In Part VI of the opinion, entitled Sentencing Charge Regarding Mitigating Circumstances, the opinion as previously published is amended by adding a footnote following the phrase "the trial court's jury instructions were not constitutionally deficient," which phrase appears at 790 F.2d at 1513. The added footnote is as follows:

At trial no objection was made challenging the lack of clarity of the instructions in this regard, providing further support for our conclusion that the instructions adequately conveyed the meaning and function of mitigating circumstances. Peek, 784 F.2d at 1493. Moreover, we note that the evidence in this case included very little in the way of mitigating circumstances. During collateral proceedings, Dobbs trial counsel testified that he had looked diligently but without success for mitigating evidence, but that at sentencing he would have argued against the death penalty, and probably had argued that the shooting was impulsive.

Although in his petition for rehearing Dobbs has not challenged Part IV of the opinion as previously published, the court sua sponte vacates Part IV and substitutes the following in lieu thereof:

IV. COMPOSITION OF GRAND AND TRAVERSE1 JURIES

The state argues that Dobbs has waived this issue for failure to make timely objection. This claim of waiver is controlled by the discussion in Part II, supra. Because at the time of Dobbs' trial in 1974, the relevant Georgia law permitted a defendant to raise federal constitutional claims in state habeas corpus proceedings in the absence of a knowing and intelligent waiver, and because the state does not assert such a waiver, Dobbs did not waive his right to assert claims concerning jury composition. Spencer v. Kemp, 781 F.2d 1458 (11th Cir.1986) (en banc).

Discriminatory selection of grand juries in state courts may be challenged under the Equal Protection Clause of the Fourteenth Amendment. Alexander v. Louisiana, 405 U.S. 625, 92 S. Ct. 1221, 31 L. Ed. 2d 536 (1972); Gibson v. Zant, 705 F.2d 1543, 1546 (11th Cir.1983). In order to make out a prima facie case of grand jury discrimination, Dobbs must generally present evidence of the following three factors:

First, the group allegedly discriminated against must be one that is a distinct class in society. Second, the group must be substantially underrepresented in the grand jury venires ... over a significant period of time. Third, the defendant must show that the selection procedure is not racially neutral or is susceptible to abuse as a tool of discrimination.

Bryant v. Wainwright, 686 F.2d 1373, 1375-76 (11th Cir.1982), cert. denied, 461 U.S. 932, 103 S. Ct. 2096, 77 L. Ed. 2d 305 (1983) (citing Castaneda v. Partida, 430 U.S. 482, 97 S. Ct. 1272, 51 L. Ed. 2d 498 (1977)).2

Dobbs challenges the selection process and composition of the grand jury venire with respect to women. Evidence of grand jury discrimination was presented at the state habeas corpus hearing. The evidence presented by Dobbs with respect to this issue was extremely sparse. In its entirety, it consisted of the testimony of Sidney Porter, Jr., who was one of the jury commissioners in Walker County, Georgia; an exhibit which listed the names of the persons on the grand jury pool which was in effect at the time Dobbs was indicted; and an exhibit showing the 1970 population of Walker County revealing that, counting persons of all ages, there were 24,735 males and 25,956 females. In addition, Dobbs' attorney stated in his place that he had gone through the names on the grand jury list and, based on the first name, determined in his own mind whether the persons were male or female, and accordingly designated each person on the list as either "M" or "F". The attorney stated that if he thought the name were ambiguous he would mark it as being female, except that when there were merely two initials in lieu of a first name, he would designate the person as male. In this manner he counted 117 women on the list, or 34.6%, and 221 men on the list, or 65.4%. Porter's brief and sometimes confusing testimony established that the jury commissioners selected persons at random (e.g., selecting every tenth name) directly from the voter registration list.

We are aware of no cases in which either the Supreme Court or this court has accepted such meager evidence as a prima facie case. First, we are concerned with the treatment of all names consisting of merely two initials as male. For example, if such names were treated as ambiguous, the system followed by the attorney would have provided for them to be marked as female. If this were done, the disparity would decrease to approximately 7%. Second, although Dobbs argues that he has established a 16.6% absolute disparity, neither the state habeas corpus judge nor the district court made a finding of fact that Dobbs proved a 16.6% absolute disparity, or a disparity of any magnitude, or approved Dobbs' method of calculation. We decline to accept the alleged 16.6% absolute disparity as a proven fact.

Moreover, we note that the evidence proffered by Dobbs pertained to the selection of only a single grand jury pool. Under the particular circumstances of this case, and in light of the sparse and inconclusive evidence adduced, we hold that Dobbs has failed to establish a prima facie case. When the jury commissioners have selected persons at random directly from the voter registration list, in such mechanical manner as to virtually eliminate the possibility of bias in that selection, we hold that it is not sufficient to establish a prima facie equal protection violation to establish that women are underrepresented on a single grand jury list by some undetermined amount.

With the above modifications of the previously published opinion, the Petition for Rehearing is DENIED and no member of this panel nor other Judge in regular active service on the Court having requested that the Court be polled on rehearing en banc (Rule 35, Federal Rules of Appellate Procedure; Eleventh Circuit Rule 26), the Suggestion for Rehearing En Banc is DENIED.