KILKENNY, Circuit Judge:
Each appellant is an Indian, who was tried by a jury and convicted of voluntary manslaughter committed on the Northern Cheyenne Indian Reservation in violation of 18 U.S.C. §§ 1112 and 1153. The cases were consolidated for trial and for the arguments on appeal. We hold that each judgment of conviction must be affirmed.
[1125]*1125FACTS
The record shows that Fred Medicine Bull, Sr. (decedent) was involved in an automobile accident on June 18, 1977, resulting in hospitalization for head injuries until June 20th of the same year. After leaving the hospital, decedent set out on a drinking spree and became intoxicated. In the early morning hours of June 21st, he ended up in a small house on the Northern Cheyenne Indian Reservation occupied by James and Emiline Limpy in Lame Deer, Rosebud County, Montana. All the occupants, the appellants, the Limpys, the decedent, and Cheryl Shoulderblades were drinking wine, and became intoxicated. In addition, there were three children and possibly other parties at the house.
Following a remark to Shoulderblades by decedent, he was assaulted and battered by each of the appellants, as a result of which he became unconscious and subsequently died. The facts surrounding the assault are not in sharp dispute. After appellants and Shoulderblades left the house, decedent was taken to Lame Deer Clinic and from there to the Crow Hospital and finally to a hospital in Billings, Montana, where he died of a subdural hematoma on June 26, 1977. Each appellant testified and conceded participation in assaulting the deceased while accusing the other of greater participation. Two doctors called by the government testified that in their opinion, decedent died from the injuries received in the fight on June 21, 1977. A more detailed treatment of the facts will be developed in our treatment of the issues before us.
PROCEDURAL BACKGROUND
On August 24, 1977, in the Billings Division of the United States District Court for the District of Montana, an indictment was returned charging appellant Waters with the crime of voluntary manslaughter in the death of decedent, in violation of 18 U.S.C. §§ 1112 and 1153. On the same date, a similar indictment was returned in the same court against appellant Brady. Following pleas of not guilty, the government moved to consolidate the causes for trial. Over appellants’ objections, the court ordered consolidation.
Later, appellant Waters, joined by Brady, filed a pretrial motion challenging the composition of the grand jury by which he was indicted and of the petit jury before which he was to be tried. He charged that the juries were selected in violation of 28 U.S.C. §§ 1861, 1862, and 1863. In support of the challenge, Waters filed an affidavit which was later supplemented by another affidavit. After a hearing on the challenge to the composition of the grand jury, the court denied Waters’ motion to dismiss the indictment.
Later, each appellant moved to sever his trial pursuant to the provisions of Rule 14, FRCrimP, contending that a joint trial would be highly prejudicial. The motion to sever was denied, and the cause proceeded to trial. Following a four day jury trial, both appellants were found guilty. Waters was sentenced to imprisonment for a period of ten years, and Brady for a period of eight years.
ISSUES
The issues before us may be summarized as follows:
I.
Each appellant contends that there was insufficient evidence to support a conviction.
II.
Each appellant contends that there was erroneous joinder of the cases for trial under Rule 13 and an improper refusal to sever the causes under Rule 14, FRCrimP.
III.
Each appellant contends that he was denied a fair trial.
IV.
Each appellant contends that there was a violation of the rules stated in Bruton v. United States, 391 U.S. 123, 126-128, 88 S.Ct. 1620, 20 L.Ed.2d 476 (1968).
[1126]*1126V.
Each appellant contends that there was a systematic exclusion of Indians from the grand and petit jury panels in violation of 28 U.S.C. §§ 1861 et seq.
SUFFICIENCY OF THE EVIDENCE
Under this heading, appellants, in general, claim: (a) that the government failed to establish beyond a reasonable doubt that the actions of the appellants, or either of them, caused decedent’s death; (b) that at most the government’s evidence showed that a brawl had occurred between appellants and decedent in which blows had been struck, but did not establish beyond a reasonable doubt the requisite intent to find that either appellant was guilty of voluntary manslaughter.
(a) The government called two doctors who testified that the injuries inflicted by appellants caused decedent’s death. That the decedent, while intoxicated, was savagely assaulted and battered by the appellants is beyond question. The witness Emi-line Limpy testified she saw Waters kick decedent at least five times. After looking away and returning to a window, Waters was still “kicking and jumping on Fred.” When this was over, Waters turned to Brady and told him it was his chance and Brady then jumped on the decedent’s stomach at least twice. Apparently this wasn’t enough for the evidence shows that Waters then continued to beat and kick the defenseless decedent.
About this time, Waters pulled the pants off decedent and proceeded to kick the helpless decedent between the legs. Shoul-derblades fully corroborated the testimony of Limpy. She testified that Brady commenced the fight by slapping decedent and that Waters joined in the attack by hitting him. After Brady started hitting him, decedent fell out of his chair. Decedent started to get up after he was knocked down the first time but fell to the floor. Shoulder-blades said that at this time both men were still beating decedent. She testified she saw both appellants repeatedly kick decedent and strike him with their fists. She said she then stopped her cousin, appellant Waters, and felt that decedent needed a doctor to save his injured eye. She then summoned the police and an ambulance. After requesting the aid, Shoulderblades left the house while decedent remained injured inside.
Brenda Brady, a twelve year old girl, was asleep when the affair started, but once awake she observed essentially the same conduct as the others. She saw Waters hit the decedent many times and after one blow, the decedent fell back. She observed appellants taking the decedent’s pants down and after he was on the floor, she saw Waters kicking the decedent. Brady did the same. Then Waters pulled the decedent back and kicked him in the shoulder, as did Brady. Later, appellants dragged the decedent to the door while administering more kicks. At the trial, each appellant, by inference at least, accused the other of striking, beating, and kicking the decedent.
The photographs of decedent’s face and body which were taken shortly after the occurrence and introduced in evidence, fully corroborate the testimony that his face and head had been cruelly battered, and that the blood, contusions, and bruises were of recent origin. Each of the doctors were acquainted with these facts when answering the hypothetical questions and based on these facts each expressed an opinion that death resulted from the blows received in the assault, rather than from head injuries suffered in the automobile accident which occurred three days prior to the attack.
We have no difficulty in holding that there was evidence sufficient to go to the jury on the cause of death and that the jury’s finding under the court’s instructions that the death was caused by the appellants’ assault is supported by substantial evidence.
Appellants argue that the doctors’ testimony did not exclude every possible hypothesis as to the cause of death. This argument is answered by United States v. Daniels, 549 F.2d 665 (CA9 1977), where it is said:
[1127]*1127“In determining the sufficiency of circumstantial evidence, the question ‘is not whether the evidence excludes every hypothesis except that of guilt but rather whether the trier of fact could reasonably arrive at its conclusion.’ ” At 668.
The doctors here could reasonably have concluded from the direct evidence that the decedent’s death was caused by appellants’ brutal attacks. Moreover, circumstantial evidence is intrinsically no different from direct evidence. United States v. Kelly, 527 F.2d 961, 965 (CA9 1976); United States v. Nelson, 419 F.2d 1237, 1240-41 (CA9 1969). On an issue such as this, we must accept the tenet that all reasonable inferences supporting the verdict are in favor of the government as the prevailing party below, Diaz-Rosendo v. United States, 357 F.2d 124, 129 (CA9 1966), cert. denied, 385 U.S. 856, 87 S.Ct. 104, 17 L.Ed.2d 83, and that it
is the exclusive function of the jury to determine the credibility of the witnesses, resolve evidentiary conflicts and draw reasonable inferences from proven facts. United States v. Nelson, supra, at 1241. Another basic rule which we must follow is that circumstantial and testimonial evidence are indistinguishable insofar as the jury’s fact finding function is concerned and that circumstantial evidence can be used to prove any fact, including facts from which another fact is to be inferred. United States v. Polizzi, 500 F.2d 856, 904 (CA9 1974), cert. denied, 419 U.S. 1120, 95 S.Ct. 802, 42 L.Ed.2d 820 (1975).
(b) The appellants advance the corresponding argument that there is no evidence to support a finding of guilty intent to commit the crime of voluntary manslaughter. We disagree. It is the jury’s exclusive function to determine the credibility of the witnesses, resolve evidentiary conflicts and draw reasonable inferences from the proven facts. United States v. Kelly, supra. It is difficult to imagine a state of facts which present a stronger case of direct participation than that here presented. The savage assaults on decedent by each of appellants, as shown by the record, beyond question form an adequate basis for a jury finding of specific intent. The court gave the jury the standard instruction on reasonable doubt, and a standard instruction on intent.1
The Supreme Court has said that our whole jury system is based upon the recognized ability of the jury to follow instructions. Opper v. United States, 348 U.S. 84, 95, 75 S.Ct. 158, 99 L.Ed. 101 (1954). Moreover, we are required to assume that the jury followed the instructions. United States v. Strand, 574 F.2d 993, 996 (9 Cir. 1978); Gray v. Shell Oil Co., 469 F.2d 742, 752 (CA9 1972), cert. denied, 412 U.S. 943, 93 S.Ct. 2773, 37 L.Ed.2d 403 (1973); Vitello v. United States, 425 F.2d 416, 422, (CA9 1970), cert. denied, 400 U.S. 822, 91 S.Ct. 43, 27 L.Ed.2d 50 (1970); Silverthorne v. United States, 400 F.2d 627, 641 (CA9 1968). Even though the appellants were intoxicated, the evidence of intent was clearly sufficient to go to the jury. These contentions are meritless.
CONSOLIDATION AND FAILURE TO SEVER
Each appellant challenges the court’s consolidation of the cases for trial and its subsequent failure to grant the motions for separate trials. These contentions are to be tested by whether a joint trial would be so prejudicial to a defendant as to require the exercise of the trial judge’s discretion in only one way; to order a separate trial. The burden of proving this prejudice is a difficult one and the ruling of the trial judge will rarely be disturbed. It is [1128]*1128not sufficient that a separate trial might offer a defendant a better chance of acquittal. Rather, the ultimate question is whether under all of the circumstances, it is within the capacity of the jurors to follow the court’s admonitory instructions and, correspondingly whether they can collate and appraise the independent evidence against each defendant solely upon that defendant’s own acts, statements, and conduct. United States v. Marshall, 532 F.2d 1279, 1282 (CA9 1976); United States v. Campanale, 518 F.2d 352, 359 (CA9 1975), cert. denied sub nom., Matthews v. United States, 423 U.S. 1050, 96 S.Ct. 777, 46 L.Ed.2d 638 (1976).
Conflicting and antagonistic defenses being offered at trial do not necessarily require granting a severance, even if . hostility surfaces or defendants seek to blame one another. United States v. Perez, 489 F.2d 51, 68 (CA5 1973); cert. denied, 417 U.S. 945, 94 S.Ct. 3067, 41 L.Ed.2d 664 (1974); Baker v. United States, 329 F.2d 786 (CA10 1964), cert. denied, 379 U.S. 853, 85 S.Ct. 101, 13 L.Ed.2d 56; cf. United States v. Kozloski, 453 F.2d 889 (CA9 1971).
[H] On the record before us, it is clear that the prejudice which either appellant may have suffered from the testimony of the other is relatively slight. It is undisputed that each appellant participated in the incident. Consequently, it would only be natural for one to try to place the blame on the other. The jury had the responsibility of assessing each of the appellant’s credibility. Moreover, the testimony of each appellant was merely cumulative of the government’s case against the other and considering the simplicity of the case, there is no sound reason to suggest that the members of the jury, being properly instructed as they were, could not realistically appraise the evidence against each appellant. Drew v. United States, 118 U.S.App.D.C. 11, 17-18, 331 F.2d 85, 91-92 (1964), is closely in point on our record.
In passing on this record, we must be guided by our general rule that joint trials of persons charged with committing the same offense expedites the administration of justice, reduces the congestion of trial dockets, conserves judicial time, lessens the burdens upon citizens to sacrifice time and money to serve on juries and avoids the necessity of recalling witnesses who would otherwise be called upon to testify only once. United States v. Gaines, 563 F.2d 1352, 1355 (CA9 1977); United States v. Patterson, 455 F.2d 264 (CA9 1972). While a great disparity in proof may be sufficient to allow a severance in certain cases, the fact that the evidence against one codefendant is more damaging than the evidence against another one is not a ground for severance. United States v. Gaines, supra, at 1355. Ultimately, the question is whether the jury can reasonably be expected to compartmentalize the evidence as it relates to separate defendants in the light of its volume and limited admissibility. United States v. Kaplan, 554 F.2d 958 (CA9 1977); United States v. Campanale, supra, at 359. Here, the trial was relatively simple, consuming only three days. Moreover, the jury was fully and carefully instructed on the limitations on the use of impeaching evidence and the use, if at all, of declarations made outside the presence of each appellant. Neither appellant has demonstrated a likelihood of prejudice sufficient to set aside the trial court’s exercise of discretion in granting the consolidation or in refusing to grant the severance.
THE BRUTON ISSUE
During the prosecutor’s cross-examination of the appellant Waters, he said he did not at any time see appellant Brady strike the deceased. In addition, he testified that he never at any time told anyone that Brady had hit or kicked decedent. On rebuttal, the government, for impeachment purposes, called Jones, an FBI agent, who testified that Waters told him during the June 21st interview that he had tried to stop Brady from striking decedent.
Appellant Brady, in his direct testimony, said: (1) that Waters hit decedent; (2) that he did not see Waters kick decedent in the face; (3) that he told Agent Jones that [1129]*1129Waters had kicked decedent in the face because he was not fully awake at the time of the interview; and (4) that he told Jones that George Whiteman had helped place decedent on the bed.
Jones’ rebuttal testimony, entered solely for the limited purpose of impeaching Brady’s testimony, was to the effect: (a) that Brady made no mention of Whiteman during their interview; and (b) that Brady had told him that Waters had hit and kicked decedent on the head.
The appellants’ Bruton claim is based entirely on these two instances of rebuttal testimony. Bruton is grounded on the Sixth Amendment right to confrontation which is effectively denied to a criminal defendant when out of court statements, made outside a defendant’s presence, are introduced at trial and the maker of the statement is not present, does not take the stand, or invokes a privilege so as to deprive the defendant an opportunity to cross-examine. Bruton, 391 U.S. at 124-26, 88 S.Ct. 1620.
In contrast to the Bruton background, here each appellant who made an extra-judicial statement was present in court, actually took the witness stand, and subjected himself to cross-examination, both by the government and by the other appellant. Additionally, there was substantial eye-witness testimony concerning the very subjects covered by the statements made to Agent Jones. Consequently, these statements failed to add “substantial, perhaps even critical, weight to the Government’s case in a form not subject to cross-examination.” Bruton at 128, 88 S.Ct. at 1623. Not to be overlooked is the government’s well founded right and need to be afforded the opportunity to impeach effectively the credibility of a defendant who testifies in his own behalf. We hold that the Bruton rule does not apply to these facts.
FAIR TRIAL ISSUES
We consolidate a number of appellants’ other contentions under this heading because they deal mainly with the fairness of the trial.
(a) It is claimed that the photographs of the decedent lying on the floor, taken after the police arrived, were highly inflammatory, presented merely cumulative evidence, and were unnecessary to prove any issue in the case. We disagree. One of the principal issues was the cause of death, Beyond question, these photographs demonstrated that someone had recently battered, bloodied, and bruised decedent’s face, and these pictures, no doubt, played a part in the doctors’ conclusions that death was caused from the injuries received in the fracas on the 21st, rather than from the injuries received in the previous automobile accident. True enough, the photographs were not for the faint hearted, but it is up to the trial judge to balance the prejudicial versus the probative effect of this type of evidence and the exercise of his discretion is
rarely disturbed. United States v. Cannon, 472 F.2d 144, 145 (CA9 1972); Maxwell v. United States,368 F.2d 735, 740 (CA9 1966). This type of photograph is inadmissible only when the picture is of such gruesome and horrifying nature that its probative value is outweighed by the danger of inflaming the jury. Rivers v. United States, 270 F.2d 435, 437-38 (CA9 1959), cert. denied 362 U.S. 920, 80 S.Ct. 674, 4 L.Ed.2d 740 (1960). We have examined the photographs and hold that the judge did not abuse his discretion in permitting their introduction in evidence.
(b) Waters argues that he was denied a fair trial by the government’s action in calling James Limpy as a witness in its case in chief when it knew that his testimony was so self-contradictory and groundless that it had no probative value. The government says that it had no knowledge that Brenda Brady was going to testify that Limpy was asleep during the entire affair, and there is no proof to the contrary. In any event, three other witnesses corroborated all, or practically all, of Limpy’s testimony. Waters also argues that the prosecution compounded this mistake by stating in his final argument that the jury should entirely disregard Limpy’s testimony. [1130]*1130Even assuming that the prosecutor should have known that Brady would testify that Limpy was asleep during the affair and further assuming that the prosecutor should not have emphasized his testimony by asking the jury to disregard Limpy’s testimony, we find that the record as a whole does not present a set of circumstances which requires a reversal. It may show inept preparation for trial and lack of good judgment in argument. At worst it presents only harmless error.
(c) Waters contends that the manner in which Brenda Brady’s testimony was presented was highly prejudicial. He does not object to Brenda’s competency as a witness, nor to the substance of her testimony. He does, however, object to the fact (1) that the prosecution was permitted to ask her leading questions on direct examination; (2) that her mother was permitted to sit beside her while she was on the witness stand; and (3) that Waters’ counsel was requested by the prosecutor to assure Brenda that he would “not be mean to her when he asked her questions.”
The record shows that the government first sought court permission to lead this twelve year old witness. This is not an unusual practice. In any event, appellant made no objection. However, he did object to the prosecutor’s request that Waters’ counsel reassure the youngster that he would “ . . . not be mean to her when he asked her questions.” The record clearly establishes that the trial judge forthwith intervened and assured the witness that she had nothing to fear. Although the government’s request of appellant’s counsel was improper and wholly unwarranted, we note that after considerable discussion about the remark, appellant’s counsel said he would be satisfied if the prosecutor would make a proper statement in open court. The judge, then speaking to the jury, said that he would call upon the prosecutor to make a statement. The statement2 apparently satisfied appellant’s counsel inasmuch as no exception was taken to it nor the court’s order to proceed. In these circumstances, we find no error.
(d) Under the “excited utterance” exception to the Hearsay Rule, FRE 803(2), Waters’ counsel, on cross-examination, attempted to question the investigating police officer regarding what the Limpys had told him about the fracas in which decedent was killed. The court sustained the government’s hearsay objection. Waters argues that he intended to impeach the Limpys’ credibility by showing that they told the police officer that there had been a stabbing.
We agree with the trial judge that the proper practice would have been for Waters to propound the impeachment questions during the cross-examination of the Limpys. The officer could then be recalled for impeachment purposes. Neither of the Limpys had been called at this time, but later took the stand as government witnesses. In sustaining the government’s objection, the court noted that the proper practice would be questioning the Limpys when they took the witness stand. Inasmuch as appellant failed to make an offer of proof, we do not know the nature of the alleged statements. Beyond that, the proper practice if he intended to use these statements as part of his own case, would have been to call the policeman in the defense and ask him the questions. Since there is no offer of proof, we have no foundation on which to decide whether the alleged statements were or were not part of the res [1131]*1131gestae or within the hearsay exception of Rule 803(2). We find no error on this contention.
(e) We have examined appellants’ other contentions, including the attack on the instructions given on inferences and presumptions, and find them groundless. The instruction on inferences is in the customary form and so is the instruction on presumptions which concludes with the language “ . . . unless and until outweighed by evidence in the case to the contrary, the law presumes that a person is innocent of crime or wrong and that the law has been obeyed.” The other fair trial contentions are without merit.
MONTANA JURY SELECTION PLAN
Each appellant filed a pretrial motion to dismiss the indictment because the grand and petit juries were selected in violation of 28 U.S.C. §§ 1861, 1862, and 1863.
In 1968 the Congress enacted the Jury Selection and Service Act, 28 U.S.C. § 1861, et seq. [the Act], The dispute in this case can be reduced to a recitation of the first sentence of the statute. “It is the policy of the United States that all litigants in Federal courts entitled to a trial by jury shall have the right to grand and petit juries selected at random from a fair cross section of the community in the district or division wherein the court convenes.” [Emphasis supplied]. Appellants contend that the Montana Jury Selection Plan [the Plan] as here applied did not meet this stated purpose, and consequently they were discriminated against. We agree with appellants that Indians form a cognizable group able to challenge the cross sectional makeup of jurors selected from the Plan.
The statutory scheme for the sections in question is actually quite simple. § 1861 states the basic policy behind the enactment, § 1862 prohibits discrimination in the selection of jurors and § 1863 provides for establishing plans for meeting both §§ 1861 and 1862.
§ 1863(b)(2) begins with the statement that either voter registration lists or lists of actual voters shall be the proper method for assembling a fair cross section of the community to serve on juries. Voter lists were chosen because they provide the best available residential lists for a community and are constantly updated. 1968 U.S.Code Cong. & Admin.News, p. 1799.
Although the statute does allow for supplementation of voter lists the legislative history is clear that additional sources can only supplement and not supplant voter lists. Moreover, the legislative history indicates that use of supplemental sources should be used only when the voter lists deviate substantially from the makeup of the local community. As the House Report noted,
“The voting list requirements, together with the provision for supplementation, is therefore the primary technique for implementing the cross sectional goal of this legislation. The bill uses the term ‘fair cross section of the community’ in order to permit minor deviations from a fully accurate cross section. The voting list need not perfectly mirror the percentage structure of the community. But any substantial percentage deviations must be corrected by the use of supplemental sources. Your committee would leave the definition of ‘substantial’ to the process of judicial decision.” U.S.Code Cong. & Admin.News 1968, p. 1794.
Thus, we can conclude that the use of supplemental sources is limited to when the court determines that the voter lists do not represent a fair cross section of the community. However, this does not answer the fundamental question left to us so clearly by the Congress; when are supplemental source lists required to be added to voter lists to implement the statutory purposes of the Act?
This question is one that seems to have escaped a concise resolution by the courts. The Ninth Circuit cases appear to be in disarray. In the earliest decision in this area, our court in United States v. Parker, 428 F.2d 488, 489 (CA9 1970), cert. denied, 400 U.S. 910, 91 S.Ct. 155, 27 L.Ed.2d 150, intimated that the oft cited holding from [1132]*1132Swain v. Alabama, 380 U.S. 202, 85 S.Ct. 824, 13 L.Ed.2d 759 (1965), that the plaintiff must meet the burden of showing a systematic exclusion of an identifiable group within the community before he has shown a prima facie case, would apply to the Act. Although the Parker decision did not impose this level of proof on challenges brought under the Act, later cases did. In United States v. James, 453 F.2d 27, 29 (CA9 1971), the court adopted the Swain language as a prerequisite to prevail under the Act. The court wrote,
“Appellant did not offer any proof of discrimination in the compilation or use of the voter registration lists or of systematic exclusion of any identifiable group within the community. There was no ‘showing of a discriminatory purpose or effect of the jury selection plan utilized in this case.’ ”
Moreover, the James court added the requirement that the plaintiff must show some type of discriminatory purpose or effect as part of a cause of action under the Act. Not surprisingly, very similar language can be found in Swain. Id. 380 U.S. at 208-09, 85 S.Ct. 824. Thus, we see the first signs of a test emerging for determining if a jury is a fair cross section of the community under the Act. As early as Parker and James the test under the statute appears to be the same as that adopted by Swain for constitutional challenges to the composition of juries.
Other Ninth Circuit authority continued to hold that proof of a systematic exclusion was essential for success under the Act. United States v. Ware, 473 F.2d 530, 537 (CA9 1973); United States v. Ross, 468 F.2d 1213, 1217 (CA9 1972), cert. denied, 410 U.S. 989, 93 S.Ct. 1500, 36 L.Ed.2d 188 (1973). Then the court seemed to require a greater burden of proof for in United States v. Fernandez, 497 F.2d 730, 733 (CA9 1974), cert. denied, 420 U.S. 990, 95 S.Ct. 1423, 43 L.Ed.2d 670 (1975), we said, “the defendants have the burden of showing, prima facie, discriminatory practices.” However, more recent Ninth Circuit authority has not followed the suggestion of Fernandez that a purposeful showing is required to prevail under the Act.
Until its decision in United States v. Potter, 552 F.2d 901, 903 (CA9 1977), this court’s standard for determining whether a voter list was a fair cross section of the community was whether the plaintiff had shown a systematic exclusion of any identifiable group in the community. However, after determining that there was no cognizable class, the Potter court erroneously said, “[w]e do not reach the question of whether systematic exclusion is required or has been established.” Id. at 903. [Emphasis supplied]. Then the court went on to create a new circuit test; whether there was a “substantial deviation” between an identifiable groups’ numbers in the community and its representation on the voting list. Thus, Potter is clearly at odds with previous Ninth Circuit authority which held that proof of systematic exclusion was required. Potter would allow mere proof of “substantial deviation” to suffice as a violation of the Act. Much the same view was taken in United States v. Kleifgen, 557 F.2d 1293, 1296-97 (CA9 1977). There the court wrote, “[o]nly when this deviation becomes substantial is a defendant deprived of his right to be judged by a grand jury chosen from a fair cross section of the community.” Id. at 1296. The Kleifgen court noted that it did not decide if the party challenging the grand jury must prove some sort of purposeful discrimination. Id. at 1296 n. 5.
Although there would appear to have been a sudden shift in Ninth Circuit authority with Potter, this can be explained by reviewing the Supreme Court’s decision in Castaneda v. Partida, 430 U.S. 482, 97 S.Ct. 1272, 51 L.Ed.2d 498 (1977). There the challenge was not statutory, but constitutional. The Court held that “[o]nce the defendant has shown substantial under-representation of his group, he has made out a prima facie case of discriminatory purpose, and the burden then shifts to the State to rebut that case.” Id. at 494 — 95, 97 S.Ct. at 1280. Thus, the Supreme Court was adding to the constitutional basis for challenging grand and petit jury selection proof of “substantial deviation.” Since Castaneda [1133]*1133was a constitutional challenge and our court in both Potter and Kleifgen has apparently applied its holding to challenges arising under the Act, the court was again demonstrating that in order for any party to prove that supplemental source lists should be added to voting lists pursuant to § 1863(b)(2) he must show that the jury selection plan produced juries constitutionally infirm. The Fifth Circuit in Foster v. Sparks, 506 F.2d 805, 816-17 (CA5 1975), held there is no distinction between the test for a constitutionally selected jury and a jury properly assembled under the Act. We choose to follow Foster as the law of this circuit.
Therefore, to prevail under the current constitutional or statutory standards, appellants must prove either a systematic exclusion of an identifiable class or “substantial deviation” between identifiable groups. This they have not done. Although the appellants couch their argument in terms of the Plan systematically excluding Indians from jury service, the record demonstrates that appellants have neither alleged nor proved any discrimination in the Plan as drafted or as applied. Thus, under the teaching of Swain and its progeny there has been no showing of systematic exclusion. Rather than resting on systematic exclusion, the appellants, in essence, rely on statistics to prove that the Plan results in a “substantial deviation” between identifiable groups. However, a close examination of the Plan as well as appellants’ statistics of “substantial deviation” demonstrate that appellants have not met their burden of proof.
Under the Plan as adopted by the United States District Court for the District of Montana, and approved by this court, the names of prospective jurors are selected at random from voter registration lists furnished by the county clerk and recorder of each county in the state. Since there is but one United States District Court in Montana and because the state is so large geographically, the district court has seen fit to divide the state into six divisions. Master jury boxes are maintained for each division in the district.
The plan operates very simply. The prospective jurors are selected on a random basis. As petit jurors are needed in each division the judge orders the, “[c]lerk to publicly draw at random from the master jury box . . . ” the names of three hundred prospective jurors. Each person whose name is drawn receives a juror qualification questionnaire. After the questionnaires are returned, the names of all qualified jurors are placed in the qualified jury box for that particular division. Names are drawn from that box for service on petit juries.
The grand jury selection process is quite similar. Since there is only one grand jury serving in Montana at any one time, each division draws grand jurors from its qualified jury box in proportion to the division’s number of registered voters as compared with the number of registered voters in the statewide district.
The indictments before us were returned by a grand jury so selected. The petit jurors were drawn from the qualified jury box of the Billings Division. Montana has 56 counties and seven Indian reservations. The Billings Division encompasses 15 counties. The court convenes only in Billings. Two of the 15 counties, Rosebud and Big Horn, include the Indian reservations for the Northern Cheyenne and Crow tribes.
Appellants’ basic argument is that although the latest census figures show that approximately 30% of the population of Rosebud county was Indian, only 6Va% of those selected from that county for inclusion in the division’s master jury box were Indian. The fundamental flaw in appellants’ argument is that the Billings Division consists of 15 counties rather than just one, and that a fair cross section of the community must be achieved within the division as a whole rather than any of the division’s component counties. The language of the statute makes this conclusion crystal clear, “litigants . . . shall have the right to grand and petit juries selected at random from a fair cross section of the community in the district or division wherein the court [1134]*1134convenes.” 28 U.S.C. § 1861. [Emphasis supplied]. Surely it was not the intention of Congress that each petit or grand jury represent a fair cross section of each county or community in all of Montana. In order to achieve a working balance between a fair representation of various groups on juries and a manageable system of selecting jurors the Congress chose a district or division based plan. 1968 U.S.Code Cong. & Admin. News, p. 1793.
Additionally, appellants have presented evidence that Indians make up only 4Vi% of the registered voters in Rosebud County while constituting 30% of the total county population. However, the appellants’ own evidence shows that 6.5% of the names from Rosebud County submitted from registered voter lists for use in the master jury box were Indians. Thus, if Indians make up only 4V<i% of Rosebud’s registered voters and yet are submitted to the master jury box at a 6.5% rate there is no indication of “substantial deviation” under the Plan or the Plan as applied on a division wide basis.
Because we have held that each appellant must show a “substantial deviation” on a district or a division wide basis before he has stated his prima facie case, we must now examine the appellants proof before us. We do know that Montana is made up of approximately 700,000 residents of which about 27,000 or 4% are Indian. Totally absent from the record is evidence of how many Indians and/or non-Indians reside in the Billings Division or how many have been included in the master jury list for that division. Before a court can make the meaningful type of comparison for determining “substantial deviation” as outlined in United States v. Potter, supra, those challenging the makeup of juries must offer statistical proof on a district or division basis.3 This appellants have not done.
Several other claims made by appellants are unconvincing. It is of no importance that Indians may have a tendency not to register to vote in non-Indian elections. The legislative history of the Act specifies that, “[n]o economic or social characteristics prevent one who wants to be considered for jury service from having his name placed in the pool from which jurors are selected.” 1968 U.S.Code Cong. & Admin.News, p. 1795.
United States v. Antelope, 555 F.2d 1378 (CA9 1977), cited by appellants is concerned with criminal law on Indian reservations and has nothing whatsoever to do with jury plans. What is there said by the court must be confined to the facts of the case. Likewise, the recent Supreme Court cases of United States v. Wheeler, 435 U.S. 313, 98 S.Ct. 1079, 55 L.Ed.2d 303 (1978), and Oliphant v. Suquamish Indian Tribe, 435 U.S. 191, 98 S.Ct. 1011, 55 L.Ed.2d 209 (1978), speak to criminal law on Indian reservations and have no application to jury selection plans.4
CONCLUSION
We conclude that appellants had a fair trial and that the judgments of conviction must be affirmed.
IT IS SO ORDERED.