(dissenting in part).
Regretfully I cannot agree with the decision reached by the majority in affirming the defendant-appellant’s convictions on the substantive counts. The trial judge, in my opinion, committed an error which requires reversal in toto and a remand for a new trial. He either should have severed the defendant-appel*737lant from the trial at the close of the Government’s case after being requested to do so, or he should have given a cautionary instruction covering the evidence of the conspiracy with which Johnson had no connection.
As Judge Stevens demonstrates, the record clearly indicates there were two separate conspiracies, one involving Joseph Altvare, Judy Altvare, Gerald Altier, Adam Jung, Kenneth Getty, and Louis Solone in receiving and selling stolen motor vehicles moving interstate and a ¿second involving only Joseph Altvare and Cecil Ray Johnson in receiving three different stolen motor vehicles. Since the indictment charged a single overall conspiracy among all the defendants and the case was tried on that basis, the court holds that the conviction of Johnson of the conspiracy cannot stand absent a cautionary instruction clearly delineating the two conspiracies and admonishing the jury not to apply evidence relating to one when considering the other. In respect to the substantive counts relating to Johnson, the majority holds that the evidence concerning the conspiracy with which Johnson was not involved did not spill over and taint the finding of guilt on those counts. I disagree.
Laying aside for the moment the question of prejudice and how it should be appraised, there is no question that there was a spillover. The spillover about which we are talking can best be demonstrated, I believe, by graphic illustrations, one when a single conspiracy is charged together with substantive counts and the other when, although a single conspiracy is charged along with substantive counts, two conspiracies are actually shown by the evidence, that is, this case.
In Illustration I a cautionary instruction telling the jury not to consider the evidence of the conspiracy when considering the substantive counts is unnecessary because the conspiracy evidence is relevant to the execution of the object of the conspiracy. A and B are engaged in a joint enterprise growing out of the conspiracy if they act together, or B and C are aiders and abettors conceptually, also the result of the conspiracy, if A acts alone. Cf. Pinkerton v. United States, 328 U.S. 640, 66 S.Ct. 1180, 90 L.Ed. 1489 (1946). In Illustration II a problem arises because the evidence relating to Conspiracy I does not have a joint enterprise or aider and abettor relationship with the substantive charges growing out of Conspiracy II. That re*738lationship exists only with respect to the evidence relating to Conspiracy II.
Since the jurors in the case at bar heard evidence involving a conspiracy in which Johnson had no part, an impermissible spillover resulted as to the substantive crimes charged against him. The critical question then arises as to its prejudicial effect on the jury and what should have been done about it. Before dealing with the quantum of the prejudice emanating from the circumstances of this case, I shall first discuss what should have been done. Two courses were open to the trial judge. He could have granted a severance at the close of the Government’s evidence or he could have given a cautionary instruction. He did neither. As to severance the Government should have known at the close of its evidence what it readily concedes now: that it had proved two separate conspiracies instead of one. The trial judge should have known it also because it was brought to his attention by the defendant’s motion for severance.
But short of granting severance the judge had a duty to admonish the jury during his charge that it should disregard the evidence of the larger conspiracy in deciding the guilt or innocence of the defendant as to the substantive charges. Instead he gave the standard instruction covering the co-conspirator exception to the hearsay rule and then in discussing the substantive counts he merely told the jurors:
Upon retiring — first of all, multiple defendants have been charged with separate crimes in each of the various counts of the indictment. You should give separate consideration and render separate verdicts with respect to each defendant and as to each count. Each defendant is, entitled to have his guilt or innocence determined as to each of the crimes charged from his own conduct and from the evidence which applies to him as if he were being tried alone.
The latter instruction was totally inadequate to meet the problem which the irrelevant evidence and co-conspirator instruction presented.
The court holds that the trial judge should have told the jury not to use the evidence of one conspiracy when considering the existence of the other conspiracy. Given that holding, for exactly the same reasons a cautionary instruc-' tion was required to neutralize, if possible, the prejudicial effect of the extraneous evidence on the substantive charges against Johnson.
There is precedential authority for the position I advance. In United States v. Bentvena, 319 F.2d 916 (1963), thirteen defendants were charged and tried for conspiracy. Three were additionally charged with substantive offenses. The defendants, including the three charged with substantive offenses, were convicted on all counts. The Second Circuit reversed the conviction of the three not only because of lack of evidence as to their participation in the conspiracy but also because of the lack of a cautionary instruction with respect to the substantive counts. There the court said: “Proper cautionary instructions . can presumably limit the prejudicial effect of the evidence presented at the trial directed towards proof of the conspiracy.” 319 F.2d at 955. The court concluded: “The jury charged with that duty could hardly have given their defense the same consideration after they were erroneously found to be narcotics conspirators as it would have had they been tried on the substantive counts alone. The mass of conspiracy evidence militated against their credibility and towards their guilt.” 319 F.2d at 955.
The obverse situation existed in Schaffer v. United States, 362 U.S. 511, 80 S.Ct. 945, 4 L.Ed.2d 921 (1960). There seven defendants were charged with conspiracy. Three substantive counts were added to the conspiracy. count charging various defendants with being the three main perpetrators of a scheme to transport stolen goods. The conspiracy count was dismissed at the close of the Government’s case for failure of proof. The *739case was submitted to the jury on the substantive counts and a conviction resulted. The Second Circuit upheld the convictions, and the Supreme Court affirmed by a five-to-four majority. The Court noted, however, that:
The case was submitted to the jury on each of these counts, and under a charge which was characterized by petitioners’ counsel as being “extremely fair.” This charge meticulously set out separately the evidence as to each of the petitioners and admonished the jury that they were “not to take into consideration any proof against one defendant and apply it by inference or otherwise to any other defendant.” 362 U.S. at 515, 80 S.Ct. at 948.
Moreover, Mr. Justice Clark, speaking for the majority, recognized the possible prejudice and the need for a trial judge to consider a severance:
It appears that not only was no prejudice shown, but both the trial court and the Court of Appeals affirmatively found that none was present. We cannot say to the contrary on this record. Nor can we fashion a hard- and-fast formula that, when a conspiracy count fails, joinder is error as a matter of law. We do emphasize, however, that, in such a situation, the trial judge has a continuing duty at all stages of the trial to grant a severance if prejudice does appear. And where, as here, the charge which originally justified joinder turns out to lack the support of sufficient evidence, a trial judge should be particularly sensitive to the possibility of such prejudice. 362 U.S. at 516, 80 S.Ct. at 948.
Thus Schaffer v. United States and United States v. Bentvena stand for the proposition that a cautionary instruction was required in the circumstances of the case before us.
For guidance in measuring the prejudice visited on the defendant Johnson, we must look to Kotteakos v. United States, 328 U.S. 750, 66 S.Ct. 1239, 90 L.Ed. 1557 (1946). There Mr. Justice Rutledge wrote extensively, in a different but analogous context, about prejudice emanating from inadmissible evidence, harmless error, and the function of appellate review. I quote in part:
Some aids to right judgment may be stated more safely in negative than in affirmative form. Thus, it is not the appellate Court’s function to determine guilt or innocence. Weiler v. United States, supra, 323 U.S. [606] at page 611 [65 S.Ct. 548, at page 551, 89 L.Ed. 495]; Bollenbach v. United States, 326 U.S. 607, 613, 614 [66 S.Ct. 402, 405, 406, 90 L.Ed. 350]. Nor is it to speculate upon probable reconviction and decide according to how the speculation comes out. Appellate judges cannot escape such impressions. But they may not make them sole criteria for. reversal or affirmance. Those judgments are exclusively for the jury, given always the necessary minimum evidence legally sufficient to sustain the conviction unaffected by the error. Weiler v. United States, supra; Bollenbach v. United States, supra.
But this does not mean that the appellate court can escape altogether taking account of the outcome. To weigh the error’s effect against the entire setting of the record without relation to the verdict or judgment would be almost to work in a vacuum. Cf. United States v. Socony-Vacuum Oil Co., supra, 310 U.S. [150] at pages 239, 242 [60 S.Ct. 811, at pages 851, 853, 84 L.Ed. 1129]. In criminal causes that outcome is conviction. This is different, or may be, from guilt in fact. It is guilt in law, established by the judgment of laymen. And the question is not were they right in their judgment, regardless of the error or its effect upon the verdict. It is rather what effect the error had or reasonably may be taken to have had upon the jury’s decision. The crucial thing is the impact of the thing done wrong on the minds of other men, not on one’s own, in the total setting. Cf. United States v. Socony-Vacuum Oil Co., supra, 310 U.S. at pages 239, 242 *740[60 S.Ct. 811, at pages 851, 853]; Bollenbach v. United States, supra, 326 U.S. 614 [66 S.Ct. 402, 406],
This must take account of what the error meant to them, not singled out and standing alone, but in relation to all else that happened. And one must judge others’ reactions not by his own, but with allowance for how others might react and not be regarded generally as acting without reason. This is the important difference, but one easy to ignore when the sense of guilt comes strongly from the record.
If, when all is said and done, the conviction is sure that the error did not influence the jury, or had but very slight effect, the verdict and the judgment should stand, except perhaps where the departure is from a constitutional norm or a specific command of Congress. Bruno v. United States, supra, 308 U.S. [287] at page 294 [60 S.Ct. 198, at page 200, 84 L.Ed. 257]. But if one cannot say, with fair assurance, after pondering all that happened without stripping the erroneous action from the whole, that the judgment was not substantially swayed by the error, it is impossible to conclude that substantial rights were not affected. The inquiry cannot be merely whether there was enough to support the result, apart from the phase affected by the error. It is rather, even so, whether the error itself had substantial influence. If so, or if one is left in grave doubt, the conviction cannot stand. 328 U.S. at 763-65, 66 S.Ct. at 1247-1248 (footnotes omitted).
There is no doubt that a large part of the total evidence submitted to the jury dealt with the conspiracy with which Johnson was not involved. For substantiation I need only refer to footnote 12 of Judge Stevens’ opinion wherein he quotes from the Government’s recital of that evidence. (It is important to note Judge Stevens’ preface to the footnote: “A major part of the facts in the government brief summarized evidence which was not admissible against appellant, but which, under the court’s instructions, the jury may have considered in finding him guilty of conspiracy.”) On the other hand, the Government’s evidence which pointed to Johnson’s guilt of receiving three stolen automobiles and selling two of them, although disputed, could fully support a verdict of guilty if that evidence were considered independently. The trouble is that it is highly probable that it was not considered independently by the jury.
The bulk of the evidence dealt with the larger of the two conspiracies. As the case was tried on a single conspiracy theory, all phases of that alleged conspiracy were lumped together. The mass of evidence concerning unrelated criminal activities of others, especially Joseph Altvare, may have influenced the jury’s decision that Johnson had knowledge of the fact that these automobiles were stolen. The instruction that the statements and acts of one defendant as a conspirator could be imputed to the other conspirators no doubt also helped to produce in the minds of the jurors a gestalt in which the substantive charges against Johnson were inextricably blended with the two conspiracies. Since neither the prosecution nor the judge did anything to disentangle the substantive charges from the mass of evidence — most of it totally irrelevant to those charges — so that they could be considered independently, there is little likelihood that they were so considered.
In this situation we should apply the criteria and admonitions contained in the quotation from Kotteakos v. United States. As appellate judges we might with justification decide from the record that the defendant was guilty. But that, as pointed out by Mr. Justice Rutledge, is not our function. Rather we must ask ourselves: Can we say with conviction that the jurors would not have decided differently had they not been subjected to the confusion which resulted from the way the indictment was drawn and the case was tried? I, as a member of the panel, do not have a firm conviction either way; admittedly, *741I find the question fraught with difficulty-
But of equal importance is the observance of form as it relates to substance. Unless we are indeed confident that a trial error was harmless, should we not insist that criminal trials be conducted so as not to be faced with that perennial hard question: Does the end justify the means? If procedural requirements are waived simply because we as appellate judges believe from an examination of the record that a defendant is guilty in fact, then the safeguards erected for fair trials crumble swiftly.
Judge Stevens indicates that any harm done the defendant by the circumstances here was harmless. Judge Hastings concurs. I respectfully do not agree for the reasons I have stated.