United States v. Allen

539 F. Supp. 296

• Court: District Court, C.D. California
• Decided: July 28, 1982
• Docket: CR 81-1013 MRP
• Status: Published

Opinion

OPINION

PFAELZER, District Judge.

In March 1980, the grand jury returned a multiple-count indictment against defendants and two other persons, charging them with one count of conspiracy to commit mail fraud, in violation of 18 U.S.C. § 371, and with several substantive counts of aiding and abetting mail fraud, in violation of 18 U.S.C. § 1341. The case proceeded to a jury trial in late June 1980. At the conclusion of the government’s case, the court acquitted defendant Richard Allen as to all counts. The jury found Richard Allen’s brother, defendant Lee Allen, guilty on the conspiracy count and most of the substantive counts. In November 1981, the grand jury returned another indictment, this one against only these two defendants, charging them again with conspiracy to commit mail fraud, in violation of 18 U.S.C. § 371, and with mail fraud, in violation of 18 U.S.C. § 1341. In the motions before the court, defendants contend that the double jeopardy clause of the Fifth Amendment bars their prosecution on the second indictment. Alternatively, they seek dismissal of the indictment pursuant to the court’s supervisory power over the administration of criminal justice. In light of the particular facts of this case, the court has concluded that the defendants’ arguments on both grounds are sound and that the motions must be granted.

I. FACTS

From 1974 to 1978, the Hughes Aircraft Company at El Segundo, California (“Hughes”) was the subject of an illegal kickback scheme conducted through one of its purchasing departments. The government commenced an investigation into the scheme in early 1978 by interviewing several of the suspected participants. On May 17, 1978, one of those participants, Kenneth Lilly (“Lilly”), provided the government with a confession and a detailed description of the scheme’s operation. Lilly, a veteran employee of Hughes, was a purchasing agent for the company’s Radar Systems Group during the operation of the kickback scheme. Lilly informed government investigators that because of pressure from his departmental supervisor, defendant Lee Allen, he had purchased materials from certain Hughes suppliers at inflated prices. In so doing, he violated company policy requiring the use of competitive bids. Among the favored suppliers named by Lilly were companies operated by Jamie Tindall (“Tindall”), Roger Rinden (“Rinden”), and Samuel Mendelow (“Mendelow”). 1 According to Lilly, at the direction of Lee Allen, each of these suppliers, after payment by Hughes, was obligated to and did forward a percentage of its profits to defendant Richard Allen’s consulting firm.

Shortly after interviewing Lilly, the government interviewed the defendants, both of whom admitted at the outset that Richard Allen had had some type of business relationship with Tindall, Rinden, and Mendelow. Indeed, Richard Allen acknowledged that his consulting firm had accepted checks from those suppliers, although he claimed not to know precisely for what purpose the payments were made.

In October 1977, toward the end of the kickback scheme’s operation and prior to the government’s investigation, Lilly was transferred out of his position as buyer for the Radar Systems Group. His responsibilities were assumed by Lerner Watkins (“Watkins”). She informed investigators in January 1978 of efforts by Tindall, Rinden, and defendant Lee Allen to obtain her assistance in continuing the kickback scheme. Watkins subsequently became a government informant, and on February 14, 1978, she tape-recorded a conversation between herself and Lee Allen in which he discussed the Tindall-Rinden-Mendelow kickback transactions.

In early March 1980, after more than two years of investigation, the government presented the grand jury with its case against the defendants. The grand jury reviewed evidence of the defendants’ transactions with Tindall, Rinden, and Mendelow. On March 19, it returned an indictment against Lilly, Tindall, and the defendants. The indictment presented to the grand jury and returned by it did not charge Rinden or Mendelow, and it did not mention their participation in the scheme.

The indictment charged in Count 1 that Lilly, Tindall, and the defendants had violated 18 U.S.C. § 371 by participating in a conspiracy to commit mail fraud. The goal of the conspiracy was allegedly “to devise a scheme (i) to defraud Hughes Aircraft Company of money and property, . . . (ii) to defraud Hughes ... of its right to the ... honest services of defendant Kenneth Lilly .. ., and (iii) to obtain money and property by false and fraudulent pretenses.” 1980 Indictment at 3, ¶ 5. According to the indictment, the scheme to defraud Hughes commenced in November 1974 and continued until sometime in early 1978. The alleged scheme contemplated that Lilly would purchase certain requested supplies for Hughes from Tindall’s Company, Ti-Con Industries, Inc. (“Ti-Con”), at excessive, noncompetitive prices. To effect this plan, Lilly would mail a purchase order to Ti-Con, and, after shipping the supplies to Hughes, Ti-Con would mail a billing invoice to Hughes. In return for this favored business, Ti-Con, after payment by Hughes, would pay a commission to defendants by periodically mailing a check to defendant Richard Allen’s consulting firm, Contemporary Associates, Inc. These payments were then channeled to defendant Lee Allen through his corporations, Greenleaf Corporation and Lee Allen and Company. The twelve overt acts alleged in the indictment specified actions taken by the conspirators to ensure Ti-Con of business with Hughes and included certain payments of Ti-Con to Contemporary Associates. The last overt act alleged was that on or about February 14, 1978, defendant Lee Allen met with Watkins, Lilly’s successor, and attempted to persuade her to join the conspiracy.

In addition to the conspiracy count, the indictment charged Lilly, Tindall, and the defendants with several substantive counts of mail fraud. Specifically, the defendants were charged with nineteen counts of aiding and abetting mail fraud, in violation of 18 U.S.C. § 1341.

Prior to trial, each of the alleged conspirators moved for a separate trial. In considering the motions for severance, the court requested that the government submit a detailed outline of the evidence it would present against each defendant at trial, including a list of witnesses. After a careful review of that evidence, the court granted Lilly’s motion to sever and denied the other motions, leaving Tindall and defendants to be tried jointly. Subsequent to the court’s ruling, the case against Tindall and the defendants proceeded to a trial by jury on June 30, 1980. One day into the trial, the government filed papers in support of calling an additional witness, Roger Rinden. The government sought to introduce his testimony to prove an act “identical” to that charged in the indictment — that Lilly and the defendants had also engaged in a mail fraud conspiracy with Rinden and Mendelow. 2 The government conceded that the decision to call Rinden was made belatedly, but attributed the delay to the fact that -it had not adequately catalogued the Rinden-Mendelow kickback transactions with the Tindall transactions. Nevertheless, the court denied the government’s request, stating that it was too late in the proceedings for the government to make this addition to its witness list. 3

At the trial, the government proceeded on the theory that defendants and Lilly had engaged in a single conspiracy with Tindall, Mendelow, and Rinden. Consistent with that theory, the court admitted evidence of the Rinden-Mendelow transactions as well as the Tindall transactions. 4 On July 8, 1980, the court granted defendant Richard Allen’s motion for judgment of acquittal as to all counts on the ground that the government had failed to prove his connection with the conspiracy and the mail fraud scheme as a matter of law. The jury found Tindall and defendant Lee Allen guilty on the conspiracy and substantive counts, 5 and Lilly, who was tried to the court, was also found guilty. Prior to sentencing, Lee Allen suffered a heart attack, and, on August 21, 1980, the court sentenced him to pay a fine and to serve a three-year probation term. Lee Allen did not appeal his convictions. Tindall and Lilly did seek review of their convictions, however, which were affirmed in a consolidated appeal on November 16, 1981.

On November 18, 1981, two days after the convictions were affirmed, the grand jury returned its second indictment against the defendants. That indictment, presently before the court, charges in Count 1 that defendants and their unindicted co-conspirators, Lilly, Rinden, and Mendelow, violated 18 U.S.C. § 371 by participating in a conspiracy to commit mail fraud. The indictment alleges that the defendants devised a scheme “to obtain money by means of false and fraudulent pretenses.” 1981 Indictment at 4, ¶ 7a. The goal of the scheme was “(i) to defraud Hughes of money ..., (ii) to defraud Hughes of certain nondisclosed profits .. ., (iii) to defraud Hughes of its right to have the company’s business . . . conducted by its employees ... honestly [and] impartially ..., and (iiii) to defraud Hughes of its right to the conscientious, loyal, [and] honest . .. services [of] ... Hughes employees.” Id. The conspiracy and scheme to defraud allegedly commenced in mid-1974 and continued until approximately April 1978. The indictment reveals that the Rinden-Mendelow scheme operated in the same manner as the scheme charged in the first indictment. 6 The forty overt acts alleged in the second indictment consist primarily of purchase orders mailed to the suppliers and payments made to Contemporary Associates, Inc. The last overt act alleged is that on or about February 14, 1978, defendant Lee Allen met with Watkins in an attempt to persuade her to join the conspiracy. In addition to the conspiracy count, the indictment also charges defendants with eleven substantive counts of mail fraud, in violation of 18 U.S.C. § 1341.

The second indictment was assigned to the Honorable Malcolm M. Lucas of the Central District of California. Defendants promptly moved to transfer the case to this court on the grounds that the indictment should be dismissed pursuant to the double jeopardy clause or the court’s supervisory power and that, because of the similarity of the indictments, this court would be in a better position to decide those issues. The government opposed the transfer on the ground that the indictments involved different transactions. Despite that opposition, however, the case was transferred to this court in mid-December 1981.

In opposing defendants’ motions to dismiss, the government points out that it now has a stronger case against defendants than it had at the time of the first trial. 7 Indeed, the government admits that soon after the court granted Richard Allen’s motion for judgment of acquittal, it subpoenaed additional documents from him in order to avoid a similar result in any future cases. 8 As the following discussion indicates, however, the double jeopardy clause and the court’s supervisory power exist for the very purpose of preventing the government from proceeding as it has in this case.

II. DOUBLE JEOPARDY

“Throughout its history, the principle of double jeopardy has restricted both the government’s ability to place the criminally accused on trial and its power to punish the guilty. . . . The general function of the double jeopardy clause is to assure that the prosecution and punishment of an individual have the degrees of finality and fairness essential to administration of the criminal law.” Note, A Definition of Punishment for Implementing the Double Jeopardy Clause’s Multiple-Punishment Prohibition, 90 Yale L.J. 632, 632, 634 (1981). In implementing the policies of the clause, 9 the Supreme Court has accorded defendants protection from multiple trials as well as multiple punishment for the same offense. North Carolina v. Pearce, 395 U.S. 711, 717, 89 S.Ct. 2072, 2076, 23 L.Ed.2d 656 (1969). In the case at bar, the prohibition against multiple trials requires dismissal of the second indictment.

A. The Conspiracy Count

The double jeopardy clause mandates that once a defendant has been convicted or acquitted of an offense, he cannot be tried again for that offense. Pearce, 395 U.S. at 717, 89 S.Ct. at 2076. It also prevents the government from subdividing a single criminal conspiracy into multiple violations of one conspiracy statute. Braverman v. United States, 317 U.S. 49, 52-53, 63 S.Ct. 99, 101, 87 L.Ed. 23 (1942). Defend ants invoke these principles in arguing that here the government’s use of two indictments impermissibly divided a single conspiracy into multiple counts and that the judgments in the first trial preclude reprosecution of that conspiracy. The critical issue in resolving this claim is whether the conspiracies in the indictments constitute the “same offense” for purposes of the double jeopardy clause. 10 As a threshold matter, however, the court must determine which same-offense test is appropriate where two indictments charge a violation of the same conspiracy statute.

“[T]he principal test for determining whether two offenses are the same for pur poses of barring successive prosecutions [is the Blockburger test].” Illinois v. Vitale, 447 U.S. 410, 416, 100 S.Ct. 2260, 2265, 65 L.Ed.2d 228 (1980). That test provides that where the same act constitutes a violation of two distinct statutory provisions, the act will nevertheless be treated as one offense unless each provision requires proof of fact that the other does not. Blockburger v. United States, 284 U.S. 299, 304, 52 S.Ct. 180, 182, 76 L.Ed. 306 (1932). Despite the prevalent use of the Blockburger formula, however, it “is not the only standard for determining whether successive prosecutions impermissibly involve the same offense.” Brown v. Ohio, 432 U.S. 161, 166 n.6, 97 S.Ct. 2221, 2226 n.6, 53 L.Ed.2d 187 (1977). Where, as here, two indictments charge a violation of the same conspiracy statute, the Blockburger analysis proves difficult of application since it assumes a violation of “two distinct statutory provisions.” 284 U.S. at 304, 52 S.Ct. at 182. Further, given that large-scale conspiracies frequently involve several individuals, numerous overt acts, and diverse illegal objectives, careful drafting of the indictments to avoid duplicative allegations would permit the government to subdivide every such conspiracy under the Blockburger test, thereby easily removing conspiracy prosecutions from double jeopardy protection. It is therefore well established that Blockburger is not the appropriate standard for evaluating double jeopardy challenges to multiple conspiracy counts charged under the same statute. See Braverman, supra, 317 U.S. at 54, 63 S.Ct. at 102 (analysis required where multiple counts charged under same conspiracy statute differs from analysis used where a single act violates two distinct statutes) (distinguishing Blockburger); Albernaz v. United States, 450 U.S. 333, 339-40, 101 S.Ct. 1137, 1142-1143, 67 L.Ed.2d 275 (1981) (approving Braverman); United States v. Phillips, 664 F.2d 971, 1006 (5th Cir. 1981) (“determining whether two conspiracies are in fact the same requires a more detailed analysis than that required ... [by] Blockburged’); United States v. Tercero, 580 F.2d 312, 315 (8th Cir. 1978) (“same evidence” test not appropriate in comparing conspiracies); United States v. Mallah, 503 F.2d 971, 985 (2d Cir. 1974) (strict same-offense test not applied to conspiracies), cer t. denied, 420 U.S. 995, 95 S.Ct. 1425, 43 L.Ed.2d 671 (1975); accord, Sanabria v. United States, 437 U.S. 54, 70 n.24, 98 S.Ct. 2170, 2181 n.24, 57 L.Ed.2d 43 (1978). 11

Another possible method of analyzing the same-offense issue in the conspiracy context is set forth in Kotteakos v. United States, 328 U.S. 750, 66 S.Ct. 1239, 90 L.Ed. 1557 (1946), and Blumenthal v. United States, 332 U.S. 539, 68 S.Ct. 248, 92 L.Ed. 154 (1947). The Kotteakos-Blumenthal analysis examines the interdependence of the co-conspirators and the knowledge that each conspirator had of the scope of the enterprise. If the actions of each conspirator are necessary to the success of the operation and each participant knew or should have known the scope of the operation, a single conspiracy is present. A relevant line of inquiry under this analysis then would be whether the “Allen-Lilly-Tindall conspiracy” could succeed without the “Allen-Lilly-Rinden conspiracy.” As was true of Blockburger, however, this method of analysis does not appear appropriate for resolving the double jeopardy issues raised in this ease.

Neither Kotteakos nor Blumenthal was a double jeopardy case, and neither addressed the issue presented here — whether a given defendant is being tried twice for the same conspiracy. The issue in both cases concerned the presentation of evidence to the jury in multi-defendant conspiracy cases. For example, in Kotteakos, the Court analogized the structure of the conspiracy before it to a wheel. A single defendant at the center of the conspiracy, the “hub” defendant, dealt separately with each of several “spoke” defendants in a loan fraud scheme. Each spoke defendant processed his loans through the hub defendant without assistance from or knowledge of the other spoke defendants. The Court held that it was prejudicial on those facts for the trial court to instruct the jury that only a single conspiracy existed. Such an instruction permitted the jury to hold each spoke defendant accountable for the actions of the other spoke defendants and improperly allowed the jury to transfer the guilt of one defendant to another. Although the Court clearly found separate conspiracies between the spoke defendants for purposes of the presentation of evidence, it did not suggest that, for double jeopardy purposes, the single hub defendant could be indicted and tried on a separate charge of conspiracy as to each spoke defendant. As to him, there was but one conspiracy. Put simply, since the gravamen of conspiracy is the illegal agreement, the hub defendant (here, the Allens) may be guilty of only one conspiracy under the double jeopardy clause. This, of course, does not contradict or negate in any way the Kotteakos rule that one set of spoke defendants cannot be held for the statements or actions of any other set of spoke defendants. See United States v. Castro, 629 F.2d 456, 464 (7th Cir. 1980) (Kotteakos not applicable in deciding double jeopardy claim of hub conspirator). Thus, in the Ninth Circuit, the Kotteakos analysis is used to determine if the proof at trial resulted in an impermissible transfer of guilt from one defendant to another, see e.g., United States v. Kenny, 645 F.2d 1323, 1334-36 (9th Cir.), cert. denied, 452 U.S. 920, 101 S.Ct. 3059, 69 L.Ed.2d 425 (1981); United States v. Thomas, 586 F.2d 123, 131-32 (9th Cir. 1978), but it is limited to such evidentiary questions. See Note, “Single vs. Multiple” Conspiracies: A Uniform Method of Inquiry for Due Process and Double Jeopardy Purposes, 65 Minn.L.Rev. 295, 299-300, 301-04 (1981) [hereinafter “Single v. Multiple” Conspiracies ] (Kotteakos-Blumenthal test applied in area of due process, not double jeopardy). 12

In applying the Braver man rule against subdividing a single conspiracy, the Ninth Circuit has adopted a “factor analysis” to determine whether two conspiracies constitute the same offense. That analysis, commonly denoted the “totality of circumstances” test, requires the court to compare “the differences in the periods of time covered by the alleged conspiracies, the places where the conspiracies were alleged to occur, the persons charged as co-conspirators, the overt acts alleged to have been committed, and the statutes alleged to have been violated.” United States v. Mayo, 646 F.2d 369, 372 (9th Cir.) (per curiam) (citing Arnold v. United States, 336 F.2d 347, 349-50 (9th Cir. 1964)), cert. denied,-U.S.-, 102 S.Ct. 979, 71 L.Ed.2d 115 (1981); accord, United States v. Jabara, 644 F.2d 574, 577 (6th Cir. 1981); United States v. Tammaro, 636 F.2d 100, 103-04 (5th Cir. 1981); United States v. Tercero, 580 F.2d 312, 315, 316 (8th Cir. 1978); “Single vs. Multiple” Conspiracies, 65 Minn.L.Rev. at 310-17. These factors are weighed in order to answer the ultimate question posed by Braverman —whether there is more than one conspiratorial agreement. 317 U.S. at 52-53, 63 S.Ct. at 101. See United States v. Stricklin, 591 F.2d 1112, 1125 (5th Cir.) (“the essence of a double jeopardy determination in a conspiracy case is whether there was more than one agreement”), cert. denied, 444 U.S. 963, 100 S.Ct. 449, 62 L.Ed.2d 375 (1979).

Defendants have carried their burden of establishing that the two indictments allege but a single conspiracy. See Sanchez v. United States, 341 F.2d 225, 227 (9th Cir.) (burden on defendant to prove sameness of offenses), cert. denied, 382 U.S. 856, 86 S.Ct. 109, 15 L.Ed.2d 94 (1965). Application of the Arnold criteria to the facts of this case clearly establishes the presence of a single conspiracy. The “two conspiracies” covered the same time period, 1974 to 1978. 13 Both were conducted within the court’s district, with the conspirators’ activities centering around the Hughes plant at El Segundo. The indictments have one overt act in common — that on February 14, 1978, defendant Lee Allen attempted to get Watkins to join the conspiracy. Given that the survival of the conspiracy was dependent upon Watkins’ participation in the scheme and that most of the overt acts in the indictments concern simple mailings, Lee Allen’s effort to get Watkins’ cooperation must be considered a substantial overt act in. furtherance of the conspiracy. And where “one or more significant overt acts are common to the alleged multiple conspiracies, a presumption should arise that only one conspiracy is involved, rebuttable only by strong evidence tending to prove that the act was in furtherance of two distinct conspiracies.” “Single vs. Multipie” Conspiracies, 65 Minn.L.Rev. at 313. That presumption has not been rebutted here. 14 With respect to the parties charged in the scheme, the core conspirators, defendants Richard and Lee Allen, are the same. Further, Lilly was the only Hughes buyer involved in the two conspiracies alleged by the government. Indeed, the only difference in the participants was the supplier to Hughes. The final Arnold factor, the underlying statutes violated, also points to the presence of one conspiracy; both conspiracies involved agreements to violate the same statute, 18 U.S.C. § 1341, the mail fraud statute.

In sum, although no single factor in the Arnold analysis should be solely determinative, in this case each factor strongly weighs in favor of finding a single conspiratorial agreement. Significantly, the indictments themselves allege the same purpose for the conspiracies — to defraud Hughes of money and the honest services of its employees. Moreover, at the first trial, the government assumed the existence of only one conspiracy, see note 4 supra, conceding that the two sets of transactions were identical but for the difference in suppliers. 15 However, a difference in suppliers does not give rise to a separate conspiracy. United States v. Castro, 629 F.2d 456, 460-64 (7th Cir. 1980). If it did, the Allens could be indicted and tried as many times as there were individual suppliers. That cannot be the law.

Because a single conspiratorial agreement encompassed both the Tindall transactions and the Rinden-Mendelow transactions, the two indictments allege the same conspiracy. 16 Since defendants have been previously tried for that conspiracy, they may not be again placed in jeopardy for that offense.

B. The Substantive Counts

Having determined that a single conspiracy existed in this case, it necessarily follows that the substantive counts of the second indictment underlie the conspiracy for which defendants have already been tried. Of course, as a general rule a defendant can first be tried on a conspiracy count and then, in a separate trial, be prosecuted for the underlying substantive offense. See United States v. Brooklier, 637 F.2d 620 (9th Cir. 1980), cert. denied, 450 U.S. 980, 101 S.Ct. 1514, 67 L.Ed.2d 815 (1981); United States v. Castro, 629 F.2d 456, 465-66 (7th Cir. 1980); United States v. Thomas, 342 F.2d 132, 135 (6th Cir.), cert. denied, 382 U.S. 855, 86 S.Ct. 105, 15 L.Ed.2d 92 (1965). Further, there can be little doubt that conspiracy to commit mail fraud and the substantive crime of mail fraud constitute separate offenses for purposes of punishment, permitting imposition of separate, consecutive sentences. Pereira v. United States, 347 U.S. 1, 11-12, 74 S.Ct. 358, 364, 98 L.Ed. 435 (1954). 17 This is not to say, however, that those offenses may always be separately tried. “Even if two offenses are sufficiently different to permit the imposition of consecutive sentences, successive prosecutions will be barred in some circumstances where the second prosecution requires the relitigation of factual issues already resolved by the first trial.” Brown v. Ohio, 432 U.S. 161, 166 n.6, 97 S.Ct. 2221, 2226 n.6, 53 L.Ed.2d 187 (1977); accord, United States v. Phillips, 664 F.2d 971, 1006 n.48 (5th Cir. 1981).

[I]f the government could have initially prosecuted a defendant for multiple offenses [in a single proceeding], further analysis is necessary if it charges him with only one and holds the others in reserve. Policies of assuring finality, sparing defendants the financial and psychological burdens of repeated trials, preserving judicial resources, and preventing prosecutorial misuse of the indictment process all come into play.

United States v. Brooklier, supra, 637 F.2d at 622. 18

Because of the substantial similarity between conspiracy to commit mail fraud and the substantive crime of mail fraud, the court finds that the double jeopardy clause also bars defendants’ prosecution on the new substantive counts.

1. Collateral Estoppel

The relationship between the statutory elements of conspiracy, 18 U.S.C. § 371, and those of mail fraud, 18 U.S.C. § 1341, is unique. 19 Conspiracy in general requires an agreement among two or more persons to commit some substantive offense. United States v. Friedman, 593 F.2d 109, 115 (9th Cir. 1979). Conspiracy to commit mail fraud in particular requires proof of an agreement to defraud. United States v. Craig, 573 F.2d 455, 485-86 (7th Cir. 1977), cert. denied, 439 U.S. 820, 99 S.Ct. 82, 58 L.Ed.2d 110 (1978). Similarly, the substantive offense of mail fraud requires proof of a scheme to defraud. United States v. Beecroft, 608 F.2d 753, 757 (9th Cir. 1979). In essence, both statutes contemplate some sort of scheme. Admittedly, the “scheme” required for mail fraud may involve only one person, unlike the “agreement” needed for conspiracy, which requires two or more persons. Nevertheless, in the case at bar, the mail fraud scheme does involve more than one person. The facts underlying the conspiracy are necessarily the facts giving rise to the substantive scheme to defraud. Indeed, an examination of the indictments in this case makes clear that the scheme alleged under Section 1341 is the conspiratorial agreement alleged under Section 371. 20 See also United States v. Cohen, 516 F.2d 1358, 1364 (8th Cir. 1975) (“In general, proof of a mail fraud scheme involving two or more persons is analogous to the nature of proof in a conspiracy.”) Since proof of the conspiratorial agreement is synonymous with proof of the scheme to defraud, the government’s failure at the first trial to link defendant Richard Allen to the conspiracy necessarily indicated a lack of proof on the substantive mail fraud counts as well, and the court accordingly entered a judgment of acquittal as to all charges. Given that this case involves a single conspiracy, it follows that the new scheme to defraud is actually the conspiratorial agreement litigated at the previous trial. Thus, even with the conspiracy count of the second indictment dismissed, the government, in trying the new substantive counts, would be relitigating the existence of the conspiratorial agreement. However, since Richard Allen has been acquitted of participating in that agreement, or scheme, trying him on the new substantive counts would permit relitigation of a fact already found in his favor. And “when an issue of ultimate fact has once been determined by a valid and final judgment, that issue cannot again be litigated between the same parties in any future lawsuit.” Ashe v. Swenson, 397 U.S. 436, 443, 90 S.Ct. 1189, 1194, 25 L.Ed.2d 469 (1970).

In Ashe, the government believed the defendant guilty of robbing each of six men engaged in a poker game. Instead of charging the defendant in a single proceeding with robbing all six players, the government brought the defendant to trial for robbing only one of them. The evidence at trial proved that a robbery had occurred and that the particular victim in question had been robbed. It was also clear that the same persons had robbed each of the players. The trial ended in a judgment of acquittal. Subsequently, the government tried the defendant for robbing a different player and obtained a conviction. On appeal, the Supreme Court reversed. The Court found that the first jury’s verdict of acquittal established that the defendant was not one of the robbers. Since that fact had been determined in the defendant’s favor at the first trial, the government could not seek to litigate the issue a second time.

In the case at bar, the government, having failed at the previous trial to prove Richard Allen’s involvement in the scheme to defraud, now seeks to relitigate that issue through the new substantive counts. Under the rule established in Ashe, it may not do so, however. The government is collaterally estopped from seeking to prove a second time Richard’s connection to that scheme. This result prevents the government from using the first trial as a dry run for subsequent prosecutions, improving its case with each successive trial. See 397 U.S. at 447, 90 S.Ct. at 1196. Although Ashe applies even where there is no indication that the prosecutor actually used the first trial to prepare for a later case, evidence of such use is present here: after the court acquitted Richard Allen, the government sought additional evidence in order to avoid a similar result in any subsequent cases.

Collateral estoppel, of course, does not bar defendant Lee Allen’s prosecution on the new substantive counts since his first trial resulted in a conviction, i.e., his connection with the agreement to defraud was proved. See Brown v. Ohio, 432 U.S. 161, 165, 97 S.Ct. 2221, 2225, 53 L.Ed.2d 187 (1977) (double jeopardy protects the accused “from attempts to relitigate facts underlying a prior acquittal”) (emphasis added) (interpreting Ashe).

2. Compulsory Joinder

The Supreme Court has recognized that two offenses may be so similar in terms of proof that they should be joined in a single proceeding. Failure to join such offenses precludes the government from trying a defendant on the charges omitted from the first trial. Here, the substantive counts in the second indictment are so similar to the offenses previously tried that the rule of compulsory joinder should apply. Thus, neither defendant can be prosecuted on those counts.

a. Comparison of the New Substantive Counts to the Conspiracy Count

As previously noted, it is generally held that a conspiracy count and its underlying substantive counts can be tried separately. However, this general rule does not appear to have been analyzed in the context of separate trials for mail fraud, 18 U.S.C. § 1341, and conspiracy to commit mail fraud, 18 U.S.C. § 371. In this case, after engaging in such an analysis, the court has concluded that the substantial identity of those two offenses compels the rejection of the general rule and instead requires application of the rule of compulsory joinder.

Since the double jeopardy issue presented here involves successive prosecutions under two distinct statutory provisions, 18 U.S.C. §§ 371, 1341, the Blockburger standard is the primary test for determining whether mail fraud must be tried with conspiracy to commit mail fraud. See Illinois v. Vitale, 447 U.S. 410, 416, 100 S.Ct. 2260, 2265, 65 L.Ed.2d 228 (1980); Brown v. Ohio, 432 U.S. 161, 166, 97 S.Ct. 2221, 2225, 53 L.Ed.2d 187 (1977). Blockburger, which examines the evidence required under each statute, would permit separate prosecution of those offenses as each offense requires proof of a fact not required by the other. See note 17 supra. However, Blockburger’s same-required-evidence test is not the sole standard for evaluating successive prosecutions under different statutes. After examining the evidence required for each offense and determining that Blockburger would permit separate trials, the court should examine the actual evidence to be presented at the second trial. Illinois v. Vitale, 447 U.S. 410, 419-21, 100 S.Ct. 2260, 2267, 65 L.Ed.2d 228 (1980); id. 447 U.S. at 426-28, 100 S.Ct. at 2270-2271 (Stevens, J., dissenting); Note, The Impact of Expanded Rules for Determining What Constitutes the “Same Offense” for Double Jeopardy Purposes: Illinois v. Vitale, 1980 B.Y.U.L. Rev. 948, 957-58; see Jordan v. Virginia, 653 F.2d 870 (4th Cir. 1980); United States v. Sabella, 272 F.2d 206, 210-12 (2d Cir. 1959) (cited with approval in Sanabria v. United States, 437 U.S. 54, 72 n.28, 98 S.Ct. 2170, 2183 n.28, 57 L.Ed.2d 43 (1978)). Under the same-actual-evidence test, two separately punishable offenses will be treated as one for purposes of trial if “the evidence required to warrant a conviction upon one of the [prosecutions] would have been sufficient to support a conviction upon the other.” Jordan, supra, 653 F.2d at 873 (quoting In re Nielsen, 131 U.S. 176, 188, 9 S.Ct. 672, 676, 33 L.Ed. 118 (1889)). 21

It is true that the evidence presented at the first trial on the conspiracy and substantive counts would not suffice to prove the new substantive counts; the latter involve mailings different from those at the first trial. However, the evidence required to prove the new substantive counts would, of necessity, reprove the conspiracy count already prosecuted. 22 Consequently, the actual-evidence test mandates that, at least on the facts of this case, the conspiracy count and the substantive counts be tried simultaneously. It follows that since defendants have already been prosecuted for the conspiracy, they cannot be tried on the mail fraud counts. The order of the trials is immaterial; the rule of compulsory joinder applies even though the conspiracy was tried first. Vitale, 447 U.S. at 421, 100 S.Ct. at 2267.

In sum, the actual-evidence test required the government to pursue all mail fraud charges against the defendants at the time it brought the initial conspiracy count. This is not to say, however, that every conspiracy to commit mail fraud must be tried with its underlying substantive counts, for the facts of another case may not require compulsory joinder. It should also be noted that the compulsory joinder rule announced here is of limited significance outside the mail fraud context. Where mail fraud is not involved, the proof required for the substantive offense will typically not include proof of the conspiracy. Thus, requiring the government to join Section 371 and Section 1341 offenses in a single proceeding simply recognizes a narrow exception to the general rule permitting separate trials on the conspiracy and substantive counts.

b. Comparison of the New Substantive Counts to the Previously Tried Substantive Counts

The preceding discussion analyzed the relationship between mail fraud and conspiracy to commit mail fraud and concluded that those offenses must be tried simultaneously. However, even if the defendants had not been tried for conspiracy at the previous trial, their prosecution on the substantive counts alone would act as a bar to trial on the new substantive counts. Since the two indictments allege but a single scheme to defraud, it follows that the nineteen substantive counts (i.e., mailings) alleged in the first indictment and the eleven alleged in the second indictment are all part of the same ongoing scheme. No doubt each of those thirty mailings is separately punishable. See, e.g., United States v. Crockett, 534 F.2d 589, 593 (5th Cir. 1976); United States v. Anderson, 466 F.2d 1360, 1361 (8th Cir. 1972). The question here, however, is whether they are each separately triable.

Mail fraud requires proof of a scheme to defraud and knowing use of the mails to execute the scheme. United States v. Beecroft, 608 F.2d 753, 757 (9th Cir. 1979). In the typical mail fraud case, the trial deals almost exclusively with the scheme to defraud, and the parties simply stipulate the mailings into evidence. For example, at defendants’ trial, the mailings were stipulated into evidence, and, as is usual in these eases, the parties devoted a negligible amount of trial time to proving use of the mails. 23 Given that every mail fraud scheme involves numerous mailings and that each mailing is separately punishable, the potential for abuse of the trial process is present in every Section 1341 prosecution. By charging only certain of the mailings in an initial indictment, the government can hold the others in reserve, awaiting the outcome of the first trial. If the prosecutor is dissatisfied with the results of that trial, additional mailings can then be charged in a second indictment. In this case for example, if the government’s argument is accepted, in theory it would have the right to as many as thirty separate trials, each one concerning a distinct mailing. 24 And at each of these trials, it would have the opportunity to relitigate the same ultimate fact — that the defendants participated in a scheme to defraud. Although the government actually seeks only two trials instead of thirty, the same double jeopardy concerns are implicated. It is not enough to rely on the belief that the government will in good faith limit itself to only two trials despite the fact that it is legally entitled to as many as thirty. The issue here is not the government’s good faith, but the defendants’ rights.

The rationale underlying the double jeopardy ban on successive trials is best described in an oft-quoted passage from Green v. United States, 355 U.S. 184, 78 S.Ct. 221, 2 L.Ed.2d 199 (1957):

The constitutional prohibition against double jeopardy was designed to protect an individual from being subjected to the hazards of trial and possible conviction more than once for an alleged offense. .. . The underlying idea ... is that the State with all its resources and power should not be allowed to make repeated attempts to convict an individual for an alleged offense, thereby subjecting him to embarrassment, expense and ordeal and compelling him to live in a continuing state of anxiety and insecurity as well as enhancing the possibility that even though innocent he may be found guilty.

355 U.S. at 187-88, 78 S.Ct. at 223 (quoted with approval in United States v. DiFrancesco, 449 U.S. 117, 128-29, 101 S.Ct. 426, 432-33, 66 L.Ed.2d 328 (1980)). It is certainly arguable that the policy considerations announced in Green would require that the separate mailings in a mail fraud scheme be tried in a single proceeding. In the usual scheme — where the mails are used frequently and repeatedly — if the ban were not invoked, a seemingly endless series of prosecutions would be possible, forcing the defendant to live in a continuing state of insecurity and anxiety from one trial to the next, not knowing which trial will be his last. 25 Additionally, since each trial would require relitigation of the same scheme to defraud, the government could no doubt strengthen its case each time, thereby increasing the chance that the defendant will be found guilty. Indeed, here the government has done just that. Yet, despite Green’s apparent relevance to successive mail fraud prosecutions, the general policies announced in that decision provide but a philosophical starting point for resolving the retrial issue presented in this case. The resolution of that issue essentially turns on the question of whether the mail fraud counts, though separately punishable, constitute a single offense for purposes of trial.

To date, the Supreme Court has employed two different tests in applying the ban on successive trials: the same-required-evidence test announced in Blockburger and the same-actual-evidence test used in Vitale. Both of those tests would permit successive trials for mail fraud. Since each mail fraud count involves a distinct mailing, each count requires proof of a fact not required by the other. Similarly, the actual evidence required to prove a particular mailing would not establish any other mailing. Application of these two tests should not exhaust the analysis, however. See Brown, supra, 432 U.S. at 166 n.6, 97 S.Ct. at 2226 n.6. The rationale of Brown and the decision in In re Nielsen, 131 U.S. 176, 9 S.Ct. 672, 33 L.Ed. 118 (1889), support a rule of compulsory joinder in the context of successive mail fraud prosecutions.

In Brown, the Court held that greater and lesser included offenses must be tried together, basing its decision on the Blockburger test. 26 This joinder rule is easily understood where the defendant has been first tried for the greater offense and convicted. In. that situation, the punishment for the greater offense necessarily includes the penalty for the lesser offense. See Westen & Drubel, Toward a General Theory of Double Jeopardy, 1978 Sup.Ct. Rev. 81, 157-59 & 158 n.332 [hereinafter Double Jeopardy Theory]. Since the lesser offense is not separately punishable, it is not separately triable. Brown, 432 U.S. at 166, 97 S.Ct. at 2225. However, the rationale of the joinder requirement is more difficult to discern where the defendant has been first tried on the greater offense and acquitted. Such an acquittal does not establish the defendant’s innocence on the lesser crime since the greater offense requires proof of a fact not required for conviction on the lesser count. Similarly, where the defendant has been initially tried on the lesser offense and convicted, his conviction should not bar separate punishment for the greater offense. See Double Jeopardy Theory, 1978 Sup.Ct.Rev. at 160-61 (discussing Jeffers v. United States, 432 U.S. 137, 97 S.Ct. 2207, 53 L.Ed.2d 168 (1977)). Yet, in Brown, the Court required joinder of greater and lesser included offenses regardless of which offense is tried first and regardless of the outcome of the first trial. The effect, then, of the joinder rule is to permit the government only one opportunity to litigate the facts underlying the lesser offense, that is, one opportunity to prove the core of facts which the two offenses have in common. Thus, in Brown, “the Court implicitly held that the Double Jeopardy Clause requires the State to join in a single prosecution all offenses that share the same factual core in common.” Id. at 163. Significantly, although the starting point in Brown was the Blockburger test, the holding in that case reaches beyond Blockburger, requiring joinder of all offenses that have a substantial core of fact in common. 27 The government is allowed only one attempt to prove the common facts, or elements, between any two offenses. See Note, The Double Jeopardy Clause as a Bar to Reintroducing Evidence, 89 Yale L.J. 962 (1980) (double jeopardy clause prevents relitigation of any fact alleged at first trial).

In the case at bar, a common core of fact — the scheme to defraud — connects all of the substantive counts. Indeed, that common core constitutes the principal element of proof in each separate count. Since the mail fraud counts in the two indictments have such a significant core of fact in common, the rationale of Brown should require their joinder in a single trial.

Additional support for a compulsory joinder rule in the mail fraud context is found in In re Nielsen, 131 U.S. 176, 187-88, 9 S.Ct. 672, 675-76, 33 L.Ed. 118 (1889), a case relied on in Brown. See 432 U.S. at 168, 97 S.Ct. at 2226. In Nielsen, the defendant was initially tried on a charge of cohabitation and was convicted. Subsequent to that trial, the government successfully prosecuted the defendant for adultery, relying on the same facts used to obtain the cohabitation conviction. The successive trials did not violate Blockburger since “conviction for adultery required proof that the defendant had sexual intercourse with one woman while married to another; conviction for cohabitation required proof that the defendant lived with more than one woman at the same time.” Brown, 432 U.S. at 166 n.6, 97 S.Ct. at 2226 n.6. Nevertheless, the Nielsen Court found the second prosecution barred by the double jeopardy clause. Noting that the “material part” of each prosecution was the same act of illegal sexual intercourse, 131 U.S. at 187, 9 S.Ct. at 675-76, the Court reversed the second conviction on the ground that “the material part of the adultery charged was comprised within the unlawful cohabitation of which the petitioner was already convicted.” Id. The Court further stated that “where ... a person has been tried and convicted for a crime which has various incidents included in it, he cannot be a second time tried for one of those incidents without being twice put in jeopardy for the same offence.” 131 U.S. at 188, 9 S.Ct. at 676 (quoted with approval in Brown, 4-32 U.S. at 168, 97 S.Ct. at 2226). Nielsen therefore stands for the proposition that offenses having a “material part” or “incident” in common must be tried together. 28 Although the Nielsen Court dealt with a prior conviction, not an acquittal, Brown makes clear that the rule of compulsory joinder applies regardless of the outcome of the first trial. Accordingly, because a material part of the new substantive counts — the scheme to defraud — has been previously tried, the government may not relitigate that issue. 29

Nielsen and Brown require that two separately punishable offenses be tried together if they have a significant core of fact in common. 30 “It has long been under stood that separate statutory crimes need not be identical — either in constituent elements or in actual proof — -in order to be the same within the meaning of the constitutional prohibition [on successive trials].” Brown, 432 U.S. at 164, 97 S.Ct. at 2224. The separate mail fraud counts, though not identical in terms of proof, are sufficiently similar to require joinder. Further, the rule of joinder applies here despite the fact that if allowed to relitigate the scheme to defraud, the government would rely on different evidence, i.e., evidence of the Rinden-Mendelow transactions instead of evidence of the Tindall transactions. “[Central to the objective of the prohibition against successive trials” is the notion that the prosecution should not be afforded “another opportunity to supply evidence which it failed to muster in the first proceeding.” Burks v. United States, 437 U.S. 1, 11, 98 S.Ct. 2141, 2147, 57 L.Ed.2d 1 (1978); see also Sanabria v. United States, 437 U.S. 54, 69-74, 98 S.Ct. 2170, 2181-84, 57 L.Ed.2d 43 (1978) (defendant’s acquittal after prosecution based on one theory of liability precludes retrial on alternative basis of liability). Thus, since the government failed to join in one proceeding all the mail fraud counts arising from the same scheme to defraud, it cannot prosecute defendants on the new substantive counts.

In sum, the double jeopardy clause requires dismissal of the second indictment. Since the conspiracies charged in the two indictments are the same conspiracy, neither defendant can be retried for that offense. Further, Richard Allen’s acquittal at the previous trial precludes the government from trying him on the new substantive counts. Finally, the rule of compulsory joinder bars both defendants’ prosecution on those counts.

C. Waiver

The government contends that even if the double jeopardy clause would otherwise bar the second trial, that is, even if the indictments allege the same conspiracy and scheme to defraud, defendants waived their double jeopardy rights during the investigative stage of this case by misleading investigators. According to this argument, even where a single conspiracy is present, the government may try a defendant repeatedly for that offense if, during the criminal investigation, he denied participation in the conspiracy and advised his co-conspirators to do the same. This contention is unsound in both law and fact.

The government argues that the defendants waived their rights not to be tried on the second indictment because they: (1) denied participation in the kickback scheme to investigators; (2) advised Lilly and Tindall to “keep quiet” during the investigation; and (3) told Rinden that the investigation might “blow over.” These actions allegedly made it more difficult for the government to gather evidence against the defendants. However, assuming that such a waiver theory is to be recognized, the facts of this case do not warrant its application to defendants. “ ‘Waiver’ is a vague term used for a great variety of purposes, good and bad, in the law. In any normal sense, however, it connotes some kind of voluntary knowing relinquishment of a right.” Green, supra, 355 U.S. at 191, 78 S.Ct. at 225. Having tried defendants on the Tindall kickback transactions under the first indictment, the government seeks through the second indictment to try them solely on the Rinden-Mendelow portion of the scheme. Thus, to waive their double jeopardy rights as to the second indictment, defendants, at a minimum, had to take steps during the investigation to cover up the Rinden-Mendelow transactions in particular. In other words, in concealing the conspiracy, the defendants themselves must somehow have evinced their understanding that there was a distinction between the Tindall portion of the conspiracy and the Rinden-Mendelow portion, with the coverup focusing on the latter. If defendants’ conduct is to transform a single offense into a multiple offense, there must be evidence that the defendants somehow contemplated the subdivision. Cf. Jeffers v. United States, 432 U.S. 137, 151-52, 97 S.Ct. 2207, 2216, 53 L.Ed.2d 168 (1977) (no double jeopardy violation to try greater and lesser included offenses separately if defendant opposes their joinder).

During the investigation, the defendants denied involvement in any kickback scheme, and their denials drew no distinctions between Tindall and Rinden/Mendelow. Similarly, any statements Lee Allen may have made to Lilly advising him to “keep quiet” were made in general terms, with no mention of any particular Hughes supplier. Further, any effort Richard Allen may have made to keep Tindall quiet does not suggest a waiver with respect to the Rinden-Mendelow transactions. Indeed, Tindall had nothing to do with those transactions. Finally, there is no evidence that defendants took any steps at all to cover up the Rinden-Mendelow portion of the scheme in particular. Richard Allen’s alleged comments to Rinden to the effect that the investigation might “blow over” is hardly an instruction that he mislead investigators. If anything, the evidence suggests that defendants concentrated their efforts on covering up the Tindall portion of the conspiracy. However, such conduct is not a basis for concluding that defendants’ actions evidenced an intent to permit a subsequent trial on the Rinden-Mendelow transactions. 31 Since defendants devoted no particular effort to concealing the Rinden-Mendelow portion of the scheme, they did not waive their right to be tried only once for that scheme 32

The government’s waiver theory misconstrues the law of double jeopardy as well as the facts of this case. The Supreme Court has consistently rejected the application of waiver theories in the double jeopardy area. See Green, supra, 355 U.S. at 191-97, 78 S.Ct. at 225-28; Burks v. United States, 437 U.S. 1, 17-18, 98 S.Ct. 2141, 2150, 57 L.Ed.2d 1 (1978); United States v. Scott, 437 U.S. 82, 99, 98 S.Ct. 2187, 2198, 57 L.Ed.2d 65 (1978). Further, there can be little doubt that acceptance of the government’s waiver theory “would unduly impair the constitutional prohibition against double jeopardy.” Green, 355 U.S. at 198, 78 S.Ct. at 229. It is anticipated that as an integral part of criminal behavior, the wrongdoer, after violating the law, will take affirmative steps to conceal his criminal act, for example, by fleeing the scene of the crime, by disposing of the tools used in the illegal endeavor, or, as here, by denying his participation in the crime. Nevertheless, the government argues that such conduct amounts to a waiver of double jeopardy protection since, in concealing his criminal activity, the defendant makes the government’s investigative task more difficult. This argument demonstrates a fundamental lack of perception of the government’s role in conducting a criminal investigation. The government alone bears the responsibility of gathering evidence against the accused, for the law places no obligation on the defendant to disclose his criminal activity. Requiring the defendant to refrain from concealing his crime would threaten important policies underlying the Fifth Amendment right against self-incrimination, see, e.g., Doyle v. United States, 426 U.S. 610, 617-19, 96 S.Ct. 2240, 2243-4, 49 L.Ed.2d 91 (1976), and the defendant should not be made to sacrifice that right in order to safeguard his double jeopardy rights. See Green, 355 U.S. at 191-92, 78 S.Ct. at 225-226. 33

Moreover, since some concealment inheres in virtually every crime, acceptance of the waiver theory would potentially deprive every defendant of the benefits of the double jeopardy clause, and for this reason alone, it must be rejected. See Grunewald v. United States, 353 U.S. 391, 77 S.Ct. 963, 1 L.Ed.2d 931 (1957). In Grünewald, the issue was whether the applicable three-year statute of limitations barred the defendants’ prosecution for conspiracy to defraud the United States. The defendants had participated in a conspiracy to prevent the Internal Revenue Service from prosecuting certain taxpayers for tax fraud. The government conceded for purposes of argument that the main goal of the conspiracy — to “fix” the tax prosecutions — was achieved more than three years prior to the return of the indictment. However, within three years of suit, in an effort to thwart the government’s discovery of the conspiracy, the defendants had destroyed certain documents, persuaded one individual to lie to the grand jury, and told several persons to “keep quiet.” The government argued that this “subsidiary cover-up conspiracy” was actually an extension of the main conspiracy to defraud, thereby bringing the latter within the limitations period. The Court rejected this argument, however, stating:

[Ajfter the central criminal purposes of a conspiracy have been attained, a subsidiary conspiracy to conceal may not be implied from circumstantial evidence showing merely that the conspiracy was kept a secret and that the conspirators took care to cover up their crime in order to escape detection and punishment. . . . For every conspiracy is by its very nature secret.... [Ejvery conspiracy will inevitably be followed by action taken to cover the conspirators’ traces. Sanctioning the Government’s theory would for all practical purposes wipe out the statute of limitations in conspiracy cases... .

353 U.S. at 401-02, 77 S.Ct. at 972. Since, as recognized in Grünewald, every conspiracy is followed by acts of concealment, 34 accepting the government’s waiver argument — that acts of concealment permit re-prosecution — would impermissibly eradicate double jeopardy protection in conspiracy and multi-party mail fraud cases. 35

Finally, acceptance of the waiver theory would encourage the government to institute prosecutions prematurely, that is, before sufficient evidence is gathered, a result which the court must avoid. See United States v. Lovasco, 431 U.S. 783, 790-95, 97 S.Ct. 2044, 2048-51, 52 L.Ed.2d 752 (1977). Where the government’s case against the defendant is weak, it could nevertheless take him to trial, hoping to find additional evidence at a later date. If the prosecutor is not satisfied with the results of the first trial and desires a second bite at the apple, he need only find additional evidence which the defendant somehow made difficult of discovery. Indeed, this appears to have happened in this case. The government now concedes that it had no evidence linking Richard Allen to the conspiracy when the first indictment was returned. 36 It claims that such evidence did not become available until one week prior to trial, in the form of Rinden’s testimony, and several months after trial, in the form of Tindall’s testimony, and it argues that Richard Allen’s alleged cover-up activities caused this delay in.discovery. Acceptance of this argument, however, would condone the premature prosecution that occurred in this case and encourage similar prosecutions in the future. For this and the other reasons stated, the court finds the government’s waiver theory to be without merit. 37

Similar to its waiver argument is the government’s claim that prosecution on the new substantive counts is permitted under the “due diligence” exception to the compulsory joinder rule. The Supreme Court has recognized that an exception to compulsory joinder “may exist where the State is unable to proceed on the more serious charge at the outset because the additional facts necessary to sustain that charge .. . have not been discovered despite due diligence.” Brown, supra, 432 U.S. at 169 n.7, 97 S.Ct. at 2227 n.7; accord, Vitale, supra, 447 U.S. at 420 n.8, 100 S.Ct. at 2267 n.8; Jeffers, supra, 432 U.S. at 151-52, 97 S.Ct. at 2216. Unlike the waiver argument, the due diligence exception does not turn on whether the defendant is responsible for the government’s failure to learn of facts underlying an offense. The exception applies as long as the government, despite its own reasonable efforts at investigation, fails to learn of an offense in time to join it with known offenses. Application of the exception here would permit the government to have a separate trial on the new mail fraud counts despite the similarity of those counts to the offenses previously tried. 38

The government rests its “due diligence” argument on the fact that it was still investigating the new substantive counts at the time it brought the first indictment; although the investigation of the Tindall transactions had concluded, investigators were continuing to gather evidence of the Rinden-Mendelow transactions. Lack of proof, however, does not invoke the due diligence exception. That exception applies only if, at the time of the first trial, the government was not aware of the offenses which it failed to join. See Vitale, 447 U.S. at 420 n.8, 100 S.Ct. at 2267 n.8. Where the government has knowledge of those offenses prior to the first indictment, it should defer prosecution on all the joinable charges until its investigation is complete.

In this case, the government unquestionably had knowledge of the facts underlying the new substantive counts at the time it sought the first indictment. The government was aware throughout its investigation that the conspiracy and mail fraud scheme included transactions with Rinden and Mendelow. Indeed, it concedes as much, stating that the prosecutor decided prior to the first indictment to prosecute defendants separately on the Rinden-Mendelow portion of the scheme. That decision, as the court holds today, was not consistent with the commands of the double jeopardy clause, and it is not made valid by the fact that the government’s investigation into that portion of the scheme was continuing. The government should have delayed prosecution on the charges in the first indictment until such time as it was prepared to pursue the substantive counts of the second indictment. 39

III. SUPERVISORY POWER

“All federal courts are endowed with certain inherent supervisory powers over the administration of justice in the courts of the United States and must utilize that power, which comprehends the power to dismiss an indictment, whenever the pursuit of truth and justice becomes tainted.” United States v. Butler, 567 F.2d 885, 893 (9th Cir. 1978) (Ely, J., concurring). Beginning with the landmark decision in McNabb v. United States, 318 U.S. 332, 63 S.Ct. 608, 87 L.Ed. 819 (1943), the federal courts have exercised their supervisory power over the administration of criminal justice to ensure that a defendant’s prosecution is conducted with fairness and integrity. Because defendants’ prosecution under the second indictment would be manifestly unjust, the court holds, as an alternate ground of decision, that the indictment must be dismissed pursuant to its supervisory power.

In McNabb, the defendants were suspected of murdering a federal agent. Without taking the defendants before a committing judicial officer, which was statutorily required, investigators detained them for two days, subjecting them to intensive questioning. During the interrogation, each of the three defendants made incriminating statements, which were subsequently used in convicting them of second-degree murder. They appealed their convictions, and the court of appeals affirmed. Before the Supreme Court, the defendants argued that the trial court violated their Fifth Amendment rights by admitting the incriminating statements. In reversing the convictions, the Court declined to reach the constitutional issue, stating that

the scope of our reviewing power over convictions brought here from the federal courts is not confined to ascertainment of Constitutional validity. Judicial supervision of the administration of criminal justice in the federal courts implies the duty of establishing and maintaining civilized standards of procedure and evidence.

... [A] decent regard for the duty of courts as agencies of justice and custodians of liberty forbids that men should be convicted upon evidence secured under the circumstances revealed here.

318 U.S. at 340, 347, 63 S.Ct. 608, 616, 87 L.Ed. 819. 40

Subsequent to McNabb, the Court has invoked its supervisory power to reverse convictions based on the discredited testimony of a government informant, Mesarosh v. United States, 352 U.S. 1, 77 S.Ct. 1, 1 L.Ed.2d 1 (1956), to reverse a conviction where the defendant was cross-examined at trial about his plea of Fifth Amendment privilege before the grand jury, Grunewald v. United States, 353 U.S. 391, 423-24, 77 S.Ct. 963, 983-84, 1 L.Ed.2d 931 (1957), to require a prosecutor to provide the defendant with government reports relating to the testimony of two informants, Jencks v. United States, 353 U.S. 657, 77 S.Ct. 1007, 1 L.Ed.2d 1103 (1957), to enjoin a federal officer from testifying in a state criminal prosecution concerning evidence obtained through an illegal search, Rea v. United States, 350 U.S. 214, 76 S.Ct. 292, 100 L.Ed. 233 (1956), and to reverse a conviction where the defendant was cross-examined at trial about his failure to provide the police with his alibi at the time of arrest, United States v. Hale, 422 U.S. 171, 95 S.Ct. 2133, 45 L.Ed.2d 99 (1975). “Whether designed to do justice in a particular fact situation or to establish general standards, these supervisory-power decisions reflect concern about the degree of fairness in the judicial process.” Note, The Supervisory Power of the Federal Courts, 76 Harv.L.Rev. 1656, 1659 (1963). See also Hampton v. United States, 425 U.S. 484, 493-94, 96 S.Ct. 1646, 1651-52, 48 L.Ed.2d 113 (1976) (Powell, J., concurring). (supervisory power should be available to bar conviction on ground of “outrageous police conduct”).

Although the case law indicates that, in using its supervisory power, the court should tailor the remedy to the injury, it is clear that dismissal of the indictment, where appropriate, is one of the remedies which may be employed. See United States v. Samango, 607 F.2d 877 (9th Cir. 1979) (dismissing indictment handed down by biased grand jury); United States v. Owen, 580 F.2d 365, 367 (9th Cir. 1978) (“Under its inherent supervisory powers, a federal court is empowered to dismiss an indictment on the basis of governmental misconduct.”); United States v. Dahlstrum, 493 F.Supp. 966, 974-75 & 975 n.16 (C.D.Cal.1980) (bad-faith use of civil summons by Internal Revenue Service to further criminal investigation warranted dismissal of indictment), appeal dismissed, 655 F.2d 971 (9th Cir. 1981), cert. denied,-U.S.-, 102 S.Ct. 1293, 71 L.Ed.2d 472 (1982); United States v. Pascal, 496 F.Supp. 313 (N.D.Ill.1979) (failure of government agents to honor non-prosecution agreement with defendant-informant warranted dismissal of indictment); United States v. Ottley, 439 F.Supp. 587 (S.D.N.Y.1977) (dismissing second count of indictment where defendant had been previously tried for similar offense and evidence supporting second count was in government’s possession at time of that trial); United States v. DeMarco, 407 F.Supp. 107 (C.D.Cal.1975) (dismissing indictment where prosecutor failed to provide defendant with exculpatory evidence).

Dismissal of an indictment pursuant to the court’s supervisory power may be appropriate where the indictment charges an offense nearly identical to one previously tried. Guido v. United States, 597 F.2d 194 (9th Cir. 1979) (per curiam). In Guido, the defendants had entered a plea of guilty to an indictment brought in the Eastern District of California charging them with conspiracy to import marijuana. A year later, a federal grand jury in Arizona returned an indictment against the defendants charging them, inter alia, with conspiracy to possess marijuana with intent to distribute. Prior to the filing of the second indictment, the Arizona prosecutor had learned of the California prosecution and its underlying facts. The defendants sought dismissal of the Arizona indictment on double jeopardy and due process grounds, claiming that it charged the same offense as the California indictment. The district court denied the motion to dismiss, and the defendants were subsequently tried and convicted.

On appeal, defendants renewed their constitutional challenges to the indictment. The Ninth Circuit declined to reach those claims, however, reversing the convictions under its supervisory power instead. The court compared the two conspiracies and concluded that they involved the same objectives, the same key participants (the defendants), the same method of operation, and the same period of time. Although the conspiracies were separately punishable, see United States v. Burkett, 612 F.2d 449, 452 (9th Cir. 1979) (discussing Guido), cert. denied, 447 U.S. 905, 100 S.Ct. 2985, 64 L.Ed.2d 853 (1980), the court nevertheless viewed the second prosecution as inherently unfair. Despite a difference in participants between the two conspiracies, the court found the offenses practically indistinguishable since they shared the same purpose, that of importing and distributing marijuana. See id. The Arizona prosecutor’s knowledge of the California indictment should have put him on notice that he was charging defendants with virtually the same offense. Because he nevertheless brought the indictment, the court invoked its supervisory power to correct such an “unjust result.”

Guido provides strong support for dismissing the indictment in this case. Even if the conspiracy offenses charged in the two indictments are separately punishable, they are virtually identical in terms of proof. The only distinction between the conspiracies charged is a difference in the supplier to Hughes, a factor which the Guido court did not consider significant. More importantly, both conspiracies involve the same purpose, to defraud Hughes, a factor which the court did find significant in Guido. Further, as in Guido, the prosecutor knew about the two conspiracies when the decision was made to seek separate prosecutions. Thus, Guido’s holding — that two separate conspiracies, identical but for a minor difference in participants, must be tried together — applies with full force here. Prosecution under the second indictment would result in unnecessary harassment of the defendants, and the court must not permit such an injustice.

As to the substantive counts, Guido also compels dismissal. The new mail fraud counts are virtually identical to the new conspiracy count; indeed, every mailing alleged under Section 1341 is also alleged as an overt act of the conspiracy. The same facts underlie both the substantive counts and the conspiracy, and therefore Guido would require joinder of those offenses. Since the new conspiracy count should have been joined in the first indictment, it follows that the new substantive counts should also have been joined.

Alternatively, the substantial similarity between the substantive counts in the two indictments warrants dismissal of the new counts. Permitting separate trials on mailings connected by the same scheme to defraud would be grossly unjust.

Although the similarity of the indictments alone is sufficient to permit dismissal of the second indictment, additional factors also make that remedy appropriate. Having carefully reviewed the procedural history of this case, the court is compelled to conclude that, from the oütset, the government viewed the previous trial as a dry run for later cases. Apparently dissatisfied with the results of that prosecution, it wasted no time in strengthening its case against the defendants. The prosecutor now has additional documents and witnesses — one of whom, Tindall, was convicted in the previous trial — to bolster his case. Further, it is clear that the previous prosecution was initiated prematurely, obviously before the government was fully prepared to try its whole case. The government’s last-minute effort to introduce Rinden’s testimony against the defendants, the lack of substantial evidence connecting Richard Allen with the conspiracy, and the circumstances surrounding the return of the first indictment 41 all support that conclusion. Given that the offenses charged in the second indictment are nearly identical to those previously tried, the court concludes that further prosecution of defendants would be oppressive. The government’s failure to “act with due regard for the integrity of the administration of justice,” United States v. Basurto, 497 F.2d 781, 793 (Hufstedler, J., concurring), makes this an appropriate case for exercising the court’s supervisory power, and, accordingly, the indictment must be dismissed. 42

CONCLUSION

The retrial prohibition of the double jeopardy clause requires dismissal of the indictment. Defendants cannot be tried on the conspiracy count since they have already been tried for that offense. Further, the doctrine of collateral estoppel precludes the government from litigating the substantive counts as to defendant Richard Allen. Finally, the rule of compulsory joinder requires dismissal of the substantive counts as to both defendants. The fact that defendants may have acted to conceal their criminal activity does not constitute a waiver of double jeopardy protection. Nor is the government excused from joining offenses under the compulsory joinder rule simply because its separate investigations into those offenses did not end simultaneously.

Alternatively, the court dismisses the indictment pursuant to its supervisory power over the administration of justice. Prosecution under the second indictment would result in manifest injustice in light of the similarity of the two indictments and the circumstances surrounding the previous trial.

Accordingly,

IT IS ORDERED that the indictment be dismissed with prejudice.

1 . Tindall was president of Ti-Con Industries, Inc. Rinden and Mendelow were co-owners of Trans American Plastics Corporation (“Trans American”), and Mendelow separately operated Diversified Engineering and a branch office of E. D. P. Industries, Inc.

2 . The government sought to introduce Rinden’s testimony pursuant to Rule 404(b), Fed.R.Evid., which provides that “[e]vidence of other crimes, wrongs, or acts . . . may ... be admissible ... [to prove] motive, opportunity, intent, preparation, plan, knowledge, identity, or absence of mistake or accident.”

3 . The government also made a belated request to introduce certain statements made by Richard Allen to a government investigator on May 19, 1978. The court denied the request after the government conceded that it had previously represented to the court that the statements would not be used:

THE COURT: [W]as there a representation by the government before that they were going to use those statements?

MR. DEIXLER: My understanding, your Honor, is that ... [the government] represented to the Court in connection with the severance motion that . . . [many] of the statements which we would like to introduce at this time would not be introduced.

THE COURT: That’s correct. MR. DEIXLER: Since that time, your Honor, the government’s view of the case has changed and at this time the government would like to introduce those statements into evidence.

THE COURT: Well, my problem, Mr. Deixler, is that I made a whole series of decisions in this case, very difficult ones, based on representations about what statements were going to be used ....

... I would have done something totally different on the severance matter, I think, if [some of these statements had been offered earlier].

MR. DEIXLER: .. . What has occurred here, [due] . . . perhaps [to] a strategy decision which the government regrets making, is that—

THE COURT: ... [I]t’s been made, and it will be lived up to. That’s it.

Trial Transcript [“Tr.”] at 752-53, 758, 759.

4 . Although the court did not permit Rinden to testify, thereby foreclosing one source of evidence of the Rinden-Mendelow transactions, the government successfully sought to introduce evidence of those transactions through Watkins’ testimony. Preliminary to the admission of that evidence, when asked by the court to explain the relevance of Watkins’ testimony about Trans American, the prosecutor stated:

That [company] is absolutely relevant. It comes up on the ... [tape recording of the February 14, 1978 conversation between Lee Allen and Watkins], your Honor, and is repeatedly discussed by Mr. Tindall and Mr. Allen to demonstrate to Miss Watkins the scope of the scheme. It is absolutely relevant to Lee Allen’s knowledge and the operation of the .. . [kickback] scheme which he describes in great detail on the tape.

It absolutely puts into context the operation of the entire conspiracy. It is completely relevant to the case.

Tr. at 871 (emphasis added).

5 . On July 9, 1980, the day before the case was submitted to the jury, the court entered judgments of acquittal in favor of Allen and Tindall on several of the mail fraud counts (five as to Allen, six as to Tindall).

6 . Of course, the indictment makes no mention of Ti-Con, Tindall’s company, but instead focuses on kickback transactions with Trans American and with Mendelow’s two companies. See note 1 supra. The overt acts of the conspiracy count make clear, however, that the indictment is principally concerned with Trans American’s role in the conspiracy as opposed to that of the other two suppliers.

7 . At the second trial, the government would call Rinden and Tindall as witnesses, neither of whom testified against defendants at the previous trial.

8 . At the previous trial, the government sought to connect Richard Allen with the conspiracy through the use of checks sent to Contemporary Associates, Richard’s company, from Ti-Con and through checks sent from Contemporary Associates to Lee Allen. The court found that those checks, without more, were insufficient as a matter of law to link Richard to the conspiracy, and it acquitted him. Approximately one month after the trial, the government subpoenaed additional documents from Richard Allen for reasons best explained by the government agent in charge of the investigation:

I realized that since the government had not yet obtained a complete record of ... [the Contemporary Associates] checks and deposits, [Richard Allen’s connection to the conspiracy] .. . was a confusing issue during the Tindall conspiracy trial and in view of this I decided to subpoena additional records from .. . [him] that could be used along with the incomplete records I had previously obtained from the banks where the Contemporary Associates accounts were located, to overcome a similar [acquittal] argument in the later cases ....

Government Declarations Re Investigation and Indictment at 19 (filed Feb. 1, 1982).

9 . The clause provides that no person shall “be subject for the same offence to be twice put in jeopardy of life or limb.” U.S.Const.Amend. V.

10 . Determining what constitutes the “same offense” has proved the most difficult and most frequently litigated double jeopardy issue. See Note, The Double Jeopardy Clause as a Bar to Reintroducing Evidence, 89 Yale L.J. 962, 963-69 (1980) (discussing various same-offense tests).

11 . In United States v. Brooklier, 637 F.2d 620, 624 (9th Cir. 1980), cert. denied, 450 U.S. 980, 101 S.Ct. 1514, 67 L.Ed.2d 815 (1981), the Ninth Circuit stated that “the Blockburger test, and nothing more, governs all post-conviction prosecutions.” Nevertheless, Brooklier, unlike the instant case, involved successive prosecutions under different statutory provisions, and its broad endorsement of Blockburger should be limited to that factual context. See United States v. Bendis, 681 F. 561, 565 n.4 (9th Cir. 1982). To apply Blockburger across the board would be contrary to the Supreme Court decisions in Braverman and Albernaz as well as to several Ninth Circuit decisions. See, e.g., United States v. Mayo, 646 F.2d 369, 372 (9th Cir.) (per curiam), cert. denied, - U.S. -, 102 S.Ct. 979, 71 L.Ed.2d 115 (1981); United States v. Johnson, 637 F.2d 1224, 1240 (9th Cir. 1980); United States v. Burkett, 612 F.2d 449, 451 (9th Cir. 1979). Moreover, Brooklier's apparent adoption of Blockburger as the sole same-offense test must be read in light of the specific issue presented in that case; the court was choosing between two alternative same-offense tests for evaluating successive prosecutions, the Blockburger test and what is commonly called the “same transaction” test. In rejecting the latter, the court simply remarked that Blockburger is the governing standard. Thus, Brooklier is not controlling where, as here, the court is not relying on the same-transaction test.

12 . It appears that the Tenth Circuit stands alone in applying Kotteakos in the double jeopardy context. Compare United States v. McMurray, 680 F.2d 695 (10th Cir. 1981) (en banc) with United States v. Bendis, 681 F.2d 561, 565 n.5 (9th Cir. 1982) (citing cases) and United States v. Jabara, 644 F.2d 574, 577 (6th Cir. 1981).

13 . Although the indictments allege that the first conspiracy ended in February 1978 and that the second conspiracy ended in April 1978, the last overt act alleged for both conspiracies occurred on February 14, 1978.

14 . It is also important that several of the overt acts alleged in the indictments, though not the same exact acts, are of similar character. For example, both conspiracies involved purchase orders mailed by Lilly as well as payments made to Contemporary Associates. Such similarity further indicates the singleness of the conspiracies. It should not be necessary for purposes of the Arnold analysis that the overt acts in the indictments be precisely the same acts. Indeed, many distinct overt acts may further a single conspiracy, as the first indictment itself demonstrates. Obviously, where the two indictments involve different suppliers, the prosecutor can easily avoid alleging the same overt acts. In that type of situation, if the Arnold factor of comparing overt acts is to have any distinction from the factor of comparing the conspirators involved, the overt acts should be compared to determine whether they are strikingly similar rather than whether they are identical.

15 . When the government attempted to add Rinden to its witness list, the prosecutor stated that “Mr. Rinden .. . would essentially say that he was involved in an identical scheme. He played the role of Mr. Tindall, the only difference between the schemes.” Tr. at 764.

16 . Having concluded that the indictments allege the same conspiracy, it follows that the two sets of transactions are simply alternative sources of evidence Underlying a single offense. Thus, the government would have faced no preindictment delay problems in deferring prosecution of the conspiracy until such time as its investigations into both sets of transactions had concluded. The fact that the investigation into the Tindall transactions concluded before all the evidence was gathered on the Rinden-Mendelow transactions does not justify separate prosecutions. See United States v. Lovasco, 431 U.S. 783, 97 S.Ct. 2044, 52 L.Ed.2d 752 (1977) (preindictment delay permissible where government defers prosecution pending end of investigation).

Additionally, the government cannot now be heard to complain that had it initially joined the two sets of transactions in one indictment, defendants might have moved for severance. Such hypothetical do not aid in the analysis of double jeopardy problems since “[ljegal consequences ordinarily flow from what has actually happened, not from what a party might have done from the vantage of hindsight.” Sanabria v. United States, 437 U.S. 54, 65, 98 S.Ct. 2170, 2179, 57 L.Ed.2d 43 (1978).

17 . If Congress intended a defendant’s actions to be punishable as two separate offenses, then the double jeopardy clause does not prevent the imposition of multiple sentences. See Albernaz v. United States, 450 U.S. 333, 344, 101 S.Ct. 1137, 1145, 67 L.Ed.2d 275 (1981). This is so because the “power to define criminal offenses and to prescribe punishments to be imposed upon those found guilty of them, resides wholly with the Congress.” Whalen v. United States, 445 U.S. 684, 689, 100 S.Ct. 1432, 1436, 63 L.Ed.2d 715 (1980). The function of the double jeopardy clause in the punishment area, then, is to determine legislative intent, which is discerned through the use of the Blockburger test. See Albernaz, 450 U.S. at 340, 101 S.Ct. at 1143. Thus, if mail fraud and conspiracy to commit mail fraud each require proof of a fact not required for proof of the other, it is presumed that Congress intended the particular defendant’s conduct to be punishable under both statutes, 18 U.S.C. §§ 371, 1341. Since mail fraud, unlike conspiracy, requires use of the mails and conspiracy, unlike mail fraud, requires the participation of more than one person, the two offenses are separately punishable. See also Ianelli v. United States, 420 U.S. 770, 785 n.17, 95 S.Ct. 1284, 1293 n.17, 43 L.Ed.2d 616 (1975) (applying Blockburger in conspiracy context to decide multiple-punishment claim).

18 . The fact that, under the double jeopardy clause, a defendant’s criminal activity may give rise to multiple sentences but only a single trial simply recognizes that the “same offense” concept changes from the multiple-punishment context to the retrial area. In the former, the clause ensures that the court imposes punishment consistent with legislative intent; in the retrial area, it serves a constitutional policy of finality for the defendant’s benefit, protecting him from the burdens of the trial process itself. See Brown v. Ohio, 432 U.S. 161, 165-66, 97 S.Ct. 2221, 2225, 53 L.Ed.2d 187 (1977); Note, Twice in Jeopardy, 75 Yale L.J. 262, 267, 277-79, 302 (1965); Note, The Protection from Multiple Trials, 11 Stan.L.Rev. 735, 736-39 (1959).

19 . The general federal conspiracy statute provides:

If two or more persons conspire either to commit any offense against the United States, or to defraud the United States, . . . and one or more of such persons do any act to effect the object of the conspiracy, each shall be fined not more than $10,000 or imprisoned not more than five years, or both.

The mail fraud statute provides;

Whoever, having devised or intending to devise any scheme or artifice to defraud, ... for the purpose of executing such scheme or artifice . .. places in any post office ... any matter or thing whatever to be sent or delivered by the Postal Service, or takes or receives therefrom, any such matter or thing, or knowingly causes to be delivered by mail . . . any such matter or thing, shall be fined not more than $1,000 or imprisoned not more than five years, or both.

20 . The government’s opening and closing statements at the previous trial also illustrate that the conspiratorial agreement is the scheme to defraud. At the outset of her opening statement, the prosecutor told the jury that “this is a conspiracy case,” Tr. at 263, and she then proceeded to describe the operation of that conspiracy. Significantly, when the prosecutor moved on to discuss the mail fraud scheme, she remarked, “[fjor .. . mail fraud . .. you have to have a scheme to defraud and you have to have something mail[ed] in the United States Mail in furtherance of the scheme to defraud. . .. The scheme to defraud, I have already outlined to you .. . . ” Tr. at 274-75. Similarly, in the closing argument, the prosecutor explained the elements of conspiracy and mail fraud. After describing Count 1, the conspiracy count, she introduced the substantive charges, saying, “[n]ow let’s break these down into . .. [their] elements. First of all, a scheme to defraud. This goes back to what I read to you in the Count 1 .... ” Tr. at 1397.

21 . Vitale, Jordan, and Sabella all relied on Nielsen in employing the actual-evidence test. Nielsen, in turn, borrowed the test from its original source, Morey v. Commonwealth, 108 Mass. 433, 434 (1871). See 131 U.S. at 188, 9 S.Ct. at 676.

22 . The elements of mail fraud are a scheme to defraud and knowing use of the mails in furtherance of the scheme. United States v. Beecroft, 608 F.2d 753, 757 (9th Cir. 1979). The elements of conspiracy are an agreement among two or more persons to commit a substantive offense and an overt act in furtherance of the conspiracy. United States v. Friedman, 593 F.2d 109, 115 (9th Cir. 1979). Here, in proving the new mail fraud counts, the government would, in establishing the scheme to defraud, also establish the agreement between Lilly and the Allens to defraud Hughes; the government cannot prove the scheme without proving the conspiratorial agreement. In establishing use of the mails for the substantive counts, the government would satisfy the overt act requirement of the conspiracy count; although the overt act element can be satisfied by some act other than use of the mails, see Ianelli v. United States, 420 U.S. 770, 785 n.17, 95 S.Ct. 1284, 1293 n.17, 43 L.Ed.2d 616 (1975), it is sufficient for purposes of the actual-evidence test that use of the mails does establish that element. Indeed, in the second indictment, every mailing giving rise to a mail fraud count is also alleged as an overt act of the conspiracy. Finally, the evidence offered on the substantive counts would satisfy the intent requirement of the conspiracy count. See United States v. Craig, 573 F.2d 455, 485 (7th Cir. 1977), cert. denied, 439 U.S. 820, 99 S.Ct. 82, 58 L.Ed.2d 110 (1978); Phillips v. United States, 356 F.2d 297 (9th Cir. 1965), cert. denied, 384 U.S. 952, 86 S.Ct. 1573, 16 L.Ed.2d 548(1966).

23 . The trial transcript, from preliminary statements to jury instructions, including transcriptions of the two tape recordings played to the jury, consists of approximately 1,485 pages. Only sixty-three of those pages concern the issue of use of the mails.

24 . An examination of the overt acts alleged in the conspiracy count of the second indictment reveals that there are a number of additional mailings which could be made the basis of still more Section 1341 charges. See 1981 Indictment at 17, 18, ¶¶ D.14.dd, ee, gg.

25 . Even now it is not clear if this will be defendants’ last prosecution on the mail fraud scheme. See note 24 supra. In this respect it is interesting to note that the second indictment, against only the defendants, was not handed down until after Lilly’s convictions were affirmed on appeal; no doubt he, too, could have been charged in the second indictment.

26 . The defendant in Brown was initially tried and convicted for the offense of joyriding — taking or operating a car without the owner’s consent. He was then charged and convicted for auto theft, which consisted of joyriding with the intent permanently to deprive the owner of possession. Thus, the only difference between the offenses was that auto theft required proof of an additional fact, the defendant’s intent. Applying the Blockburger test, the Court determined that joyriding and auto theft were but one constitutional offense for purposes of trial:

As is invariably true of a greater and lesser included offense, the lesser offense — joyriding — requires no proof beyond that which is required for conviction of the greater — auto theft. The greater offense is therefore by definition the “same” for purposes of double jeopardy as any lesser offense included in it.

432 U.S. at 168, 97 S.Ct. at 2226.

27 . Blockburger is not an end in itself but is simply a means of resolving the particular double jeopardy issue before the court. For example, in the multiple-punishment context, where Blockburger’s purpose is to discern legislative intent, see notes 17 & 18 supra, the Court has recently noted that that test need not be followed if Congress has otherwise made its purpose known. See Albernaz v. United States, 450 U.S. 333, 340, 101 S.Ct. 1137, 1143, 67 L.Ed.2d 275 (1981) (“because it [Blockburger] serves as a means of discerning congressional purpose the rule should not be controlling, where, for example, there is a clear indication of contrary legislative intent”); United States v. Hartley, 678 F.2d 961, 991-92 (11th Cir. 1982) (finding separately punishable offenses where Blockburger indicated one offense) (applying Albernaz). In the retrial area, Blockburger serves to determine whether two offenses are so similar in terms of proof that the defendant’s interest in finality requires their joinder. Thus, where, as here, the offenses at issue are as similar as those in Brown, the double jeopardy clause should require their joinder even though Blockburger does not. To hold otherwise would exalt form over substance, which is to be avoided. United States v. DiFrancesco, 449 U.S. 117, 142, 101 S.Ct. 426, 440, 66 L.Ed.2d 328 (1980).

28 . Although the Court in Brown relied on Nielsen, it based its decision on Blockburger:

Because we conclude today that a lesser included and a greater offense are the same under Blockburger, we need not decide whether the repetition of proof required by the successive prosecutions against Brown would otherwise entitle him to the additional protection offered by ... Nielsen.

432 U.S. at 166 n.6, 97 S.Ct. at 2226 n.6; see also Jeffers v. United States, 432 U.S. 137, 151 n.17, 97 S.Ct. 2207, 2216 n.17, 53 L.Ed.2d 168 (1977) (reserving decision on whether double jeopardy clause prevents repetition of proof). The Court therefore drew a distinction between the Blockburger rule of compulsory joinder and the Nielsen rule, characterizing the latter as a protection against “repetition of proof.” Of course, as discussed above, the Blockburger joinder requirement also has the effect of preventing a repetition of proof. Actually, both rules are simply alternative methods of answering the ultimate question posed by any joinder claim — whether the offenses at issue are sufficiently similar to require joinder. Blockburger offers an indirect means of resolving that issue, focusing on the differences between the offenses; Nielsen provides a direct approach, examining the similarity of the offenses.

It should be noted that in applying Blockburger in the multiple-punishment area, the Court has stated that “[i]f each [offense] requires proof of a fact that the other does not, . .. [multiple sentences are permissible] notwithstanding a substantial overlap in the proof offered to establish the crimes.” Ianelli v. United states, 420 U.S. 770, 785 n.17, 95 S.Ct. 1284, 1293 n.17, 43 L.Ed.2d 616 (1975) (emphasis added) (quoted with approval in Brown, 432 U.S. at 166, 97 S.Ct. at 2225). This statement simply recognizes that the question of whether Congress has authorized separate punishments is not controlled by the overlap in proof. In the compulsory joinder context, however, as Brown and Nielsen make clear, that overlap is precisely what triggers the defendant’s interest in finality.

29 . The first indictment charged defendants with aiding and abetting mail fraud under 18 U.S.C. §§ 2, 1341, while the second indictment charges them simply with mail fraud under 18 U.S.C. § 1341. Nevertheless, the second trial would still expose defendants to a repetition of proof, since, in proving aiding and abetting at the previous trial, the government had to, and did, litigate the scheme to defraud. See United States v. Ruffin, 613 F.2d 408, 412 (2d Cir. 1979) (conviction for aiding and abetting requires proof that someone committed principal offense).

30 . The compulsory joinder rule announced here is to be distinguished from the “same transaction” test, which has never been explicitly approved or disapproved by the Supreme Court. That test requires “prosecution in one proceeding ... of ‘all the charges against a defendant that grow out of a single criminal act, occurrence, episode, or transaction.’ ” Brown, 432 U.S. at 170, 97 S.Ct. at 2227 (Brennan, J., concurring) (quoting Ashe v. Swenson, 397 U.S. 436, 453 54, 90 S.Ct. 1189, 1199, 25 L.Ed.2d 469 (1970) (Brennan, J., concurring)). Unlike a joinder rule based on Brown and Nielsen, the same-transaction test requires joinder without reference to the statutory elements of the offenses and without analysis of the overlap in proof.

31 . The government’s argument assumes that, absent waiver, a second trial would be impermissible under the double jeopardy clause. The basis of the waiver theory is that a defendant, by actively misleading investigators, increases the government’s burden of proving his guilt and that a guilty defendant should not be permitted to escape punishment under such circumstances. Thus, the theory appears to have no application to Lee Allen since he was convicted at the previous trial.

32 . Since the Rinden-Mendelow transactions underlie both the conspiracy count and the substantive counts, the court’s conclusion on the waiver issue applies to the entire indictment.

33 . The government’s suggestion that Richard Allen waived his double jeopardy rights by denying his participation in the conspiracy seems most odd in light of the fact that he was acquitted at the previous trial.

34 . While Grünewald dealt only with conspiracy, its observations apply equally to multi-party mail fraud schemes.

35 . Of course, a defendant’s acts of concealment at some point may so interfere with the administration of justice that they will become criminal acts themselves. See 18 U.S.C. § 1503 (obstruction of justice). Indeed, in Grünewald, one of the defendants was convicted for attempting to influence witnesses appearing before the grand jury.

36 . The government has admitted that “[t]he only such evidence [linking Richard Allen to the conspiracy] which arguably was available in 1980 was the testimony of Rinden. ... [He] did not disclose to the government until June 25, 1980 that he could testify to direct conversations with Richard Allen in which Allen delineated the corrupt and illicit purposes of the payments from suppliers.” Government’s Petition for Rehearing at 33-34 (filed Feb. 26, 1982).

37 . As support for its waiver theory, the government relies on cases permitting a second trial where the defendant’s conduct has necessitated the early termination of the first trial. See, e.g., United States v. Mastrangelo, 662 F.2d 946 (2d Cir. 1981) (first trial terminated where defendant participated in murder of government’s only witness), cert. denied - U.S. -, 102 S.Ct. 2236, 72 L.Ed.2d 847 (1982); McKissick v. United States, 398 F.2d 342 (5th Cir. 1968) (first trial terminated where court learned that defendant, who testified, had committed perjury). Those cases, all of which are mistrial cases, simply recognize that retrial is permissible if a defendant’s initial trial is aborted prior to judgment for “manifest necessity.” See Arizona v. Washington, 434 U.S. 497, 98 S.Ct. 824, 54 L.Ed.2d 717 (1978) (discussing mistrial standard). Such necessity may be caused by the prosecutor’s conduct, Illinois v. Somerville, 410 U.S. 458, 93 S.Ct. 1066, 35 L.Ed.2d 425 (1973) (failure to properly draft indictment), the defendant’s conduct, cf. Arizona v. Washington, supra (defense counsel made improper and prejudicial comment in opening statement), or the jury’s inability to reach a verdict, United States v. Perez, 22 U.S. (9 Wheat.) 579, 6 L.Ed. 165 (1824). In those circumstances, the second trial is permitted in order to provide the government with “one complete opportunity to convict those who have violated its laws.” Arizona v. Washington, 434 U.S. at 509, 98 S.Ct. at 832. However, the Supreme Court has never applied the “manifest necessity” standard outside the mistrial area, that is, where the first trial has proceeded to judgment. This is so because “[a] primary purpose [of the double jeopardy clause is] to preserve the finality of judgments.” Crist v. Bretz, 437 U.S. 28, 33, 98 S.Ct. 2156, 2159, 57 L.Ed.2d 24 (1978). Thus, where, as here, the initial trial resulted in conviction or acquittal, the mistrial cases are of no import. Moreover, where a defendant’s acts of concealment occur prior to trial and the government nonetheless proceeds to trial believing its evidence is sufficient, “the prosecutor [has been afforded] one full and fair opportunity to present his evidence [to the trier of fact].” Arizona v. Washington, 434 U.S. at 505, 98 S.Ct. at 830.

38 . The exception, even if applicable, would not permit a second prosecution on the conspiracy. The cases suggesting the existence of the due diligence exception have done so only in the context of compulsory joinder, that is, where the offenses at issue are separately punishable. Indeed, where the government is seeking to try a defendant a second time for the same punishable offense, the double jeopardy clause bars the second trial primarily to prevent the government from presenting facts which it did not use at the first trial. See Burks v. United States, 437 U.S. 1, 11, 98 S.Ct. 2141, 2147, 57 L.Ed.2d 1 (1978). Since the conspiracy counts in the indictments allege the same punishable offense, the due diligence exception applies only to the compulsory joinder of the new substantive counts with the offenses previously tried. See Part II.B.2 supra. Similarly, the exception does not apply to the collateral estoppel protection afforded Richard Allen as to the new substantive counts; even if those offenses were belatedly discovered by the government, an ultimate fact necessary for conviction on those counts has been previously litigated between the parties and found in Richard’s favor.

39 . Since the separate mail fraud counts constitute a single offense for purposes of trial, the government would not have faced any preindictment delay problems in trying all thirty mailings at the time of the second indictment. The government’s continuing investigation into some of those mailings permitted a delay in prosecution on those counts already investigated. See Lovasco v. United States, 431 U.S. 783, 97 S.Ct. 2044, 52 L.Ed.2d 752 (1977).

An exception to joinder in the mail fraud context may exist where the statute of limitations is about to run on some of the mailings. Such an exception would not justify the separate prosecutions in this case, however. Indeed, the mailings in the first indictment covered the same time period as those in the subsequent indictment.

40 . As noted in McNabb, the scope of the court’s supervisory power is not confined to the reach of the Constitution. Accord, United States v. Cortina, 630 F.2d 1207, 1216 (7th Cir. 1980) (construing United States v. Payner, 447 U.S. 727, 100 S.Ct. 2439, 65 L.Ed.2d 468 (1980)). Thus, the fact that the double jeopardy clause might not bar the second trial does not preclude the court from dismissing the indictment pursuant to its supervisory power. United States v. Ottley, 439 F.Supp. 587 (S.D.N.Y.1977) (although double jeopardy clause permitted prosecution on second count, court would bar prosecution on that count under supervisory power).

41 . The prosecutor in charge of the conspiracy case, after deciding to seek multiple indictments, presented the first indictment to the grand jury on the final day of its sitting. The government has explained the decision to seek multiple indictments and the resulting rush to the grand jury as follows:

[The prosecutor], who was pregnant at the time [the first indictment was sought] and due to deliver, had a choice to make. .. . Should she indict ... [the Ti-Con] case, which was completed, ... or should she defer until the time when the other investigations were done.

I mean, she might have had to restart with another grand jury, but it could have been done. 1 think that is simply a decision that prosecutors come to. I don’t read anything vindictive into that. She felt it was a separate case.

Transcript of Proceedings on the Motion to Dismiss at 30-31.

42 . Nothing in United States v. Payner, 447 U.S. 727, 100 S.Ct. 2439, 65 L.Ed.2d 468 (1980), is contrary to this court’s dismissal of the indictment under the supervisory power. Payner did “not limit the traditional scope of the supervisory power in any way,” id. 447 U.S. at 735 n.8, 100 S.Ct. at 2446 n.8, but held only that the supervisory power is not to be exercised either “at the instance of a party who ... is not the victim of the challenged practices,” id. 447 U.S. at 735, 100 S.Ct. at 2446, or where the injured party is not before the court, id. 447 U.S. at 735 n.7, 100 S.Ct. at 2446 n.7. Since neither of those exceptions applies here, Payner is not controlling. See United States v. Cortina, 630 F.2d 1207, 1215-16 (7th Cir. 1980) (discussing Payner); The Supreme Court, 1979 Term, 94 Harv.L.Rev. 75, 187-96 (1980) (same).

Nor does the court accept the government’s contention that defendants’ alleged acts of concealment deprive them of the protection afforded by the supervisory power. The court’s resolution of this waiver argument in the double jeopardy context, see Part II.C. supra, applies equally to the exercise of the supervisory power.