David Hanon sued Dataproducts Corporation, a manufacturer of computer printers, and its chief executive officer, Jack C. Davis, under section 10(b) of the Securities Exchange Act of 1934 and Rule 10b-5 promulgated thereunder, 17 C.F.R. § 240 .10b-5. Hanon owned ten shares of Datapro-ducts stock. He alleged Dataproducts misled the market about one of its new product lines and concealed strategic decisions to discontinue two printers and to close a production facility.

The district court denied Hanon's motion for class certification ; Subsequently, after approximately thirty depositions had been taken and thousands of documents had been produced, the district court granted Dataproducts’ motion for summary judgment with respect to Hanon’s individual claims.

Hanon appeals from the district court’s denial of class certification and its summary judgment. We have jurisdiction under 28 U.S.C. § 1291 . We affirm the denial of class certification, affirm the summary judgment in part, reverse in part and remand for further proceedings.

STANDARD OF REVIEW

Summary judgment is reviewed de novo. In re Apple Computer Sec. Litig., 886 F.2d 1109, 1112 (9th Cir.1989), cert. denied sub nom. Schneider v. Apple Computer, 496 U.S. 943 , 110 S.Ct. 3229 , 110 L.Ed.2d 676 (1990). We may affirm on any basis supported by the record. Id.

Summary judgment may be granted if “there is no genuine issue as to any material fact” and “the moving party is entitled to a judgment as a matter of law.” Fed.R.Civ.P. 56(c). This rule does not require that there be no factual dispute. “[T]he mere existence of some alleged factual dispute between the parties will not defeat an otherwise properly supported motion for summary judgment; the requirement is that there be no genuine issue of material fact.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 247-48 , 106 S.Ct. 2505, 2510 , 91 L.Ed.2d 202 (1986). A material fact is genuine if “the evidence is such that a reasonable jury could return a verdict for the nonmoving party.” Id. at 248 , 106 S.Ct. at 2510 . Conversely, “[wjhere the record taken as a whole could not lead a rational trier of fact to find for the non-moving party, there is no ‘genuine issue for trial.’ ” Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 587 , 106 S.Ct. 1348, 1356 , 89 L.Ed.2d 538 (1986).

Although “[mjateriality and scienter are both fact-specific issues which should ordinarily be left to the trier of fact,” “summary judgment may be granted in appropriate cases.” Apple Computer, 886 F.2d at 1113 . Summary judgment may be defeated in a securities fraud derivative suit “only by showing a genuine issue of fact with regard to a particular statement by [the company] or its insiders.” Id. at 1118 .

FEDERAL SECURITIES FRAUD

A. Material Misrepresentations and Omissions

“[A]n issuer’s public statements cannot be analyzed in complete isolation.” In re Convergent Technologies Sec. Litig., 948 F.2d 507, 512 (9th Cir.1991). In order to survive summary judgment, Hanon “must demonstrate that a particular statement, when read in light of all the information then available to the market, or a failure to disclose particular information, conveyed a false or misleading impression.” Id.

1. Disclosures Regarding Solid Ink Technology

Hanon contends Dataproducts misled the market by issuing two overly optimistic press releases that neglected to reveal severe problems with solid ink technology. The first press release, issued May 3, 1988, states that “Solid ink jet is widely endorsed as the future technology for office and color printing.”

“[Projections and general expressions of optimism may be actionable under the federal securities laws.” Apple Computer, 886 F.2d at 1113 ; see also Virginia Bankshares, Inc. v. Sandberg, — U.S. -, -, 111 S.Ct. 2749, 2756-61 , 115 L.Ed.2d 929 (1991) (knowingly false statements of reasons, opinion, or belief, even though conclusory in form, may be actionable as misstatements of material fact). In this circuit, a projection or statement of belief may be actionable to the extent that one of three implied factual assertions is inaccurate: “(1) that the statement is genuinely believed, (2) that there is a reasonable basis for that belief, and (3) that the speaker is not aware of any undisclosed facts tending to seriously undermine the accuracy of the statement.” Apple Computer, 886 F.2d at 1113 .

There is no evidence in the record to show that Dataproducts did not genuinely believe its statement regarding solid ink technology, or that its optimistic belief in solid ink was unreasonable. Hanon has not shown that solid ink technology as a whole, or with respect to color printing, encountered problems which would seriously undermine the accuracy of the May 3 press release. Furthermore, the statement does not guarantee success of solid ink color printers, but recognizes only that Datapro-ducts’ research is progressing.

The other alleged misrepresentation as to solid ink technology occurred on May 4, 1989, when Dataproducts publicly stated: “Solid Ink is the best way to deliver high quality, low cost color printing to the market and our development program in this area is progressing.”

We need not determine whether this statement is misleading because it was issued after Hanon bought his stock and thus could not have affected his stock’s April 14, 1989 market price or his decision to buy on that date. See Williams v. Sinclair, 529 F.2d 1383, 1389 (9th Cir.1975) (allegedly fraudulent acts which occur after a plaintiff’s purchase of stock cannot form the basis of a section 10(b) claim because the acts were not performed in connection with the purchase or sale of securities), cert. denied, 426 U.S. 936 , 96 S.Ct. 2651 , 49 L.Ed.2d 388 (1976); Raschio v. Sinclair, 486 F.2d 1029, 1030 (9th Cir.1973).

2. Disclosures Regarding the SI 480 Printer

Hanon contends Dataproducts misled the market by improperly touting its SI 480 printer when it was aware the printer had severe technological problems. Hanon relies on a press release issued by Datapro-ducts on May 20, 1988:

Hanon also points to statements made in Dataproducts’ first quarter report, dated June 15, 1988: “The SI 480 is compact. It’s quieter than dot matrix and daisy wheel printers and faster than most. The print quality is superior to laser printers, ... Unlike other non-impact technologies, ... solid ink is insensitive to media, handling all types of papers, ... with virtually no variation in print quality.”

Hanon argues he has raised a genuine issue of material fact as to whether the May 20, 1988 press release and the statements in the June 15, 1988 first quarter report praising the SI 480 were false and misleading because they were made when Dataproducts knew the printer had severe technical problems. We agree.

This case is similar to Apple Computer in which we held that genuine issues of material fact existed with respect to whether a new disk-drive’s technical problems constituted material facts tending to undermine the unqualified optimism of Apple’s positive statements about the product. See Apple Computer, 886 F.2d at 1115-16 . In Apple Computer, Apple issued a press release stating that Twiggy, a newly developed computer disk-drive, “ ‘ensures greater integrity of data than the other high density drives by way of a unique, double-sided mechanism designed and manufactured by Apple.’ ” Id. at 1115. Apple also claimed that Twiggy " 'represents three years of research and development and has undergone extensive testing and design verification during the past year.’ ” Id. At the time these statements were made, internal tests conducted by Apple indicated some slowness and unreliability in Twiggy’s information-processing capabilities. Id. Further evidence showed that, two weeks before the press release was issued, top executives were warned Twiggy’s reliability problems would require Apple to delay the introduction of another new product, the Lisa computer. Id.

Here, there is at least a triable issue of whether technical problems with the SI 480 were material facts tending to undermine the optimism of Dataproducts’ public statements. In March 1988, approximately two months before the press release was issued, Robert Kilcullen, senior vice president and business manager for line printers, stated in his corporate diary, “SI 480— cannot build it reliable [sic].” On September 16, 1988, Kilcullen wrote, “Si 480— reputation in field bad.” On a page of the diary dated October 14, 1988, Kilcullen listed various aspects of a particular “cold start problem” with the SI 480. On October 8, 1988, Kilcullen noted an ink durability problem with the SI 480, which caused the ink on printed pages to crack and fall off when the page was folded.

From this evidence, a reasonable jury could conclude that Dataproducts publicly released optimistic statements about the SI 480 at a time when it knew the printer could not be built reliably. In addition, several high level executives within Data-products testified that they knew, by late 1987 and early 1988, the SI 480 was suffering from cold start and ink durability problems. These problems were confirmed by statements in Kilcullen’s notebook which showed that they continued at least until the end of 1988. From this evidence it could be inferred that the SI 480 suffered from serious ink durability and cold start problems, and these problems were known to Dataproducts at the time it issued its optimistic statements. See id. at 1113 (generally optimistic statements are actionable if the company did not believe the statements when they were made).

Dataproducts argues that even if the SI 480 had problems which were not disclosed, the problems were not significant enough to mandate public disclosure. We cannot conclude this as a matter of law. “There is a difference between knowing that any produet-in-development may run into a few snags, and knowing that a particular product has already developed problems so significant as to require months of delay.” Id. at 1115.

Production of the SI 480 was stopped several times in 1988 due to performance deficiencies. The printer experienced a severe ink durability problem, a problem which undermined the product’s essential function. Kilcullen stated it could not be built reliably. A jury could find that Data-products’ failure to disclose the SI 480’s technical problems had a misleading effect on the market price of Dataproducts stock. See id. (genuine issue of fact presented as to whether optimistic statements were misleading, because tests conducted inhouse indicated slowness and unreliability in information-processing capabilities).

Dataproducts asks us to conclude as a matter of law that the market was aware of the problems and risks associated with the SI 480 because the printer had been introduced to the market approximately five months prior to the release of the statements and had been tested by industry analysts, who noted some of its problems. We cannot make this conclusion.

“The investing public justifiably places heavy reliance on the statements and opinions of corporate insiders.” Id. at 1116. “[Cjorporate insiders are not relieved of their duty to disclose material information where that information has received only brief mention in a few poorly-circulated or lightly-regarded publications.” Id. “In order to avoid Rule 10b-5 liability, any material information which insiders fail to disclose must be transmitted to the public with a degree of intensity and credibility sufficient to effectively counterbalance any misleading impression created by the insiders’ one-sided representations.” Id.; see also Basic Inc. v. Levison, 485 U.S. 224, 249 , 108 S.Ct. 978, 992 , 99 L.Ed.2d 194 (1988) (in order to rebut the presumption that a fraudulent statement has been transmitted through the market price defendant must show that “corrective statements” have “credibly entered the market”).

Dataproducts has produced no evidence to show that the SI 480’s technical problems were “made known to the market by the press or by anyone else,” Apple Computer, 886 F.2d at 1115 , or that “the market as a whole was aware that there was a problem with [the SI 480].” Id. Although its optimistic statements were not as exuberant as those expressed by Apple with regard to the Lisa computer, Dataproducts has not presented evidence of any publication, poorly-circulated or otherwise, detailing the SI 480’s ink durability, reliability or cold start problems. It has failed to show that the press stressed the risks Datapro-ducts was taking with this product, or detailed the underlying technical problems producing those risks. Cf. id. at 1115-16 (at least twenty articles of intense, sustained press disclosure of adverse information served to render even statements of “unqualified exuberance” unactionable under Rule 10b-5). A rational jury could find “a ‘substantial likelihood’ that full disclosures by [Dataproducts] would have ‘significantly altered the total mix of information made available.’ ” Id. at 1115 (quoting TSC Indus., Inc. v. Northway, Inc., 426 U.S. 438, 449 , 96 S.Ct. 2126, 2132 , 48 L.Ed.2d 757 (1976)).

We also reject Dataproducts’ argument that it is beyond genuine dispute that it effectively cautioned the market about risks involved with the SI 480 by stating in its annual and 10-K reports that the SI 480 would experience higher costs and lower profit margins because it was a new product.

In its 1988 annual report, Dataproducts stated: (1) “[u]ntil full production levels [of the SI 480] are attained, manufacturing costs on these products will be higher and margins lower than the products they have replaced”; (2) “[production in Fiscal 1988 [for the SI 480] was not yet up to the level of market demand”; and (3) engineering and development costs increased in fiscal 1988 primarily because of the “continuing development costs of solid ink, thermal laser and new line printer programs.” Whether these disclosures effectively warned the market during the relevant time period of specific risks associated with the SI 480 is a question of fact for the trier of fact to resolve. Although the statements effectively warn investors of the increased costs associated with the SI 480, there is no mention of the printer’s technical problems, development and manufacturing risks or other relevant risk factors. Cf. Convergent Technologies, 948 F.2d at 515-16 (“more than generalized statements of risk” sufficiently alerted the market to problems with the company’s new product line); Moorhead v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 949 F.2d 243, 245-46 (8th Cir.1991) (securities claim cannot be based on misrepresentations in feasibility study which was addressed by “repeated, specific warnings of significant risk factors and the disclosures of underlying factual assumptions”).

Dataproducts argues it was not required to criticize the SI 480 because a company is not required to denigrate its own products. We reject this argument. Denigrating or not, Rule 10b-5 imposes a duty to disclose material facts that are necessary to make disclosed statements, whether mandatory or volunteered, not misleading. See 17 C.F.R. § 240 .10b~5.

We conclude that Hanon produced sufficient evidence to raise genuine issues of material fact as to whether the statements made in the May 20,1988 press release and in the June 15, 1988 first quarter report, when read in light of all the information then available to the market, including the problems with the SI 480 which the evidence suggests existed, conveyed a false or misleading impression. Cf. Convergent Technologies, 948 F.2d at 512 .

3. Omissions Regarding Decisions to Discontinue Printers

Hanon contends Dataproducts decided to discontinue the SI 480 and ATB printers in March 1989, but that it misled the market by failing to disclose this fact until September 1989, six months later. In support of this argument, Hanon points to four pieces of evidence: (1) a note written by Davis, dated March 8, 1989, stating, “[kjill some loss leader products SI 480, ATB,” (2) a memorandum written by Robert Doretti, senior vice president of sales and marketing, dated March 3, 1989, discussing lowering the price of the SI 480 in order to drive down its inventory; (3) a note from Davis to Kilcullen, dated March 3, 1989, concerning the SI 480: “what we need is a complete ‘end of life’ program,”; and (4) testimony by Moore, Dataproducts’ strategic and investment advisor, that it was known within Dataproducts as of March 1989 “that the Si 480 would be discontinued at some point.”

No reasonable jury could conclude from this evidence that Dataproducts knew in March 1989, with sufficient certainty, when it was going to discontinue the SI 480 and ATB product lines. Davis’s March 8 note was contained in a document identified as a preliminary list of ideas to be considered at a planning meeting the following day. In a March 10, 1989 memo recording final decisions made at the March 9 meeting, no mention was made of plans to discontinue either the SI 480 or the ATB. This memo is consistent with Davis’s sworn testimony that the company did not decide to discontinue the SI 480 or ATB in March 1989, which is uncontroverted. The most Ha-non’s evidence shows is that Dataproducts knew in March 1989 that “the SI 480 would be discontinued at some point.”

This evidence does not meet the level of certainty required to make the statements material. See Basic, 485 U.S. at 232 , 108 S.Ct. at 983 (where an “event is contingent or speculative in nature, it is difficult to ascertain whether the ‘reasonable investor’ would have considered the omitted information significant at the time”); Vaughn v. Teledyne, Inc., 628 F.2d 1214, 1221 (9th Cir.1980) (disclosure not required when there is no evidence that estimates were made with such reasonable certainty even to allow them to be disclosed to the public).

4. Disclosures Regarding the ATB Printer

The ATB printer is an automated ticketing and boarding pass printer developed and introduced by Dataproducts with Memorex-Telex in 1989. The ATB was designed in anticipation of the airline indus try standardizing some industry norms, such as the way tickets and boarding passes are produced. In May 1988, the ATB was in development at Dataproducts’ Milford facility. When the Milford facility was closed in 1988, the ATB program was moved to Dataproducts’ Woodland Hills facility.

By September 1989, Dataproducts realized that if the ATB was to be successful, significant additional funds would have to be invested for market development and. product customization. Shortly thereafter, Dataproducts decided to discontinue the ATB and sold the technology to Datamax, which is now successfully marketing the printer.

Dataproducts’ 1988 10-K report summarized the status of the ATB printer:

Hanon argues this description of the ATB printer is false ■ and misleading because Dataproducts portrayed it as ready for production in 1988 when it was evident that the airline industry was not standardizing as forecasted and Memorex-Telex, Dataproducts’ partner in the ATB venture, desired to terminate the ATB program. We disagree.

The 1988 annual report states only that Dataproducts expected to start shipments of the ATB in the second quarter of fiscal 1989. Dataproducts had every reason to expect shipments of the ATB to begin at this time because by May 1988, when the annual report was issued, it had entered into a contract to sell the ATB to Memo-rex-Telex, who was supposed to market the printer to airlines. Unfortunately for Dataproducts, Memorex-Telex terminated this contract in the fall of 1988, at which time Dataproducts began to get into the market alone.

Hanon argues Dataproducts failed to inform the public of major problems in marketing the ATB printer to airlines. We reject this argument. Hanon has not identified any marketing problem other than the facts that the market for the ATB was limited to the airline industry, and that the success of the ATB was dependent upon the airline industry standardizing its methods for issuing tickets and boarding passes.

As to the first factor, Dataproducts publicly disclosed that the market for the ATB was limited to the airline industry. As to the second, it was a known fact that the ATB was designed to provide automatic ticketing and boarding passes to the entire airline industry. It was obvious that to fulfill this function, the industry had to have standardized methods for printing tickets and boarding passes, which it did not have. With the lack of standardized methods for issuing tickets and boarding passes a known condition, it was not Data-products’ obligation to inform the public that the airline industry did not adopt procedures to change this condition. See Wielgos v. Commonwealth Edison Co., 892 F.2d 509, 515 (7th Cir.1989) (no duty to disclose regulatory conditions).

5. Disclosures Regarding the Woodland Hills Facility

Hanon argues when Dataproducts announced its decision to close the Milford operating facility, and described one time costs for relocating the serial printer business, there was a consensus within the company that the Woodland Hills facility would also be closed, requiring a substantial. charge against earnings. He contends that in early 1988 Dataproducts decided to close the Woodland Hills facility, but it concealed that decision until May 1989 when a press release announced the facility was up for sale. He argues concealment of this information artificially inflated the price of Dataproducts’ stock between early 1988 and May 1989.

To support his argument that a decision had been made to close the Woodland Hills facility in early 1988, Hanon relies on a summary report prepared February 28, 1988 by Dataproducts’ auditors, Arthur Young & Co., which states that the consensus of Dataproducts’ senior staff “is to close Milford now and Woodland Hills in 3 to 5 years.”

The Arthur Young report, however, shows only that Dataproducts’ senior staff believed the Woodland Hills facility should be closed sometime in 3 to 5 years. Discussion among management staff regarding potential action to be taken sometime in the distant future is not an item appropriately made a part of a public disclosure because of its speculativeness. See TSC Indus., 426 U.S. at 448-49 , 96 S.Ct. at 2131-32 (securities laws do not require management to bury shareholders in an avalanche of trivial information); Marx v. Computer Sciences, Corp., 507 F.2d 485, 491 (9th Cir.1974) (a company must disclose material facts but it “need not detail every corporate event, current or prospective”).

B. Reliance Under a Fraud-on-the-Market Theory

“In the usual claim under Section 10(b), the plaintiff must show individual reliance on a material misstatement. Under the fraud-on-the-market theory, the plaintiff has the benefit of a presumption that he has indirectly relied on the alleged misstatement, by relying on the integrity of the stock price established by the market.” Apple Computer, 886 F.2d at 1113— 14. “An investor’s reliance on the market, therefore, is equivalent to reliance upon statements made to the market, or the nondisclosure of material information.” Convergent Technologies, 948 F.2d at 512 n. 2.

The district court granted summary judgment with respect to Hanon’s 10b-5 claim on the issue of reliance:

Hanon argues the district court erred in concluding that an investor’s sophistication renders inapplicable the fraud-on-the-market presumption of reliance. We agree.

“Differences in sophistication, etc., among purchasers have no bearing in the impersonal market fraud context, because dissemination of false information necessarily translates through market mechanisms into price inflation which harms each purchaser identically.” Blackie v. Barrack, 524 F.2d 891, 905 (9th Cir.1975), cert. denied, 429 U.S. 816 , 97 S.Ct. 57 , 50 L.Ed.2d 75 (1976); see In re MDC Holdings Sec. Litig., 754 F.Supp. 785, 802 (S.D.Cal.1990) (if the defendants perpetrated a fraud on the market, even the most sophisticated investors would be deceived). Sophisticated investors are as entitled to rely on the fraud-on-the-market theory as anyone else.

Dataproducts argues that, apart from Hanon’s sophistication, the undisputed facts show he did not rely on the market in buying his ten shares of stock. Dataproducts points out that Hanon could not have expected to profit from his investment of $171.25 because he would have had to pay at least $90.00 in commissions and other charges in order to sell his stock. It also points to the uncontroverted evidence that Hanon previously brought four derivative suits against publicly held corporations with the help of some of the same lawyers involved here; and that he also bought ten shares of stock in each of seven other high technology companies around the same time he bought his ten shares of Datapro-ducts. Dataproducts contends the proof of Hanon’s unusual buying strategy and his experience in securities litigation support a strong inference that he bought his stock in order to bring this lawsuit. This is sufficient, according to Dataproducts, to rebut the fraud-on-the-market presumption.

It is true that one way to rebut the fraud-on-the-market theory is to show that the plaintiff would have bought his stock at the same price had he known the information that was not disclosed or misrepresented. Basic, 485 U.S. at 248-49 , 108 S.Ct. at 992-93 . Here, however, Hanon testified in his deposition that he bought his ten shares of Dataproducts stock to make money, receive more information about the company, “test the waters,” and have an incentive to follow the company’s progress. Substantial evidence in the record suggests these were not his true motives, but it is for the trier of fact to determine the credibility of Hanon’s testimony.

Where the nonmoving party in a motion for summary judgment has come forward with direct evidence contrary to that offered by the movant, a credibility issue is raised. McLaughlin v. Liu, 849 F.2d 1205, 1207-08 (9th Cir.1988) (sworn statements are direct evidence of a fact in dispute). Credibility determinations are for the trier of fact and, therefore, are not appropriately resolved by summary judgment. Id. at 1207 ; City of Vernon v. S. Calif. Edison Co., 955 F.2d 1361, 1370 (9th Cir.1992). Although in reviewing summary judgment, we are entitled to consider the plausibility and reasonableness of inferences arising from Hanon’s circumstantial evidence, we must treat Hanon’s direct evidence as if it were true. Anderson, 477 U.S. at 255 , 106 S.Ct. at 2513 ; T.W. Elec. Serv., Inc. v. Pacific Elec. Contractors Ass’n, 809 F.2d 626 , 631-32 (9th Cir.1987) (where direct evidence produced by the moving party conflicts with direct evidence produced by the nonmoving party, the court must assume the truth of the evidence set forth by the nonmoving party with respect to that fact). Treating Hanon’s deposition testimony as true, he has raised a genuine issue of material fact as to his reliance on Dataproducts’ alleged fraud on the market.

C. Scienter

Section 10(b) liability requires a showing of scienter: a mental state embracing intent to deceive, manipulate or defraud. Hollinger v. Titan Capital Corp., 914 F.2d 1564, 1568 (9th Cir.1990) (en banc), cert. denied, — U.S.-, 111 S.Ct. 1621 , 113 L.Ed.2d 719 (1991). Scien-ter may be satisfied by either proof of actual knowledge or recklessness. Id. at 1568-69. Reckless conduct necessary to satisfy the scienter requirement is:

Id. at 1569 (citations omitted).

Dataproducts argues its statements regarding the SI 480 were made with a good faith belief in their accuracy. As evidence of its good faith, Dataproducts asserts that it continued to manufacture, sell and invest millions of dollars in the SI 480 from May 1988 to April 14, 1989, and that its officers did not sell their stock during the class period.

This may all be true. However, Hanon produced evidence that Dataproducts, through its responsible employees, knew about the cold start and ink durability problems with the SI 480 in late 1987 and early 1988. Dataproducts produced no evidence to show that these problems were corrected by May 1988, when the allegedly misleading statements were made, and Kilcullen’s notebook reflects that the problems continued at least until late 1988. See Apple Computer, 886 F.2d at 1117 (“significant technological flaws in Twiggy raise a genuine issue of material fact as to [the optimistic] statements’ recklessness”).

In sum, there is a genuine issue of material fact on the issue of scienter.

CLASS CERTIFICATION

Although the district court offered no explanation for its order denying Ha-non’s Rule 23(a) motion for class certification, and made no findings, we may sustain the court’s ruling on any ground supported by the record. See Inda v. United Air Lines, Inc., 565 F.2d 554, 562-63 (9th Cir.1977) (when a basis for denial of class certification appears from the record with clarity the appellate court may determine in the first instance whether or not a ground for denial exists in the record), cert. denied, 435 U.S. 1007 , 98 S.Ct. 1877 , 56 L.Ed.2d 388 (1978).

For a named plaintiff to obtain class certification, the court must find: (1) nu-merosity of plaintiffs; (2) common questions of law or fact predominate; (3) the named plaintiff’s claims and defenses are typical; and (4) the named plaintiff can adequately protect the interests of the class. Fed.R.Civ.P. 23(a).

Dataproducts argues Hanon’s request for class certification should be denied because he has failed to establish Rule 23(a)’s typicality requirement. See Mantolete v. Bolger, 767 F.2d 1416, 1424 (9th Cir.1985) (plaintiff bears the burden of establishing the four prerequisites of Fed.R.Civ.P. 23(a)).

The purpose of the typicality requirement is to assure that the interest of the named representative aligns with the interests of the class. Weinberger v. Thornton, 114 F.R.D. 599, 603 (S.D.Cal.1986). “ ‘Typicality refers to the nature of the claim or defense of the class representative, and not to the specific facts from which it arose or the relief sought.’ ” Id. (citation omitted). The test of typicality “is whether other members have the same or similar injury, whether the action is based on conduct which is not unique to the named plaintiffs, and whether other class members have been injured by the same course of conduct.” Schwartz v. Harp, 108 F.R.D. 279, 282 (C.D.Cal.1985).

Several courts have held that “class certification is inappropriate where a putative class representative is subject to unique defenses which threaten to become the focus of the litigation.” Gary Plastic Packaging Corp. v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 903 F.2d 176, 180 (2d Cir.1990), cert. denied, — U.S. -, 111 S.Ct. 675 , 112 L.Ed.2d 667 (1991); see also J.H. Cohn & Co. v. American Appraisal Assoc., Inc., 628 F.2d 994 , 998-99 (7th Cir.1980); Hoexter v. Simmons, 140 F.R.D. 416, 422-23 (D.Ariz.1991) (plaintiffs claims atypical of class because unique defense could be asserted against them); Rolex Employees Retirement Trust v. Mentor Graphics Corp., 136 F.R.D. 658, 664 (D.Or.1991) (“The certification of a class is questionable where it is predictable that a major focus of the litigation will be on an arguable defense unique to the named plaintiff or to a subclass.”).

We agree that a named plaintiff’s motion for class certification should not be granted if “there is a danger that absent class members will suffer if their representative is preoccupied with defenses unique to it.” Gary Plastic Packaging Corp., 903 F.2d at 180 . Here, Hanon’s unique background and factual situation require him to prepare to meet defenses that are not typical of the defenses which may be raised against other members of the proposed class. Hanon’s reliance on the integrity of the market would be subject to serious dispute as a result of his extensive experience in prior securities litigation, his relationship with his lawyers, his practice of buying a minimal number shares of stock in various companies, and his uneconomical purchase of only ten shares of stock in Dataproducts. See Hoexter, 140 F.R.D. at 422-23 (named plaintiffs may not serve as class representatives because their prior litigation experience as plaintiffs was so extensive it would inevitably subject them to unique defenses concerning their reliance on the market’s integrity in purchasing stock); Shields v. Smith, [1991 Tr. Binder] Fed.Sec.L.Rep. (CCH) P 96,449 (N.D.Cal.1991) (plaintiff’s reliance on the integrity of the market would be “subject to serious rebuttal as a result of [her] professional practice of buying a few shares in troubled companies in order to later serve as class representative”).

We emphasize that the defense of non-reliance is not a basis for denial of class certification. See Blackie, 524 F.2d at 901 n. 17; In re Pizza Time Theatre Sec. Litig., 112 F.R.D. 15, 22 (N.D.Cal.1986) (lack of reliance is a defense; it goes to the merits of the case and cannot be considered in a certification motion); see also Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 177-78 , 94 S.Ct. 2140, 2152 , 40 L.Ed.2d 732 (1974) (a motion for class certification is not the appropriate point at which to resolve the merits of a plaintiffs claim). Nevertheless, we are “at liberty to consider evidence which goes to the requirements of Rule 23 even though the evidence may also relate to the underlying merits of the case.” In re Unioil Sec. Litig., 107 F.R.D. 615, 618 (C.D.Cal.1985). A class may only be certified if we are “satisfied, after a rigorous analysis, that the prerequisites of Rule 23(a) have been satisfied.” General Tel. Co. Southwest v. Falcon, 457 U.S. 147, 161 , 102 S.Ct. 2364, 2372 , 72 L.Ed.2d 740 (1982). Because of Hanon’s unique situation, it is predictable that a major focus of the litigation will be on a defense unique to him. Thus, Hanon fails to satisfy the typicality requirement of Rule 23(a).

CONCLUSION

Hanon has raised genuine issues of material fact as to whether Dataproducts’ optimistic statements about the SI 480 in its May 20, 1988 press release and in its June 15, 1988 first quarter report were materially misleading, whether the evidence establishes scienter, and whether he relied on the market in buying his shares. We reverse the district court’s grant of summary judgment in favor of Dataproducts on these issues. There are no triable issues of material fact as to the remainder of Ha-non’s claims, and as to those we affirm the district court’s summary judgment. We also affirm the district court’s denial of Hanon’s motion for class certification because he fails to meet the typicality requirement of Rule 23(a).

AFFIRMED in part, REVERSED in part and REMANDED to the district court for further proceedings consistent with this opinion.

. We refer to the defendants Dataproducts Corporation and Jack C. Davis jointly as "Datapro-ducts." . Solid ink technology is a non-impact printing process which uses ink in a solid form. When applied, the ink is melted in the printer at a very high temperature and then jetted onto paper. Ink attaches to the paper and resolidifies in the form of letters or other characters, resulting in a raised texture. . Hanon does not allege, nor has he come forward with any evidence to show, that he relied on the May 4, 1989 statement in holding his Dataproducts stock. . The uncertainty of this decision is made even more evident by the fact that Dataproducts ultimately decided not to close the facility, but instead sold it and leased it back. . Dataproducts contends the fraud-on-the-market presumption is not applicable here because Hanon's odd lot purchase required him to pay a 50% premium and, thus, he did not buy at the "price set by the market.” We disagree. The purchase price of the stock was set by the market. The $9.00 per share commission was a transactional fee required due to the small number of shares Hanon purchased. .Although these cases discuss the sophistication of investors in the context of granting class certification, their reasoning is equally applicable to determining whether a sophisticated investor is entitled to rely on the fraud-on-the-market theory.