ORDER AND JUDGMENT ACCEPTING MAGISTRATE JUDGE'S REPORT AND RECOMMENDATION AND DISCOVERY ORDER AS THE OPINION AND ORDER OF THE COURT

          ROBERT HOLMES BELL, District Judge.

         Before the court is plaintiff's motion for class certification pursuant to Federal Rule of Civil Procedure 23(c)(1). A Report and Recommendation was issued by United States Magistrate Judge Joseph G. Scoville, denying this motion. Plaintiffs filed objections to the magistrate judge's report and recommendation. Pursuant to Rule 72(b) of the Federal Rules of Civil Procedure, this court has made a de novo determination upon the record, has considered plaintiffs' objections to the report and recommendation and has concluded that no portion of the latter is contrary to law.

          The court does add the following. In plaintiffs' objections to the report and recommendation, they raise for the first time the argument that even if they are not entitled to a presumption of reliance pursuant to the Fraud-on-the-Market theory, they are entitled to a presumption of reliance under Affiliated Ute Citizens v. United States, 406 U.S. 128, 92 S.Ct. 1456, 31 L.Ed.2d 741 (1972). Pursuant to the exhaustion doctrine that is applied with regularity to issues raised on appeal, the court need not review this argument as it was not raised before the magistrate judge. See Paterson-Leitch v. Mass. Mun. Wholesale Elec. Co., 840 F.2d 985, 990-91 (1st Cir.1988) (" We hold categorically that an unsuccessful party is not entitled as of right to de novo review by the judge of an argument never seasonably raised before the magistrate." ); Greenhow v. Secretary of Health & Human Servs., 863 F.2d 633, 638-39 (9th Cir.1988) (holding that the district court acted within its discretion in refusing to consider arguments that were never presented to the magistrate judge).

         In addition, the court emphasizes the report and recommendation's conclusion that, because proof of individual reliance on the alleged misrepresentations is necessary to prevail on a section 10(b) securities fraud action, Central Bank v. First Interstate Bank, 511 U.S. 164, __, 114 S.Ct. 1439, 1448, 128 L.Ed.2d 119, 141 (1994), no claim could be made against Price Waterhouse before its April 15, 1993, certification of Embrace's financial statements was released. Plaintiffs object that this conclusion ignores the fact that they have alleged a civil conspiracy count against all defendants. The magistrate judge, however, recognized that under Michigan law a conspiracy standing alone without the commission of underlying wrongdoing is not actionable. Since the underlying wrongdoing in this case requires individual proof of reliance, Id., common questions do not predominate and the plaintiff class cannot be certified with respect to this count.

         In summary, the court denies plaintiffs' motion to certify the plaintiff class for the following reasons:

(1) Plaintiffs have not borne their burden of showing that any putative class is so numerous that joinder of all members is impractical;

(2) plaintiff Ginsberg's claims are not typical of with regard to classes A (purchasers of Embrace common stock between May 21, 1992, and February 15, 1994, asserting claims against Embrace and the individual defendants) and A/PW (purchasers of Embrace common stock between April 15, 1993, and February 15, 1994, asserting claims against the individual defendants and Price Waterhouse);

(3) plaintiffs have failed to demonstrate that they are adequate representatives for class A or A/PW because of the possible existence of unresolved conflicts of interest; plaintiff Babcock has failed to demonstrate that he is an adequate representative of Class B (purchasers of Embrace series D common stock warrants during 1993, asserting claims against Embrace and the individual defendants), or B/PW (purchasers of warrants after April 15, 1993, asserting claims against the individual defendants and PW), as he has not pleaded the existence of a viable claim against Price Waterhouse, the principle defendant;

(4) with respect to counts one, two, four, and five, common questions do not predominate because individual proof of reliance, causation and damages will be necessary; it is unlikely that plaintiffs will have the benefit of the fraud-on-the-market theory for counts one and two, and no rebuttable presumption is available on state-law counts four and five; and, finally,

(5) a class action is not a superior method of adjudicating this controversy.

         In accordance with the opinion entered this date, IT IS HEREBY ORDERED that the magistrate judge's report and recommendation is ACCEPTED and ADOPTED as the opinion of the court with the additional conclusion and emphasis added.

         Plaintiffs have also filed objections to the same magistrate judge's denial of its motion to compel the production of a videotape and certain personnel records of Price Waterhouse. In accordance with Federal Rule of Civil Procedure 72(a), the court has considered plaintiffs' objections to the magistrate judge's discovery order and has concluded that no portion of the latter is clearly erroneous or contrary to law.

         In addition, IT IS HEREBY ORDERED that the discovery order issued by the magistrate judge is ACCEPTED and ADOPTED as the order of the court.

          REPORT AND RECOMMENDATION ON MOTION TO CERTIFY PLAINTIFF CLASS

          SCOVILLE, United States Magistrate Judge.

          This is a securities case alleging misrepresentation in connection with the sale of common stock and stock purchase warrants of Embrace Systems Corporation (" Embrace" ). In their second amended complaint, all plaintiffs allege that they purchased Embrace common stock at various times between May 21, 1992, and February 15, 1994; plaintiff Babcock alleges that he purchased Embrace warrants in 1993. Defendants are Embrace (now in Chapter 7 liquidation), six individuals who were officers, directors, or employees of Embrace, and Embrace's former accountant Price Waterhouse. The complaint purports to be brought on behalf of two plaintiff classes: (1) persons who purchased common stock of Embrace between May 21, 1992 and February 15, 1994 (" Class A" ) and (2) persons who purchased Embrace's Series " D" common stock purchase warrants issued during 1993 (" Class B" ). Both putative classes exclude Embrace itself, its officers, directors, and other affiliates. (Sec.Am.Comp. ¶ 11).

         Plaintiffs' second consolidated and amended class action complaint states five counts. Count I alleges that all of the defendants have violated section 10(b) of the Securities Exchange Act of 1934, 15 U.S. C. § 78j(b) [Manipulative and deceptive devices], and Rule 10b-5 promulgated thereunder, as to both putative classes. Count II alleges that the individual defendants are liable under section 20 of the 1934 Act, 15 U.S. C. § 78t [Liability of controlling persons], to both putative classes. Count III alleges that Embrace and the individual defendants (but not Price Waterhouse) have violated section 12(2) of the Securities Act of 1933, 15 U.S. C. § 77 l (2) [Civil liabilities arising in connection with prospectuses and communications], as to Putative Class B. Count IV alleges that all of the defendants have committed common law fraud, as to both putative classes. Finally, Count V alleges that all defendants have committed civil conspiracy, as to both putative classes.

         This matter is presently before the court on plaintiffs' motion for class certification, pursuant to FED.R.CIV.P. 23(c)(1). After the motion was filed, the parties were allowed to conduct discovery for four months, addressed to the class certification issues. Only Price Waterhouse filed a brief in opposition to the motion. After plaintiff's reply brief was filed, District Judge Robert Holmes Bell referred the motion to me for a report and recommendation. 28 U.S. C. § 636(b)(1)(B). I conducted a hearing on the motion on August 25, 1995, at which counsel presented legal arguments and each side presented the testimony of an expert witness in economics, addressing plaintiff's fraud-on-the-market theory as it relates to class certification issues. After considering the record and governing authorities, I conclude that plaintiffs have failed to carry their burden of establishing the propriety of a class action in this case. I therefore recommend that the motion for class certification be denied and that the case proceed on the claims of the named plaintiffs only.

          Proposed Findings of Fact

          A. Parties

         1. Defendant Embrace is a Delaware corporation with principal offices in Holland, Michigan. The company was formed on April 29, 1992, as a result of a merger between Haben Industries, Inc., a small publicly held company, and Embrace Technology Systems, Inc., a small privately held corporation. At the time of its formation, Embrace was involved in the manufacture of plastics, making various plastic containers by the " rotational molding" method. Additionally, the company planned to develop and market a new product called " Puffibre," an insulation product made of plastic. Until February of 1994, Embrace common stock was listed on the National Association of Securities Dealers Automated Quotation (NASDAQ) system as a " SmallCap" stock. Embrace is now in Chapter 7 bankruptcy proceedings.

         2. Each named plaintiff alleges that he purchased common stock of Embrace during the Class A period. (Sec.Am.Comp. ¶ 5). The second amended complaint does not disclose whether any or all of the plaintiffs disposed of any of their stock before February of 1994, when Embrace stock was delisted from the NASDAQ system. Consequently, the record does not disclose whether any of the plaintiffs remain shareholders of Embrace. Additionally, plaintiff Babcock alleges that he purchased $37,500 of Embrace's Series D common stock warrants in April or May of 1993. ( Id., ¶ 5(e)).           3. The individual defendants are former directors, officers, and employees of Embrace, who were allegedly responsible for the false public statements giving rise to plaintiffs' claims. Defendant Price Waterhouse acted as Embrace's certified public accountant from May 12, 1992, through March 2, 1994, when it resigned. Plaintiffs' principal claim against Price Waterhouse is that it issued an unqualified opinion on Embrace's 1992 financial statements, in circumstances that make Price Waterhouse liable for the material misrepresentations contained therein. It is undisputed that April 15, 1993, was the earliest possible date the Price Waterhouse opinion was publicly available. Plaintiffs also claim that Price Waterhouse participated in preparing false SEC filings thereafter.

          B. Plaintiffs' Claims

         4. As noted above, plaintiffs assert three federal claims and two state-law claims. Count I, the only federal count brought against Price Waterhouse, is under section 10(b) of the 1934 Act and Rule 10b-5 brought on behalf of both classes. Count II is brought on behalf of both classes against the individual defendants as controlling persons under section 20 of 1934 Act. Count III, brought on behalf of Class B only, concerns the common stock warrants and arises under section 12(2) of the 1933 Act. In count III, plaintiffs seek recision of the warrant purchases, among other relief. The last two counts assert state-law claims of fraud and conspiracy on behalf of both classes against all defendants.

         5. Plaintiffs' claims arise from a series of public announcements and disclosures, alleged to be false, fraudulent, and misleading. The allegations concerning these disclosures consume approximately thirty pages of the Second Amended Complaint (¶ ¶ 19-90). For present purposes, the allegations of false, incomplete and misleading statements may be summarized as falling into four categories. First, plaintiffs allege that Embrace misrepresented the status of its " Puffibre" product, falsely stating that manufacturing operations had begun and making a series of baseless predictions about future sales and profits. Second, Embrace made misstatements about its capital structure, including falsely claiming that it had reacquired 5 million shares, which resulted in an understatement of net loss per share. Third, the company improperly accounted for the value of certain assets, causing over half of its asset value to be seriously overstated. Fourth, Embrace falsely announced that an institutional investor had agreed to purchase at least $20 million of its common stock. Finally, plaintiffs attack numerous financial disclosures made in Embrace's audited financial statements, Form 10-Q's, Form 10-K's, Form 8-K's, Form S-8's, and the like. These allegedly false statements were made over nearly a two-year period, commencing with press releases issued on May 21, 1992, continuing through Forms 10-Q issued November 13 and December 3, 1992, through audited financial issued April 15, 1993, through press releases and SEC filings issued during 1993 and early 1994.

         6. On February 16, 1994, the company announced that serious misstatements had been made by former management officials in 1992 and 1993. One of these consolidated actions was filed on the same day, February 16, 1994, in the Western District of Michigan. New SEC filings by Embrace disclosed the existence of most of the allegedly fraudulent statements set forth in the complaint. The price of the company's stock fell and the stock was ultimately delisted from the NASDAQ system.

          C. Embrace Common Stock

          7. The record does not presently indicate the number of shares of Embrace common stock outstanding at May 21, 1992, the beginning of the putative class period. The stock had no trading history before April 29, 1992, as the company did not exist before that date. By November 12, 1992, there were 8,102,625 shares outstanding, according to the then-current form 10-Q. On December 31, 1992, Embrace had 346 shareholders. Plaintiffs have not provided the court with any information concerning the number of persons who purchased shares during the Class A period, despite the availability of discovery for this purpose. Consequently, the number of class members of Class A, although apparently ascertainable, is not presently known. Furthermore, plaintiffs have not supplied information concerning the number of purchasers during the class period who are excluded from class membership because of their insider status with Embrace.

         8. Throughout the relevant period, Embrace common stock was listed on the NASDAQ system as a " SmallCap" company. NASDAQ is a system of computers linking brokers and financial institutions around the country. The system displays information concerning trades of listed shares, as well as the prices at which market makers in the shares are willing to buy or sell the particular stock. Substantial publicly held companies such as Microsoft are listed on the NASDAQ National Market System, or " NASDAQ/NMS." Such companies make up approximately two-thirds of the shares traded on the NASDAQ system, but represent more than 96% of the capitalization of NASDAQ companies. (Kangas Aff., ¶ 13). Smaller companies are listed on " NASDAQ SmallCap." Such companies represent the " bottom tier of NASDAQ securities in terms of capitalization, number of shareholders and activity." (Kangas Aff., ¶ 13). Although they make up one-third of the system's shares, such companies represent only about 3% of the capitalization of NASDAQ companies. ( Id. ).

         9. Embrace common stock was actually traded for less than two years, from April 29, 1992, to February 14, 1994. For the great majority of this time, the stock was very thinly traded. The stock averaged only seven trades per day. Exhibit E to the Kangas affidavit shows daily volume for trading in Embrace stock during this period. Defendants' Hearing Exhibit B portrays the same information in graphic form. These exhibits show that from April of 1992 through the end of September, 1993, trading in the stock was minuscule to nonexistent. On thirty days during the purported class period, no trades took place at all, the non-trading days being equally divided between 1992 and 1993. On numerous other days ( e.g., 7/2/92, 7/7/92, 9/22/92, 10/7/92, 10/18/92, 10/26/92, 11/10/92, 11/18/92, 12/9/92, 1/5/93, 5/10/93, 6/29/93), trading volume was only 1,000 or 2,000 shares per day. Then in late September, 1993 (September 23 and 24), two days of relatively heavy trading occurred, followed by another period of virtually nonexistent trading. On October 25, 1993, a period of what can only be termed " frenzied" trading began, with daily volumes in the millions of shares. This lasted for less than four months, at the end of which time the stock was delisted. The mean weekly trading volume from May 21, 1992, through October 22, 1993, was only 39,627 shares. Only when the final four months of atypical trading are included does the mean weekly trading volume for the entire period exceed 200,000 shares. The stock's median trading volume as a percentage of outstanding shares was only .08%, as compared to the average NASDAQ percentage of .8%. (Cox Aff., ¶ 17).

         10. The market for the purchase and sale of Embrace common stock during the putative class period exhibits none of the attributes of an " efficient" market, that is, a market in which the price of the security promptly, accurately, and without bias reflects available information about the value of the company. The affidavit and testimony of Dr. Charles C. Cox, a former commissioner and acting chairman of the SEC, clearly indicate that Embrace stock was too thinly traded and its price too insensitive to changes in market information to be considered efficient. Dr. Cox compared Embrace stock to 99 randomly selected, publicly traded companies of all sizes (Cox Aff., ¶ 10) and concluded that the market for Embrace stock compared poorly with regard to important indicia of efficiency, such as level of capitalization, percentage of shares held by insiders, bid-ask spread, coverage in financial reports, stock trading volumes, volatility of price, and interest in short selling.

          11. The affidavit and testimony of Mark Kangas, plaintiff's economic expert, took issue with that of Dr. Cox and argued that the market for Embrace stock was efficient. Mr. Kangas based this conclusion principally on the fact that Embrace stock was listed on NASDAQ. There is no evidence that mere listing on NASDAQ, especially on the SmallCap tier, is an indicia of efficiency, let alone a conclusive one. Mr. Kangas also criticized Dr. Cox for comparing Embrace to much larger companies, accusing him of unfair comparison. Kangas's criticism misses the point. Comparison of Embrace to larger, actively traded companies is critical, because the economic studies establishing the attributes of market efficiency were all conducted on larger companies. To the extent that a smaller company such as Embrace can be shown to resemble these " model" companies, one might safely conclude that the market for that stock is relatively efficient. However, as the attributes of a small company begin to diverge from those of the model companies, a finding of market efficiency becomes unwarranted.

         12. On the present record, there is little, if any, support for a finding that the market for Embrace shares was efficient at any time during the company's existence.

          D. Common Stock Purchase Warrants

         13. During 1993, Embrace offered and sold Series D common stock purchase warrants to fifty-six investors. The price of the warrants was established by Embrace, not by any market mechanism. The warrants were sold in fixed blocks of 12,500 warrants (1 unit) at the set price of $1.00 per warrant. Plaintiffs allege that a total of $2,249,500 in warrants was ultimately sold. (Sec.Am.Comp. ¶ 89(a)). The warrants entitled purchasers to buy Embrace common stock at $8.00 per share between May 1, 1993, and December 31, 1996.

         14. The complaint alleges that the sale of warrants violated the securities laws because of the false and misleading statements concerning the company, as summarized above. Additionally, they allege specific impropriety peculiar to the sale of the warrants, in that the company failed to file a registration statement for the underlying securities and failed to deliver the warrants until late 1993 or early 1994.

         15. The Embrace warrants were not sold on the open market. Rather, they were offered to certain individuals who met minimum financial requirements. The offering was made by Embrace either directly to prospective purchasers or through agents.

         16. The warrants were not registered, and plaintiffs do not allege that registration was required. Embrace relied upon the exemption from registration created by section 4(6) of the 1933 Securities Act and Regulation D for limited offerings to " accredited investors." The warrants were offered pursuant to offering materials and a subscription agreement dated February 1, 1993. (See Babcock Dep., Ex. 3A). The " Purchaser Statement" included in those materials contained the purchaser's acknowledgment that there was " no market" for the warrants and that the warrants were not transferable by the purchaser.

         17. Plaintiff Babcock received offering materials in late February, 1993, receiving some items from Embrace and others from a brokerage house. (Babcock Dep., 47-62). He testified that the warrant offering was a " private placement." ( Id., 65). Babcock ultimately purchased $37,500 (three units) of warrants in " April or May" of 1993. (Sec.Am.Comp. ¶ 5(e)). Herbert Eldean, who was originally a named plaintiff but has since withdrawn, purchased $300,000 (24 units) worth of warrants. Consequently, the smallest investment by the fifty-six purchasers of Embrace warrants was $12,500 (1 unit), while investors such as Babcock and Eldean invested far more.

         18. At no relevant time was the market for Embrace stock purchase warrants " open," " impersonal," or efficient. In fact, it was a closed market with a single seller, Embrace. The warrants were not listed on any exchange, but were sold in individual transactions between purchasers and Embrace or its agents.

          Discussion

         I. Class Definitions

          As noted above, plaintiffs purport to represent two classes: Class A, composed of all purchasers of Embrace common stock from May 21, 1992, through February 15, 1994, and Class B, all purchasers of Embrace's Series D common stock purchase warrants. It is obvious at the outset that plaintiffs' definitions are not defensible with regard to its claims against Price Waterhouse. As Judge Bell determined in his opinion of April 26, 1995 (docket # 107), Price Waterhouse is only liable under section 10(b) for actually making material misstatements. See Central Bank of Denver v. First Interstate Bank of Denver, 511 U.S. 164, __, 114 S.Ct. 1439, 1448, 128 L.Ed.2d 119 (1994). Secondary liability for aiding and abetting a section 10(b) violation was expressly rejected by the Central Bank Court. Id. at 1439, 114 S.Ct. at 1447. The earliest possible statement by Price Waterhouse that would qualify under the Central Bank case was its opinion certifying Embrace's 1992 financial statements. This opinion was not released to the public until April 15, 1993.

         Consequently, it is patent that no class can be certified in this case with regard to Price Waterhouse containing members whose only purchase of Embrace stock predated April 15, 1993. It is also clear that some of the stock purchases by plaintiffs Gary Reuter, Harry Hoffman, David Babe, and Robert Esna, cannot form the basis for an individual claim, let alone a class-based claim, against Price Waterhouse, as those stock purchases predated April 15, 1993. ( See Sec.Am.Comp. ¶ ¶ 5(b), (c), (f), (g)).

         If purchasers of common stock are to be accorded class treatment in this case, two classes must be defined: purchasers between May 21, 1992, and February 15, 1994, asserting claims against Embrace and the individual defendants (original Class A); and purchasers between April 15, 1993, and February 15, 1994, asserting claims against those defendants and Price Waterhouse (" Class A/PW" ).

         Putative Class B, composed of purchasers of Series D common stock warrants " during 1993," suffers from the same problem. Warrant purchasers who bought before April 15, 1993, have no possible claim against Price Waterhouse. Consequently, two classes of warrant purchasers must be defined: purchasers of the warrants issued in 1993 asserting claims against Embrace and the individual defendants (original Class B); and warrant purchasers on or after April 15, 1993 asserting claims against those defendants and Price Waterhouse (" Class B/PW" ).

         II. Factors Under Rule 23

         Rule 23 of the Federal Rules of Civil Procedure sets forth the requirements for maintenance of a class of persons as a party to an action. That rule provides in part:

(a) PREREQUISITES TO A CLASS ACTION. One or more members of a class may sue or be sued as representative parties on behalf of all only if (1) the class is so numerous that joinder of all members is impracticable, (2) there are questions of law or fact common to the class, (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class, and (4) the representative parties will fairly and adequately protect the interests of the class.

FED.R.CIV.P. 23(a). Plaintiffs in the present case seek to maintain a Rule 23(b)(3) class action, which requires the additional finding that

the questions of law or fact common to the members of the class predominate over any questions affecting only individual members, and that a class action is superior to other available methods for the fair and efficient adjudication of the controversy.

FED.R.CIV.P. 23(b)(3).

          " The class action device was designed as an exception to the usual rule that litigation is conducted by and on behalf of the individual named parties only." General Telephone Co. of the Southwest v. Falcon, 457 U.S. 147, 155, 102 S.Ct. 2364, 2369, 72 L.Ed.2d 740 (1982). District courts have broad discretion in certifying and managing class actions. See Califano v. Yamasaki, 442 U.S. 682, 703, 99 S.Ct. 2545, 2558, 61 L.Ed.2d 176 (1979); Weaver v. University of Cincinnati, 970 F.2d 1523, 1531 (6th Cir.1992), cert. denied, 507 U.S. 917, 113 S.Ct. 1274, 122 L.Ed.2d 668 (1993). A class action may not be certified unless the trial court is satisfied, after " rigorous analysis," that the prerequisites of Rule 23(a) have been met and that the action falls within one of the three categories of Rule 23(b). See General Telephone Co., 457 U.S. at 161, 102 S.Ct. at 2372. Plaintiffs bear the burden of proving that they satisfy all prerequisites for class certification. Thompson v. County of Medina, 29 F.3d 238, 241 (6th Cir.1994); Reid v. White Motor Corp., 886 F.2d 1462, 1471 (6th Cir.1989), cert. denied, 494 U.S. 1080, 110 S.Ct. 1809, 108 L.Ed.2d 939 (1990). Having performed the analysis required by the Supreme Court, I find that plaintiffs have not borne their burden of showing that this case is appropriate for class action treatment.

          A. Numerosity

          The first prerequisite under Rule 23(a) is that " the class is so numerous that joinder of all members is impracticable." According to the Supreme Court: " The numerosity requirement requires examination of the specific facts of each case and imposes no absolute limitations." General Tel. Co. of Northwest, Inc. v. EEOC, 446 U.S. 318, 330, 100 S.Ct. 1698, 1706, 64 L.Ed.2d 319 (1980). To satisfy the numerosity requirement, the plaintiffs must show that joinder is impracticable, not impossible. Daigle v. Shell Oil Co., 133 F.R.D. 600, 603 (D.Colo.1990). In determining whether joinder is impracticable, the court must consider the size of the proposed class, the class's geographical dispersion, the ability of claimants to bring individual suits, and whether the members' names are easily ascertainable. Id.; see Robidoux v. Celani, 987 F.2d 931, 936 (2d Cir.1993); see also, Hum v. Dericks, 162 F.R.D. 628, 634 (D.Haw.1995) (applying these factors, noting there is " no magic number," declining to certify 200 member class and collecting cases where class certification was denied to larger proposed classes).

         In the present case, the court cannot begin to perform the " examination of the specific facts" required by the Supreme Court, as plaintiffs have failed to provide the court with any information upon which to base a finding concerning the numerosity of putative Classes A or A/PW. The number of purchasers should be easily ascertainable from transfer records of Embrace. Although this case was commenced over one year ago, plaintiffs have not consulted the stock transfer records to determine the number of purchasers of common stock during the putative class period. Instead, plaintiffs ask the court to engage in guesswork by presuming a large number of purchasers (" possibly hundreds" ) on the basis of the existence of ten million shares of outstanding common stock. Plaintiffs rely on cases holding that plaintiffs need not identify each class member with specificity and allowing the court to rely on " common sense assumptions" in order to show numerosity. E.g., Wolgin v. Magic Marker Corp., 82 F.R.D. 168, 171 (E.D.Pa.1979).

         Plaintiffs' suggested approach is unacceptable. Where the size of a proposed class is purely speculative, the numerosity requirement of Rule 23(a) is not met. Marcial v. Coronet Ins. Co., 880 F.2d 954, 957 (7th Cir.1989); Kohn v. Mucia, 776 F.Supp. 348, 352-53 (N.D.Ill.1991). Although a presumption of the type suggested by the cited decisions may be warranted in some cases, especially those involving NYSE companies, the facts of this case do not warrant any presumption concerning the number of purchasers of common stock. At oral argument, plaintiffs conceded that shareholders existed before the beginning of the putative class period, as a result of the original merger. The only information concerning the number of shareholders provided to the court is that at the end of 1992, there were 346 shareholders of record. There is no indication concerning how many of these persons held shares before the beginning of the class period and how many acquired them during the class period. Reliance upon the mere number of outstanding shares as indicative of a large number of purchasers could be misleading in this case, because of the huge concentration of shares in the hands of insiders, who are expressly excluded from the class. The second amended complaint itself (¶ ¶ 37-38) discloses that five million shares, half of the total outstanding, were owned by D. Mark Meade, Embrace's chairman and CEO who resigned to take another position on or about October 27, 1992. Another 2.8 million shares were owned by Donald W. Freeman, a defendant (¶ 6). Both expert witnesses alluded to the large number of shares held by insiders. Consequently, the record is devoid of any direct evidence concerning the number of purchasers of common stock during any period, and the facts in this case admit of no presumption concerning a large number of purchasers.

          Plaintiffs assert the principle that difficulty in identifying all class members should not deter a court from certifying a class. See Roman v. Korson, 152 F.R.D. 101, 105-06 (W.D.Mich.1993). Plaintiffs are not entitled to rely on this principle. There has been no showing of any difficulty in ascertaining the identity of class members. Plaintiffs have simply failed to make any effort to do so. Where, as here, a plaintiff has made no attempt to ascertain the size of the class, courts have refused to engage in speculation on the issue of numerosity. See, e.g., Makuc v. American Honda Motor Co., 835 F.2d 389, 394-95 (1st Cir.1987); Vigue v. Ives, 138 F.R.D. 6, 8 (D.Me.1991); Waller v. International Harvester Co., 98 F.R.D. 560, 563 (N.D.Ill.1983).

         In summary, plaintiffs have presented absolutely no evidence concerning the number of potential members of Class A or Class A/PW, nor have they provided any excuse, such as unavailability of information, for their failure to do so. The facts presented in this case do not allow any " common sense assumptions" concerning the number of common stock purchasers. Plaintiffs have therefore failed in their burden of showing that the number of common stock purchasers during either putative class period A or A/PW is so numerous that joinder is impracticable.

         The B classes, composed of purchasers of Embrace Series D warrants, suffer from a different affliction. The named plaintiffs have ascertained the identity of the members of Class B, identifying fifty-six purchasers of the Embrace warrants. Despite the relatively small number of class members, plaintiffs argue that a class should nevertheless be certified, citing cases in which classes with as few as twenty-five members have been found to satisfy the numerosity requirement. See, e.g., Philadelphia Elec. Co. v. Anaconda Am. Brass Co., 43 F.R.D. 452 (E.D.Pa.1968).

          Plaintiffs' analysis is superficial. The size of the putative class is not determinative of the numerosity question. The " real inquiry" under Rule 23(a)(1) is " whether joinder would be impractical." Daigle v. Shell Oil Co., 133 F.R.D. at 603. " A relatively small class may be certified if joinder is impractical. However, the converse must also be true for relatively large classes; certification may be denied if all plaintiffs can be joined." Id.

          Determination of practicability of joinder depends on " all the circumstances surrounding a case, not on mere numbers." Robidoux v. Celani, 987 F.2d at 936. Relevant considerations include judicial economy arising from the avoidance of a multiplicity of actions, geographic dispersion of class members, financial resources of class members, the ability of claimants to institute individual suits, and the like. Id. These factors tend to militate against class certification. The number of members for Class B (fifty-six) is finite and relatively small; membership in Class B/PW could be even smaller. More important, each class member has a large individual investment and therefore a strong motivation for individual action. The facts presently available to the court indicate that the smallest investment in warrants was $12,500 and the largest $300,000. Named plaintiff Babcock, whose claims are alleged to be typical of other Class B members, had a $37,500 investment in warrants. These facts indicate that warrant purchasers have strong financial resources and the ability to institute their own actions. The B classes do not present a situation where numerous small investors, each with a relatively insignificant loss, would have no practical impetus for maintaining an individual action, such that a class action is the only means of affording a remedy. See, e.g., Deposit Guar. Nat'l Bank v. Roper, 445 U.S. 326, 339, 100 S.Ct. 1166, 1174, 63 L.Ed.2d 427 (1980) (where alternatives are " multiplicity of small individual suits" or class action, class action may be the only effective remedy). Nor is this a situation where the class is composed of ADFC claimants or other members whose poverty would preclude individual action. See, e.g., Robidoux, 987 F.2d at 936. The factual situation presented by the B classes involves a small number of relatively well-to-do, sophisticated investors, each with a sizable claim for damages, and a resulting strong individual motivation to seek relief.

          Although the warrant purchasers are alleged to reside in eleven different states, this in itself does not appear to provide an obstacle to joinder, either in the present case or in a finite number of individual actions in which warrant purchasers are grouped together as plaintiffs. In this regard, the present case closely resembles McMerty v. Burtness, 72 F.R.D. 450 (D.Minn.1976), a federal securities case in which the court refused to certify a class numbering 79 members located in seven states. The court noted that each putative member had a substantial monetary interest in excess of $4,000. 72 F.R.D. at 453. In finding that the putative class did not meet the numerosity requirement of Rule 23(a), the court made the following comment, which applies with equal force to the present case: " This case does not present the situation in which the small magnitude of each claim relative to the high cost of individual litigation makes a class action the only effective mechanism for seeking redress." Id. Other courts faced with small class membership and high stakes have reached the same conclusion. See, e.g., Liberty Lincoln Mercury v. Ford Marketing, 149 F.R.D. 65, 74 (D.N.J.1993) (refusing to certify class of 38 to 123 car dealers, commenting that " a class action is not appropriate when proposed class members are able to protect and defend their own interests." ); Daigle v. Shell Oil Co., 133 F.R.D. at 603 (78 class members, each with substantial claim).

         Courts have repeatedly recognized that conclusory or speculative allegations that joinder is impracticable are not sufficient to satisfy the numerosity requirement of Rule 23(a). See Marcial v. Coronet Ins. Co., 880 F.2d at 957; Liberty Lincoln Mercury, 149 F.R.D. at 74. In the present case, plaintiffs' presentation concerning the alleged numerosity of the B classes does not rise above the merely conclusory. The balance of relevant factors indicates that joinder of the fifty-six warrant holders in a single action would not be impracticable in the circumstances of this case. I therefore conclude that the requisite showing of numerosity has not been made for any of the putative classes.

          B. Commonality

         Rule 23(a)(2) requires that " questions of law or fact common to the class" be present as a prerequisite to any class action. This provision does not require that all questions of law and fact raised in the case be common. See Forbush v. J.C. Penney Co., 994 F.2d 1101, 1106 (5th Cir.1993). Of the issues identified by plaintiff, the following appear to be common:

(a) whether defendants violated federal securities laws and state common law;

(b) whether the press releases, SEC filings, and other public statements made by defendants misrepresented material facts;

(c) whether all defendants participated in a common course of conduct;

(d) whether all defendants acted willfully or recklessly; and

(e) whether the market price of the common stock was artificially inflated.

( See Sec.Am.Comp. ¶ 16). The issues identified by plaintiffs clearly are common to the claims of all class members and easily satisfy the commonality requirement of Rule 23(a)(2).

         Defendants contest commonality, arguing that the presence of oral representations and the need for each plaintiff to show reliance undermine a finding of commonality. While defendants raise substantial issues, they pertain more properly to the typicality or predominance issues, discussed below. The existence of one or more individualized questions does not defeat a finding of commonality, as Rule 23 does not require that all questions of law or fact be common. See Rodriguez v. Berrybrook Farms, Inc., 672 F.Supp. 1009, 1015 (W.D.Mich.1987).

         I therefore conclude that plaintiffs have established the commonality requirement for putative Classes A, A/PW, B, and B/PW.

          C. Typicality

          The third prerequisite in Rule 23(a) allows a class action to be maintained only if " the claims or defenses of the representative parties are typical of the claims or defenses of the class." In order to satisfy the typicality requirement, the class representatives' claims must have the same essential characteristics as the claims of the other members of the class. De La Fuente v. Stokely-VanCamp, Inc., 713 F.2d 225, 232 (7th Cir.1983). A representative's claim is typical if it arises from the same event or course of conduct giving rise to the claims of other class members and the claims are based on the same legal theory. Id. The test for typicality, like commonality, is " not demanding." Forbush v. J.C. Penney Co., 994 F.2d 1101, 1106 (5th Cir.1993).

         In Classes A and A/PW, all claims are essentially based upon allegedly false information disseminated over a period of months concerning the future of the Puffibre product, Embrace's capital structure, and the value of certain of its assets. Although the claims of purchasers who bought their shares early in the period may differ somewhat from those who purchased later, depending on the amounts of information or misinformation then known to the market, these differences do not now appear material. A finding of typicality is even stronger for the B classes in count 3, as the common stock underlying the Series D warrants was allegedly never registered. According to plaintiffs' theory, this failure rendered the entire sale of warrants void. If this is so, then the claim of every Class B plaintiff is the same as the claim of every other.

          In arguing against a finding of typicality, defendants rely upon the doctrine that a plaintiff subject to unique defenses can be rendered atypical and not representative of the class. See Ballan v. Upjohn Co., 159 F.R.D. 473, 479-481 (W.D.Mich.1994). A defense that is peculiar to the named plaintiff can destroy typicality, because such a defense " can distract the named plaintiff to such an extent that his or her representation of the interests of the rest of the class will suffer." Id. at 480. In arguing this point, defendants assert that a number of the plaintiffs are susceptible to unique defenses because they had access to non-market and insider information. Defendants assert that several plaintiffs had direct contact with officers and employees of Embrace or that they talked to stock brokers, who themselves had contact with corporate officials. Having reviewed the deposition excerpts relied upon by defendants, I determine that defendants' contention is not borne out by the record, except as to Ginsberg. Most plaintiffs appear from the record to have been rather typical investors who relied upon their brokers for advice. In some instances, they also spoke to officers of the company. At best, the facts relied upon by defendants indicate that these plaintiffs may have had an opportunity to receive oral representations concerning the stock and that their decision to purchase the stock could have been based upon the oral information. At this point, however, their receipt of information and possible reliance upon it remain speculative. Each plaintiff has filed an affidavit indicating that he did not rely upon any information that was not a matter of public record. It often happens that purchasers rely upon statements by brokers, and that brokers have some level of access to corporate officials. This, without more, is insufficient to find a class representative to be atypical. See Kirkpatrick v. J.C. Bradford & Co., 827 F.2d 718, 724 (11th Cir.1987), cert. denied, 485 U.S. 959, 108 S.Ct. 1220, 99 L.Ed.2d 421 (1988).

          Plaintiff Ginsberg presents different circumstances. Unlike the other plaintiffs, he was not merely an investor in the company. Rather, he was a paid consultant, hired to help the company with its corporate strategic planning. In this capacity, he met with corporate officers in private, spending twenty to thirty minutes each with certain individuals in November of 1993. In addition, he reviewed corporate documents, including a " business segment planning" document, which addressed the company's strengths, weaknesses, markets, and performance. ( See Deposition Exhibits, attached to Def. Ex. F). Mr. Ginsberg purchased 3,800 shares of Embrace stock on November 17, 1993, a few days after his meetings with numerous Embrace officials. (Dep. at 49-50; Sec.Am.Comp. ¶ 5(i)). Mr. Ginsberg's purchase of an additional 2,000 shares on December 2, 1993, immediately followed his holding of a two-day workshop on December 1 and 2. (Dep. at 54-55; Sec.Am.Comp. ¶ 5(i)).

          On the basis of these facts, it is clear that Mr. Ginsberg was in no sense a " typical" investor in Embrace stock and that his situation differs markedly from that of class members. At this point in the litigation, it is not necessary for this court to find with finality that Mr. Ginsberg in fact relied upon non-public information in making his stock purchases. To negate the typicality of a representative's claim, it is only necessary that the defense be " unique, arguable and likely to usurp a significant portion of the litigant's time and energy." McNichols v. Loeb Rhoades & Co., 97 F.R.D. 331, 334 (N.D.Ill.1982); accord, J.H. Cohn & Co. v. American Appraisal Assoc., Inc., 628 F.2d 994, 999 (7th Cir.1980) (presence of arguable defense peculiar to named plaintiff sufficient to destroy typicality).

         With regard to the B classes, defendants again point to some contact between plaintiff Babcock (the only named plaintiff) and two brokers from Paragon Capital who were marketing Series D warrants. As is true with the majority of the plaintiffs, Mr. Babcock appears to be a typical investor who merely had discussions with a broker in connection with purchasing shares. The courts have not disqualified named plaintiffs who may have relied upon brokers, for the obvious reason that the brokers were probably themselves relying upon public information. Defendants have not made a showing that Mr. Babcock's decision was probably made on inside information or that his situation was atypical. Unlike the situation presented by plaintiff Ginsberg, there is no real danger that the trial will be sidetracked by an exposition of personal defenses peculiar to Mr. Babcock.

         In summary, I recommend a finding of typicality for Classes A and A/PW for all plaintiffs except Ginsberg and for plaintiff Babcock with respect to Class B and B/PW.

          D. Adequacy of Class Representatives

         Two separate considerations are relevant with regard to the question of the named plaintiffs' adequacy as representatives of the putative classes under FED.R.CIV.P. 23(a)(4). The first is the experience and ability of class counsel. See Cross v. National Trust Life Ins. Co., 553 F.2d 1026, 1031 (6th Cir.1977). Defendants have raised no concerns in this regard, and plaintiffs' attorneys appear to be experienced in the securities-law area.

          The second consideration involves the existence of possible antagonistic interests between named plaintiffs and absent class members. See Cross, 553 F.2d at 1031. One possible source of such antagonism is the existence of defenses unique to a named plaintiff. " Where it is predictable that a major focus for litigation will be on an arguable defense unique to the named plaintiff or a small sub-class, then the named plaintiff is not a proper class representative." Fradkin v. Ernst, 98 F.R.D. 478, 488 (N.D.Ohio 1983). In this regard, the adequacy prong of Rule 23 overlaps with considerations of typicality. As noted above, none of the named Class A or A/PW plaintiffs appears to be subject to a unique defense so distracting that it could create an antagonistic interest, except plaintiff Ginsberg. Consequently, to the extent that plaintiff Ginsberg's claims are not typical, he also must be deemed an inadequate representative of the members of Classes A and A/PW. See J.H. Cohn, 628 F.2d at 999 (presence of arguable defense peculiar to named plaintiff undermines both typicality and adequacy).

          With respect to the B classes, the adequacy of the only named plaintiff, Craig Babcock, is subject to a major, unanswered question. As noted above, no purchaser of stock or warrants has any section 10(b) claim against Price Waterhouse until, at the earliest, April 15, 1993, the date of publication of the auditor's opinion. The class period for putative Class B/PW therefore can begin no earlier than April 15, 1993. Babcock alleges that he purchased warrants in " April or May" of 1993. (Sec.Am.Comp. ¶ 5). This vague allegation fails to establish that Babcock has any claim against Price Waterhouse, as a purchase in early April would not be actionable under Central Bank. The bedrock requirement of adequacy is that the named representative be a member of the class. See General Telephone Co., 457 U.S. at 156, 102 S.Ct. at 2369; Reid v. White Motor Corp., 886 F.2d at 1471. It would clearly be unwarranted to certify a class represented by a single named plaintiff who has no claim against the only solvent defendant. Until Babcock clarifies the date of his purchase, his adequacy as class representative will not be established.

          A class representative may also be found inadequate where the representative's interests actually conflict with those of absent class members. Because of the binding effect of any judgment on absent class members, the adequacy of representation and the absence of conflicts of interest involves the due process rights of absent members. See Hansberry v. Lee, 311 U.S. 32, 45, 61 S.Ct. 115, 119, 85 L.Ed. 22 (1940). The adequacy of class representatives in this regard is primarily a factual issue for determination by the trial court. See Ballan v. Upjohn Co., 159 F.R.D. at 482.

         Because of the theory of liability that plaintiffs have chosen to espouse, serious potential conflicts of interest may arise. Plaintiffs are proceeding on a theory of " fraud on the market." The essence of the fraud-on-the-market theory is that a plaintiff seeking to recover under section 10(b) for material misstatements need not prove personal reliance upon the misstatement. Instead, reliance is to be presumed. See Basic, Inc. v. Levinson, 485 U.S. 224, 108 S.Ct. 978, 99 L.Ed.2d 194 (1988). Such presumption is warranted if the market for the stock is open, impersonal and efficient, such that any material misrepresentation made by the issuer will quickly and accurately be reflected in the market price of the security. 485 U.S. at 244, 108 S.Ct. at 990. The linchpin of the fraud-on-the-market theory is the concept of " price inflation." The presumption of reliance recognized in Basic assumes that the price of the security has transmitted the misrepresentation. Indeed, in the instant case, plaintiffs expressly allege that the price of Embrace common stock was " artificially inflated" throughout the class periods as a result of the alleged frauds. (Sec.Am.Comp. ¶ 1).

         The courts are just beginning to recognize the conflicts of interest that are virtually inherent when a fraud-on-the-market theory is pressed on behalf of a plaintiff class. The seminal case in this area is In re Seagate Technology II Securities Litigation, 843 F.Supp. 1341 (N.D.Cal.1994). The Seagate court identified two major conflicts of interest that could serve to make class representatives inadequate. The " most serious" type of class conflict is that between stock traders who sell securities on a particular day and those plaintiffs who purchase on the same day. 843 F.Supp. at 1359. Selling plaintiffs benefit from maximizing the difference between price inflation at purchase and inflation at sale. 843 F.Supp. at 1359. The interest of those claimants lies in minimizing the degree of price inflation existing at the date of sale. Those plaintiffs buying on that date have " an exactly opposite interest." Id. Such a person would " seek to maximize the degree of price inflation existing on that date. In other words, different plaintiffs have conflicting incentives in shaping the evidence." Id. The Seagate court recognized that previous decisions tended to dismiss this problem, but criticized these decisions and rejected their use of sub-classes as a false solution to this very real problem. Id. at 1359-1362.

          In Ballan v. Upjohn Co., 159 F.R.D. 473 (W.D.Mich.1994), Judge Hillman followed Seagate in denying class certification, on grounds of adequacy, to plaintiffs asserting a fraud-on-the-market theory. 159 F.R.D. at 482-84. Other courts have refused to follow Seagate. Compare Ballan, 159 F.R.D. at 485 ( Seagate is a " scholarly and thoughtful exposition of two potential sources of significant class conflict that go beyond mere calculation of damages." ) with In re Proxima Corp. Sec. Lit., 1994 WL 374306, at *19 (S.D.Cal.1994) (declining to follow Seagate ). In my opinion, those cases declining to at least examine the issue abdicate their responsibility to absent class members under Rule 23 and have not confronted the implications of the fraud-on-the-market theory for class actions. The comment of the Seagate court is apt: " Driven perhaps by an overarching desire to effectuate the securities laws and provide relief to small-claim plaintiffs, the courts have failed to address the importance that the fraud-on-the-market theory gives to class conflicts over price inflation. These conflicts are not only serious, but they are pervasive and enough to preclude satisfaction of the adequacy of representation element of [FedR.Civ.P.] 23." Seagate, 843 F.Supp. at 1359.

         Plaintiffs in the present case have completely glossed over this troubling issue. The complaint itself is inadequate to determine whether and to what extent such a conflict exists, as it does not allege whether some or all of the class representatives sold stock during the class period. Plaintiffs' legal presentation does not mention the issue or raise possible solutions for any problems that may exist. " [P]laintiffs must provide evidence establishing that to the extent conflicts exist in the plaintiff class, they are not so serious as to preclude a finding of adequacy of representation." Seagate, 843 F.Supp. at 1366. Plaintiffs in the instant case have made no effort in this regard. At this point, they have not met their burden of showing adequacy of representation.

          E. Predominance of Common Questions

          In addition to meeting all prerequisites of Rule 23(a), a plaintiff seeking class certification must satisfy at least one of the prerequisites of subsection (b) of the Rule. In the present case, plaintiffs rely upon Rule 23(b)(3), which allows a class action to be maintained if " the court finds that the questions of law or fact common to the members of the class predominate over any questions affecting only individual members, and that a class action is superior to other available methods for the fair and efficient adjudication of the controversy." FED.R.CIV.P. 23(b)(3). " These requirements reflect the fact that special caution must be exercised in class actions of this type because of the loose affiliation among the class members, which is thought to magnify the risks inherent in any representative action. Thus, the draftsmen of the rule have imposed an obligation on the court to make findings that will demonstrate the utility and propriety of employing the class action device in the case before it." 7A CHARLES A. WRIGHT, ARTHUR R. MILLER, AND MARY KAY KANE, FEDERAL PRACTICE AND PROCEDURE: CIVIL, § 1777 at 518 (2d ed. 1986).

         The predominance test involves " an attempt to achieve a balance between the value of allowing individual actions to be instituted so that each person can protect his own interests and the economy that can be achieved by allowing a multiple party dispute to be resolved on a class action basis." Id. at 519. In determining whether common issues predominate, the court " is under a duty to evaluate the relationship between the common and individual issues," taking a pragmatic approach to determine the relative weight and importance of the common and individual issues. Id., § 1778.

         The bulk of the parties' factual presentations and legal arguments was devoted to the predominance issue. Defendants assert that plaintiffs will be unable to rely upon the fraud-on-the-market theory and that individual issues of reliance, proximate cause and damages will therefore overwhelm the common issues. The testimony of Dr. Cox, defendants' expert witness, was devoted to the proposition that the securities at issue in this case were not traded in an efficient market, and that the rebuttable presumption of reliance as recognized by the Supreme Court in Basic v. Levinson, 485 U.S. 224, 108 S.Ct. 978, 99 L.Ed.2d 194 (1988), is not available to plaintiffs. In response, plaintiffs have raised three essential points. First, they argue that an examination of the strength of their fraud-on-the-market theory represents an impermissible inquiry into the merits of the case in violation of the so-called " Eisen rule," which purportedly prohibits examination of the merits of the plaintiff's case in connection with class certification. (Citing Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 178, 94 S.Ct. 2140, 2152, 40 L.Ed.2d 732 (1974)). Second, they argue that, even if plaintiffs do not have the benefit of the fraud-on-the-market presumption, common questions will continue to predominate because a class action may not be denied merely on account of the need for individual proof of reliance. Finally, they assert that the fraud-on-the-market theory does indeed apply to this case, because the market for Embrace common stock and warrants was " efficient" within the meaning of Basic. In support of this final proposition, plaintiffs offered the testimony of Mark Kangas.           The case law governing the foregoing questions is not harmonious. The fraud-on-the-market theory is of recent vintage, and courts have come to conflicting conclusions concerning its effect on the maintenance of class actions. I conclude, for the reasons set forth in the following sections, that the Sixth Circuit would resolve the issues raised in this case in the following fashion: (1) a preliminary factual inquiry into the merits of the fraud-on-the-market theory is permissible and even necessary, in order for the court to fulfill its obligations under Rule 23; (2) individual questions of reliance, proximate cause, and damages will predominate in this case, in the absence of assumed reliance under the fraud-on-the-market theory; and (3) the fraud-on-the-market theory will very likely be unavailable to plaintiffs under the facts of this case.

         1.

         Plaintiffs argue vociferously that no inquiry into the strength of their fraud-on-the-market case is permissible in connection with a motion to certify a plaintiff class. Plaintiffs rely on the so-called " Eisen rule," which rests upon the following sentence found in the Eisen case: " In determining the propriety of a class action, the question is not whether the plaintiff or plaintiffs have stated a cause of action or will prevail on the merits, but whether the requirements of Rule 23 are met." Eisen, 417 U.S. at 178, 94 S.Ct. at 2153. Plaintiffs are supported in this effort by numerous district court opinions, which have likewise interpreted Eisen to preclude at this stage of the case any careful examination of the nature of plaintiffs' claims or the proof necessary to support them. See, e.g., In re First RepublicBank Sec. Lit., 1989 WL 108795, at *9 (N.D.Tex. Aug. 1, 1989) (examining whether the defendants could rebut the fraud-on-the-market presumption goes to the merits of the case and cannot be considered on motion for class certification); accord, In re Rospatch Sec. Lit., 1991 WL 427890, at *6 (W.D.Mich. July 22, 1991) (mere allegation of theory sufficient for class determination purposes).

         This conclusion is made possible only by a misreading of a single sentence in Eisen, coupled with a disregard of subsequent Supreme Court authority. Eisen was a combined antitrust and securities action in which the named plaintiff had maximum damages of $70. The case's torturous path to the Supreme Court consumed eight years. The district court had initially denied class certification. On a second appeal, the court of appeals had indicated that certification might be available, but remanded the matter for further factual findings. The district court thereupon conducted a preliminary hearing to decide the merits of the case and found that the plaintiff was " more than likely to succeed on the merits." 417 U.S. at 168, 94 S.Ct. at 2147. The district court then imposed upon the defendants 90% of the cost of notifying class members. On the third appeal the circuit court held that the case could not be maintained as a class action. The Supreme Court held that defendants could not be forced to bear the burden of plaintiff's cost in notifying class members. The Court found the district court's method of conducting a hearing, making a finding that plaintiff was " more than likely to prevail on the merits," and then attempting to shift the cost of notifying class members to the defendants, was improper. It was in the context of disapproving the trial court's allocation of costs on the basis of likelihood of success on the merits that the Supreme Court made the statement that has blossomed into the " Eisen rule."

          The Eisen rule represents a gross over reading of the Supreme Court's comment concerning examination of the merits of a class action case. Since Eisen, the Supreme Court has repeatedly recognized that, " the class determination generally involves considerations that are enmeshed in factual and legal issues comprising the plaintiff's cause of action." General Tel. Co. v. Falcon, 457 U.S. at 155, 102 S.Ct. at 2369. " ‘ Evaluation of many questions entering into determination of class action questions is intimately involved with the merits of the claims. The typicality of the representative's claims or defenses, the adequacy of the representative, and the presence of common questions of law and fact are obvious examples. The more complex determinations required in Rule 23(b)(3) class actions entail even greater entanglement with the merits....' " Coopers & Lybrand v. Livesay, 437 U.S. 463, 469 n. 12, 98 S.Ct. 2454, 2458, n. 12, 57 L.Ed.2d 351 (1978) (quoting 15 C. WRIGHT, A. MILLER, & E. COOPER, FEDERAL PRACTICE & PROCEDURE, § 3911, p. 485 n. 45 (1976)).

         Consequently, numerous courts recognize the virtual necessity of examining issues related to the merits of a case when making a class determination. The Eisen language " should not be talismanically invoked to artificially limit a trial court's examination of the factors necessary to a reasoned determination of whether" plaintiffs have met their " burden of establishing each of the Rule 23 class action requirements." Love v. Turlington, 733 F.2d 1562, 1564 (11th Cir.1984). The Sixth Circuit has likewise rejected reliance on Eisen as a reason to limit the rigorous inquiry required of the court to determine whether class certification is appropriate. See Thompson v. County of Medina, 29 F.3d 238, 240-41 (6th Cir.1994); accord, Liberty Lincoln, 149 F.R.D. at 73 (collecting cases).

         Moreover, courts have recognized a special need for factual scrutiny when class representatives seek to proceed on a fraud-on-the-market theory. In the seminal case of Basic, Inc. v. Levinson, in which the Supreme Court first recognized the validity of the fraud-on-the-market theory, the Court itself noted the centrality of the fraud-on-the-market theory on class certification question. The very purpose of the grant of certiorari was " to determine whether the courts below properly applied the presumption of reliance in certifying the class...." 485 U.S. at 230, 108 S.Ct. at 982. In that regard, the Court observed that " [r]equiring proof of individualized reliance from each member of the proposed plaintiff class effectively would have prevented respondents from proceeding with the class action, since individual issues then would have overwhelmed the common ones." Id. at 242, 108 S.Ct. at 989. Consequently, if an examination of the merits of the fraud-on-the-market theory were inappropriate at the class certification stage, the Supreme Court would have had no occasion to grant certiorari or to decide the Basic case. To the extent that Eisen presents some general bar to an examination of the merits, it has been modified by Basic in fraud-on-the-market cases:

Given the relevant recency of the decision in Basic, the court concludes that the Eisen rule prohibiting examination of the merits during the class certification process must yield. By adopting the fraud-on-the-market theory, the Basic court implied, though perhaps unintentionally, that the district courts must examine some of the merits.

In re Seagate Technology II Sec. Lit., 843 F.Supp. at 1367; accord, In re MDC Holdings Sec. Lit., 754 F.Supp. 785, 804-05 (S.D.Cal.1990); Sheftleman v. Jones, 667 F.Supp. 859, 867 (N.D.Ga.1987) (a " reasoned decision on class certification necessitates some inquiry into the reliance question." ); cf., Cammer v. Bloom, 711 F.Supp. 1264, 1287-1293 (D.N.J.1989) (availability of fraud-on-the-market presumption is a question of law for the court, suitable for pretrial evidentiary hearing).

         The commentators have likewise recognized that the advent of the fraud-on-the-market theory requires the court to make an initial assessment of the theory's applicability in connection with the class certification issue:

It is important to have a fairly simple way of resolving-at least tentatively-at an early stage in a case whether the market for a particular security is open, developed and efficient, since this affects whether a claim has been stated and whether a class may be certified. For statement of a claim, the allegations of the complaint are normally sufficient. But their sufficiency for class determination is open to question since the ruling for or against the class in practice often determines the fate of the case. Thus, some sort of evidentiary base is appropriate for class determination.

3 ALAN R. BROMBERG & LEWIS D. LOWENFELDS, SECURITIES FRAUD AND COMMODITIES FRAUD, § 8.6 at 8:814 (2d ed. 1988) (emphasis added).

          It is one thing to argue that a court should refuse class certification because it thinks plaintiffs have a weak case or should be more inclined to certify a class because of the perceived strength of plaintiffs' case. That course would clearly be improper. It does not follow, however, that the court must accept uncritically the class representatives' assertion that a class action will be manageable and that individual issues will not predominate because plaintiffs will have the benefit of a presumption relieving them of the need for individual proofs. In such a case, the court can and must ask whether the case will truly be as simple as plaintiffs contend. The alternative is unacceptable, as it entails putting the court and the parties to the burden of class action treatment of the case on the unexamined assumption that plaintiffs will have the benefit of a presumption that may never materialize. Plaintiffs making such a contention should be prepared to make at least a facial showing that the theory will be available to them in the particular case.

         The court is " 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. Lit., 107 F.R.D. 615, 618 (C.D.Cal.1985). Although the issues may sometimes be " plain enough from the pleadings," " sometimes it may be necessary for the court to probe behind the pleadings before coming to rest on the certification question." General Tel. Co., 457 U.S. at 160, 102 S.Ct. at 2372. This is such a situation. In the circumstances of this case, I conclude that the court can and must conduct a preliminary inquiry into the likely availability of the fraud-on-the-market theory in order to determine whether common issues truly will predominate in this case.

         2.

          Alternatively, plaintiffs argue that common issues will predominate even if they do not have the benefit of the fraud-on-the-market presumption and therefore must prove individual reliance. At oral argument, plaintiffs' counsel starkly stated that no court had ever failed to certify a class because of the presence of individual reliance issues. This statement was not merely hyperbolic; it was plainly wrong.

         Analysis of this issue again must start with Basic v. Levinson, and the Supreme Court's square holding that " reliance is an element of a Rule 10b-5 cause of action." 485 U.S. at 243, 108 S.Ct. at 989. In Basic and later cases, the Court made it pellucid that proof of reliance is part of the plaintiffs' burden. See, e.g., Central Bank, 511 U.S. at __, 114 S.Ct. at 1449 (" A plaintiff must show reliance on the defendant's misstatement or omission to recover under 10b-5." ). The Basic Court expressly recognized the potential dependence of the class certification question on the availability of a rebuttable presumption of reliance: " Requiring proof of individualized reliance from each member of the proposed plaintiff class effectively would have prevented respondents from proceeding with the class action since individual issues would have overwhelmed the common ones." 485 U.S. at 242, 108 S.Ct. at 989. Opinions of the lower courts have recognized the dramatic effect that the fraud-on-the-market presumption has on the predominance question. If the presumption applies, a class action is much easier to maintain, because individual proof of reliance is not necessary. In the absence of such a presumption, individual issues are more likely to predominate. See, e.g., Seagate, 843 F.Supp. at 1356-57; In re LTV Sec. Lit., 88 F.R.D. 134, 141-42 (N.D.Tex.1980).

          Furthermore, contrary to plaintiffs' argument, district courts have refused to certify class actions for the very reason that individual proof of reliance would be necessary, thus creating an unmanageable lawsuit in which numerous " mini-trials" on the issue of reliance and damages would overwhelm the common issues. Zlotnick v. T.I.E. Communications, Inc., 123 F.R.D. 189 (E.D.Pa.1988), is instructive. In Zlotnick, the case was before the court pursuant to remand from the Third Circuit. The Third Circuit had held that plaintiff was not entitled to rely upon the fraud-on-the-market theory in proving his federal securities case. 836 F.2d 818 (3d Cir.1988). On remand, plaintiff moved to certify a class. Relying upon Basic, the trial court determined that, in the absence of presumed reliance, common issues would not predominate, because all class members would be required to submit proof of reliance:

Moreover, each class member's proof of reliance will be highly individualized-including each investor's knowledge of and experience in the market; an accounting of the information, advice, or rumors each investor received; and testimony as to each member's actual state of mind at the time he or she covered and why that differed from his or her state of mind at the time he or she sold short.

Based on the reasons stated above, I find that individualized questions of fact would predominate over common issues at trial and that a class action is not superior to other available methods for the fair and efficient adjudication of controversy.

123 F.R.D. at 194. Numerous other courts, both before and after Basic, have declined to certify classes where individual proof of reliance would be necessary. See, e.g., In re Synergen, Inc. Sec. Lit., 154 F.R.D. 265, 267 (D.Colo.1994) (declining to certify state-law fraud claims for class-action treatment because of necessity for separate trials on reliance); In re Rospatch Sec. Lit., 1991 WL 427890, at *11 (same); Wall v. Merrill Lynch, Pierce, Fenner & Smith, 1992 WL 245540 (N.D.Ill. Sept. 21, 1992) (individualized issues of reliance and causation predominated in RICO case); Angelastro v. Prudential-Bache Sec. Inc., 113 F.R.D. 579, 584-85 (D.N.J.1986) (in absence of presumption of reliance, 10b-5 claim not certified for class treatment); Beebe v. Pacific Realty Trust, 578 F.Supp. 1128, 1151 (D.Or.1984) (declining to certify for class treatment plaintiffs' claims for common-law fraud and negligent misrepresentation because individual reliance issues predominate); Westlake v. Abrams, 98 F.R.D. 1, 4 (N.D.Ga.1981) (reliance issues predominated in 1933 Act case).

         Plaintiffs cite several cases in which courts have certified class actions regardless of the need for individual proof of reliance. Most of the cases predate Basic. Some of them rely upon the proposition that lack of reliance is a defense and not part of the section 10b claim. See, e.g., Gorsey v. I.M. Simon & Co., 121 F.R.D. 135, 138-39 (D.Mass.1988); In re Pizza Time Theatre Sec. Lit., 112 F.R.D. 15, 22 (N.D.Cal.1986). Such decisions are clearly bad law after Basic. where the Court made it clear that proof of reliance is part of plaintiff's case-in-chief, in the absence of a presumption. 485 U.S. at 243, 108 S.Ct. at 989. The only post- Basic Court of Appeals case cited by plaintiffs for this proposition is Hanon v. Dataproducts Corp., 976 F.2d 497 (9th Cir.1992). The statement is pure dictum, far removed from the holding of the case. The Hanon court affirmed the trial court's denial of a motion for class certification, finding that the named plaintiff failed to meet the typicality requirement of Rule 23(a). In passing, the court said that the defense of nonreliance should not be a basis for denying class certification, citing pre- Basic cases for the proposition that lack of reliance is a defense. 976 F.2d at 509. This statement was extraneous to the holding of the case and, with all due respect, not very well thought out in light of Basic.

         Although some authority can be found to support plaintiff's arguments, the better-reasoned authority is to the contrary. Where numerous mini-trials are necessary to resolve individual questions of reliance and causation, the benefits of a class action disappear. If, therefore, a fraud-on-the-market theory is not reasonably available to plaintiffs and individual proof on these issues is necessary, common questions will not predominate.

         3.

          The final issue relating to predominance is whether plaintiffs will likely be entitled to invoke the presumption of reliance under a fraud-on-the-market theory. As stated above, the court is not required at this point to decide the merits of plaintiffs' fraud-on-the-market theory. The court should, however, make a preliminary determination concerning the likely strength of the theory, so that an intelligent decision can be made concerning the probable issues for resolution at trial. On the basis of the evidence now of record, I conclude that plaintiffs have virtually no chance of succeeding on this theory.

         Although the fraud-on-the-market theory had been applied by lower federal courts since about 1976, the Supreme Court did not recognize the theory until Basic, Inc. v. Levinson, 485 U.S. 224, 108 S.Ct. 978, 99 L.Ed.2d 194 (1988). By endorsing the fraud-on-the-market theory, the Supreme Court has allowed an economic hypothesis to stand in the place of proof. The Basic Court authorized a rebuttable presumption of reliance in cases involving securities traded on an open, developed, and efficient market. 485 U.S. at 247-48, 108 S.Ct. at 992. The Court drew upon economic studies espousing the " efficient capital market hypothesis," a theory of financial economics which says that information regarding a company's expected future value is quickly and accurately incorporated into the price at which the company's securities actively trade on an open, efficient, and well-developed market. 485 U.S. at 246 n. 24, 108 S.Ct. at 991 n. 24 (citing Daniel R. Fischel, Use of Modern Financial Theory in Securities Fraud Cases Involving Actively Traded Securities, 38 BUS.LAW 1, 4 n. 9 (1982); Roger J. Dennis, Materiality and the Efficient Capital Market Model: A Recipe for the Total Mix, 25 WM. & MARY L.REV. 373, 374-381 (1984)). Relying on these economic studies, the Basic Court reached its conclusion through a three-step analysis. First, the efficient capital market hypothesis allows the court to assume that any material representation made by the issuer would quickly and accurately be reflected in the market price of the security, so long as the market involved is " efficient." 485 U.S. at 244, 108 S.Ct. at 990. Next, an investor's reliance upon the integrity of the market price of such a security would be reasonable. Id. at 246-47, 108 S.Ct. at 991-92. Finally, because an investor who trades in a particular security can be presumed to have done so in reliance on the market price, if that price reflects some misrepresentation made by the issuer, the purchaser can be presumed to have relied on the misrepresentation itself, although indirectly. Id. at 247, 108 S.Ct. at 991.

         The linchpin of the fraud-on-the-market theory is the existence of an efficient market. In the absence of an efficient market, any presumption of reliance becomes unfounded. The Sixth Circuit has repeatedly underscored the necessity for a finding of efficiency as a prerequisite to invocation of the fraud-on-the-market theory. For example, in Freeman v. Laventhol & Horwath, 915 F.2d 193 (6th Cir.1990), the court flatly stated that the fraud-on-the-market theory " cannot be extended to securities traded on markets that are not efficient." 915 F.2d at 197. The court pointed out that an inefficient market " by definition, does not incorporate into its price all the available information about the value of a security." Id. at 198. The court ultimately held that a primary market for newly issued securities is not an efficient market, as a matter of law. Id.; accord, Ockerman v. May Zima & Co., 27 F.3d 1151, 1158-59 (6th Cir.1994).

         The Sixth Circuit has given the lower courts some guidance in determining whether a market is developed and efficient. In Freeman, the Sixth Circuit described a developed market as having a high level of trading activity, with trading information widely available. An efficient market is one in which the price of the stock rapidly reflects new information. An open market " is one in which anyone, or at least a large number of persons, can buy or sell." 915 F.2d at 198-99. The Freeman court turned to an earlier district court decision, Cammer v. Bloom, 711 F.Supp. 1264 (D.N.J.1989), for five factors " that would be useful in proving that a security was traded on an efficient market, justifying the application of the fraud-on-the-market theory" :

(1) a large weekly trading volume; (2) the existence of a significant number of reports by securities analysts; (3) the existence of market makers and arbitrageurs in the security; (4) the eligibility of the company to file an S-3 Registration Statement; and (5) a history of immediate movement of the stock price caused by unexpected corporate events or financial releases.

Freeman, 915 F.2d at 199. The facts of record strongly indicate that plaintiffs will not be able to show that the market for Embrace common stock was developed or efficient. The market for Embrace stock purchase warrants was not open, developed or efficient.

         a. Common Stock

         As stated in the proposed findings of fact, Embrace common stock was publicly traded for a period of less than two years on the NASDAQ SmallCap system. Even plaintiff's expert conceded that such stocks fall within the " bottom tier of NASDAQ securities in terms of capitalization, number of shareholders and activity." (Kangas Aff., ¶ 13). Consideration of the relevant economic factors discloses little evidence that the market for Embrace stock was efficient enough to warrant a presumption of reliance.

          Large Weekly Trading Volume.

          The weekly trading volume of Embrace common stock was minuscule to nonexistent for most of the class periods, except for a four-month flurry of activity just before the company collapsed. The mean weekly trading volume for a great portion of the class periods (May 21, 1992, through October 22, 1993) was only 39,627 shares. The median trading volume as a percentage of shares was only .08%, a fraction of the NASDAQ average. Somewhat misleadingly, Mr. Kangas reported to the court that the average weekly trading volume during the entire class period exceeded 200,000 shares. As defendants' Exhibit 2 graphically shows, however, this is an example of a fallacious statistic, in that the frenzied four-month period at the end of Embrace's trading life heavily skews the average. Viewed as a whole, the trading in the class period can only be characterized as extremely thin. Embrace averaged only seven trades per day, had thirty days in the class period when the stock did not trade at all, and experienced numerous days in which trading amounted to no more than 1,000 to 3,000 shares. ( See Proposed Findings, ¶ 9). Certainly, a period of only four months of substantial trading, followed by a crash, cannot be viewed as evidence of a developed or efficient market.

         A large weekly volume of stock trades suggests that there is " significant investor interest in the company." Cammer, 711 F.Supp. at 1286. Such interest, in turn, implies a likelihood that many investors are executing trades on the basis of newly available corporate information. Id. The trading history of Embrace stock shows just the opposite. This factor weighs heavily against the finding of efficiency.

          Existence of a Significant Number of Reports By Securities Analysts.

          The more securities analysts that follow a security, the greater the likelihood that all information disseminated by a corporation is being relied upon. See Cammer, 711 F.Supp. at 1285-87. Major corporations are closely followed by hundreds of securities analysts throughout the country. It is undisputed that no analysts followed Embrace stock.

         In an effort to fill this gap, plaintiffs point to sporadic news reports concerning Embrace in sources as disparate as The Wall Street Journal, the Buffalo News, and the Baton Rouge Press. Such random commentary by news reporters cannot substitute for serious attention to stock by professional securities analysts. No logical inference concerning widespread knowledge of a company's affairs follows from such sporadic reporting. There are few publicly traded companies that escape passing mention in the press, especially in their home towns. Such haphazard media attention is hardly indicative of an efficient market.

          The Existence of Market Makers and Arbitrageurs in the Security.

          For NASDAQ stocks, a market maker is merely a brokerage house that announces itself ready to buy or sell a specified number of shares at specified prices. At various times during the class periods, Embrace's common stock had seven to twenty-eight market makers. This information, without more, is virtually meaningless. As Dr. Cox testified, it is impossible to accord much significance to the number of market makers, until one knows the volume of shares that they committed to trade, the volume of shares they actually traded, and the prices at which they did so. The economic literature has criticized reliance upon the number of market makers as an indicator of efficiency. See Victor L. Bernard, Christine Botosan & Gregory D. Phillips, Challenges To the Efficient Market Hypothesis: Limits To the Applicability of Fraud-on-the-Market Theory, 73 NEB.L.REV. 781, 799, 805 (1994); Brad M. Barber, Paul A. Griffin & Baruch Lev, The Fraud-on-the-Market Theory and the Indicators of Common Stocks' Efficiency, 19 J.CORP.L. 285, 307 (1994). The Barber study found no empirical correlation between the number of market makers and the efficiency of the market. The Bernard article notes that, in contrast to analysts, market makers generally do not analyze and disseminate information about the stock that they make a market for and therefore do not contribute to the efficiency of the stock's price. Market makers behave this way because acting as a market maker commits one to purchasing or delivering only a limited number of shares at the posted quotes.

         Consequently, the mere number of market makers, without much more information, is not a very important indicia of market efficiency. This factor does not contribute to a finding of efficiency in the present case.

          Eligibility of the Company to File an S-3 Registration Statement.

          Although not dispositive, this fact is extremely important. Companies filing the S-3 form are required to disclose the least amount of information. Other companies file more detailed reports. Only corporations whose stocks are actively traded and widely followed are allowed to use the S-3 form by the SEC. See Harman v. LyphoMed, Inc., 122 F.R.D. 522, 525 (N.D.Ill.1988); see also, Barbara Black, Fraud on the Market: A Criticism of Dispensing With Reliance Requirements In Certain Open Market Transactions, 62 N.C.L.REV. 435, 468 (1984). The SEC allows companies to file the abbreviated S-3 form only on the premise that the stock is already traded on an open and efficient market, such that further disclosure is unnecessary. Black, Fraud on the Market, at 468. Commentators thus have suggested a bright line test, under which only companies eligible to file Form S-3 would be eligible for fraud-on-the-market treatment. E.g., Carol R. Goforth, The Efficient Capital Market Hypothesis-An Inadequate Justification for the Fraud-on-the-Market Presumption, 27 WAKE FOREST L.REV., 895, 933-36 (1992). Although courts have declined to adopt such a bright line test excluding companies not eligible to file the form S-3, they nevertheless consider this " an important factor" in fraud-on-the-market analysis. See, e.g., Cammer, 711 F.Supp. at 1284-85.

         It is undisputed that Embrace was never eligible to file form S-3. Dr. Cox testified that Embrace's capitalization fell far short of the minimum threshold for S-3 treatment. Furthermore, the stock was not actively traded for the requisite period.

         Eligibility to file the S-3 form is predicated upon the SEC's belief that the market for the company's stock already operates efficiently and that further disclosure is unnecessary, as the market price has already accounted for relevant information. See Cammer, 711 F.Supp. at 1284. For companies ineligible to file Form S-3, no such assumption can be made. Although ineligibility does not automatically disqualify a company from a finding of an efficient market, this factor weighs heavily against a finding of efficiency.

          A History of Immediate Movement of the Stock Price Caused by Unexpected Corporate Events or Financial Releases.

          If a plaintiff can empirically demonstrate that stock prices regularly rose or fell in prompt response to market information, this fact would be significant in establishing an efficient market. Cammer, 711 F.Supp. at 1287. In an effort to make such a showing, Mr. Kangas attempted to show instances of a fall in the price of Embrace stock in response to the dissemination of adverse information. Dr. Cox demonstrated, however, that within days thereafter, the price rose almost to its original level, without any dissemination of " good news" that could account for the increase. In other words, the price history of Embrace stock appears to have been random and volatile. ( See Def.Ex. 3). Plaintiffs have not shown any reliable relationship between changes in price and public announcement of information to which the price changes can reasonably be attributed.

          Other Factors.

          The economic literature suggests the relevance of other factors that tend to demonstrate market efficiency, including capitalization, bid-ask spread, interest of institutional investors, percentage of stock not held by insiders (called " float" ), and other economic variables. See Barber, et al., The Fraud-on-the-Market Theory, at 291-92; Robinson, Fraud-on-the-Market Theory and Thinly Traded Securities Under Rule 10b-5: How Does A Court Decide If a Stock Market is Efficient, 25 WAKE FOREST L.REV. 223 (1990). The economic literature posits that the most important of these factors is the size of the firm, measured in terms of market capitalization, coupled with trading volume. See Bernard, et al., 83 NEB.L.REV. at 792-93; Victor L. Bernard & Jacob K. Thomas, Evidence That Stock Prices Do Not Reflect the Implications of Current Earnings for Future Earnings, 13 J.ACCT.ECON. 305, 307 (1990). Large firm size and dollar trading volume tend to reflect the magnitude of economic incentive to eliminate mispricing. " Market capitalization may be relevant to market efficiency because market participants have greater incentives to invest resources in assessing the value of larger capitalization firms." (Cox Aff., ¶ 11). Conversely, pricing inefficiencies are much more likely to be found in small companies, especially when their stock is not widely held by institutional investors. Bernard, et al., 73 NEB.L.REV. at 792-93.

         Dr. Cox analyzed Embrace common stock on a number of these criteria, comparing Embrace to ninety-nine other publicly traded corporations of all sizes. His analysis shows that Embrace compared poorly. Its capitalization and volume were relatively small and holdings by institutional investors were low. The percentage of shares held by the public rather than insiders (" float" ) was low-another indicator of inefficiency. Cox concluded that these and other factors clearly showed an inefficient market.

         Plaintiffs, through Mr. Kangas, vigorously criticized Dr. Cox's method, charging him with making an unfair comparison between Embrace and much larger corporations, including Ford Motor Co. The criticism is unwarranted. In recognizing the efficient capital market hypothesis underlying the fraud-on-the-market theory, the Supreme Court relied upon economic studies performed on New York Stock Exchange and other major companies. See Goforth, 27 WAKE FOREST L.REV. at 933 (" Only the largest and most organized capital markets have been studied." ). Since Basic, courts have recognized this fact, remarking that the theory is well-suited to stocks " traded in national secondary markets such as the New York Stock Exchange." See, e.g., Freeman, 915 F.2d at 199. In light of this, some courts have drawn a bright line, refusing as a matter of law to accord fraud-on-the-market treatment to stocks traded over-the-counter, as such stocks generally do not exhibit hallmarks of efficiency. See, e.g., Epstein v. American Reserve Corp., No. 79C4767, 1988 WL 40500, at *5 (E.D.Ill. April 21, 1988); see also, Bernard, et al., 73 NEB.L.REV. at 805 (fraud-on-the-market theory should be limited to NYSE, AMEX, and largest NASDAQ companies). Other courts, while refusing to automatically rule out smaller companies, have emphasized the need to scrutinize the economic evidence closely for such companies. See, e.g., Cammer, 711 F.Supp. at 1280-87.           Consequently, it is critical to compare Embrace's stock to those large, heavily traded corporations for which the fraud-on-the-market theory has been approved by the Supreme Court. Even though Embrace stock was traded on the " lower tier" of the NASDAQ system, if the market for its shares closely resembled that of the larger companies in its economic attributes, it would be reasonable to treat Embrace as if it were a larger company in assessing the efficiency of the market for its stock. Conversely, to the extent that the market for Embrace stock diverges in these economic attributes from that of these model companies, the theoretical underpinnings of the fraud-on-the-market theory begin to crumble. This is precisely what Dr. Cox has demonstrated. Embrace common stock bore little relevant similarity to those stocks found to be efficient in the studies upon which the fraud-on-the-market theory is based.

         In an effort to support a finding of efficiency, plaintiffs' argue that " to the extent plaintiffs are required to plead an efficient market, they have done so by alleging that the common stock of Embrace was traded on a well-developed national exchange, NASDAQ, a highly efficient and automated market." Mr. Kangas, plaintiffs' expert, also based his conclusion of efficiency primarily on the fact that Embrace stock was traded on the NASDAQ SmallCap system, arguing that all stocks so traded are efficient. Plaintiffs' argument is completely unsupported in law. The issue is not whether NASDAQ is efficient. The market system upon which a particular stock trades provides some insight as to the likelihood that the market for that stock is efficient, but it is not dispositive. See Cammer, 711 F.Supp. at 1281. To the extent that a NASDAQ SmallCap listing says anything about the market for Embrace stock, it would indicate an inefficient rather than an efficient market. The inquiry, however, " remains on the market for that stock, and not the location where the stock trades." Harman, 122 F.R.D. at 525. No court has ever held that a finding of efficiency can be based on the mere fact that a stock is traded on NASDAQ, let alone on the " lower tier" of the NASDAQ system. Consequently, even the allegations in plaintiffs' complaint concerning market efficiency are too conclusory and fail to set forth facts tending to show an efficient market. See Alter v. DBLKM, Inc., 840 F.Supp. 799, 804-05 (D.Colo.1993) (court need not accept conclusory allegation of market efficiency); In re MDC Holdings, 754 F.Supp. at 805 (rejecting conclusory allegations of market efficiency on motion to certify class); Greenberg v. Boettcher Co., 755 F.Supp. 776, 781-82 (N.D.Ill.1991) (plaintiff must plead facts showing efficiency). By focusing primarily upon the attributes of the NASDAQ system, rather than upon the economic attributes of Embrace stock, plaintiffs' factual presentation at the hearing was likewise deficient.

         Plaintiffs place great reliance on Simpson v. Specialty Retail Concepts, 823 F.Supp. 353 (M.D.N.C.1993), in which the court denied a motion for summary judgment on the question of market efficiency. The common stock at issue in Specialty Retail arguably displayed even fewer indicia of efficiency than Embrace common stock. Specialty Retail must be rejected as an aberration. The common stock examined by the court in that case clearly did not trade in a developed and efficient market. As a district court in the Fourth Circuit, the Specialty Retail court was not bound by the pronouncements of the Sixth Circuit in Freeman, which make it clear that the fraud-on-the-market presumption is simply inapplicable to inefficient markets. The best the Specialty Retail court could say was that the stock showed " some degree of market efficiency." 823 F.Supp. at 355. As the Freeman court made clear, reliance will be presumed " only when it is logical to do so." " The fraud-on-the-market theory cannot be applied logically to securities that are not traded in efficient markets." 915 F.2d at 198. " Some degree" of market efficiency is simply not enough to meet the Sixth Circuit standard.

          In summary, the economic factors identified by the courts or suggested by the literature point strongly toward a finding that the market for Embrace common stock was not efficient enough to warrant invocation of the fraud-on-the-market theory. Consequently, I conclude that no presumption of reliance will be available to plaintiffs in pursuing their 1934 Act claims set forth in counts 1 and 2, arising from their purchase of Embrace common stock. Likewise, plaintiffs will not have the benefit of any presumption in proving their state-law claims in counts 4 and 5. Count 4 alleges common law fraud under Michigan law. Reliance is a necessary element of plaintiffs' claim. See Kassab v. Michigan Prop. Ins. Co., 441 Mich. 433, 442, 491 N.W.2d 545, 548 (1992). The Michigan Supreme Court has never adopted a fraud-on-the-market theory or other presumption that would relieve a plaintiff of the burden of proving individual reliance in this case. This being so, class certification is inappropriate on the common-law fraud claim for the same reason that it is unworkable on the 10b-5 claim. See In re Synergen Sec. Lit., 154 F.R.D. at 267; In re Rospatch Sec. Lit., 1991 WL 427890, at *11; In re One Bancorp Sec. Lit., 136 F.R.D. 526, 532-33 (D.Me.1991).

         The fifth count, for civil conspiracy under Michigan law, is derivative of count 4. Under Michigan law, a conspiracy standing alone without the commission of underlying wrongdoing is not actionable. See Fenestra v. Gulf Am. Land Corp., 377 Mich. 565, 593-94, 141 N.W.2d 36, 48-49 (1966). Consequently, the gist of the action is not the pursuance of the conspiracy, but the wrongful act itself. See Roche v. Blair, 305 Mich. 608, 613-14, 9 N.W.2d 861, 863 (1943). An allegation of conspiracy under Michigan law is merely a device by which to hold liable persons who cooperated with the principal actor in committing a civil wrong. Consequently, proof of the conspiracy count would require proof of all elements of the underlying fraud claim. Proof of individual reliance would therefore be necessary in connection with count 5 as well.

         Although I do not recommend that the court enter a definitive judgment on the issue at this stage in the litigation, I do conclude that it is unlikely plaintiffs will be successful in establishing that the market for the sale of Embrace common stock was efficient. The strong probability is that individual questions of reliance will overwhelm the common issues in counts 1, 2, 4 and 5.

         b. Stock Purchase Warrants

          The contention that the market for Embrace stock purchase warrants was " open, developed and efficient" is not even arguable. According to the Sixth Circuit, an open market " is one in which anyone, or at least a large number of persons, can buy or sell." Freeman, 915 F.2d at 198. It is uncontested that Embrace warrants were only sold by Embrace itself. The offer was made only to specific, targeted purchasers who met certain investment criteria. Resale by the purchasers was restricted. Consequently, the market was not open on either the seller's or the purchaser's side.

         Furthermore, the price was set completely by the offeror, and not by the market. In Freeman, the Sixth Circuit held, as a matter of law, that the market for the sale of newly issued municipal bonds was not efficient or developed. 915 F.2d at 199. Consequently, the court determined that the fraud-on-the-market theory was not available with regard to such financial instruments. The same factors cited by the Freeman court to disqualify the initial public offering of municipal bonds serve to disqualify Embrace warrants from fraud-on-the-market treatment. As in Freeman, the Embrace warrants were not sold on any exchange, the price was unilaterally set by the issuer, and the trades were made in private transactions conducted by the company or its agents, not " actively in an impersonal market." " This decreases the probability that the price could be relied on reasonably as an accurate reflection of their true value," as the price is not set by the market. Freeman, 915 F.2d at 199; accord, Ockerman, 27 F.3d at 1158-59; Eckstein v. Balcor Film Inv., 8 F.3d 1121, 1130 (7th Cir.1993), cert. denied, 510 U.S. 1073, 114 S.Ct. 883, 127 L.Ed.2d 78 (1994) (initial offering sold at fixed price ineligible for fraud-on-the-market presumption); Platsis v. E.F. Hutton & Co., 642 F.Supp. 1277, 1301 (W.D.Mich.1986) (since limited partnership interests were sold directly by issuer and not on open market, fraud-on-the-market theory inapplicable), aff'd, 829 F.2d 13 (6th Cir.1987), cert. denied, 485 U.S. 962, 108 S.Ct. 1227, 99 L.Ed.2d 427 (1988).           The Freeman court decided as a matter of law that fraud-on-the-market does not apply to initial public offerings. During his examination in court, Mr. Kangas (plaintiff's own expert) conceded that the Embrace stock purchase warrants were like initial public offerings in these relevant respects. Such instruments are simply ineligible for a presumption of reliance, which presumes active, high volume trading on secondary established markets and not a limited offering to a handful of people at a set price with prohibitions against resale.

         In an effort to avoid this result, plaintiffs argue that the price of the stock purchase warrants was dependent on the price of the underlying common stock, such that any fraud-on-the-market for the common stock would be somehow transferred to the market for the warrants. Plaintiffs' theory of " derivative" fraud-on-the-market is unsupported. Neither the Supreme Court nor the Sixth Circuit has ever extended the fraud-on-the-market theory to a derivative type of security. Nor do plaintiffs cite any economic studies showing an empirical relationship that might support such an extension. Plaintiffs cite only a single district court case, Deutschman v. Beneficial Corp., 132 F.R.D. 359, 377 (D.Del.1990), in which the court allowed purchasers of call options to invoke the fraud-on-the-market theory. The court rationalized that " the premium, or price of the option contract, is directly responsive to the market price of the underlying security and to information affecting that price." 132 F.R.D. at 370-71. The federal appellate courts have never endorsed such a sweeping generalization. The Deutschman court cited no empirical studies to support its extension of the theory and was forced to distinguish weakly a case from its own reviewing court, Zlotnick v. T.I.E. Communications, 836 F.2d 818 (3d Cir.1988), in which the Third Circuit denied fraud-on-the-market treatment to short sales of securities. The Deutschman case is not persuasive. Furthermore, the price of the warrants was set by Embrace and was in no sense " responsive" to the stock price. In this crucial respect, the warrants are distinguishable from the call options examined in Deutschman. The price of the Deutschman call options floated with the market; the price of Embrace warrants was fixed.

         In summary, plaintiffs have failed to show that common questions predominate in counts 1, 2, 4, and 5. They will likely be required to show individual reliance and causation as an element of each count. The necessity of numerous " mini-trials" involving the facts peculiar to every absent plaintiff will overwhelm the common issues. For count 3, in which reliance is not an element, this problem does not exist and the common issues will predominate.

          F. Superiority

          Finally, a representative plaintiff seeking to invoke Rule 23(b)(3) must show that a class action is " superior to other available methods for the fair and efficient adjudication of the controversy." FED.R.CIV.P. 23(b)(3). The drafters of the rule point the court to several factors to be considered regarding this issue, including the interest of class members in controlling separate actions, the extent and nature of other lawsuits already commenced, the desirability of concentrating the litigation in a particular forum, and difficulties likely to be encountered in the management of the class action. FED.R.CIV.P. 23(b)(3)(A)-(D). These factors naturally are interdependent and overlapping both among themselves and with the prerequisites to a class action prescribed in Rule 23(a). See, 7A WRIGHT, MILLER & KANE, FEDERAL PRACTICE AND PROCEDURE § 1780 at 562. These issues, and especially the manageability inquiry, encompass the whole range of practical problems that may render the class action format inappropriate in a particular case. Id.; see Eisen v. Carlisle & Jacquelin, 417 U.S. at 164, 94 S.Ct. at 2146.

          In arguing for a finding of superiority, plaintiffs cite numerous decisions indicating that securities cases are particularly well-suited for class action treatment. See, e.g., Mader v. Armel, 402 F.2d 158, 161 (6th Cir.1968), cert. denied, 394 U.S. 930, 89 S.Ct. 1188, 22 L.Ed.2d 459 (1969). With this innocuous proposition, there can be little dispute. The fact remains, however, that courts do not certify classes in every securities case, as a class action is not invariably a superior method of proceeding. This is one such instance. Where individual hearings are required on questions of reliance, causation, and damages, the case " is hardly the picture of judicial economy envisioned by Rule 23." Kurczi v. Eli Lilly & Co., 160 F.R.D. 667, 681 (N.D.Ohio 1995).

         Denial of class certification in this case does not mean that the parties and the judicial system must tolerate numerous, duplicative lawsuits, all designed to adjudicate the liability of Price Waterhouse and the other defendants for the allegedly fraudulent statements made concerning Embrace. Those issues should be tried but once. As Price Waterhouse correctly points out, the doctrine of collateral estoppel is sufficient to prevent relitigation of the common liability issues in case after case. The named parties in the present case certainly have the means and the incentive to pursue their individual claims. If groups of other investors are so inclined, they can intervene here or begin their own lawsuits. Discovery and other pretrial proceedings can easily be coordinated under the provisions of rules governing multi-district litigation. See 28 U.S. C. § 1407. The common liability issues may also be consolidated for trial, so that only one jury determination is necessary on the common issues. If they are successful on the common liability issues, the plaintiffs could then return to their home courts for individual trials on the questions of reliance, causation, and damages.

         Consequently, I conclude that the class action device is not a superior method of adjudicating this case, nor is it the only method of preventing duplicative lawsuits and inconsistent verdicts.

         III. Proposed Ultimate Findings and Recommended Disposition

         For the foregoing reasons, I recommend that court adopt the following findings with regard to the relevant issues under Rule 23:

         A. The appropriately defined putative classes in this case are as follows:

-Class A, composed of purchasers of Embrace common stock between May 21, 1992, and February 15, 1994, asserting claims against Embrace and the individual defendants;

-Class A/PW, composed of purchasers of Embrace common stock between April 15, 1993, and February 15, 1994, asserting claims against those defendants and Price Waterhouse;

-Class B, composed of purchasers of Embrace series D common stock warrants during 1993 asserting claims against Embrace and the individual defendants;

-Class B/PW, composed of purchasers of warrants after April 15, 1993, asserting claims against those defendants and Price Waterhouse.

         Excluded from all classes are defendants, the members of their immediate families, any entity in which any of the defendants has a controlling interest, and the legal representatives, heirs, successors or assigns of any of the defendants.

         B. Plaintiffs have not borne their burden of showing that any putative class is so numerous that joinder of all members is impractical.

         C. Common liability questions of law and fact exist with regard to all counts.

         D. With regard to classes A and A/PW, the claims of all representative plaintiffs are typical of the claims of the class, except for named plaintiff Ginsburg, whose claims are not typical. The claims of plaintiff Babcock are typical of those of the class with respect to Class B and B/PW.

         E. Plaintiffs have failed to demonstrate that they are adequate representatives for class A or A/PW, because of the possible existence of unresolved conflicts of interest that plaintiffs have failed to address. Plaintiff Babcock has failed to demonstrate that he is an adequate representative of Class B or B/PW, as he has not yet clearly pleaded the existence of a viable claim against Price Waterhouse, the principal defendant.

          F. With regard to counts 1, 2, 4, and 5, common questions do not predominate, because individual proof of reliance, causation, and damages will be necessary. It is unlikely that plaintiffs will have the benefit of a rebuttable presumption of reliance under the fraud-on-the-market theory for counts 1 and 2, and no rebuttable presumption is even theoretically available on state-law counts 4 and 5. Common questions predominate on count 3, as proof of reliance is not part of plaintiffs' case.

         G. A class action is not a superior method of adjudicating this controversy.

         I further recommend that plaintiffs' motion for class certification be denied.

         Dated: September 28, 1995

          NOTICE TO PARTIES

         Any objections to this Report and Recommendation must be filed and served within ten days of service of this notice on you. 28 U.S. C. § 636(b)(1)(C); Fed.R.Civ.P. 72(b). All objections and responses to objections are governed by W.D.Mich.L.R. 13(b). Failure to file timely objections may constitute a waiver of any further right of appeal. United States v. Walters, 638 F.2d 947 (6th Cir.1981); see Thomas v. Arn, 474 U.S. 140, 106 S.Ct. 466, 88 L.Ed.2d 435 (1985).