Appellant was convicted of conspiracy to manufacture methaqualone. There was abundant evidence and, indeed, defendant admitted that he had at various times helped produce methaqualone. His major arguments on appeal, however, are that there was insufficient evidence of any overt act in furtherance of the charged conspiracy occurring within the five year limitations period prior to the indictment and that he withdrew from the conspiracy prior to the five year limitations period. He also challenges the jury instructions.

Defendant was indicted on November 25, 1985. We review the evidence with particular attention to events after November 25, 1980 to determine whether prosecution was barred by the statute of limitations, 18 U.S.C. Sec. 3282, or whether, instead, as the government contends, overt acts in furtherance of the conspiracy were performed after November 20, 1980 and defendant's association with it continued into the limitations period.

There was evidence of the following. According to the testimony (given under immunity) of Gerard Stackhouse, he and one Francis Foley discussed manufacturing drugs sometime in 1976. Thereafter, Stackhouse researched how to produce methaqualone, and in 1977 a facility was rented in Worcester, Massachusetts. Foley financed the operation and helped to renovate the place while Stackhouse, along with one Leonard Lewis, tried to produce methaqualone. Also present doing carpentry work was Tilman Lukas who, although originally unaware of any wrongdoing, guessed something illegal was afoot and was let in on the secret. After nine months to a year of effort, Stackhouse and Lewis succeeded in producing methaqualone powder. The next step was to turn the powder into tablets. Toward that end, in the spring of 1978, Foley and Stackhouse went to California where they met with Gene Wise and defendant. Defendant, according to his own testimony, had constructed a pill machine for his employer, Wise, and this machine was supplied to Foley and Stackhouse so that they could produce Quaaludes, the pill form of methaqualone. Stackhouse had trouble with the machine, so in May 1978 defendant went to Massachusetts. Defendant said he got the machine running and helped in other aspects of the Worcester methaqualone operation for about four months. Defendant became uncomfortable, however, because he felt operations were not sufficiently concealed and Foley was ordering chemicals too openly. Consequently, he said, he returned to California in September 1978. During the following twelve months, defendant came to Massachusetts on several errands for Wise, sometimes carrying money to Foley and pills back to Wise.

Meanwhile, the Worcester group had two mishaps. First, in late 1978, they lost their lease. Second, at around the same time, Stackhouse and others (but not Foley) were stopped on their return trip from New York after purchasing there the direct chemical precursors to methaqualone. The agents took the chemicals, but allowed Stackhouse to go. Stackhouse had no further association with Foley's enterprise after that.

After Stackhouse's departure, Foley moved the business to South Boston. In June 1979, a building at 377 West First Street, South Boston was acquired and renovations were begun.

Defendant next appeared on the scene in the summer of 1979. Defendant said he returned to Massachusetts at that time and rented an apartment in Hingham (starting September 1, 1979) in order to participate in a venture Foley and Wise were contemplating. The precise nature of the Foley-Wise venture was never decided upon, defendant said, but among the matters discussed was manufacturing the precursors to maethaqualone, an operation defendant thought would be perfectly legal. Defendant's job, he said, was to set up the South Boston space so that it could be used for any number of chemicals, a general purpose facility so to speak. Defendant connected a condenser, checked seals on reactor vessels, and tested equipment, but, he said, never manufactured anything. In December 1979, at a time when defendant was running tests on equipment and wondering what would happen next, Wise was lost at sea. The demise of Wise, whom defendant regarded as his protector, coupled with friction with Foley and defendant's feeling that he was in over his head with larger equipment than he had ever used before, convinced defendant to leave Massachusetts, he said. On January 15, 1980, he turned in the keys to his Hingham apartment and informed the rental manager he was leaving. The lease was formally terminated April 30, 1980, but the rental manager testified defendant was not in the apartment after January 15, 1980. Defendant said he never told Foley about being in over his head for he was afraid Foley would take it badly. Instead, he simply packed and left. After returning to California, defendant put his knowledge to use and made methaqualone himself. There is no contention that defendant's production of methaqualone in California was part of the conspiracy charged in the indictment.

In contrast to defendant's testimony that he left Boston in January 1980, never again to provide any aid to Foley's business, was the testimony of William Garstang. In early 1979, Garstang moved into the house defendant rented in California. Defendant, who shared the residence with Garstang, would be absent from California approximately two weeks out of every six, according to the latter. In the spring of 1980, defendant told Garstang that he was involved in a drug manufacturing operation in Boston or a suburb, that the operation was having difficulties (including management problems), and that he was making curative efforts to try to get the operation to run smoothly and get paid. Defendant's periodic absences from California did not cease until summer 1980, Garstang said, when defendant and Garstang began manufacturing their own methaqualone in California.

The gas records for the South Boston facility from September 1979 to June 12, 1980 were introduced into evidence. They show a peak usage in January 1980 ($309.70). Thereafter, the bills dropped ($171.95 February; $68.65 March; $57.97 April; $22.25 May), and service was terminated after June 12, 1980 for nonpayment.

There is no direct evidence of what, if anything, transpired at the South Boston site after January 1980 (when defendant says he left) through summer 1980 (when the gas was turned off and defendant's periodic trips east ceased) through November 25, 1980, the date five years prior to defendant's indictment. Robert Hanley, the building's previous owner who had a business next door, said he never saw a sign of life after the renovations.

We now come to events occurring within the limitations period. In late May 1981, Federal Drug Enforcement Administration (DEA) agents searched the South Boston premises. They found a dismantled laboratory. Some, but not all, of the chemicals needed to make methaqualone were present. Samples of residue found in various places along with dust and other sweepings from the floor were taken and analyzed. They were found to contain methaqualone and other chemicals from which methaqualone is produced. While the DEA chemist could not say when the lab had last been operative, he felt it indeed had been used to produce methaqualone in view of the methaqualone residue found on the floor and in drawers.

After the May 1981 search, the agent padlocked the door and put seals on the door and building indicating a search warrant had been executed. Thereafter, surveillance was conducted on three occasions. On July 28, 1981 and July 30, 1981, Foley was observed removing a glass condenser and other equipment from the premises.

Tilman Lukas (the carpenter at the Worcester operation to whom Stackhouse had eventually acknowledged that methaqualone was being produced) testified that in the summer of 1981, Foley called and asked if Lukas would store equipment for him. Not long thereafter, Foley made two or three trips to Lukas's business in Amherst, Massachusetts with glassware, mantels, chemical drums, drying ovens, rheostats, pipe fittings, and electrical and building equipment which Lukas stored for him.

With this evidentiary background, we turn to the indictment, the district court's rulings, and the arguments.

The indictment, returned November 25, 1985, charged that "[f]rom on or about June 7, 1979 [the date Daniel Shea, an alleged co-conspirator acquitted at trial, acquired the South Boston facility] until on or about July 30, 1981 [the date Foley was observed removing equipment] at Boston and Amherst and elsewhere," Foley, Shea, and defendant conspired to manufacture methaqualone. The acts upon which the government relies for its position that the conspiracy continued to exist within the limitations period--November 25, 1980 to November 25, 1985--are Foley's removal of equipment from the South Boston facility and transportation of it to Tilman Lukas's business in Amherst in the summer of 1981.

In denying defendant's motion for a judgment of acquittal, the district court had no hesitancy concluding withdrawal from the conspiracy had not been established as a matter of law, for withdrawal requires more affirmative action then just dropping out. The difficult question, the court stated, was whether the removal of the materials from the South Boston facility in July 1981 was an act in furtherance of the charged conspiracy. The court concluded that it was, stating as follows:

"The Worcester evidence I think is probative of the fact that this was like a floating crap game, in the sense that it was a floating laboratory. It was a pretty big float, because this was a big operation, but still, it is like any moonshine operation, floating crap game, any illegal operation that requires premises and special equipment. It moves around, and this was moved once, and I think the transfer to Amherst at the end of July by Mr. Foley of key ingredients or equipment is not only consistent with manufacture but very probative of a plan to continue it.

"So there is the slim reed on which this case passes the motion [for acquittal]."

Defendant repeats these same two arguments--insufficient evidence of any overt act in furtherance of the conspiracy within the limitations period and withdrawal from the conspiracy--on appeal. Logically, perhaps, we should first determine whether the charged conspiracy existed within the limitations period, for if it did not, then we would not need to address withdrawal. However, because the stringent requirements the law imposes for withdrawal--mere cessation of activity is not enough--somewhat color our view as to whether the charged conspiracy continued into the limitations period, we turn first to withdrawal.

Defendant, in arguing withdrawal, points out that there is no evidence he did anything in furtherance of the charged conspiracy after the summer of 1980, the date Garstang said defendant returned permanently to California. The government argues that neither defendant's moving back to California nor the absence of evidence that defendant actually helped Foley in some way after the summer of 1980 required a finding of withdrawal. In the prosecution's view, the evidence supports the conclusion that defendant's role was not only to set up the Boston laboratory, a task which may have been completed in January 1980 when defendant terminated his Hingham lease, but also to serve as a consultant as needed.

The law with respect to withdrawal has been frequently stated. "[M]ere cessation of activity in furtherance of the conspiracy does not constitute withdrawal." See, e.g., United States v. Dunn, 758 F.2d 30, 37 (1st Cir.1985) (quoting United States v. Phillips, 664 F.2d 971, 1018 (5th Cir.1981), cert. denied, 457 U.S. 1136, 102 S. Ct. 2965, 73 L. Ed. 2d 1354, 459 U.S. 906, 103 S. Ct. 208, 74 L. Ed. 2d 166 (1982)); United States v. Mardian, 546 F.2d 973, 978 (D.C.Cir.1976). In order to withdraw, a conspirator must act affirmatively either to defeat or disavow the purposes of the conspiracy. Dunn, 758 F.2d at 38; Hyde v. United States, 225 U.S. 347, 369, 371-72, 32 S. Ct. 793, 803-04, 56 L. Ed. 1114 (1912) (withdrawal requires affirmative action; a disclosure to authorities may not suffice if the disclosure is incomplete and defendant continues to acquiesce in the conspiracy). Typically, there must be evidence either of a full confession to authorities or a communication by the accused to his co-conspirators that he has abandoned the enterprise and its goals. United States v. Steele, 685 F.2d 793, 803-04 (3d Cir.), cert. denied sub nom. Mothon v. United States, 459 U.S. 908, 103 S. Ct. 213, 74 L. Ed. 2d 170 (1982).

Defendant does not rely on confession, and hence the question is whether the evidence supports only the view that defendant permanently severed his ties and communicated that fact to Foley (or another co-conspirator) prior to November 25, 1980.

Admittedly, defendant said he did not tell Foley he felt unqualified for the job or that he was leaving. Rather, defendant said, he simply returned to California. One inference may be that by leaving so abruptly, defendant graphically communicated (in a nonverbal manner) that he was quitting the enterprise. That, however, the government argues, is not the only permissible conclusion the jury could draw. Defendant had once before returned to California after his active (four month) daily participation in the Worcester operation had ceased, yet he had remained available and had even made a few deliveries of money and pills for Foley and Wise. Similarly, the jury could conclude that defendant's surrender of the keys to his Hingham apartment did not denote withdrawal. Rather, defendant's primary task of setting up the lab had been completed and defendant was no longer needed on a daily basis. When needed, however, defendant would periodically come to Boston up to the summer of 1980. That defendant may not have been called upon for his particular expertise--defendant has degrees in electrical engineering--after the summer of 1980 does not necessarily establish either that defendant had quit the enterprise or that Foley had retired him, the government would asseverate. Arguably, Foley's energies may have been concentrated on more basic problems--Wise, a source of financing, had died and the gas had been disconnected for nonpayment--which would need to be surmounted before defendant's specialized talents could be put to use. Foley had once before overcome setbacks when he relocated from Worcester to Boston. He might have hoped to do so again and then to call on defendant to put the lab back into operation.

We think, under the stringent standards for withdrawal, that the district court was probably correct that defendant could not be said as a matter of law to have withdrawn from the conspiracy and that the issue was one for the jury. Nevertheless, the case is troubling. For all that appears, there had been a substantial hiatus (approximately a year, at a minimum) in operations and in defendant's direct participation when Foley removed the equipment in July 1981. (Gas had been turned off in June 1980, defendant's trips east had ceased at about that time, the lab's neighbor had observed no activity, and the lab had been discovered by the DEA in May 1981.) Absent either active involvement or active withdrawal, for how long is defendant to be held responsible as a conspirator for whatever methaqualone operation Foley might contemplate or eventually undertake?

There are parallels between the present case and United States v. Borelli, 336 F.2d 376 (2d Cir.1964), cert. denied sub nom., Cinquegrano v. United States, 379 U.S. 960, 85 S. Ct. 647, 13 L. Ed. 2d 555 (1965). There, defendants were indicted for a drug importation conspiracy alleged to have taken place from 1950 to 1959. There was evidence that a core group of conspirators repeatedly imported and distributed heroin. The only evidence, however, as to a particular conspirator, Tantillo, was that he accepted two deliveries of heroin in 1951 (beyond the limitations period) and that in 1960 (within the limitations period), at a chance meeting with another conspirator, Tantillo said he was sorry he had been compelled to stop buying from the other. In claiming withdrawal, Tantillo relied on the absence of activity on his part after (as well as long before) August 15, 1957, the date five years prior to the return of the indictment. The court, however, in an opinion written by Judge Friendly, pointed to the "rigorous requirements for making out the defense of withdrawal" the Supreme Court had laid down in Hyde v. United States, 225 U.S. at 369, 32 S.Ct. at 803, and concluded that although Tantillo had done nothing actively in furtherance of the conspiracy after 1951, he had not as a matter of law withdrawn; his 1960 conversation raised a jury issue whether he had sufficiently manifested an intention to withdraw. Borelli, 336 F.2d at 385, 388-89.

The court briefly considered whether Hyde 's affirmative action requirement was too severe when a defendant "simply engaged in the supply, transportation, or purchase of contraband before the period of limitations and the only effects realistically continuing into the period were the profits the central core had realized and its hope of further assistance." It concluded, however, it did not need to reach that question. Instead, Judge Friendly focused on whether Tantillo's agreement with the core group continued into the period not barred by the statute of limitations. Judge Friendly explained as follows. The gist of a conspiracy offense is the agreement. Consequently, it was "essential to determine what kind of agreement or understanding existed as to each defendant." Borelli, 336 F.2d at 384 (emphasis added). The agreement would not necessarily be the same with respect to each defendant in a multifaceted drug conspiracy; the agreement among the core members might embrace more extensive operations than the agreement between core participants and lower echelon personnel or helpmates whose involvement was more peripheral. To warrant conviction, "the Government must present evidence justifying the jury in finding beyond a reasonable doubt that the particular agreement into which a defendant entered continued into the period not barred by limitation ... and the scope of his agreement must be determined individually from what was proved as to him." Borelli, 336 F.2d at 385. Following this approach, the court concluded Tantillo's agreement had been less extensive in time and scope than that of the core importers.

We find this approach--focusing upon the particular agreement the defendant entered--instructive here. Utilizing Judge Friendly's framework, we turn to the question whether the government proved beyond a reasonable doubt that an overt act in furtherance of the particular agreement defendant made occurred within the limitations period.

2. Existence of the Conspiracy within the Limitations Period

Here, as in Borelli, the accused was admittedly involved actively with the core mover (Foley) at one time, but thereafter ceased active participation. According to the government's theory of the case, defendant's agreement with Foley was to set up the South Boston lab and then to serve as a consultant on mechanical or chemical matters as needed. But is there any basis for concluding beyond a reasonable doubt that defendant agreed, whether implicitly or explicitly, to serve Foley wherever Foley might choose to relocate? That is, did defendant, by once agreeing to be associated with Foley's Boston methaqualone operations thereby, absent meeting the stringent requirements for withdrawal, become responsible as a conspirator in whatever future methaqualone operations Foley might undertake? We think not. To be sure, defendant had once followed Foley from Worcester to South Boston. As a result of that relocation, however, there had been a considerable interruption in production (since a new facility had had to be found and renovated), there were changes in personnel and in defendant's role, and the government has not contended that the South Boston and Worcester operations were all part of one continuing conspiracy. We see much less support for viewing Foley's removal of equipment from the South Boston lab and storing of it in Amherst in the hope of resuming operations at some, as of then, unknown time and place as part of or a logical consequence of the particular agreement defendant entered into when he threw in with Foley in South Boston.

Continuity is particularly attenuated in this case in view of a number of relevant factors. Wise, defendant's preceptor, was dead. There had been a substantial hiatus in operations. While it can be concluded from the DEA chemist's testimony that the South Boston lab was used at some time to produce methaqualone, there is no basis to conclude production occurred close to Foley's July 1981 removal of equipment. The DEA had found the lab dismantled, gas had been terminated in June 1980, defendant's trips east had ceased at about that time, Foley's group by then had no known person other than defendant with the knowledge to produce methaqualone, and the lab's neighbor saw no activity. In May 1981, the DEA sealed and posted the lab, thus alerting the lab's former workers of the government's suspicion of wrongdoing. Hence, Foley's removal of lab equipment was not closely similar to a simple dismantling of a still followed by quick resumption in a new location by a group with a repeated history of such evasions. Foley's conduct can, we think, far more logically be regarded as a unilateral grasp to salvage what he could of a moribund operation.

It may be that Foley did hope to resurrect the operation. It may be that, had Foley found a suitable new facility, defendant would have joined him. But the burden of proving beyond a reasonable doubt the existence of the particular conspiracy--as determined by defendant's agreement--within the limitations period was upon the government. In light of the cessation of both operations generally and defendant's active participation particularly by the summer of 1980, and having in mind the intervening DEA raid, we conclude that the fact that defendant had once reassociated with Foley after Foley had changed locations was not sufficient to show that defendant had agreed to do it again. Consequently, the government failed to prove that Foley's removal of equipment in July 1981 was an overt act in furtherance of the particular agreement defendant entered. On this record the statute of limitations barred the prosecution.

The judgment of conviction is reversed.