OPINION

GOETTEL, District Judge:

This Title VII case is before the Court on remand from the Second Circuit Court of Appeals. The facts are set forth at length in our earlier decision, Sobel v. Yeshiva University, 566 F.Supp. 1166 (S.D.N.Y. 1983) (“Sobel I”), familiarity with which is assumed. In brief, the two named plaintiffs brought this discrimination action on behalf of themselves and other female physicians on the faculty of defendant Yeshiva University’s Albert Einstein College of Medicine (“Einstein”). The Equal Employment Opportunity Commission (“EEOC”) joined the suit as a plaintiff-intervenor. Following a bench trial, we dismissed the plaintiffs’ claim of pay discrimination, and reserved decision on a claim of discrimination in pension benefits. 1 The latter claim was ultimately settled by the parties. The plaintiffs appealed the dismissal of the pay claims; 2 the EEOC did not. 3

In its brief, per curiam opinion, the Second Circuit directs us to reconsider this case in light of the Supreme Court’s recent decision in Bazemore v. Friday, — U.S. -, 106 S.Ct. 3000, 92 L.Ed.2d 315 (1986), “particularly with respect to the significance of preact discrimination and the evidentiary weight to be afforded multiple regression analysis.” Sobel v. Yeshiva University, 797 F.2d 1478, 1479 (2d Cir.1986).

Discussion

I. The Significance of Pre-Act Discrimination

The parties were directed to brief the issue of whether this suit is affected by *589 Bazemore ’s holding on the significance of pre-Act discrimination. They have failed to do more than state the obvious. The plaintiffs assert that Bazemore is directly in point and requires a verdict in their favor. The defendant contends that, since no preAct discrimination was established by the evidence, Bazemore is inapplicable.

The initial problem in evaluating this case in light of Bazemore is that the facts of the two cases differ markedly. In Bazemore, prior to August 1, 1965, the North Carolina Agriculture Extension Service (the “Extension Service”) had been racially divided into two branches, one white and one black, with differing pay scales. In response to the Civil Rights Act of 1964, the two branches were merged, but disparities in salaries continued. The plaintiffs sued in 1971 alleging racial discrimination in employment and in the provision of services by the Extension Service. The District Court declined to certify a class action and found for the defendants on all issues. The Court of Appeals affirmed. Although the appellate court acknowledged that some pre-existing salary disparity remained, 4 it held that the defendants were not obligated to eliminate salary disparities that originated prior to 1972. The Supreme Court reversed, holding that pre-Act salary discrimination did not excuse perpetuating the discrimination after the Extension Service became covered by Title VII. “To hold otherwise would have the effect of exempting from liability those employers who were historically the greatest offenders of the rights of blacks.” 106 S.Ct. 3006.

In Bazemore, the pre-Act racial discrimination affected all blacks and, undoubtedly, was illegal. 5 In the instant action, little evidence was offered to establish if, or how, the defendant discriminated against women prior to 1972, the year in which Title VII became applicable to universities. 6 For female faculty members hired after 1972, the evidence showed an absence of statistically significant differences in salary during the relevant period of 1974 to 1979. 7 Sobel I, supra, 566 F.Supp. at 1182. Any salary differences during that period seemed to result from lower salaries initially paid to some female faculty members hired prior to 1972. Id. However, the evidence showed that this probably resulted not from discrimination, but from several gender-neutral factors. 8

*590 Throughout the lengthy pretrial proceedings and most of the trial of this case,. the plaintiffs endeavored to prove discrimination on a theory of disparate treatment of female faculty at Einstein during the years 1974 to 1979. Despite seven years of discovery and elaborate statistical analyses by notable experts, the evidence failed to support this claim. 9 When this became evident at trial, the plaintiffs switched to a theory of disparate impact, claiming that discriminatory acts prior to the relevant period created a continuing discriminatory effect on the plaintiffs’ salaries.

At the time this case was tried, claims of disparate impact required proof of a present violation of Title VII, not merely a past act of discrimination with ongoing effects. United Airlines, Inc. v. Evans, 431 U.S. 553, 97 S.Ct. 1885, 52 L.Ed.2d 571 (1977). Although Bazemore does not explicitly overrule Evans, it seems to call into question the continued validity of Evans’ approach to proving disparate impact claims. Had we relied solely on Evans in dismissing the plaintiffs’ disparate impact claim on the merits, our prior ruling might require reconsideration. However, we dismissed that claim not only on the merits, but also on the procedural ground of unfairness to the defendant. Sobel I, supra, 566 F.Supp. at 1186-89. The latter is an adequate independent ground for dismissal, regardless of whether Bazemore overruled Evans sub silencio. In dismissing the plaintiffs’ disparate impact claim on procedural grounds, we stated as follows:

For seven years prior to the trial and throughout much of the trial itself, the defendant was confronted solely with a claim of disparate treatment. Much of the Court’s purpose in permitting the parties so much time to prepare this case was to ensure full and fair discovery and complete development of all the legal issues that might arise at trial. Thus, to allow the plaintiffs to adopt a completely different theory of liability at such a late date, particularly when in the years preceding the trial they gave little if any indication that they were claiming disparate impact, would be to work a gross injustice on the defendant.

Id. at 1187 (footnote omitted).

The plaintiffs not only waited until the last minute to assert their disparate impact claim, they presented scant evidence to support such a claim, and even that was flawed. See infra section II.A. It would be highly prejudicial to the defendants at this point to reopen the case and allow further discovery by the plaintiffs on the unlikely chance that they could now develop a viable disparate impact claim tailored to Bazemore’s holding. 10 We see no rea *591 son to change our decision dismissing this claim on procedural grounds. Accordingly, we need not consider whether, after Bazemore, the claim also fails on the merits.

II. The Weight to be Afforded Multiple Regression Analysis

It is not clear why the Court of Appeals felt that Bazemore might require any change in this Court’s original assessment of the weight to be afforded multiple regression analyses. On appeal, the plaintiffs argued that this Court “rejected” the multiple regression analysis they offered in evidence. Plaintiffs’ Appellate Reply Brief at 6. That is incorrect; 11 with a single exception, discussed below in subsection A, all of the plaintiffs’ multiple regressions were received in evidence. This Court, however, did not find them to be conclusive and, indeed, found serious defects in the approaches used by the plaintiffs’ experts. On this remand, the plaintiffs shift their argument from one of admissibility to one of acceptance. They contend that Bazemore requires that multiple regression analyses not only be admitted and considered by the Court, but also be accepted. Bazemore does not stand for that proposition.

In Bazemore, the Court of Appeals affirmed the District Court’s ruling that the plaintiffs’ regressions were unacceptable as evidence of discrimination because they omitted a number of variables that could have affected salary levels. The Supreme Court, reversing, held that

a regression analysis that includes less than “all measurable variables” may serve to prove a plaintiff’s case. A plaintiff in a Title VII suit need not prove discrimination with scientific certainty; rather, his or her burden is to prove discrimination by a preponderance of the evidence. Whether, in fact, such a regression analysis does carry the plaintiffs’ ultimate burden will depend in a given case on the factual context of each case in light of all the evidence presented by both the plaintiff and the defendant.

106 S.Ct. at 3009 (citation omitted). This holding instructs us to consider multiple regression analyses regardless of the omission of arguably relevant variables, but, to weigh them in light of all evidence presented by both sides. This is precisely the consideration we gave the plaintiffs’ regressions.

At trial, the plaintiffs relied heavily on statistical proof. We noted that this approach was acceptable as a means of proving discrimination, and stated,

In such a case, the court’s task is to determine whether the plaintiff’s statistics make out a prima facie case of a practice or pattern of discrimination, and, if so, whether that case is fatally undercut by a showing that the plaintiff’s “proof is either inaccurate or insignificant.” [Int ernational Brotherhood of] Teamsters v. United States, supra, 431 U.S. at 360, 97 S.Ct. at 1867. Of course, in determining whether the plaintiff has established a prima facie case, the court should not ignore the defendant’s relevant evidence of nondisparate treatment. “Prima facie evidence means the ‘net of the evidence;’ that is, the court must consider the evidence presented by both parties.” Underwood v. Jefferson Memorial Hospital, 639 F.2d 455, 457 (8th Cir.1981).

Sobel I, supra, 566 F.Supp. at 1173.

Unfortunately, the plaintiffs’ statistical proof was seriously flawed. Id. at 1178-85. The EEOC concedes that the regressions had “some unqualifiable productivity variables that could not be controlled for. The omission of these variables always is a basis for legitimate concern.” EEOC Reply Memorandum at 19. We previously commented as follows:

Although the plaintiffs correctly argued that, in order to make out a prima facie case, they were not required to produce a perfectly designed or specified model, *592 but rather one that only roughly controlled for the necessary considerations, the plaintiffs failed to produce a model that met even this lesser standard.

Sobel I, supra, 566 F.Supp. at 1182. When the plaintiffs incorporated additional data and variables suggested by the defendant, the plaintiffs’ regressions made only a borderline showing of discrimination (slightly more than two standard deviations), and then only for certain years. Id. at 1176, 1181.

Moreover, the plaintiffs’ regressions were virtually the only evidence presented with respect to the class. The named plaintiff, Edna H. Sobel, testified about her experiences. However, her situation was somewhat unique. Indeed, by being a “squeaky wheel,” she got some pay increases that others did not. See id. at 1185, 1188 n. 56. Her testimony, standing alone, did little or nothing to assist the classwide claims. The other named plaintiff did not testify.

There was, therefore, none of the direct evidence of discrimination with respect to any particular class member that might have provided an acceptable means of proving discrimination in a case of this nature____ It seems reasonable to assume that, if there had been class-wide discrimination, the plaintiffs should have been able to demonstrate some instances of discrimination against individuals. The absence of any such anecdotal evidence casts considerable doubt on the plaintiffs’ claims. This is particularly true in a case such as this where discovery was undertaken over a seven-year period, with all of the resources of the EEOC available for the last several years.

Id. at 1186 (footnote and citation omitted). See Rossini v. Ogilvy & Mather, Inc., 798 F.2d 590, 604 (2d Cir.1986); EEOC v. Sears, Roebuck & Co., 628 F.Supp. 1264, 1280-87 (N.D.Ill.1986). This is in sharp contrast with Bazemore, in which “petitioners offered an impressive array of evidence to support their contention [of salary discrimination].” 106 S.Ct. at 3010. See also International Brotherhood of Teamsters v. United States, 431 U.S. 324, 339, 97 S.Ct. 1843, 1856, 52 L.Ed.2d 396 (1977) (“[individuals who testified about their personal experiences ... brought the cold numbers convincingly to life.”).

In considering this issue in Bazemore, the Supreme Court reversed because the lower court had focused solely on the inadequacy of the regression analysis and had ignored additional evidence of discrimination. 106 S.Ct. at 3002. We did not similarly limit our consideration of the evidence in the instant suit. Rather, we concluded, upon examination of all of the evidence from all parties, “that the plaintiffs have failed to prove a prima facie case of disparate treatment with respect to salary. They have failed to carry their burden of persuasion for both the individual and the class claims.” Sobel I, supra, 566 F.Supp. at 1186. We find that Bazemore requires no change in this conclusion.

A. Exhibit 111

In remanding this case, the Court of Appeals made no reference whatever to the trial court’s refusal to receive plaintiffs’ Exhibit 111 in evidence. Both the Court and the parties are uncertain whether this is an issue on remand. However, since Exhibit 111 was the only instance in which a regression was not received in evidence, and since it constituted virtually the only proof of a disparate impact claim, prudence dictates a reconsideration of our ruling in that regard.

Exhibit 111 was rejected on two separate grounds, one procedural and one substantive. The substantive ground concerned our interpretation of United Airlines, Inc. v. Evans, supra, 431 U.S. 553, 97 S.Ct. 1885, 52 L.Ed.2d 571 (1977). As noted above in section I, while not specifically overruling Evans, Bazemore at least calls into question the continued validity of its approach to proof of disparate impact claims. However, this need not alter our ruling, which was based on the separate and completely independent procedural basis of untimeliness.

As already mentioned, this case went through many years of extensive discovery. *593 It was developed and tried as a disparate treatment case. The evidence available to the defendants at trial related to the period from 1974 to 1979. At the very conclusion of the plaintiffs’ rebuttal case, they offered Exhibit 111 to support a claim of disparate impact. The exhibit contained the statistical flaws mentioned above, without addressing the major criticisms raised by the defendants’ experts. The exhibit was of little probative value since the plaintiffs had failed to assemble an adequate data base regarding faculty employed prior to 1974. 12 However, the major reason for refusing to receive Exhibit 111 in evidence was its untimely submission. 13 Sobel I, supra, 566 F.Supp. at 1182 n. 40. The defendant’s experts indicated that they would require a lengthy adjournment of the almost completed trial as well as substantial additional trial time to respond to the exhibit. This Court stated at trial that,

if this [Exhibit 111] were vital, new, different evidence, I think I would undergo the inconvenience to the rest of the trial schedule. But it is just an attempt to deal with some of the criticism, and to my way of thinking not the major criticism. I think the major, criticism of the multiple linear regression studies is the inability to deal with productivity in a meaningful sense and to find proxies that will capture the missing productivity-

Trial Transcript at 2713-14.

The EEOC offers a curious argument that, although this Court’s rejection of Exhibit 111 at trial was perhaps proper, it should now be admitted to “assist the Court of Appeals in assessing the merits of this case if a subsequent appeal should be taken.” EEOC Reply Memorandum at 13. Evidence is received in the trial court. To justify introducing new evidence and a new legal theory after seven years of discovery, and a three-week trial, a party must establish both that it was not responsible for this delay (usually by establishing that the other side caused it) and that the evidence is highly probative and warrants the extraordinary exercise of the Court’s discretion in accepting it. Here, the plaintiffs failed in both regards. They neither justified their tardy change of direction and proof, nor produced a document that clearly, on its face, would have changed the trial results. Consequently, the Court declines to reconsider its evidentiary ruling excluding Exhibit 111, or to direct that it be added to the record for purposes of appellate factual review.

Conclusion

The Court, having reconsidered the case pursuant to the direction of the Court of Appeals, adheres to its earlier decision and directs entry of judgment for the defendant.

SO ORDERED.

. Prior to trial, the plaintiffs withdrew claims of discrimination in promotions and other aspects of employment.

. Both the plaintiffs and the defendant appealed our denial of their respective applications for attorneys fees. Sobel v. Yeshiva University, 619 F. Supp. 839 (S.D.N.Y.1985). However, since the recent remand by the Second Circuit makes no mention of the fee appeals, that issue is not presently before us.

. Since the EEOC did not appeal, it may not benefit from this remand. See Cruickshank & Co. v. Dutchess Shipping Co., 805 F.2d 465 (2d Cir.1986); 9 J. Moore, B. Ward, & J. Lucas, Moore's Federal Practice ¶204.11[4]-[5] (2d ed. 1986). The EEOC has nevertheless submitted papers on the instant issues and also moves to vacate the final judgment so that it may participate in further proceedings in this matter. Since we conclude that no further proceedings are necessary, we need not rule on the EEOC’s motion to vacate judgment. We have, however, read and considered the EEOC’s submissions on the remanded matters.

. Indeed, the defendants apparently admitted that they had been unable to eliminate all such disparities. Bazemore, supra, 106 S.Ct. at 3006.

. Discrimination on the basis of race was made illegal under the thirteenth and fourteenth amendments to the Constitution, and the Civil Rights Act of 1866, 42 U.S.C. §§ 1981, 1983. Long before Title VII was enacted, blacks could assert employment discrimination claims under these existing laws. Today, pay and employment discrimination claims by blacks are often brought both under Title VII and the Civil Rights Act. Sexual pay discrimination, however, is solely a product of Title VII, and claims in that regard did not exist for pre-Act periods.

. Women were certainly not discriminated against in hiring, since Einstein had a higher percentage of women on its faculty than most of the other medical schools in the country. See Sobel I, supra, 566 F.Supp. at 1169 n. 8.

. The plaintiffs filed this suit in 1975. The "relevant period” for the plaintiffs’ claims began in 1974, because that was the earliest year open under the applicable statute of limitations. Although the plaintiffs contend that we failed to consider proof of matters occurring prior to 1974, their accusation is unfounded. Evidence of facts and events occuring beyond the limitations period may be used to establish the motive or intent of actions taken within the open period. When such evidence was offered, we accepted it for that purpose.

. For example, the female faculty members at Einstein were generally younger and less experienced than their male counterparts, probably reflecting women’s later entrance into the medical field, and women more often specialized in less lucrative specialties, such as pediatrics and psychiatry. Since experience and area of specialization were factors in establishing salaries, the net result was often lower pay for women. Pay increases, however, were found to be neutral "both in principal and in operation.” Sobel I, supra, 566 F.Supp. at 1188. In fact, the guideline system for pay increases may have favored women. Id. at 1183-84. Also, as we noted in our original decision,

During the late 1960’s and early 1970’s, Einstein was in difficult financial straits caused, at least in part, by the opening of its own hospital in the Bronx at a time when the need for a new hospital was not economically apparent. Consequently, it appears that most of Einstein’s faculty members were not paid what they were worth. Raises tended to go only to those who made a great fuss about the *590 failure to receive an adequate salary. In most instances, those who protested were men who were the sole sources of income for their families. (Although no statistics were presented on this matter, we take it as an accepted sociological fact that the percentage of men who were the sole wage earners for families with children exceeded the percentage of married women who were such.) This tendency to favor the sole working members of families appears to have disappeared in the 1970’s as circumstances changed.

Sobel I, supra, 566 F.Supp. at 1184 n. 49. This policy, which, in the late 1960’s and early 1970’s, was considered not only socially acceptable but also socially desirable, more often favored men. However, to the extent that a woman was the sole wage earner for a household she benefitted rather than suffered from the policy.

. A disparate treatment claim requires proof that the defendant intended to discriminate, and that discrimination was a regular, rather than unusual, practice in the treatment of the class on behalf of which the action is brought. See International Brotherhood of Teamsters v. United States, 431 U.S. 224, 335 n. 15, 336, 97 S.Ct. 1843, 1854 n. 15, 1855, 52 L.Ed.2d 396 (1977). Besides the weakness of the statistical evidence and the almost total absence of anecdotal proof of discrimination, the evidence presented by the defendant revealed a decentralized administration with responsibility for salary decisions delegated to department heads, and an absence of motive or opportunity to discriminate against highly trained professionals in a competitive market, all of which tended to disprove the plaintiffs' discrimination claims. See Sobel I, supra, 566 F.Supp. at 1186.

. In view of the evidence presented, we question whether such a case could ever have been successful. Had the Bazemore decision been available in 1983 when this case was tried, or, better yet, in 1975 when the suit was filed, the plaintiffs might have focused more specifically on a disparate impact claim. However, this is speculative hindsight since Bazemore was not decided until three years after this case was *591 tried. And, as indicated above, the factual differences between this case and Bazemore make the latter’s application here both difficult and questionable.

. This error is not surprising, however, since the plaintiffs were represented on appeal by their third attorney, who was not involved in the trial of this action.

. In addition, the pre-1974 regressions did not contain variables that the plaintiffs’ experts conceded at trial were indispensable to a properly specified model and which they used for the 1974-1979 period.

. The pretrial discovery orders required that data base objections be completed in the spring of 1982, that statistical reports be exchanged in June 1982, and rebuttal reports in July 1982.