This appeal arises out of an alleged securities fraud under section 10(b) of the Securities Exchange Act of 1934, 15 U.S.C. § 78j(b), and Rule 10b-5 of the Securities Exchange Commission, 17 C.F.R. § 240.10b-5. The district court granted summary judgment on the statute of limitations and under the “bespeaks caution” doctrine. This appeal followed.1 We reverse the grant of summary judgment and remand for further proceedings.
Background
On October 31,1984, the defendants (hereinafter “Winthrop”) began soliciting investment in a limited partnership to purchase and manage an office building located at 353 Sacramento Street in San Francisco, California. Winthrop solicited accredited investors who were interested in a long-term investment. Winthrop touted its experience in managing real estate projects and said the partnership would provide investors with favorable tax benefits, substantial cash proceeds beginning after 1990, and substantial property appreciation to be realized upon the building’s ultimate sale. In the short term, Winthrop anticipated operating losses (which would produce the tax advantages) and provided for a $15 million operating reserve to cover the losses. In the event that the $15 million reserve fund was not adequate, Winthrop pledged to make a $2 million loan to the partnership, to seek refinancing of the building, and to solicit additional investment from other limited partners if necessary.
The prospectus and a public accountant’s report provided numerous general warnings that the forecast benefits were not guaranteed. More specifically, the prospectus cautioned:
[Winthrop]. ha[s] been advised that the rent-up of the building has been adversely affected by a variety of factors including the fact that the building had a change in ownership prior to completion, the color of the exterior of the building, the floor size of the building which is not attractive to large users of space, and a general market slowdown in the San Francisco real estate market. Consequently, the Project was 20% leased by the end of 1983. During 1984 through September 30, 1984 an additional 33.3% of the Project was leased, which rate is consistent with the General Partners’ assumption in the Financial Forecast that the Project will be fully leased by September 30, 1985. The terms of tenant leases that the Operating Partnership will be able to obtain in the future will depend on a number of factors, including competitive conditions in the San Francisco commercial real estate market and the attractiveness of the Project to prospective tenants.
Appellees’ Supp. Excerpt of Record (“ASER”) at 35. The prospectus also encouraged investors to request information from Winthrop, but told them Winthrop “believe[d] that historical financial statements would not necessarily be indicative of future operating results of the Project and, accord*880ingly, [Winthrop did] not include any such statements in the [prospectus].” Id. at 45.
By the close of the offering period in March 1985, more than $28 million had been invested in the partnership. From the outset, the partnership’s earnings were less than expected.
In letters in May and September 1985, Winthrop told investors tax losses were higher than forecast due in part to “lower-than-expeeted rental income,” a “longer than anticipated lease-up period,” and delays in selling partnership units. Winthrop also said the San Francisco office space market had “soften[ed]” and “slowed down during the first quarter of 1985,” but seventy-three percent of the space was leased due to Winthrop’s “marketing efforts.” Based on its “strong momentum” in leasing, Winthrop projected ninety-six percent leasing by the end of 1985 and expressed “confidence of its ability to meet the new demands posed by the San Francisco office market.” See Appellants’ Excerpt of Record (“AER”) at 494, 498-99.
On March 20, 1986, Winthrop reported that tax losses were again higher than expected, rental rates were thirty percent “less than the original forecast,” and “higher-than-forecasted tenant improvement costs and leasing commissions have been incurred.” As a result, Winthrop said, the partnership was suffering “an extended period of negative cash flow” and “approximately $11,300,-000 of the $15,000,000 reserve had been expended” and would likely be exhausted by the end of 1986. Along with this news of poor financial performance, however, Winthrop also told investors of two positive.results. First, there was a “surge in leasing activity” due to Winthrop’s “successful” marketing program, resulting in leases for ninety-six percent of the building’s rentable space. Second, Winthrop had refinanced the building’s original mortgage and expected “positive cash flow” by August 1986. Id. at 501-02.
On September 11, 1986, Winthrop told investors “we are extremely pleased to report that property operations are now stabilized and that the Partnership is well positioned for future property appreciation.” Winthrop credited this good financial news to the refinancing of the building and the fact that ninety-seven percent of the building was now leased. Winthrop blamed the partnership’s past troubles on a city plan to limit new downtown office space, the anticipation of which led to the acceleration of construction projects and the weak real estate market in 1985 and 1986. However, Winthrop said, “we have come through this difficult period and the Partnership is now positioned to benefit from the limited construction required by the ordinance.” Id. at 507-08. Similarly, on March 17, 1987, and again on June 5, 1987, Winthrop told investors the partnership was progressing “smoothly” and had high occupancy rates. Id. at 511-12.
On December 28, 1987, however, Winthrop told investors the partnership had suffered new set-backs. Specifically, Winthrop said, the occupancy rate had fallen to ninety-four percent due to the bankruptcy of four tenants and the project needed to be refinanced again. In addition, Winthrop explicitly told investors the partnership had failed to achieve the projections contained in the original prospectus because the market “unexpectedly entered a period of weakness” beginning in 1985. Id. at 513.2 A memorandum circulated to Winthrop’s officers in 1988 said the December 1987 letter “gives investors bad news for [the] first time.” Id. at 520.
On June 21,1990, the partnership declared bankruptcy. Plaintiffs filed suit on September 10, 1990, alleging that Winthrop made unreasonable projections about the partnership’s expected performance based on assumptions that were contrary to fact as of October 81,1984, in violation of federal securities law. The plaintiffs also brought pendent state law claims.
Statute of Limitations
Applying the pre-Lampf statute of limitations, this case is governed by Califor*881nia’s three-year limitations period for fraud. See Mosesian v. Peat, Marwick, Mitchell & Co., 727 F.2d 873, 876 (9th Cir.), cert. denied, 469 U.S. 932, 105 S.Ct. 329, 83 L.Ed.2d 265 (1984). The limitations period commences when the plaintiff “has actual or inquiry notice that a fraudulent misrepresentation has been made.” Volk v. D.A Davidson & Co., 816 F.2d 1406, 1412 (9th Cir.1987). “[I]f a prudent person would have become suspicious from the knowledge obtained through the initial prudent inquiry and would have investigated further, a plaintiff will be deemed to have knowledge of facts which would have been disclosed in a more extensive investigation.” Briskin v. Ernst & Ernst, 589 F.2d 1363, 1367 (9th Cir.1978). Summary judgment is appropriate only ‘“when uncontroverted evidence irrefutably demonstrates plaintiff discovered or should have discovered the fraudulent conduct.’” Davis v. Birr, Wilson & Co., 839 F.2d 1369, 1370 (9th Cir.1988) (per curiam) (quoting Kramas v. Security Gas & Oil Inc., 672 F.2d 766, 770 (9th Cir.), cert. denied, 459 U.S. 1035, 103 S.Ct. 444, 74 L.Ed.2d 600 (1982)); see also S.E.C. v. Seaboard Corp., 677 F.2d 1301, 1310 (9th Cir.1982) (moving party has “extremely difficult burden to show that there exists no issue of material fact regarding notice”).
It is undisputed the plaintiffs had no actual knowledge of the alleged fraud prior to September 10, 1987, the date three years prior to the plaintiffs’ filing of their suit. Nonetheless, the district court held the plaintiffs had sufficient notice of facts to require them to investigate the possibility of fraud prior to that date, but they failed to do so, commencing the limitations period. Plaintiffs assert the district court erred because, at the earliest, the first indication of fraud they received was the December 1987 letter to investors which Winthrop characterized as “giv[ing] investors bad news for [the] first time.” '
The plaintiffs contend that at least until the December 1987 letter, they had notice only of the partnership’s poor financial performance resulting from general market conditions. It is well settled that poor financial performance, standing alone, does not necessarily suggest securities fraud at the time of the sale, but could also be explained by poor management, general market conditions, or other events unrelated to fraud, creating a jury question on inquiry notice. See Mosesian, 727 F.2d at 878; Seaboard Corp., 677 F.2d at 1310; Briskin, 589 F.2d at 1368.3 In Briskin, for example, this court held that a missed dividend, an eighty-five percent decline in stock value, and documents showing a “shortage of capital and increasing debt” were not sufficient as a matter of law to place the plaintiffs on inquiry notice of fraud as to misrepresentations of the defendant’s financial condition at the time of a merger. Id. The worst financial news the investors received in this case was that leasing was taking “longer than anticipated,” rental rates were thirty percent below projections, tenant improvement costs and leasing commissions were “higher-thanforecasted,” and the reserve fund was seventy percent depleted and could be exhausted in 1986. This financial news is not obviously worse than the eighty-five percent decline in stock value and the “shortage of capital” at issue in Briskin, especially because Winthrop, in the prospectus, had anticipated losses and told investors about its contingency plans in the event that the reserve fund was exhausted. Thus, the extent of the bad news in this case provides no basis for distinguishing Briskin and its progeny, and thus we find the ordinary rule applicable.4
*882In Davis, on which the district court relied, the plaintiff, a highly sophisticated investor who followed his investments closely and recommended investment strategies to the defendant, “was aware that [his] investments resulted in losses.” 839 F.2d at 1370. Under those circumstances, the evidence was “irrefutabl[e]” that the investor should have known about the fraud, and summary judgment was thus proper. Id. By contrast, the investors in this case were limited partners who could not direct the activities of the partnership and had no reason to maintain regular contact with Winthrop. The investors also knew they had made a long-term investment which was not readily transferable and which would not produce significant cash proceeds until after 1990, nor other non-tax benefits until the ultimate sale of the building. Moreover, Winthrop repeatedly touted its experience and ability to monitor the daily operations of the partnership.5 In light of these facts, we think a reasonable jury could find the investors acted reasonably in not investigating the possibility of fraud until after 1987.
Winthrop argues, however, that it informed the investors of specific financial information “in direct conflict with what plaintiffs complain they were led to believe” about the rosy state of San Francisco’s rental market. See Cook v. Avien, Inc., 573 F.2d 685, 697 (1st Cir.1978). It is true Winthrop told investors in May 1985 that the office real estate market had “slowed down during the first quarter of 1985,” which was contemporaneous with the close of the offering period. Winthrop maintains that this suggests reasonable investors would have made inquiries about when the market decline began and what Winthrop knew about it. This argument may be persuasive to a jury, but we cannot say as a matter of law that a generalized statement a market has “slowed down” during the offering period requires investors to begin an immediate investigation into the possibility of fraud on Winthrop’s part.6
Winthrop also argues the investors were told “right after they signed their promissory notes that the very projections about which they waited five and one-half years to complain were way off target.” Brief for Appellees at 11. We agree that the specificity of the information given is a relevant factor in determining when a reasonable investor should begin to investigate the possibility of fraud in an investment. See, e.g., DeBruyne v. Equitable Life Assur. Soc. of the U.S., 920 F.2d 457, 466-67 (7th Cir.1990) (finding inquiry notice as a matter of law where defendants repeatedly disclosed specific information which plaintiffs alleged was concealed). In this case, Winthrop told the investors that the partnership was experiencing “longer than anticipated lease-up period,” “higher-than-forecasted tenant improvement costs and leasing commissions,” and was achieving rental rates thirty percent below original projections. Not until after September 1987, however, did Winthrop tell investors the original projections were sufficiently “off target” that the investment was likely to fail. In other words, investors could reasonably believe that Winthrop, as it claimed in the prospectus, could refinance the project, lend money to the project, or solicit additional investors in order to achieve success in the long term despite shortcomings in the partnership’s financial performance in 1985 and 1986.
It is also significant that Winthrop always east the poor financial performance of the partnership in a positive light. We agree with the plaintiffs that a jury could find they reasonably relied on such optimistic statements.7 Winthrop told investors of a “surge *883in leasing activity” due to Winthrop’s successful” marketing program, and of a refinancing that would lead to “positive cash flow.” Most importantly, following the news of a depleted reserve fund and the “longer than anticipated” lease-up of the building, Winthrop gave investors a rosy report in September 1986, stating that , the budding was ninety-seven percent leased, property operations had “stabilized,” and the partnership had “come through this difficult period” and was “well positioned for future property appreciation.” This optimistic statement was consistent with Winthrop’s contingency plans in the prospectus, as well -as one of the central goals of the partnership, namely, long-term appreciation. A reasonable jury could find that any concerns which investors had prior to September 1986 were alleviated by this letter to investors. Thus, we hold the plaintiffs’ claims are not barred as a matter of law by the statute of limitations.
The “Bespeaks Caution” Doctrine.
In granting the defendant’s motion for summary judgment the district court ruled, in the alternative, that plaintiffs case on the merits must -be rejected as a matter of law under the bespeaks caution doctrine. This court has recognized the bespeaks caution doctrine as “a mechanism by which a court can rule as a matter of law ... that defendants’ forward-looking representations contained enough cautionary language or risk disclosure to protect [them] against claims of securities fraud.” In re Worlds of Wonder Sec. Litig., 35 F.3d 1407, 1413 (9th Cir.1994) (quotation omitted) (emphasis added), cert. denied, — U.S.-&-, 116 S.Ct. 185 & 277, 133 L.Ed.2d 123 & 197 (1995). As this court explained in Fecht v. Price Co., 70 F.3d 1078 (9th Cir.1995), cert. denied, — U.S.-, 116 S.Ct. 1422, 134 L.Ed.2d 547 (1996), granting summary disposition under tile bespeaks caution label represents a conclusion that, as a matter of law, a securities prospectus as a whole is not misleading due to the risks disclosed and the nature and extent of the other cautionary language employed. Id. at 1082.
The plaintiffs argue the bespeaks caution doctrine is inapplicable to their claims of misrepresentations about “historical facts.”They contend they have produced expert testimony showing a number of misrepresentations by Winthrop that constitute securities fraud. We focus on three central allegations, which are either undisputed, or if disputed, would necessarily defeat a motion for summary judgment. First, the prospectus said 33.3 percent of the building had been leased during the first nine months of 1984, when in fact only 12.8 percent of the building had been leased. Second, the prospectus said 1984 tenant improvement costs, or the cost of redesigning and decorating the interior of the building to suit new tenants, would average $21 per square foot, when such costs had averaged roughly $26 per square foot through October 31, 1984. Third, the prospectus said the average net effective rental rate for leases in the building would be $31 per square foot in 1984, when the net effective rent for leases executed as of October 31, 1984 was only $25.70 per square foot. Overall, the plaintiffs allege that these misrepresentations about the building’s track record as of October 31, 1984 led to projections for future rental income and profitability which had no reasonable basis in fact.8
Although we acknowledge the bespeaks caution rule is not applicable to misrepresentations of historical facts, we cannot say the district court clearly erred in finding the doctrine a relevant concern. The historical facts alleged to be false formed the basis of Winthrop’s projections of profitability, and thus the bespeaks caution doctrine is rele*884vant in determining whether the prospectus as a whole was misleading as a matter of law.
Nevertheless, we conclude that the district court erred in holding as a matter of law that the disclosures in Winthrop’s prospectus were sufficient to make the prospectus as a whole not misleading. First, most of the risks disclosed in the prospectas were so generalized in nature that a reasonable jury could nonetheless find the prospectus misleading.9 Second, several disclosures affirmatively vouched for the accuracy of the historical data underlying the projections, which plaintiffs allege to be untrue. For example, in cautioning investors about the possibility that occupancy projections would not be met, the prospectus stated that “[t]here is no assurance _ that tenants executing leases in the future ... will be willing to commit to rents equal to or in excess of rents agreed to by lessees that already have executed leases.” ASER at 18 (emphasis added). Elsewhere, the prospectus said “assumptions as to future events ... are subject to interpretation and change.” Id. at 105 (emphasis added). Third, other disclosures were coupled with allegedly unreasonable projections that would have softened the impact of Winthrop’s disclosures. Where Winthrop disclosed the real estate market downturn in 1983 and 1984, it also said that notwithstanding the “market slowdown” it would fully lease the building by September 1985 (based on the allegedly false representation that 38.3 percent of the building’s space was leased during the first nine months of 1984). Id. at 35. We have found no disclosure which clearly and specifically cautioned investors that Winthrop’s projections could be in error due to uncertainty in the building’s historical track record, and in fact Winthrop expressly told investors such historical information was not relevant to their investment decision. Thus, we hold that a jury could reasonably find the document misleading despite the risks disclosed and the cautionary language employed.
Other Defenses
Notwithstanding the bespeaks caution doctrine, Winthrop argues we should affirm the grant of summary judgment on other grounds because the plaintiffs have failed to produce sufficient evidence of scienter or loss causation. A successful securities fraud claim under Rule 10b-5 requires (1) a misstatement or omission (2) of material fact (3) made with scienter (4) on which the plaintiff justifiably relies (5) that proximately causes the plaintiffs investment loss. See McGonigle v. Combs, 968 F.2d 810, 817 (9th Cir.), cert. dismissed, 506 U.S. 948, 113 S.Ct. 399, 121 L.Ed.2d 325 (1992). The district court noted a lack of evidence of scienter, but disavowed its reliance on this issue as a basis for summary judgment. The court also declined to reach the issue of loss causation.
In order to meet the scienter requirement, plaintiffs must show either knowing or reckless conduct on the part of defendants. See Hanon v. Dataproducts Corp., 976 F.2d 497, 507 (9th Cir.1992); see also Ernst & Ernst v. Hochfelder, 425 U.S. 185, 214, 96 S.Ct. 1375, 1391, 47 L.Ed.2d 668 (1976) (negligent conduct is not actionable under Rule 10b-5). “[C]ircumstantial evidence can be more than sufficient” to prove scienter, however, due to the “difficulty of proving the defendant’s state of mind.” Herman & MacLean v. Huddleston, 459 U.S. 375, 390 n. 30, 103 S.Ct. 683, 692 n. 30, 74 L.Ed.2d 548 (1983).
Winthrop’s basic argument is that it would not have disclosed so many risks to investors or “repeatedly told them that its projections and market analysis were incorrect” if Winthrop had intended to defraud investors. Brief for Appellees at 26. We do not think this issue is resolved so easily. It is plausible that the perpetrators of a sophisticated fraud would make sufficient disclosure to gain the trust of investors while concealing more adverse information from them. And, as discussed above, Winthrop did not tell investors its projections were “incorrect” but rather that the partnership’s earnings were *885less than the forecast due to general market conditions.
Winthrop also argues that the evidence on which the plaintiffs rely to show scienter is inadmissible. The plaintiffs offer evidence that Winthrop knew by October 31, 1984 that its prospectus contained misrepresentations, including an October 31, 1984 memorandum from an outside lawyer to Winthrop’s lawyer which said it would be “difficult to justify assumptions for 1984 which are contrary to fact.” AER at 436. Plaintiffs also rely on a July 1984 report by the building’s former owners which showed higher tenant improvement costs, higher leasing commissions, and a slower lease-up period than forecast in the prospectus. Id. at 413. The district court did not pass on the admissibility of this evidence, and we decline to do so on appeal, except to note that the evidence is not clearly inadmissible. First, the October 31, 1984 memorandum appears to be admissible to show notice to Winthrop of assumptions that are “contrary to fact.”10 Second, the July 1984 report appears to be admissible because there is evidence that the report, or at least the, documents from which the report was created, were available to Winthrop prior to October 31, 1984.11
The plaintiffs also rely on evidence that was available- to Winthrop after the offering was made, but before the offering period closed, to show scienter. Winthrop had a continuing duty to disclose material information that it received prior tó the close of the offering period. See Shawmut Bank, N.A v. Kress Assocs., 33 F.3d 1477, 1488 (9th Cir.1994); Greenfield v. Heublein, Inc., 742 F.2d 751, 758 (3d Cir.1984), cert. denied, 469 U.S. 1215, 105 S.Ct. 1189, 84 L.Ed.2d 336 *886(1985); cf. Backman v. Polaroid Corp., 910 F.2d 10, 16-17 (1st Cir.1990) (en banc). The plaintiffs’ evidence consisted of their expert’s analysis of leases Winthrop entered into during the offering period, a marketing proposal from a real estate broker in November 1984, and an analysis by one of Winthrop’s property managers, all of which show that achievable rental rates were below those assumed or projected by Winthrop in October 1984. The analysis by the plaintiffs’ expert showed actual rates obtained were as much as forty percent less than assumed for 1984. AER at 575. None of this information was .provided to the investors during the offering period. We think a jury could reasonably find this information was material and should have been disclosed to investors at a time when investors still had the opportunity to withdraw their investment in the partnership.12
We also find the plaintiffs have offered sufficient evidence to demonstrate causation. The requirement of loss causation means the plaintiffs must show not only that Winthrop’s fraud induced them to make the investment, which is not an issue on appeal, but also that disclosure of the truth would have “reduced the proper valuation of their investment.” McGonigle, 968 F.2d at 820. Winthrop relies primarily on Bastian v. Petren Resources Corp., 892 F.2d 680 (7th Cir.), cert. denied, 496 U.S. 906, 110 S.Ct. 2590, 110 L.Ed.2d 270 (1990), in which the Seventh Circuit found that the plaintiffs failed to produce any evidence that the defendant’s misrepresentations about its personal integrity and experience had caused the loss of the plaintiffs’ investment in oil and gas limited partnerships. The court reasoned the plaintiffs “wanted to invest in oil and gas limited partnerships” managed by competent and honest people, but even if they had done so, they would have lost money due to an industry-wide recession. Id. at 684. In this case, by contrast, the plaintiffs object to Winthrop’s representations of profitability, which shows an intent to invest in a profitable sector of the economy, whether it was San Francisco real estate or some other sector of the economy. According to the plaintiffs’ evidence, honest representations in 1984 would have shown this investment was not a profitable one. Thus, if Winthrop had not made the alleged misrepresentations, the value of the plaintiffs’ long-term investment would have been reduced, and the plaintiffs would have made other investments rather than this one. This is sufficient evidence of causation to withstand summary judgment.13
REVERSED and REMANDED.
. This is the second appeal in this case. The district court originally dismissed the case under the uniform national statute of limitations provided in Lampf, Pleva, Lipkind, Prupis & Petigrow v. Gilbertson, 501 U.S. 350, 111 S.Ct. 2773, 115 L.Ed.2d 321 (1991). In 1992, Congress enacted legislation adopting the pre-Lampf limitations periods for plaintiffs who filed suit before Lampf was decided. See 15 U.S.C. § 78aa-1. The plaintiffs sought to have their claims reinstated, but the district court found § 78aa-l unconstitutional. This court reversed because the case was ‘‘properly construed as ‘pending’ ... for'separation of powers purposes” because an appeal was still available to plaintiffs when Congress enacted § 78aa-1. Gray v. First Winthrop Corp., 989 F.2d 1564, 1571 (9th Cir.1993); see also Plant v. Spendthrift Farm, Inc.,- U.S.-, -, 115 S.Ct. 1447, 1457, 131 L.Ed.2d 328 (1995). After plaintiffs’ claims were reinstated, a new motion for summary judgment was granted from which the present appeal was taken.
. In the December 1987 letter, Winthrop also described its efforts to obtain new tenants and reminded investors that "the original projection of their benefits through 1987 was almost exclusively in the form of tax losses.” Id. at 517.
. The district court distinguished the ordinary rule favoring jury resolution of the issue of inquiry notice on the basis that all of the communications to investors were contained in undisputed documents and no credibility determinations were required. We agree these factors generally weigh in favor of summary adjudication, but they do not necessarily preclude the need for a jury to evaluate "investor reasonableness” in light of the documents and other relevant evidence. See id.
. The ordinary rule arguably applies with special force in this case because Winthrop repeatedly attributed the partnership's poor financial performance to the "slowdown" or “softening” in the general San Francisco real estate market. Poor market conditions is the type of explanation for poor financial performance that this court recognized as defeating summary judgment on the issue of inquiry notice in Mosesian, Seaboard Corp., and Briskin.
. For example, the prospectus said "[t]here is no assurance that the Management Agent will successfully manage or rent-up the Project_ It should be noted, however, that [Winthrop’s] two San Francisco based asset management employees ... [and its] Boston based asset management department ... will continuously monitor the operation of the Project.” ASER at 19.
. It was not until the December 1987 letter that Winthrop told investors the market “unexpectedly entered a period of weakness" beginning in 1985, when in fact the prospectus said there was a "general market slowdown” beginning in 1983. Compare AER at 513 (emphasis added) with ASER at 35.
.Winthrop warns that if the statute of limitations does not bar this lawsuit, then investors will always be able to sit on their rights in the hopes that a troubled investment will rebound. See Tregenza v. Great American Communications *883Co., 12 F.3d 717, 722 (7th Cir.1993) ("If the stock rebounded from the cellar [the plaintiffs] would have investment profits, and if it stayed in the cellar they would have legal damages."), cert. denied, - U.S. -, 114 S.Ct. 1837, 128 L.Ed.2d 465 (1994). We think this case is distinguishable, however. A jury could reasonably excuse the plaintiffs for not investigating the poor financial performance because Winthrop, in the prospectus, told investors the partnership could weather the type of set-backs the plaintiffs encountered and, in its letters to investors, repeatedly said the investment was stabilizing.
. Plaintiffs also allege that Winthrop misrepresented its experience with properly management, the location of the building, and various facts relating to the status of the San Francisco real estate market in 1984. Given our disposition of this case, we need not address the merits of these allegations.
. For example, the public accountant’s report stated “some assumptions will inevitably not materialize" and the prospectus said the investment was “subject to the risks inherent in the ownership of real estate." ASER at. 114, 18 (emphasis added). The prospectus also stated "[n]o assurance can be given that the forecasted results will be achieved.” Id. at 105.
. The October 31, 1984 memorandum was written by Robert Fisher to Hal Brown, Winthrop’s lawyer. The relevant passage of the memorandum stated:
I understand from Larry Wolk that the assumptions used at least for 1984 in the Financial Forecast with respect to office, and retail rental rates and leasing commissions do not correspond to the weighted rates in the leases currently in effect. Although I recognize that the forecast denominates the figures contained therein as assumptions, I find it difficult to justify assumptions for 1984 which are contrary to fact. Let’s discuss this further.
AER at 436. Winthrop argues this memorandum is inadmissible because it is hearsay and neither Wolk nor Fisher were deposed to authenticate the document. This argument overlooks the fact that it would be admissible not to prove the truth of the matter asserted — i.e., that the assumptions were "contrary to fact” — but rather to show scienter on the part of Brown, Winthrop’s agent. See Fed.R.Evid. 801(c) (statement not hearsay if not "offered in evidence to prove the truth of the matter asserted"). Moreover, Brown acknowledged receiving the memorandum and testified he could not recollect what actions he took in response to the memorandum. AER at 230-31.
Winthrop also argues that it is undisputed that the concerns expressed in the memorandum were in error. Citing to a table in a report by the plaintiffs’ real estate expert, Winthrop argues that the net effective rents for five-year leases signed in 1984 ($26.60) was higher than forecast in the prospectus for 1984 ($26.25). Id. at 586. However, the plaintiffs’ expert states "the actual net effective rent, adjusted for over-standard tenant improvements exceeding $5/sq.ft. was $25.56 for five-year leases, or only 97.37% of the lease up assumption for five-year leases in the [prospectus].” Id. at 571. The discrepancy appears to be the result of the exclusion of certain leases from the average net effective rents. This type of dispute over the meaning of facts seems to us to be suited for the jury as a trier of fact. More importantly, this is only one assumption of many which the plaintiffs challenge. Winthrop must recognize the factual dispute over whether it misrepresented the lease-up rate and tenant improvement costs for the building.
. The July 1984 report was prepared by Miller, Klutznick, Davis & Gray ("MKDG”), the building’s prior owners. Winthrop argues it is not admissible because there is no evidence the document was seen by Winthrop prior to October 31, 1984, nor any evidence showing it conflicted with the assumptions in the prospectus. We disagree. First, Stephen Elpem, who was responsible for providing the prior owners' documents to Winthrop, testified he sent all documents relating to the building’s operational history to Winthrop prior to October 1984, and that in any event Winthrop could have learned about the average tenant improvement costs by "reading the leases and reading the expense contracts.” AER at 658, 661. Second, one of Winthrop’s employees, Philip Brannigan, who concluded that “our projections were more conservative than MKDG’s,” ASER at 436, nonetheless acknowledged that MKDG’s projections of tenant improvement costs and leasing commissions were higher than those used by Winthrop. Id. at 430. Moreover, on its face, the MKDG report conflicts with the lease-up projections in the prospectus, not expecting full leasing until January 1987. See AER at 413. Thus, the MKDG report could provide relevant and admissible evidence of Winthrop’s scienter.
. The plaintiffs also rely on evidence generated after the offering period as circumstantial proof that Winthrop knew its assumptions were false and its projections were unreasonable prior to or during the offering period. In light of our holding based on the plaintiffs' other evidence, we need not address the admissibility and relevance of this evidence.
. The trial court did not pass upon the issue of causation. On the record made on the motion for summary judgment we find scant evidence offered by either party to allow the trial court, or this court, to rule as a matter of law on the issue of causation. Suffice it to say, our brief discussion of this issue demonstrates sufficient evidence to at least avoid summary judgment on the question of causation. It is important to recognize that vacating the grant of summary judgment moves the case to a new level of factual inquiry at trial. As in all cases, the plaintiffs must still produce sufficient facts to show causation in order to avoid a judgment as a matter of law and to submit the issue to a trier of fact. Nothing we say here is intended to limit either party as to evidence relating to the claim or defense at trial on the issue of causation.