PER CURIAM:

This appeal is from a final decree in an action by the United States against the City of Buffalo, which has been pending for some time in the District Court for the Western District of New York, concerning employment discrimination by the City's police and fire departments against blacks, Spanish-surnamed Americans (SSA's) and women. Finding that the City had engaged in a pattern or practice of discrimination in violation of Title VII of the Civil Rights Act of 1964, as amended by the Equal Employment Opportunity Act of 1972, 42 U.S.C. § 2000e et seq., Chief Judge Curtin in November, 1979, entered a final decree granting injunctive and monetary relief. The City appealed from certain aspects of the decree.

After studying the briefs and appendix and hearing argument on May 5, 1980, we were satisfied that Chief Judge Curtin's findings of fact were not erroneous and that his conclusions of law were proper, for reasons which are stated in his careful and persuasive opinions, notably United States v. City of Buffalo, 457 F.Supp. 612 (W.D.N.Y. 1978), and need not be repeated here. Our only concern was with Part II of his decree, entitled "GOALS", the most pertinent portions of which are set out in the margin. Since several other cases involving the use of racial quotas were then under advisement by other panels, we decided to defer decision until the results of their deliberations were available, although, as matters turned out, only Guardians Association of the New York City Police Department, Inc. v. Civil Service Comm'n of the City of New York, 630 F.2d 79 (2d Cir. 1980), proved to have real significance.

As appears from the decree, the district court set "interim hiring goals" of 50% for minorities in the Police and Fire Departments and of 25% for women in the Police Department. The figures compare with a City work force roughly 18% black, 4% SSA and 41% female. In contrast, according to appellants' brief, p. 26, of the persons who actually appeared to take the written police examination in 1973, 6.08% were black males, 1.98% Hispanic males, 12.76% white females, 5.39% black females, and .68% Hispanic females. The corresponding figures for the firefighters' examination in 1973 were 15.78% black males, 1.65% Hispanic males, 1.53% white females, 1.25% black females, and .13% Hispanic females. The percentages for those applying to take the examinations were for the most part somewhat higher. The City urges that "Since these applicants were recruited after an intensive campaign, these figures provide a more realistic measure of minority and female interest in police and firefighter positions than do general population figures."

We entertain no doubt that, in light of the evidence in this case and our prior decisions on the subject, the district court was justified in setting interim hiring requirements for blacks and SSA's at a percentage higher than their representation in the City's total work force or the applicant pool, and for women at a percentage higher than their representation in the applicant pool. We agree that, as Judge Newman stated in Guardians Association, supra, 630 F.2d at 109, even interim hiring ratios "greater than the minority percentage in the applicant pool or the relevant work force, should be imposed only upon clear evidence and appropriate findings of the need to redress demonstrated prior discrimination of long standing that has had a significant impact on minority employment." However, those criteria were met here. The district judge found, on ample evidence, that the proven violations were serious and that little or no progress had been made in remedying them. Moreover, he concluded that the percentages of minorities and women applying for or taking the 1973 examinations could not be relied upon because the history of discrimination in hiring by both departments suggested that those percentages did not reflect the actual interest of those groups in employment as police or fire department officers.

We cannot quarrel with the district court's requirement that, on an interim basis, minorities receive 50% of all entry level positions, police officerships, and firefighter appointments and that women receive 25% of all entry level positions and police officerships. The choice of any particular interim hiring ratio is necessarily somewhat arbitrary. Here, as in Vulcan Society of New York City Fire Department v. Civil Service Comm'n, 490 F.2d 387, 399 (2 Cir. 1973), the ratio chosen was appropriate in light of "the resentment of non-minority individuals against quotas of any sort and of the need of getting started to redress past wrongs." The important thing is that the duration of such hiring quotas should be no longer than is necessary. The figures chosen here were not unreasonably high in light of the finding of serious discrimination and lack of previous progress, the slow rate of hiring projected in the police department, and the likelihood that prior discrimination had discouraged minorities and women from applying for jobs.

With respect to duration, the court said in Guardians Association, supra, 630 F.2d at 109 (footnote omitted):

If a hiring ratio is imposed beyond the interim period in which a valid selection procedure is developed in order to reach a required long-term target, the justification for its use must be especially compelling.

If the district court had contented itself with imposing hiring ratios until it had found that proper selection procedures had been established and that no further hiring requirements were appropriate, with the United States having the right at that time to advocate such further requirements on the ground that the continuing effects of past discrimination had not been eliminated, the City would have had no cause to complain. The City likewise can have no rational basis for objecting to the provision in paragraph 7, see note 1, that the interim hiring ratios should cease at an earlier date in the event that the minority composition of the uniformed personnel of the Police Department and the Fire Department should become equal to or greater than the percentage of minorities in the City's labor force as determined by the most recent census. What gives concern is the provision in paragraph 2 requiring the City "to seek to achieve the long-term goal of reaching a minority composition in the ranks of uniformed personnel within the Police and Fire Departments comparable to that of the work force within the city as a whole, according to the most recent census."

The United States does not attempt to defend this portion of the decree on the merits but asserts rather that, in view of paragraphs 7 and 8, it serves no operative purpose and can be disregarded. Brief, p. 37, n. 14. We cannot accept this contention. Paragraph 2 must be read as meaning what it says. It would afford a basis, in addition to the appropriate one mentioned in the last sentence of paragraph 7, for arguing that hiring ratios must be continued even after valid selection procedures had been approved. Indeed, so long as it remains in the decree, the City, even after having been relieved of interim hiring ratios because these were found to be no longer appropriate, might be required to take other measures to achieve "the long-term goal" prescribed in paragraph 2.

If, as the United States contends in its brief, paragraph 2 has no effect, it should be stricken as confusing surplusage. If, as we believe, it does have an effect, it should likewise be. While disparities between the proportion of minorities or women in the work force or the applicant pool and those who pass examinations or are employed have evidentiary value with respect to discrimination, once proper selection procedures have been established judicial prescription of further goals or the use of quota hiring to meet them is only justified if necessary to eliminate the effects of past discrimination. There is no requirement, either constitutional or statutory, that each identifiable group of employees should exactly reflect the racial or sexual composition of the work force. We consider all this to be the teaching of our cases, see Kirkland v. New York State Department of Correctional Services, 520 F.2d 420, 430 (2 Cir. 1975), rehearing in banc denied with three judges dissenting, 531 F.2d 5 (2 Cir. 1975), cert. denied, 429 U.S. 823, 97 S.Ct. 73, 50 L.Ed.2d 84 (1976); Chance v. Board of Examiners, 534 F.2d 993, 998 (2 Cir. 1976), cert. denied, 431 U.S. 965, 97 S.Ct. 2920, 53 L.Ed.2d 1060 (1977); Equal Employment Opportunity Comm'n v. Local 638, 532 F.2d 821, 831 (2 Cir. 1976). Once valid selection procedures have been instituted, the determinations whether it is necessary to prescribe ultimate goals to eliminate the effects of past discrimination, whether such goals should be a proportion of the total work force or some smaller and, in some cases, more representative unit, and whether attainment of such goals must be through continued hiring quotas or may be by less abrasive measures such as recruitment programs raise questions of the utmost difficulty. Such questions should not be faced sooner than need be and should then be determined after the most careful consideration of detailed evidence. Here the record is barren of evidence that with properly validated selection procedures minorities would have applied for and achieved posts in the uniformed services in a ratio comparable to their numbers in the City's work force as a whole. Indeed, the court made no specific finding that interim relief would be inadequate to eradicate the effects of past discrimination and that it was therefore necessary to prescribe a long-term goal. Moreover, the entire issue can be better considered, on a fuller record, when the City has satisfied the court that it has developed valid selection procedures, if the United States then argues that quota hiring requirements should nevertheless be continued. Without prejudice to what, if anything, may then be decided with respect to the need for an ultimate goal, we think the decree should be modified by striking paragraph 2 and, as so modified, is affirmed. The United States may recover two-thirds of its costs.