WINTER, Circuit Judge:

In two class actions, the contention that women and blacks were systematically excluded from grand and petit juries in Dillon and Marion Counties, South Carolina, respectively, was asserted. In No. 72-2097 (Blackwell v. Thomas, involving Dillon County), the district court dismissed the case as moot on the basis of stipulated facts. In No. 72-2065 (Vandross v. Lane, involving Marion County), the district court gave judgment for defendants on the ground that plaintiffs had failed to make out a prima facie case of discrimination. The appeals were briefed and argued seriatim, but since they involve common questions of law, we decide them together. In No. 72-2097 (Dillon County), we affirm. In No. 72-2065 (Marion County), we vacate the judgment and remand for further proceedings.

I.

Both cases involve South Carolina's jury selection scheme, the statutory basis of which passed constitutional muster in Franklin v. South Carolina, 218 U.S. 161, 30 S.Ct. 640, 54 L.Ed. 980 (1910). See also Carter v. Green County, 396 U.S. 320, 334-335, 90 S.Ct. 518, 24 L.Ed.2d 549 (1970). By virtue of § 38-52, Code of Laws of South Carolina (1971 Cum.Supp.), the jury commissioners of each county must prepare annually a list of electors, between the ages of twenty-one and sixty-five, "of good moral character" and free from all legal exception, from which jurors are drawn. § 38-61, Code of Laws of South Carolina (1962 Ed.). While § 38-52 requires only that "not less than two from every three of such electors qualified" shall be included on the jury list from which jurors are drawn, the statute is silent as to when, in what manner, and by whom the reduction in number is to be made if the commissioners conclude to use less than the whole. Since there is no fixed neutral formula for a reduction in numbers, there may or may not be deliberate or inadvertent systematic exclusion depending upon how the statute is administered in various counties of the state.

II. [3] No. 72-2097

The statistics stipulated in Blackwell disclosed a "substantial disparity" between the portion of presumptively qualified blacks and women and the number of representatives of these two groups who actually served on juries in Dillon County from 1969 through June, 1971. Since this disparity coincided with the opportunity for discrimination inherent in the South Carolina statute, a prima facie case would have been established so that the state would have the burden of explaining how the disparity occurred and proving that it was not the result of actual discrimination. Stephens v. Cox, 449 F.2d 657 (4 Cir. 1971).

However, after plaintiffs filed their complaint in Blackwell, the Solicitor of Dillon County petitioned Judge James A. Spruill, Jr., of South Carolina's Fourth Judicial Circuit, for an order requiring that a new jury list be prepared. This relief was sought because new yearly lists had not been prepared in December, 1969 and December, 1970, as required by state law. Judge Spruill granted relief and ordered the preparation of a new list.

It was stipulated that this new list was prepared by the county's jury commissioner:

. . . using the most current registration list of electors between the ages of twenty-one and sixty-five years. Said registration list reflected a total of approximately 6,269 such electors in Dillon County, of whom 4,469 (71%) were white, 1,800 (29%) were nonwhite, 2,988 (48%) were male, and 3,281 (52%) were female. The new jury list was comprised of all those names appearing on that registration list. A copy of the new jury list was then prepared so that each individual name could be inserted into a separate capsule of the type appended hereto and made a part hereof, and each of those capsules thus prepared was placed in the new jury box. (emphasis added).

Since the box from which the grand jury and petit juries were drawn contained the names of all of the electors of Dillon County, no opportunity to discriminate existed; and, not surprisingly, the petit jury panels drawn for the rest of 1971 consisted of 27% blacks and 42% women, and the grand jury consisted of 22% blacks and 55% women.

We agree with the district court that, as a result of the selection procedure actually followed and the statistical result which obtained with reference to the new list, no justiciable case or controversy remained to be settled. Plaintiffs, however, contend that equitable relief should have nevertheless been granted to insure against systematic exclusion in the future selection of jurors in Dillon County. Nothing in the statutory scheme, they point out, mandates that the names of all registered voters be placed in the jury box from which jury lists are drawn; nor can it be said with absolute certainty that the pre-June 1971 pattern of discrimination will not reappear.

S.E.C. v. Medical Committee for Human Rights, 404 U.S. 403, 406-407, 92 S.Ct. 577, 30 L.Ed.2d 560 (1972); United States v. Phosphate Export Asso., 393 U.S. 199, 202-204, 89 S.Ct. 361, 21 L.Ed.2d 344 (1968); and United States v. W. T. Grant Co., 345 U.S. 629, 632-633, 73 S.Ct. 894, 97 L.Ed. 1303 (1953) govern our ruling with respect to this contention. They establish the proposition that voluntary cessation of illegal conduct may render a claim for equitable relief moot, but only "if subsequent events made it absolutely clear that the allegedly wrongful behavior could not reasonably be expected to recur." Phosphate Export Asso., 393 U.S. at 203, 89 S.Ct. at 364. Without that assurance the case will not be moot, because "[t]he defendant is free to return to his old ways." W. T. Grant Co., 345 U.S. at 632, 73 S.Ct. at 897.

Whatever discrimination may have been practiced in the selection of juries in Dillon County prior to June 1971, it is evident that prompt and effective corrective action was achieved by the intervention of a state judicial officer as soon as deficiencies were brought to his attention. In this regard, it is not without significance that prior to the filing of plaintiffs' suit in the district court, neither plaintiffs nor anyone else sought corrective action in the state court. In the light of these facts, we cannot hypothesize that either the jury commissioners will not comply strictly with state law in the future as they have been directed to do by a state judge, or that the state judiciary will not immediately rectify any discrimination in the selection of future juries if complaint is made to them. Since we conclude that there is no "cognizable danger of recurrent violation," W.T. Grant Co., 345 U.S. at 633, 73 S.Ct. at 898, we think the case is moot, and we affirm its dismissal by the district court.

III. [11] No. 72-2065

The jury selection procedures in Marion County have never been the subject of state court investigation and possible state court corrective action. Although the district court found that jurors were selected from all qualified electors of the county and predicated its opinion on that finding, the record is devoid of any evidence or stipulated fact to support it. Thus, the case comes to us with the questions unanswered as to whether the jury commissioners exercised their statutory authority to select the names of two out of three electors to place in the box from which jurors are drawn, and if so, how the selection was made. It is conceded that the voting lists showing names of electors which were furnished them contained racial designations. We must therefore consider the case as one in which deliberate or inadvertent systematic exclusion might have occurred since, as we have noted, (a) the South Carolina statute requires that only the names of two out of three electors shall be used and makes no provision as to how the reduction is to be made, and (b) the lists of electors furnished to the jury commissioners contained racial designations.

In Marion County, blacks comprise 51% of the aggregate population, 44% of the population between the ages of twenty-one and sixty-five, and 41% of the registered voters. Women comprise 53% of the aggregate population and 54% of the population between the ages of twenty-one and sixty-five. The percentage of women among registered voters is apparently unknown.

The membership of the 1970 grand jury was 22% black and 28% women. The membership of the 1971 grand jury was 33% black and 28% women, while that of the 1972 grand jury was 28% black and 33% women. With regard to membership of petit juries, in 1970, 35% were black and 52% were women; in 1971, 32% were black and 53% were women; and in 1972, 29% were black and 52% were women. Thus, while the percentage of women on petit juries throughout the period was almost exactly the same as the percentage of women in the aggregate population and the population between the ages of twenty-one and sixty-five (the only figures with which a comparison can be made), there was a not insubstantial deviation of blacks on grand and petit juries throughout the period compared to the percentage of blacks among registered voters of the county. Particularly with respect to petit juries, the underrepresentation of blacks became progressively more marked from year to year within the three-year period.

The district court relied on Swain v. Alabama, 380 U.S. 202, 85 S.Ct. 824, 13 L.Ed.2d 759 (1965) in concluding that the underrepresentation of blacks and women was not sufficient to prove racial discrimination; and on its authority, defendants urge us to find that there was not a "substantial disparity" so as to warrant relief or even further investigation. As to women, we agree that the disparity was not substantial, even though we do not know the percentage of women among the electorate. With respect to blacks, we conclude otherwise. As to them, the stipulated underrepresentations of 6% in 1970, 9% in 1971, and 12% in 1972 are not insubstantial, although we do not presently conclude that they constitute a "substantial disparity." While Swain concluded, on the facts of that case, that "[w]e cannot say that purposeful discrimination based on race alone is satisfactorily proved by showing that an identifiable group in a community is underrepresented by as much as 10%," 380 U.S. at 208-209, 85 S.Ct. at 829, Swain was not a case in which the same opportunity for exclusion as exists under the South Carolina jury selection statutes was present. Where, as here, the opportunity for systematic exclusion exists, we think the disparity sufficiently great to warrant an evidentiary exploration of how the jury selection statutes are administered to see if the two elements which would warrant relief under Stephens v. Cox, supra, coalesced. We will vacate the judgment and remand for that purpose.

Affirmed in No. 72-2097; vacated and remanded in No. 72-2065.