dissenting:
I respectfully dissent.
In 1979, another panel remanded this case to take further evidence in the hope that the present appeal would return “with reasonable speed.” 598 F.2d at 722. Five years later, we remand again in order to have the district court repeat the precise exercise ordered by the earlier panel.
In my view, we should reverse and enter judgment for the Boárd of Education. The prior opinion held that the de facto segregation at Andrew Jackson High resulted solely from residential demographics and was thus not unconstitutional. Therefore, the Board was not then, and is not now, under any federal legal obligation to do anything to alter the racial composition of Andrew Jackson High.
Notwithstanding the lack of any legal obligation, the Board has sought to create an opportunity for students at Andrew Jackson to pursue their education in integrated schools through a transfer plan that utilizes racial quotas. In the prior opinion, we held that such quotas are justified by a compelling governmental interest where they result in greater integration than would exist under a racially neutral (and *584constitutionally valid) system of attendance zones based on residence. We remanded, however, because of the lack of evidence in the record as to why the various formulae used by the plan to fashion the quotas were chosen. Given the nature of the underlying legal issue, there is no federal constitutional imperative to require more than a showing that the factual basis for the quotas’ formulae is sufficient to ensure that the plan actually decreases segregation.1
Although the majority appears to assume the present plan does reduce segregation, it creates a legal rule that such a plan be clearly demonstrated to be the best available to reduce segregation. Since the Board need not offer any plan, much less one judged by us to be the most desirable, the constitutional source of this requirement is not evident to me. It is, moreover, a most unwise ruling, for two reasons.
First, demographic variables are too numerous and too imponderable to locate soealléd “tipping points” with any precision. They will vary from school to school, neighborhood to neighborhood, and ethnic group to ethnic group. Moreover, unpredictable events occurring elsewhere which affect the degree of racial tension in the society can quickly change behavior. One cannot determine a “tipping point” predicting human behavior as scientists predict the movement of planetary objects. I fear the majority’s quest will result only in this case returning to us in 198? with more recent but equally equivocal statistics.
Second, the majority’s ruling will have the unintended consequence of increasing racial imbalance in education. While the plaintiffs here may be confident of obtaining a better plan through political efforts if the present one is invalidated, the undeniable precedential effect of this ruling is to throw substantial roadblocks onto the paths of school boards which voluntarily undertake plans to reduce racial imbalance. Lawyers who counsel such boards will surely advise them that today’s decision guarantees protracted and unpredictable litigation arising out of these plans and that such a voluntary effort is simply not worth the cost. Given the controversy inevitably generated by these plans, this additional obstacle can only decrease voluntary efforts.
For the reasons stated, I respectfully dissent.
. The prior panel, in a footnote, indicated, without further discussion or elaboration, that each component of the plan had to be shown to be “necessary.” Since, as I discuss infra, I believe there is no single best formula, I would not read "necessary" to mean "essential” but rather to mean "appropriate." See McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316, 413-14, 4 L.Ed. 579 (1819) (Marshall, C.J.).