnover rate of only 1.9. The Solitron defendants admitted such overstatements of inventory to the Renegotiation Board in 1974.

There was also direct evidence that accounting procedures at the Riviera Beach, Florida plant, where Solitron carried on its government business, were inadequate to cost inventory accurately. As Barry stated in a letter to renegotiation counsel, the accounting department was "understaffed" and "poorly managed," in keeping with the company's policy at that time: "get the shipment out the back door, get the invoice in the mail to the customer, operate with a minimum overhead, and profits will be generated." Inability to cost certain "highly reliable" and commercial inventory resulted in overstatements in 1967 of $556,000, in 1968 of $1.35 million, and in 1969 of $904,000; there were also raw-material and piece-part costing and other errors of lesser magnitude. Mathematical calculation and addition errors in the year 1968 amounted to $1.05 million.

The evidence of each individual defendant's knowledge of and involvement in the preparation and publication of false financial statements was sufficient to support findings that they each violated Rule 10b-5. Friedman, the head of the company, was in touch with the Florida plant by telephone each day, received regular summaries and reports from the plant, and frequently visited it. He testified at trial that he knew the effect of inventory valuation on profits. There was thus evidence from which the jury could have concluded that he was aware of the company's failure to cost its inventory properly.

Trager, who had been employed by Sternbach until 1971 and who had worked on Solitron audits every year from 1964 to 1969, became comptroller of Solitron in the spring of 1969, assistant treasurer in 1970, and treasurer in 1974. He prepared the reports to the Renegotiation Board showing that profits had been inflated because of overvaluation of inventory, and testified that he knew where to look for the evidence of this overvaluation.

Barry, who was liable for 1970 only, spent fifteen years as an accountant in the Intelligence Division of the Internal Revenue Service before joining the Solitron accounting department at Riviera Beach in July 1968. He became head of that accounting department in 1969 and treasurer of Solitron in 1970. He went back to the 1967-70 inventory figures for Riviera Beach and showed the inventory overstatements. His documentation helped Price Waterhouse reach its conclusions about inventory overstatements and helped Trager prepare his reports to the Renegotiation Board. Barry admitted to the Renegotiation Board the inadequacies of the Riviera Beach accounting procedures. His testimony at trial showed that he had not verified the 1970 inventory sheets that he certified as properly costed.

Without the overstatements, not only would the earnings per share have been less during the years 1966 through 1970, but the gradient of earnings growth would have been flatter. Indeed in 1970 there would have been a decline in earnings. There was expert testimony before the jury about the effect that the overstatement of earnings and the gradient of earnings growth have on stock prices generally. Overall, we believe there was substantial evidence to support the jury's verdict that the Solitron defendants all had the necessary scienter; the court's refusal to set aside the verdict was proper.

b. Consignments Reported as Sales

The Solitron defendants claim further that there was no evidence they committed fraud in reporting three consignment transactions-with Best & Raynor, JFD Electronics Co., and AEL Israel, Ltd.-as sales. We find Judge Brieant's conclusion to the contrary correct. Trager, in preparation for the Renegotiation Board proceedings, went back to documents that had been in his possession in 1969 and 1970, and showed that the transactions classified as sales had in fact been consignments. Friedman negotiated and signed the written contract with AEL Israel, which was specified as a consignment arrangement in accord with the understanding of the principal officers of AEL. Yet over $700,000 of consignments were carried on the books as sales. Although Friedman said he thought the first draft of the agreement provided that the risk would be on AEL, the jury did not have to credit his testimony that he believed the contract-which he signed and which placed the risk of loss on Solitron-provided for a sale.3

On the cross-appeal, the plaintiffs object both to the district court's finding that there was insufficient evidence to support the verdict against Sternbach, and to the court's failure to give the same "recklessness" charge with respect to Sternbach that it gave with respect to the Solitron defendants. The court instructed that actual knowledge by Sternbach of the fraud by Solitron and by the individual defendants was required. The jury found scienter, but the judge set that finding aside. The jury also found, with respect to a pendent state law count, that Sternbach had been negligent, a finding which Judge Brieant set aside and which is not on appeal.

This court has held that proof of reckless conduct meets the requirement of scienter in a section 10(b) claim, IIT v. Cornfeld, 619 F.2d at 923. For the imposition of aider and abettor liability under section 10(b), however, we have held that recklessness satisfies the scienter requirement where "the alleged aider and abettor owes a fiduciary duty to the defrauded party," Rolf v. Blyth, Eastman Dillon & Co., 570 F.2d at 44. We have not decided whether recklessness is sufficient in the absence of such a relationship. Id. at 44 n.9. At least two district court judges have concluded that recklessness may be sufficient, at least in the case of accountants who know or can reasonably foresee that third parties will rely on their audit or opinion letter. See Oleck v. Fischer, (1979 Transfer Binder) Fed.Sec.L.Rep. (CCH) P 96,898 (S.D.N.Y.1979), aff'd, 623 F.2d 791 (2d Cir. 1980); In re Investors Funding Securities Litigation, 1980 Fed.Sec.L.Rep. (CCH) P 97,717 at 98,763 (S.D.N.Y.1980).

We have said in dicta in connection with accountants' liability that it is helpful to refer to Judge Goldberg's statement in Woodward v. Metro Bank of Dallas, 522 F.2d 84, 97 (5th Cir. 1975):

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.

See Edwards & Hanly v. Wells Fargo Securities Clearance Corp., 602 F.2d 478, 484-85 (2d Cir. 1979). See also IIT v. Cornfeld, 619 F.2d at 924-25. The dicta also seem to refer to a sliding scale of intent; Judge Gurfein wrote that "something closer to an actual intent to aid in a fraud" is necessary to find a person liable for aiding and abetting, absent a fiduciary relationship, than is necessary to find the principal liable. Edwards & Hanly, 602 F.2d at 485. See also IIT v. Cornfeld, 619 F.2d at 925. The evidence here at the least would support a finding that Sternbach's conduct-with respect to mathematical errors, consignments recorded as sales, and inventory overvaluation-was highly reckless. In light of our holding below, however, we need not reach the question whether Sternbach, in view of whatever duty it had to disclose, should have been judged under a recklessness standard.

The court granted judgment to Sternbach notwithstanding the verdict on the count of aiding and abetting Solitron's 10b-5 violation, finding "no evidence from which a reasonable juror acting reasonably could conclude that Sternbach aided and abetted the primary securities law violation with the requisite scienter." The court, interpreting the requisite scienter to be actual knowledge of the fraud since Sternbach did not owe a fiduciary duty to the plaintiff investors, found that the evidence at most showed that Sternbach was negligent. The proof did not show, the court wrote, that Sternbach's partner Edward Cole knew that the Best & Raynor, JFD Electronics, and AEL Israel transactions were consignments, even though "no confirmations were received by Sternbach when requested, no payments were made, no purchase orders were produced and shipments were made close to the end of the fiscal year," and even though a letter from AEL Israel to Cole said the transaction was a consignment. The court found also that even if, as plaintiffs sought to prove, Sternbach performed inadequate and erroneous audits, this did not constitute knowing assistance in the fraudulent overstatement of inventory.

Recalling the stringent standard for granting judgment notwithstanding the verdict, we disagree with the court's conclusion that no reasonable juror could have reached a verdict against Sternbach. First, the court's conclusion that Sternbach lacked actual knowledge is inconsistent with the jury's finding, upheld by the court on defendants' post-trial motion, that Trager-who was employed by Sternbach in 1967, 1968, and early 1969-had actual knowledge of the fraud. Second, evaluation of Cole's credibility, in the face of evidence suggesting he was on notice that the transactions he certified as sales were consignments, was for the jury, not the court. And in light of Sternbach's claims to have performed various costing procedures and tests on Solitron's 1967-70 inventory, the jury might well have concluded that Sternbach knew that its certification of reports overstating inventory was fraudulent. The combination of these facts leads us to the conclusion that there was sufficient evidence for the jury to infer that Sternbach had actual knowledge of fraud in violation of Rule 10b-5.

3. Fraud by the Solitron Defendants, 1972-74

On the cross-appeal, plaintiffs also argue that the district court erred in finding that there was no evidence to support the verdict against the Solitron defendants with respect to the financial statements for 1972-74. While these financial statements stated that the Renegotiation Board had determined that Solitron had made excess profits, they did not state that the amounts involved were.$3.2 million for the fiscal year ending February 28, 1967 and $4.4 million for the fiscal year ending February 28, 1968. They stated only that the company was contesting the determinations and that management believed that no significant refunds would be required after renegotiation. Plaintiffs argued below that these statements were materially misleading to investors because they suggested that the renegotiation proceedings had minimal significance.

The judge overturned the jury verdict in plaintiffs' favor because he found that the evidence did not support any conclusion other than that Solitron management, which never believed that the renegotiation assessment was meritorious, had truthfully disclosed to the public. The defendants knew from the Price Waterhouse reports, which they themselves commissioned, that Solitron inventory and income had been materially overstated; they therefore believed that there were no excess profits. Their renegotiation representative, Alexander Kirk, had advised them that the renegotiation assessment had no merit.

While we find in reviewing the record that the question is a close one, we do not think the court erred in setting aside the verdict on this count. There was considerable evidence supporting the view that the defendants reasonably believed that the "box car" numbers used in the government's tentative renegotiation assessment were truly unfounded. The court also noted that to this date the renegotiation claims have never been resolved, suggesting that the government either is extremely inefficient or agrees that the renegotiation claims have no merit. Plaintiffs presented no particular evidence to rebut the defendants' view, but merely argued that the Solitron defendants would not have made the reports they did to the Renegotiation Board unless they expected to be held liable. It is more plausible, however, to infer that they made the reports in the expectation of avoiding liability, in which case their statements to the public were not misleading.

Expert witness Gorkiewicz testified for the plaintiffs that when earnings per share were corrected for overvaluation of inventory and sales, Solitron shares were overpriced by 44% in 1967, 72.1% in 1968, not at all in 1969, and 69.6% in 1970. The court submitted to the jury special interrogatory 6a as follows:

State the percentage, if any, by which you find false or misleading financial statements inflated the market price of Solitron stock.

Period of            Financial
Overstatement,       Statements For  Percentage of
if any               Year Ending     Inflation, If Any
------------------------------------------------------
   5/5/67 - 6/5/68        2/27/67    _________________
   6/6/68 - 5/22/69       2/27/68    _________________
   5/27/70 - 6/2/71       2/28/70    _________________

The jury responded by filling in for each of the above three periods, respectively, 33%, 54.2% and 52.2%. The judgment entered on April 9, 1981 limited recovery by the members of the class by stating:

However, the recovery of any purchaser, exclusive of interest, shall not exceed the actual loss of that purchaser based on the market price of Solitron stock on October 31, 1979, that being the date on which the class was closed and following which no opt-out was authorized, or the actual loss based on actual selling price of the stock if sold prior to October 31, 1979.

All parties are agreed that the cut-off date of October 31, 1979 is wrong and that the cut-off date for opting out is irrelevant. The date should be December 16, 1970, when the defendants ceased the "pollution" of the market place. Harris v. American Investment Co., 523 F.2d 220 (8th Cir. 1975), cert. denied, 423 U.S. 1054, 96 S. Ct. 784, 46 L. Ed. 2d 643 (1976). Thus the plaintiffs are precluded from claiming a larger award based on declines in price after that date and the defendants from awarding a smaller amount based on increases in price thereafter. See Voege v. Ackerman, 364 F. Supp. 72 (S.D.N.Y.1973).

The Solitron defendants argue that plaintiffs' damage theory is illogical and has no support in this circuit because it does not relate the alleged overpricing either to actual market behavior or to disclosure events, and because it treats the stock as overvalued by a constant percentage for each year. But as the district court pointed out, the year-long periods presented to the jury in Special Interrogatory 6a were stipulated to be accurate time periods, corresponding to the figures outlined by Gorkiewicz. And while it seems odd that the calculations are for full years and vary sharply from the last day of one to the first day of the next, these are only hypothetical constructs which had to be made rather arbitrarily since there were no disclosures until December 1970. Each year was an era and the point in time between years an epoch.

We agree with the court below that with respect to 1967 and 1968, the damage question was one of fact, resolved satisfactorily by the jury through 25% reductions in Gorkiewicz's estimates. But we think that the Solitron defendants are correct that the jury's finding that the stock was overvalued by 52.2% in 1970 cannot stand. After the December 1970 disclosure of the inventory write-downs, the market in the stock declined immediately only 11% and then rose. This actual response of the market must be taken into account in determining the damages for all or part of the year 1970. See Shapiro v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 495 F.2d 228 (2d Cir. 1974); SEC v. Texas Gulf Sulphur Co., 401 F.2d 833 (2d Cir. 1968) (en banc), cert. denied, 394 U.S. 976, 89 S. Ct. 1454, 22 L. Ed. 2d 756 (1969). This is a matter we leave to the sound discretion of the trial judge, because there might have been other favorable developments in the company or in the market place offsetting the harmful effects of the disclosure.

Solitron also argues that the district court erred because it failed to consider the drastic market decline of 1970, citing Rolf v. Blyth, Eastman Dillon & Co. and Feit v. Leasco Data Processing Equipment Corp., 332 F. Supp. 544, 586 (E.D.N.Y.1971). But these cases and the rule they announced have no bearing in a case like this in which a stock purchaser has purchased his stock at a price higher than he would have paid had true financial statements been made. Rolf involved the decline in value of a portfolio of one investor improperly advised by an investment adviser who was recklessly aided and abetted by the investor's broker. Feit involved a fraudulent registration statement suit under section 11, with statutorily determined damages expressly excluding damages caused by "independent forces," 332 F. Supp. at 586. Here the issue is the amount by which each class member was defrauded on the date of his purchase. Any subsequent decline in the market had no effect on that fraudulent sale.

Sternbach argues that if liable, it is entitled to contribution from the Solitron defendants. We agree. Courts in this circuit have permitted contribution in section 10(b) cases even though section 10 of the Securities Exchange Act, unlike sections 9 and 18, does not expressly provide therefor. See Tucker v. Arthur Andersen & Co., 646 F.2d 721, 727 n.7 (2d Cir. 1981); Seymour v. Bache & Co., 502 F. Supp. 115, 119 (S.D.N.Y.1980); Alexander & Baldwin, Inc. v. Peat, Marwick, Mitchell & Co., 385 F. Supp. 230 (S.D.N.Y.1974); Globus, Inc. v. Law Research Service, Inc., 318 F. Supp. 955, 958 (S.D.N.Y.1970), aff'd, 442 F.2d 1346 (2d Cir.), cert. denied, 404 U.S. 941, 92 S. Ct. 286, 30 L. Ed. 2d 254 (1971). In Tucker, a section 10(b) case, this court noted that "under the securities laws, a person who has defrauded the plaintiff in violation of those laws may be liable for contribution to another person who has similarly defrauded the plaintiff." 646 F.2d at 727 n.7. The amount of contribution we leave to the sound discretion of the trial judge.

Judgment affirmed in part, reversed in part, and remanded.