After a jury trial Geoffrey Honneus was convicted in the district court upon charges arising from his participation in a scheme to buy marihuana in bulk in Jamaica and transport it back to New England in an auxiliary yacht chartered for the purpose. His contentions on appeal, and the facts material to them, are discussed below.

* Honneus contends that it was improper to convict and sentence him under three conspiracy counts all stemming from the same conspiracy. In the six count indictment Honneus and others were charged with three substantive offenses and three counts of conspiring to accomplish the substantive offenses. The substantive counts were for importing marihuana in violation of 21 U.S.C. 952(a); distributing and possessing marihuana with intent to distribute in violation of 21 U.S.C. 841(a)(1); and smuggling marihuana in violation of 18 U.S.C. 545. The count charging a conspiracy to import and that charging a conspiracy to distribute and possess were brought under separate drug conspiracy statutes, respectively 21 U.S.C. 963 and 21 U.S.C. 846. The count charging a conspiracy to smuggle was brought under the general federal conspiracy statute, 18 U.S.C. 371.

Identical overt acts were listed in the indictment under each conspiracy count; and Honneus argues that since the evidence disclosed but a single agreement, the fact that the objects of the agreement amounted to three distinct crimes did not transform one conspiracy into three. In Braverman v. United States, 317 U.S. 49, 53, 63 S. Ct. 99, 102, 87 L. Ed. 23 (1949), the Supreme Court said:

'Whether the object of a single agreement is to commit one or more crimes, it is in either case that agreement which constitutes the conspiracy which the statute punishes. The one agreement cannot be taken to be several agreements and hence several conspiracies because it envisages the violation of several statutes rather than one.'

In Braverman the defendant had been sentenced upon each of several conspiracy counts brought under 37 of the then Criminal Code charging a conspiracy to violate different provisions of the Internal Revenue laws. The Supreme Court held that only a single sentence could be imposed and remanded for resentencing.

In Braverman, where all counts violated but a single conspiracy statute, it could without difficulty be said that the single conspiracy differed 'from a single act which violates two statutes.' Id. at 54, 63 S.Ct. at 102. In the present case, each conspiracy count was brought under a different federal conspiracy statute. In addition, the statutory penalties vary slightly as between statutes, and there appears to be some difference as to standards of proof respecting overt acts. The Government contends that these factors turn the one criminal agreement into a trilogy of crimes.

We disagree. While undoubtedly Congress meant to attack the drug trade with severity, see Gore v. United States, 357 U.S. 386, 78 S. Ct. 1280, 2 L. Ed. 2d 1405 (1957), we doubt that it meant to authorize, or could authorize, a court to impose three punishments for one conspiracy. Congress may treat different aspects of the same conduct as separate crimes only if there is a meaningful distinction between the elements constituting each offense. See id.; American Tobacco Co. v. United States, 328 U.S. 781, 66 S. Ct. 1125, 90 L. Ed. 1575 (1946); Blockburger v. United States, 284 U.S. 299, 304, 52 S. Ct. 180, 76 L. Ed. 306 (1931). Here, unlike in American Tobacco, the agreement was identical in all three alleged conspiracies; the Government concedes as much. In Braverman the Court held that all separate criminal objects are 'embraced' within the single continuing agreement. 'The conspiracy is the crime, and that is one, however diverse its objects.' 317 U.S. at 54, 63 S. Ct. at 102, quoting from Frohwerk v. United States, 249 U.S. 204, 210, 39 S. Ct. 249, 63 L. Ed. 561 (1918) (Holmes, J.). We hold that charging a single illicit agreement under several statutes does not make it more than one conspiracy. United States v. Noah, 475 F.2d 688 (9th Cir. 1973); United States v. Mori, 444 F.2d 240 (5th Cir. 1971).

We do not question Congress' power to enact the three statutes nor the Government's power to charge under them; but we find only one crime for which only a single sentence could be imposed. We therefore disagree with Honneus that he is entitled to a new trial. Cf. Braverman, supra, 317 U.S. at 55, 63 S. Ct. 99; Mori, supra, 444 F.2d at 246. An election was required in connection with sentencing, but not before. For practical reasons, we think the Government was entitled to request separate verdicts under all three statutes. The conspiracy with its several objects could, it is true, have been charged in a single count, Braverman, supra, 317 U.S. at 54, 63 S. Ct. 99. But had this been done, the court would not have known whether Honneus was found by the jury to have conspired to achieve all or only some of the illegal objects, and thus would not have known which of the statutory sentences were available. Other problems could perhaps have arisen had the court provided for special verdicts or asked special questions. See United States v. Spock, 416 F.2d 165 (1st Cir. 1969). Alternatively, if the Government had been forced to elect before trial, it would have had to gamble on proving one illegal object when there were, in fact, several, any one of which would have supported a finding of the agreement's illegality.

We thus find no error, nor do we believe there was prejudice, in submitting all three counts to the jury. But since there was but one crime, only one sentence could be imposed. We accordingly vacate the separate concurrent sentences under Counts 1, 3 and 5 and remand for sentencing upon any of the three counts the United States may select. Surplus counts are to be dismissed.1

Respecting Count 4,2 which charged the distribution and possession of marihuana with intent to distribute, Honneus makes several arguments about venue and jurisdiction. First he says that the indictment was defective for failure to allege where the offense took place. But it is well established that an indictment is not legally insufficient for failure to include such an allegation. See, e.g., United States v. Branan, 457 F.2d 1062 (6th Cir. 1972); Carbo v. United States, 314 F.2d 718 (9th Cir. 1963), cert. denied, 377 U.S. 953, 84 S. Ct. 1625, 12 L. Ed. 2d 498 (1964). Honneus would have been entitled to the information had he sought a bill of particulars. He did not do so. Our attention is directed to a co-defendant's motion for particulars which was allowed and to which, Honneus now argues, the Government responded inadequately. This is of no assistance to Honneus. His sole connection with the motion is a paper filed later by his attorney^2A purporting to join in 'each and any and all of the co-defendant's motions, requests and petitions.' Apart from the fact that it was 'denied' by the court, the shotgun joinder obviously did not alert the court that Honneus was seeking information as to the place of possession, or that he was dissatisfied with information in the Government's bill of particulars filed in response to a co-defendant's motion.

Also in conjunction with Count 4, Honneus argues that he was entitled to an instruction on 'venue'. There was evidence that Honneus, who lived in Massachusetts, flew to Jamaica and, with others, purchased the marihuana there and arranged to have it loaded on a vessel which had been sailed from Duxbury, Massachusetts to Jamaica expressly to make the pickup. After the vessel was loaded, it sailed back to Massachusetts, where its owner contacted Honneus and arranged to have the marihuana off-loaded in Maine coastal waters. Finally there was evidence from which the jury could have found that Honneus had in his possession and sold some of the marihuana in Massachusetts prior to his arrest in that state. The latter evidence of illicit possession and distribution within Massachusetts was ample to support both the jury's verdict under Count 4 and venue within the District of Massachusetts, F.R.Crim.P. 18; see 18 U.S.C. 3237.

But Honneus argues that since the jury may have rested its conviction only upon the evidence of possession within Jamaica, the court erred in not instructing that possession had to be found in Massachusetts. The premise of this argument is doubtful. One might question whether a jury which found Honneus guilty on all counts, including ones for smuggling and conspiracy, was likely to have rejected the evidence of possession and distribution within Massachusetts. It is also quite possible, although we do not decide the question, that possession in Jamaica by a U.S. resident intending to distribute within the United States and later doing so, is within the reach of United States criminal jurisdiction and was prosecutable in the district of Massachusetts where Honneus was arrested.3 18 U.S.C. 3238.

[9,10] But even if it is assumed that Honneus would have been entitled upon appropriate request to an instruction requiring a finding of possession within the geographical confines of Massachusetts, we are not convinced that it was plain error for the court not to have given one. A failure to instruct on venue is not normally plain error. See United States v. Guy, 456 F.2d 1157, 1163 (8th Cir. 1972); Bellard v. United States, 356 f.2d 437 (5th Cir. 1966). Honneus was tried in the district both where he was living and where the criminal plans were laid and took effect. Cf. Travis v. United States, 364 U.S. 631, 81 S. Ct. 358, 5 L. Ed. 2d 340 (1969). We see in this instance no reason that would exempt him from the usual, salutory requirement that one complaining about an omitted instruction must have tendered a request and objected to its omission. United States v. Lachman, 469 F.2d 1043 (1st Cir. 1972). We note that the Government had proposed an instruction permitting conviction for possession outside the jurisdiction. The court declined to give it, and Honneus might have concluded that any instructions on the subject could only be to his detriment. Not only did he not assert lack of venue or jurisdiction as a ground for acquittal in motions made during the trial, but his only request for instruction arguably raising the matter seemed to focus on a different issue4 -- namely, whether the marihuana had to be Cannabis sativa L. See Part IV infra. This request might have been enough had Honneus objected after the court had instructed the jury. Instead, Honneus' counsel, although he raised other objections, indicated no dissatisfaction an the ground at issue. The case bears little resemblance to one cited by appellant, United States v. Rodriguez, 465 F.2d 5 (2d Cir. 1972), in which the counsel's failure to object to the charge was offset by his other actions and by the court's own recognition of venue as a prime issue.

We hold that no instruction was required. We find no merit in the different venue argument addressed to Count 2.

Appellant argues that rulings of the court excluding certain responses of witnesses were erroneous in that they reflected a mechanical application of the so-called 'federal' rule limiting cross-examination to matters presented during direct. See 6 J. Wigmore, Evidence 1885-91 (3d ed. 1940). The most serious complaint of this nature stems from circumstances relating to Honneus' claim that he was approached before trial by one Martha Snyder on behalf of a government witness, William Helliesen, with an offer that Helliesen would keep silent for $10,000. Honneus' counsel reported the alleged solicitation to the Government, whose agents thereafter interrogated Helliesen. It was after these events, which were made known to the court, that, prior to his cross-examination of Helliesen, Honneus' counsel announced an intention to ask Helliesen 'if he knew one Martha Snyder, in fact, his girl friend.' Although counsel said the question was for impeachment purposes, the court insisted that the 'scope of the cross-examination is set by the scope of the direct examination,' and that 'your alleged purpose does not set the rules of the ballgame.' From the court's remarks, counsel could understand that he was to make no inquiry of Helliesen about the purported bribe solicitation. The district court erred. Counsel was entitled to inquire whether Helliesen knew Martha Snyder and whether Helliesen had asked her to communicate an offer on his behalf to Honneus. The 'federal' rule, even is its most draconian form, does not deny to a cross-examiner reasonable latitude to inquire into relevant matters that may show a witness's bias or prejudice or otherwise impeach his credibility. Alford v. United States, 282 U.S. 687, 51 S. Ct. 218, 75 L. Ed. 624 (1931); Smith v. Illinois, 390 U.S. 129, 88 S. Ct. 748, 19 L. Ed. 2d 956 (1967). Whether Helliesen had sought to sell his testimony was such a matter. The court had broad discretion, of course, to control the extent of examination and to decide when a particular subject was exhausted. Once Helliesen denied communicating any offer through Martha Snyder, it could have curtailed further inquiry, and dampened any improper effort by counsel to ask questions framed so as to inform the jury of the supposed bribe attempt. But it should not have shut off all inquiry.

The denial of cross-examination upon a proper subject for cross-examination is a ground for reversal 'if the denial appears to have been harmful.' Wheeler v. United States, 351 F.2d 946, 947 (1st Cir. 1965). In Wheeler, defense counsel made an offer of proof that the witness would, if permitted to answer, have given an answer favorable to the defense. Here, on the other hand, counsel never outlined the questions he proposed to ask, nor did he say that he had reason to believe that Helliesen would admit participating in the bribe solicitation. This fact alone does not discharge the Government's burden. The right having been erroneously denied, the burden was upon the Government to establish the harmlessness of the denial; Honneus need not show 'that the cross-examination, if pursued, would necessarily have brought out facts tending to discredit . . ..' Alford, supra, 282 U.S. at 692, 51 S. Ct. at 219. On the other hand the error, serious as it was, is not to be viewed in isolation in determining whether it was harmful. Later in the trial the Assistant United States Attorney, in the presence of defense counsel, detailed the steps taken to investigate Honneus' story. He represented that Helliesen, when interrogated by federal agents, had flatly denied having made any such solicitation and had termed the story a defense ploy; he also stated that Martha Snyder had declined to discuss the matter and would claim her privilege against self incrimination. Honneus' attorney, who was free to interview Helliesen, never challenged the representation that Helliesen denied the story nor suggested any reason that Helliesen would admit to it or that there were facts with which to confront him. And later, when Martha Snyder was subpoenaed by the defense, her attorney informed the court that she would claim her privilege against self incrimination and she was excused without testifying. Bearing in mind also that Helliesen's direct testimony was strongly corroborated by other witnesses and evidence, and that Honneus' counsel was permitted an otherwise searching cross-examination of Helliesen who, the jury learned, was in jail and had received inducements from the Government to testify, we are satisfied that the error was harmless.

Honneus' other complaints of limitations upon cross-examination are less substantial. It is true that the so-called 'federal' rule, if useful as a general guide, should never be rigidly applied. A cross-examiner has a right to reasonable leeway for inquiry into facts which modify or explain away other facts developed on direct examination. But most of the disputed rulings were justified apart from a mechanistic application of the 'rule'. Some questions were of marginal relevancy or went into matters already explored. The court had discretion to exclude them, at least without an offer of proof or other indication where they would lead. In several instances the excluded questions were asked on recross, not cross. More strictly than with cross-examination, a court may limit recross to the subject matter of redirect and may exercise extensive discretion over its scope.

Honneus further attacks the exclusion of questions put to Richard Thurlow, a former confederate truned government witness. Counsel was allowed to bring out that Thurlow had been a user of many types of drugs, but was stopped when he inquired, 'And you told me that you had a nervous breakdown, didn't you?' Also excluded was the question, 'Were you in the Jackson Memorial Hospital in Miami, Florida?' The court indicated that it would not permit such impeachment without a 'hospital record'.

The court acted within its discretion. By revealing the dates and condition for which treatment was provided, a hospital record would at least have provided a basis for gauging the materiality of the inquiry. While insanity or abnormality at the time of observing the facts testified to or at the time of testimony are provable, a 'nervous breakdown' at some unspecified time is too remote. Cf. Sinclair v. Turner, 447 F.2d 1158, 1162-1163 (10th Cir. 1971); 3A J. Wigmore, Evidence 935 (3d ed. 1970). No offer of proof was made that Thurlow's condition could be shown to relate to his competency or qualifications as a witness. The court was entitled to weigh the potential unfairness of a free wheeling inquiry intended to stigmatize the witness against whatever materiality the evidence might have. While of course a hospital record is not the only proper way to establish mental impairment, see 3A Wigmore, supra, 878, 879, the court could impose the requirement in this instance since it could exclude the inquiry, as made, altogether.

Appellant complains that the court's impatience created a hostile atmosphere for the cultivation of a successful cross-examination. We are not persuaded that this was so. The Government presented a strong case, and a factual defense was virtually nonexistent. In such circumstances it may be especially hard to detect the threshold beyond which legitimate cross-examination becomes an idle fishing expedition. A court must allow inquiry that is germane and probative, but it need not bend over backwards to encourage the structuring of fanciful defenses. While the court's rulings here were on the side of strictness, we cannot say (except in the one respect already mentioned) that they improperly cut off any material avenues of inquiry, nor does it appear that the court's deportment was lacking in essential fairness. We therefore find no abuse of discretion.

Honneus has persistently insisted that the Government never proved him guilty of dealing in 'Cannabis sativa L.'. The offenses in Counts 1-4 all have to do with dealings in 'marihuana', listed as a controlled substance in 182 schedule I(c)(10). Marihuana is defined in 802(15) as certain parts of 'the plant Cannabis sativa L.'.

Honneus sought to introduce testimony by Dr. Richard Schultes5 that the Cannabis plant is a genus having three species, sativa, indica, and ruderalis, and that both sativa and indica presently grow in Jamaica. As the Government recovered no marihuana for analysis, there was no proof that the type or types of Cannabis Honneus purchased in Jamaica and later shipped to the United States were not indica rather than sativa.

Dr. Schultes testified during a voir dire that the disputed nomenclature could be traced back to Linnaeus' famous eighteenth century botanical compendium which listed under the genus Cannabis only a single species, sativa. A generation later, Lamarck named a Cannabis specimen collected in India, 'indica', and in the 1920's Russian botanists published in their native country a study naming a third variant 'ruderalis'. Nonethless, according to Dr. Schultes, the 'usually accepted view' (which he himself 'echoed') until a few years ago was that there was only one species of Cannabis. Now, after more study, Dr. Schultes believes there are three. Compare Schultes, Random Thoughts and Queries on the Botany of Cannabis, in C. Joyce & S. Curry, The Botany and Chemistry of Cannabis 11-38 (1970), with Schultes, Klein, Plowman & Lockwood, Cannabis: An Example of Taxonomic Neglect, 23 Harv. Botanical Museum Leaflets 325 (1974). See also R. Schultes & A. Hoffman, The Botany and Chemistry of Hallucinogens 58 (1973).

The district court, urling in a Memorandum and Order that Congress meant to include any and all marihuana-producing Cannabis in specifying 'Cannabis sativa L.', excluded Dr. Schultes' testimony and denied Honneus' motions for acquittal and for jury instructions. Courts in three circuits have similarly held, based upon the statutory history of the 1970 Act's predecessor, the marihuana Tax Act of 1937, 26 U.S.C. 4761, from which the definition of marihuana was taken. United States v. Rothberg, 351 F. Supp. 1115 (E.D.N.Y. 1972), aff'd, 480 F.2d 534 (2d Cir. 1973), cert. denied, 414 U.S. 856, 94 S. Ct. 159, 38 L. Ed. 2d 106; United States v. Moore, 330 F. Supp. 684, 686 (E.D.Pa.1970), aff'd, 446 F.2d 448 (3d Cir. 1971), cert. denied, 406 U.S. 909, 92 S. Ct. 1617, 31 L. Ed. 2d 820; United States v. Gaines, 489 F.2d 690 (5th Cir. 1974). In testimony on the Marihuana Tax Bill before the House Committee on Ways and Means, Commissioner of Narcotics Anslinger in 1937 stated that 'Cannabis indica, or marihuana, has as its parent the plant known as Cannabis sativa'. The Commissioner went on to say,

'It's popularly known in India as Cannabis indica; in America, as Cannabis American; in Mexico as Cannabis Mexicana, or marihuana.

'It is all the same drug, and is known in different countries by different names. It is scientifically known as Cannabis sativa . . ..' Hearings on H.R. 6385 before the Comm. on Ways & Means, 75th Cong., 1st Sess. at 37-38 (1937).

Dr. Lyster Dewey, a botanist formerly in charge of the Agriculture Department's fibre division, also testified that there was 'only one species known as hemp.' Id. at 55.6 Honneus argues that this certainty as to only one species is misleading, that others concerned with the 1937 Act were much less clear that sativa included other types if there were any others. There is indeed evidence of confusion over terminology,7 but none that Congress meant to exclude from regulation any type of the plant producing the hallucinogenic material popularly known in this country as 'marihuana'. See H.R.Rep.No.792, 75th Cong., 1st Sess. at 3-4 (1937); 81 Cong.Rec. 1440-41 (June 11, 1937) (appendix); 81 Cong.Rec. 5689-90 (June 14, 1937). Appellant's suggestion that sativa was selected because of an intention to deal only with the type most commonly grown in America is not borne out by the legislative record. Such an approach would have amounted to exemption of material from easily procurable plants having the same properties as the ones regulated. See generally 81 Cong.Rec. 5690 (June 14, 1937). We are persuaded that Congress adopted 'Cannabis sativa L.' believing it to be the term that scientists used to embrace all marihuana-producing Cannabis; the other named sorts were not seen as separate Cannabis species. Linnaeus had listed but the one species,8 and appellant's own expert concedes that until recently the monotypic view has been the 'usually accepted' one. This is not to try to refute Dr. Schultes' present view. The issue is not whether marihuana is monotypic or polytypic but what Congress meant when it used the term 'Cannabis sativa L.'. We hold that the district court's urling and exclusion of Dr. Schultes' testimony were correct.

A remaining question is whether the definition was adequate to forewarn Honneus of the crimes for which he was later charged. We think it was. Right or wrong, the monotypic classification has had considerable currency until very recently at least; use of 'Cannabis sativa L.' was, we think, sufficient to put a procurer on notice that he acted at his peril, see Rothberg, supra, and of course there is no evidence whatever that Honneus sought only 'indica' rather than 'sativa' or believed that he was acting legally. To the contrary, the venture was clandestine, and its participants clearly recognized its illegality. Had the question arose in connection with a drug less widely publicized than marihuana, the definition might take on greater importance. Here we find it too obvious for discussion that Honneus understood in advance that the importation and distribution were crimes.

Honneus argues that his former associates, Robert Puffer and Helliesen, should not have been permitted to give opinion testimony concerning a resemblance between the material they handled in 1972 and marihuana. Puffer testified that a sack brought from Jamaica contained darkish-green, leafy plant material that appeared similar to marihuana (which he testified to having smoked and seen growing before). Helliesen said the material was 'greenish material resembling marihuana'. (Helliesen further testified to having smoked marihuana on numerous occasions.) A third associate, Richard Thurlow, gave similar testimony to which appellant does not, however, object, since it merely described the material as 'a greenishbrown vegetable matter', similar to 'material' he had previously smoked. The witnesses, though without scientific expertise, had personal experience with marihuana. Their comparisons with the material at issue, based on personal observations, did not involve inferences which the jury could have equally well made from the witnesses' descriptions absent the comparison to marihuana. See generally 7 J. Wigmore, Evidence 1918 (3d ed. 1940). Nor, on the other hand, did the witnesses' testimony purport to be definite representations that the material was marihuana; we thus need not consider whether they would have been competent to express such a definite opinion. Cf. People v. Kenney, 30 N.Y.2d 154, 331 N.Y.S.2d 392, 282 N.E.2d 295 (1972). We think it was within the discretion of the court to have allowed the comparisons that were made.

The more basic question is whether, absent a report of laboratory analysis, there was ample evidence to support findings that the material was marihuana. We believe there was. The circumstantial and direct evidence when added together was extremely strong. This is not a case where the nature of a drug is sought to be established merely by a naive subject's reaction to a puff or two on a cigarette. It was testified that the parties, including Honneus, constantly spoke of the material as 'marihuana' or 'grass'. The sole object was to purchase and distribute marihuana. There was extensive direct evidence of Honneus' negotiations with sellers in Jamaica resulting in the purchase of what Honneus plainly believed to be marihuana. The same substance was later resold in Massachusetts at prices of $175-$250 a pound and Honneus admitted it was 'grass' in conversations. The material was smoked, produced a high, and had the outward appearance of marihuana.

We need not consider whether in different cases the absence of a specific and conclusive laboratory report would be fatal. Much depends on the totality of the evidence. A user's report, standing alone, might in some cases be a shaky reed, but that question is not before us. The evidence here was compelling that the substance was marihuana; indeed, barring an unlikely hoax fooling Honneus and his experienced associates, and those to whom they later sold, the substance had to be marihuana.

Honneus makes one final complaint, namely, the court's failure to give promptly adequate limiting instructions with respect to hearsay conspiracy testimony. Early in the trial (as a precedent for a number of occasions thereafter) the Government asked a witness whether he had had a conversation with one of the defendants (in this instance Honneus) relating to the others. Counsel for another defendant objected. Before receiving the testimony the court instructed the jury that declarations affecting other alleged members of the conspiracy, not present at the time the statements were made, were not to be considered against them unless the jury was satisfied that they were in fact members of the conspiracy. The court failed, however, to instruct as to the second aspect of this matter, that the existence of the conspiracy and defendant's participation therein had to be established by independent non-hearsay evidence before the jury could consider against defendant a co-conspirator's hearsay statement. No counsel pointed out this omission, although later Honneus made a request-- of an incorrect nature-- as to the first aspect. The court did not cover the second aspect until the charge, and Honneus says that this was too late. To sustain this contention he must show plain error. F.R.Crim.P. 52(b).

The limiting instruction that the court did give initially is perhaps the more important of the two conditions. Cf. Lutwak v. United States, 344 U.S. 604, 618-619, 73 S. Ct. 481, 97 L. Ed. 593 (1953). The other, however, is of importance as well. It is to be borne in mind that a jury has no experience in, or presumably knowledge of, the hearsay rule. Unguided, it would be most natural for the jury, in endeavoring to ascertain whether the defednants were in fact members of the conspiracy, to consider all evidence that apparently bore thereon. We must agree with United States v. Apollo, 476 F.2d 156, 163 (5th Cir. 1973), where the court said that there is a 'minimum obligation on the trial judge in a conspiracy case in which extrajudicial statements of alleged co-conspirators are proffered to give a cautionary instruction on the limited uses of hearsay testimony, explaining clearly to the jury the requirement that the conspiracy itself and each defendant's participation in it must be established by independent nonhearsay evidence which must be given either prior to the introduction of any evidence or immediately upon the first instance of such hearsay testimony.'

No previous case in this circuit has established that it may ever be plain error not to charge, a priori, on this second aspect. Moreover, in the present case the independent non-hearsay evidence tying Honneus to the conspiracy was adequate by any standard. We thus do not find plain error. The Apollo rule should, however, be observed in conspiracy cases hereafter tried in this circuit. Failure to do so will result in reversal in any case where we believe the omission to have affected substantial rights.

We affirm the judgment of the district court except for the separate concurrent sentences under Counts 1, 3, and 5. We vacate those sentences and remand the case for sentencing upon any one of the three conspiracy counts the United States may select, and for dismissal of the other two conspiracy counts.

So ordered.