OPINION

WRIGHT, C.J.

Defendants Paul Zamora, Milo P. Saling, and Michael B. Szymanski, contend on appeal that charges resulting in their convictions for conspiracy and grand theft are barred by the applicable statute of limitations. We conclude that such contentions are meritorious and reverse the judgments.

On June 22, 1972, the Santa Barbara County Grand Jury returned a five-count indictment against defendants. The charges stemmed from a fire which partially destroyed the interior of a residence on April 10, 1968. Count I alleged a conspiracy to commit arson (Pen. Code, §§ 182, 447a), [1] count II charged a conspiracy to burn insured property with intent to defraud the insurer (§§ 182, 548), and count III alleged a conspiracy to commit grand theft (§§ 182, 487). The indictment further charged the commission of 19 overt acts beginning in late March 1968 and continuing up to and including June 12, 1972, each of which was alleged to have been committed in furtherance of one or more of the three conspiracies charged. Counts IV and V charged two separate acts of grand theft (receipt of insurance proceeds) occurring on May 3, 1968, and September 16, 1968, respectively, and further alleged that neither of the thefts was discovered until February 22, 1972.

Defendant Zamora filed a motion to set aside the indictment (§ 995) on the ground, inter alia, that all five charges were then barred by the three-year statute of limitations (§ 800) and also interposed demurrers to counts I through III on the same ground. Defendants Saling and Szymanski filed a separate motion to dismiss (§ 995) and demurrers to all counts of the indictment also raising statute of limitations issues. [2] After a hearing and the submission of extensive points and authorities by both the People and defendants, the motions to dismiss were denied and the demurrers were overruled.

During the course of the subsequent 31-day jury trial, allegations of overt acts committed on June 12, 1972, and April 17, 1972, were stricken on the motion of the People and the indictment was amended to allege that the most recent overt act in furtherance of each of the three conspiracies occurred in late December 1970 or early January 1971. Each defendant was found guilty on each of the five counts of the indictment. Motions by all three defendants for judgment notwithstanding the verdict or in the alternative for a new trial based, inter alia, on the statute of limitations, were denied. On appeal defendants once again contend, in a variety of different ways, that the limitation periods had run and that therefore the trial court committed reversible error in not sustaining the demurrers and in failing to grant the post-trial motions and the motions to set aside the indictment. [3]

In 1967 defendant Saling owned several companies known as the M.S. Group, a real estate management and development firm involving all facets of real estate operations. Saling and defendant Szymanski directed various aspects of the brokerage, escrow, renovation and management services provided by the company. During October 1967, Saling, Szymanski, and Guido Hanak, a painting foreman for the M.S. Group, allegedly conspired and attempted to burn down a residential structure owned by the M.S. Group on Clifford Street in Santa Barbara. Only minor damage was done, however, because closed doors within the structure acted as firebreaks. Fire Department Inspector Nielsen suspected that an act of arson had been committed and believed that Hanak was the perpetrator, but Neilsen was unable to develop a case for prosecution and the investigation was terminated.

Defendant Zamora was the manager of Oakdale Manor, Inc., a family corporation which invested in real property, and he had previously worked for Saling as a real estate salesman. Other principals in Oakdale Manor included Daryl Skare, Zamora's brother-in-law who was a Santa Barbara police officer, and Mr. and Mrs. Wood, parents of Zamora's wife. In late March 1968, Zamora approached Guido Hanak through Saling and asked Hanak to plan a fire in a residential structure at 1010 Garcia Road, a building owned by Oakdale Manor. The prime motivation was not perpetration of an insurance fraud. Instead, Zamora was primarily concerned with the most inexpensive method of removal of the Garcia building because it sat across the property line of two adjoining lots and prevented development of a large canyon area behind the two parcels. Fred Cobler, a painter who had worked for both Zamora and Saling was hired to spray the interior of the house with a generously thinned oil base paint, the purpose of which was to increase the flammability of the structure. Although Hanak could have done such work and had done similar work preparatory to the Clifford Street fire, the employment of Cobler was intended to avoid a second connection between Hanak and a destructive residential fire. Cobler suspected that an arson was planned when Zamora instructed him to use the unusual oil base paint and to spray over electrical fixtures and unrepaired wall cracks. Zamora also told Cobler that "an accident might happen to the house" but stated that the latter would be warned in time to remove his painting equipment.

Hanak was driven to the Garcia Road residence by Zamora on the night of April 10, 1968. Zamora left for a prearranged meeting place and Hanak spread five gallons of highly flammable lacquer thinner, which he had purchased earlier in the day, throughout a basement kitchen and hallway. Unknown to him, Szymanski had already saturated the area with other flammable liquid and when Hanak struck a match an explosion occurred. Hanak was severely burned and sustained a number of cuts when he threw himself out a window to escape the flames. Hearing approaching sirens, he ran from the yard leaving his eyeglasses, smoking pipe, and flashlight in the kitchen. He telephoned Zamora and went to Zamora's house where an attempt was made to treat his severe burns and cuts.

Later that evening Hanak, Zamora, Saling, and Szymanski met at Saling's residence to devise a plan to aid Hanak. They agreed that Hanak would be taken to receive medical attention the next morning and that Saling and Szymanski would report that Hanak had fallen into Saling's backyard barbeque pit while holding a gallon of paint thinner during a wild party on the preceding evening. Zamora agreed to pay Hanak's medical bills secretly through Saling in order to obscure any connection between Hanak and Zamora or the fire on his property. The conspirators also planned to place Hanak in a private hospital on the far side of Santa Barbara in an attempt to conceal Hanak's burns from any police or fire department investigators. This agreement to conceal the source of Hanak's burns, the cause of the Garcia Road fire and Zamora's connection with Hanak, subsequently became known as the "barbeque-pit agreement" and plays a significant role in the resolution of defendants' statute of limitations contention as to the conspiracy convictions.

Notwithstanding the conspirators' efforts, investigators quickly learned of Hanak's burns and suspected that he was the perpetrator of the fire on Garcia Road, particularly because they already knew that he had been the prime suspect in the Clifford Street fire investigation. Saling and Szymanski were believed to be a part of a conspiracy to commit an arson because they supplied Hanak's alibi and because both of them had been suspected of involvement in the Clifford Street incident. Once fire department investigators had gathered initial evidence implicating Saling, Szymanski, and Hanak, the task of preparing the case for prosecution was largely undertaken by the Santa Barbara Police Department. Several police officers participated in the initial investigations, but the ultimate responsibility for the investigation was assigned to Officer Lenz. Although he suspected arson and conspiracy, Lenz was unable to connect Hanak with the eyeglasses which had been left at the scene of the fire and he was unable to uncover a motive for arson. Lenz discussed the case with Officer Skare, Zamora's brother-in-law, but he learned little to aid in the investigation. The possibility of insurance fraud was not pursued and the insurer paid about $15,000 for restoration work on the structure. The investigation was soon relegated to Lenz' spare time. Because of lack of progress the fire department offered to do additional investigative work but their efforts were unfruitful. No further investigation was made by Lenz after mid-1968.

In February 1972, Fred Cobler, the painter, informed an investigator from the district attorney's office that he knew that the fire on Garcia Road was the result of arson and that Zamora had been involved. The lead was pursued and in June Hanak was offered immunity from prosecution in exchange for his testimony. Hanak had not been fully reimbursed for his medical expenses and, in fact, Saling had sued Hanak during the previous year and had won a judgment against him. Hanak accepted the offer of immunity and recounted all the details of the arson plan and the coverup scheme. He also took part in efforts to gain further evidence incriminating Zamora, Saling, and Szymanski. These disclosures resulted in the indictment which was returned on June 22, 1972. All five counts of the indictment arise out of the planning for, commission and concealment of only the fire on Garcia Road.

There appears to be little dispute that People v. Crosby, supra, 58 Cal.2d 713 , sets forth the controlling principles of law for determining whether the instant conspiracy convictions are barred by the statute of limitations. [4] However, the People suggest that we should go beyond Crosby and analogize to our recent decision in People v. Leach (1975) 15 Cal.3d 419 [ 124 Cal. Rptr. 752 , 541 P.2d 296 ].

In Leach we clarified the views set forth in People v. Saling (1972) 7 Cal.3d 844 [ 103 Cal. Rptr. 698 , 500 P.2d 610 ], [5] wherein we interpreted the scope of Evidence Code section 1223, California's codification of the coconspirator's exception to the hearsay rule. The primary issue considered in Leach and Saling was the termination point of a conspiracy when analyzed in light of policy considerations underlying the rule of evidence there involved. We held that under certain limited circumstances declarations of coconspirators uttered after the attainment or abandonment of the principal objective of a conspiracy may properly be admitted under the coconspirator's exception to the hearsay rule. The contention that we should now analogize to Leach and Saling in order to resolve the instant case bespeaks a misunderstanding of the principles expressed in those cases and is but one more example of the confusion and imprecision which continues to plague and further muddy the already far too murky waters of conspiracy law. As hereinafter appears, resolution of the statute of limitations issue in the present case turns on a different set of considerations from those which were determinative of the results reached in Leach and Saling .

Initially, we note that there is a point at which the major evidentiary consideration, that of reliability, overlaps with some of the concerns which have motivated the enactment of statutes of limitation. To some extent, the concept of a period of limitation developed in recognition of the ever increasing difficulty faced by both the government and a criminal defendant in obtaining reliable evidence (or any evidence at all) as time passes following the commission of a crime. However, this consideration cannot be viewed as the singular justification behind statutes of limitation. Unlike the balancing approach utilized for consideration of the admissibility of marginally reliable evidence (Evid. Code, § 352) or on the question of a defendant's constitutional right to a speedy trial ( Jones v. Superior Court (1970) 3 Cal.3d 734, 740 [ 91 Cal. Rptr. 578 , 478 P.2d 10 ]), the government cannot overcome the bar of a statute of limitations by demonstrating a lack of prejudice to the defendant. Since a conviction based on acts occurring outside the statutory period is invalid irrespective of whether the delay has resulted in any prejudice to the defendant, it can hardly be said that reliability of evidence is determinative on the limitation issue. (1) Furthermore, in California the statute of limitations constitutes a substantive rather than a procedural right which is not waived by failure to assert it at the pleading stage. If reliability of evidence was the sole factor, certainly a conviction based on a guilty plea would be valid notwithstanding the running of a limitation period. Yet, it is now well settled that a conviction, even if based on a plea of guilty, is subject to collateral attack if the charge was originally barred by the applicable limitation period. ( In re Demillo (1975) 14 Cal.3d 598 [ 121 Cal. Rptr. 725 , 535 P.2d 1181 ]; People v. McGee (1934) 1 Cal.2d 611, 613 [ 36 P.2d 378 ].) [6]

Other policy considerations which underlie the concept of a period of limitation vary in purpose. The possibility of self-reformation by the criminal offender may lessen the need for society to impose corrective sanctions and society's impulse for retribution may correspondingly diminish as time passes. Statutes of limitation also encourage the swift and effective enforcement of the law, hopefully producing a stronger deterrent effect. They tend to limit the chance that the first offense will spawn blackmail of the offender by others threatening disclosure crime breeding more crime. Finally, adoption of a period of limitation represents a legislative recognition that for all but the most serious of offenses (such as murder or kidnaping) a never-ending threat of prosecution is more detrimental to the functioning of a civilized society than it is beneficial. (See generally, Model Pen. Code, § 1.07, com. (Tent. Draft No. 5, 1956); Note, The Statute of Limitations in Criminal Law: A Penetrable Barrier to Prosecution (1954) 102 U.Pa.L. Rev. 630, 632-634; Note, Conspiracy, Concealment and the Statute of Limitations (1961) 70 Yale L.J. 1311 , 1334-1335.) In light of these differing policies which underlie the limitation concept, it should be clear that Leach and Saling , with their grounding in evidentiary concerns, are of little import to the instant case. (Cf. Oakley, From Hearsay to Eternity: Pendency and the Co-Conspirator Exception in California Fact, Fiction, and a Novel Approach (1975) 16 Santa Clara L. Rev. 1 , 46, fn. 174.) Furthermore, past failures to grasp the differences between limitation issues and those of an evidentiary nature have only confounded attempts to clarify conspiracy law. In this case the question is not what sort of evidence should have been admitted at trial, but whether defendants should have had to stand trial at all. (Note, supra, 70 Yale L.J. 1311 , 1323-1324.) We, therefore, decline the People's invitation to utilize Leach and Saling in resolving the limitation issues in the present case.

(2) It has long been the rule in conspiracy cases that a limitation period begins to run from the time of the last overt act committed in furtherance of the conspiracy. ( People v. Crosby, supra, 58 Cal.2d 713, 727-729 ; Bompensiero v. Superior Court (1955) 44 Cal.2d 178, 184-185 [ 281 P.2d 250 ].) [7] Our task, therefore, in the present case is to ascertain whether any overt acts in furtherance of the three conspiracies were committed within the three-year period prior to June 22, 1972, the date on which the indictment was returned and filed. [8]

The amended indictment charged that an overt act (No. 16) in furtherance of the conspiracies occurred on June 30, 1970, [9] and that the most recent overt act (No. 17) was committed in late December 1970 or early January 1971. [10] On the face of the indictment, these two "overt acts" fall within the three-year period preceding the filing of the indictment. Thus, if either of them can be deemed to have been an overt act committed in furtherance of the conspiracies and if there is sufficient evidence to support the jury's implied finding that at least one of the acts took place as charged, the conspiracy convictions are not barred by the statute of limitations.

In an effort to make the requisite showing, the People analogize to the holding in People v. Leach, supra, 15 Cal.3d 419, 423-424 , an approach which we have already disavowed in the present context. We are told that, "All three conspiracies had one major aim: to burn the house down, get the money from the insurance company, and equally important, keep the matter quiet so that the parties could retain the benefit they had reaped by reason of the fire." The People next contend that there was evidence presented at trial, independent of the statements of the coconspirators themselves, which showed the continuing nature of the conspiracy. This evidence included the events surrounding the so-called "barbeque-pit agreement" and the subsequent meetings and communications between the various coconspirators, overt acts Nos. 16 and 17 falling in the latter group. Thus under this approach, Hanak is seen as an "arsonist-for-hire" still seeking payment from the other conspirators for his criminal services on their behalf (see People v. Leach, supra, 15 Cal.3d 419, 431 ) thereby purportedly showing that the final two "overt acts" alleged in the indictment were committed in furtherance of the conspiracies and within the statutory period.

Preliminarily, it should be noted that we perceive three "aims" rather than the "one major aim" mentioned in the People's contention, for we fail to comprehend how arson, grand theft, and concealment of those offenses can be characterized as one "major aim" of the conspiracies. In any event, while this contention illustrates the confusion which conspiracy cases seem to engender, it does bring us to the crux of the issue. Simply stated, the crucial question is: can acts of concealment committed by conspirators subsequent to the completion of the substantive offense which is the object of the conspiracy be construed as overt acts in furtherance of the conspiracy so as to delay the commencement of the running of the limitation period? We note that the United States Supreme Court was presented with a similar issue in Grunewald v. United States (1957) 353 U.S. 391 [ 1 L.Ed.2d 931 , 77 S.Ct. 963 , 62 A.L.R.2d 1344 ], and an examination of that case provides a useful framework within which to consider the case at hand. Before beginning our discussion of Grunewald , however, it is necessary to briefly set the context in which that case came before the high court.

Grunewald was preceded to the Supreme Court by two other conspiracy cases, Krulewitch v. United States (1949) 336 U.S. 440 [ 93 L.Ed. 790 , 69 S.Ct. 716 ], and Lutwak v. United States (1953) 344 U.S. 604 [ 97 L.Ed. 593 , 73 S.Ct. 481 ]. However, both of those cases, unlike Grunewald and the instant case, dealt not with the commencement of the running of a limitation period but rather with the duration of a conspiracy for purposes of the coconspirator's exception to the hearsay rule. Thus the context in which Grunewald arose was strikingly similar to the present situation in which the Leach and Saling cases have been followed in this court by a conspiracy case raising a statute of limitations issue.

In Krulewitch the defendants were charged with conspiracy to transport a woman in interstate commerce for the purpose of prostitution. At trial the prosecution sought to introduce against Krulewitch hearsay statements of a coconspirator who had attempted to conceal Krulewitch's part in the scheme. The statements were admitted under the coconspirator's exception as having been made in furtherance of the conspiracy despite the fact that they were made one and one-half months after the woman had been successfully transported from New York to Miami. The Supreme Court reversed the resulting convictions on the ground that the hearsay statements could not have been in furtherance of the charged conspiracy since the government had failed to plead that the scheme had encompassed a subsidiary conspiracy aimed at preventing detection and punishment. Significantly, the court chose not to base its holding on the broader ground that the conspiracy, as a matter of law, could not have continued once the woman was no longer in interstate commerce. Mr. Justice Jackson, in a frequently cited concurring opinion, further suggested that the crucial flaw in the government's position was its failure to appreciate the difference between an implied and an explicit agreement to conceal reached before commission of the substantive offense which was the object of the conspiracy. Although he strongly condemned the doctrine of implied conspiracies, noting that an implicit conspiracy to conceal could be imputed to almost every conspiracy thereby impermissibly creating a conspiracy of indeterminate duration, Mr. Justice Jackson invited the government to allege and prove an explicit agreement to conceal as a means of extending the duration of such a conspiracy. ( Krulewitch v. United States, supra, 336 U.S. 440, 455-457 [ 93 L.Ed. 790, 800-801 ].)

In Lutwak v. United States, supra, 344 U.S. 604 [ 97 L.Ed. 593 ] the government unsuccessfully sought to overcome the objections posed by the Krulewitch majority. Lutwak and his codefendants were convicted of conspiring to obtain the illegal entry of three aliens into the United States by arranging to have the aliens marry three honorably discharged veterans who would then return to the United States with their newly acquired "war brides." Once the aliens were successfully admitted the unconsummated marriages were to be dissolved. Various acts and statements by the conspirators which occurred after the last alien had entered the United States were admitted against the defendants at trial. This time, however, the government specifically alleged that two subsidiary but integral goals of the conspiracy included the subsequent concealment of the conspiracy and the subsequent separation and divorce of the spouses. The government did not simply argue that these goals were implicit in the conspiracy but presented evidence of falsified information which had been supplied by the entering couples in order to conceal the real nature of the marriages. The Supreme Court nevertheless refused to deem the conspiracy to have extended beyond the date of the last alien's entry. Although the court affirmed the convictions, invoking the harmless error rule, it once again held that the government's proof was lacking because the evidence was nothing more than an attempt to imply a subsidiary agreement to conceal. ( Lutwak v. United States, supra, 344 U.S. 604, 616-617 [ 97 L.Ed. 593, 602-603 ].) The court, however, declined to rule out the possibility that proof of an express agreement to conceal the conspiracy could have resulted in the continuance of such conspiracy beyond the date of the last alien's entry.

Finally in Grunewald the court was once again confronted with the effect of acts of concealment on the duration of a conspiracy. Significantly, while the court noted that Grunewald was not an evidence case as had been Krulewitch and Lutwak ( Grunewald v. United States, supra, 353 U.S. 382, 399-401 [ 1 L.Ed.2d 931, 940-942 ]), it did not employ any different analysis from that which it had used in the earlier cases. The defendants in Grunewald were executives of two business firms which were investigated by the Treasury Department for tax evasion. It was alleged that such defendants conspired with Treasury and other governmental officials to obtain "no prosecution" rulings from the Bureau of Internal Revenue in 1948 and 1949. Charges of an improper coverup were not brought until the defendants were indicted in 1954. In an attempt to avoid the difficulties which had arisen in Krulewitch and Lutwak the government alleged conspiracies on three different theories, only the first of which is here relevant. The government sought on that theory to bring the conspiracy within a three-year limitation period by arguing that even if the main objective of the conspiracy was to "fix" the tax investigations, there was proof of an actual subsidiary agreement by the conspirators to conceal the conspiracy in order to escape detection and avoid punishment. The government charged within the three years prior to 1954 the commission of overt acts of concealment pursuant to the agreement and submitted proof thereof at trial.

The Supreme Court was no more impressed with this attempt to extend the duration of the conspiracy than it had been with the previous efforts in Krulewitch and Lutwak. Examining the government's evidence the court stated, "We find in all this nothing more than what was involved in Krulewitch , that is, (1) a criminal conspiracy which is carried out in secrecy; (2) a continuation of the secrecy after the accomplishment of the crime; and (3) desperate attempts to coverup after the crime begins to come to light; and so we cannot agree that this case does not fall within the ban of those prior opinions." ( Grunewald v. United States, supra, 353 U.S. 382, 403 [ 1 L.Ed.2d 931, 942-943 ].) Although the court never indicated what quantum of evidence would be sufficient direct proof of an agreement to conceal which would extend the duration of the conspiracy, it seemed to require at the very least hearsay evidence of an actual agreement to conceal the conspiracy supported by proof of overt acts. ( Id., at p. 404, fn. 16 [1 L.Ed.2d at p. 943].) [11]

Application of the Grunewald analysis to the conspiracy counts in the present case would require reversal of count I (conspiracy to commit arson of a dwelling house) and count II (conspiracy to burn insured property with intent to defraud the insurer). As we interpret Grunewald there is insufficient evidence in the present record to support a finding that the conspirators explicitly agreed to conceal those conspiracies prior to the commission of the arson which was their central object. Since that object was completed at the instant that the Garcia Road building was set ablaze, [12] no agreement subsequently entered into with the object of concealing those conspiracies could extend the time period within which the People might have charged those violations. (See Grunewald v. United States, supra, 353 U.S. 382, 405-406 [ 1 L.Ed.2d 931, 943-944 ].)

The charge of conspiracy to commit grand theft, as alleged in count III, requires a different analysis. The substantive offense which was the object of that conspiracy was the receipt of the insurance proceeds and, therefore, that object was not achieved until the last such payment was made. As we read Grunewald we are invited to conclude that direct evidence of an explicit agreement to conceal the grand theft conspiracy entered into prior to the receipt of final payment would constitute direct evidence of the sort which would extend the duration of the conspiracy beyond the date of final payment. We see the testimony of Hanak relative to the making of the so-called "barbeque-pit agreement" as just such evidence. Viewed in this manner the present case is almost identical to the first theory proposed by the government in Grunewald. (See Grunewald v. United States, supra, 353 U.S. 382, 398 [ 1 L.Ed.2d 931, 939-940 ].) We thus direct our attention to the question which the Supreme Court left unanswered in that case, i.e., can such an explicit agreement to conceal extend the duration of the conspiracy and thereby forestall the commencement of the running of the limitation period? If we are unwilling to accept the invitation of Grunewald, however, the limitation period must be deemed to have commenced running with the last overt act of grand theft, and thus to constitute a bar to the charges contained in count III.

Central to a resolution of this question is the distinction first raised by Mr. Justice Jackson's concurring opinion in Krulewitch between implied and explicit agreements to conceal a conspiracy. ( Krulewitch v. United States, supra, 336 U.S. 440, 455-457 [ 93 L.Ed. 790, 800-801 ].) Although such a distinction is superficially appealing, closer analysis reveals that the distinction is one more of form than of substance.

Criminal liability for conspiracy, separate from and in addition to that imposed for the substantive offense which the conspirators agree to commit, has been justified by a "group danger" rationale. The division of labor inherent in group association is seen to encourage the selection of more elaborate and ambitious goals and to increase the likelihood that the scheme will be successful. Moreover, the moral support of the group is seen as strengthening the perseverance of each member of the conspiracy, thereby acting to discourage any reevaluation of the decision to commit the offense which a single offender might undertake. And even if a single conspirator reconsiders and contemplates stopping the wheels which have been set in motion to attain the object of the conspiracy, a return to the status quo will be much more difficult since it will entail persuasion of the other conspirators. ( Callanan v. United States (1961) 364 U.S. 587, 593-594 [ 5 L.Ed.2d 312, 317-318 , 81 S.Ct. 321 ]; Goldstein, Conspiracy to Defraud the United States (1959) 68 Yale L.J. 405 , 413; Developments in the Law: Criminal Conspiracy (1959) 72 Harv.L.Rev. 920, 924.) In response to these evils and as a deterrent to such group activity special tactical advantages, unavailable in other contexts, have been granted to prosecutors when pursuing conspiracies. (See Krulewitch v. United States, supra, 336 U.S. 440, 445-454 [ 93 L.Ed. 790, 795-800 ], Jackson, J., concurring; Note, supra, 70 Yale L.J. 1311 , 1312-1314; Goldstein, supra, 68 Yale L.J. 405 , 409-412.) As an additional measure aimed at deterrence, conspirators have also been charged with responsibility for any acts of a coconspirator, even if unintended or expressly forbidden, as long as they are in furtherance of the common purpose. ( People v. Kauffman (1907) 152 Cal. 331, 335 [ 92 P. 861 ].) [13] Needless to say these rules and advantages peculiar to conspiracy cases have made it a popular theory with prosecutors, leading Judge Learned Hand to refer to it as "that darling of the modern prosecutor's nursery." ( Harrison v. United States (2d Cir.1925) 7 F.2d 259, 263 .) [14]

The attempt to extend the duration of the conspiracy in the present case can be seen as a further attempt to deter future concerted criminal activity by the use of a special theory unique to conspiracy law. Although we support this salutary goal, we cannot agree with the means proposed to achieve it. It can be argued that group activity increases both the likelihood that the criminal objective will be attained and the possibility that the perpetrators can more easily escape subsequent detection and punishment ( Pinkerton v. United States, supra, 328 U.S. 640, 644 [ 90 L.Ed. 1489, 1494-1495 ]; United States v. Rabinowich (1915) 238 U.S. 78, 88 [ 59 L.Ed. 1211, 1214-1215 , 35 S.Ct. 682 ]), but it may be just as likely that "a large number of participants will increase the prospect that the plan will be leaked as that it will be kept secret." (Goldstein, supra, 68 Yale L.J. 405 , 414.) Because we perceive little distinction between concealment of a crime by conspirators and the concealment usually undertaken by a single criminal offender, we conclude that the context in which a conspiracy is undertaken affords little reason to create a special conspiracy exception which would afford, in effect, an extended limitation period.

Adoption of an implicit-explicit concealment distinction would also result in the effective elimination of the rule that a limitation period runs as to a particular conspirator from the time he withdraws from a conspiracy. (See People v. Crosby, supra, 58 Cal.2d 713, 730 .) "A defendant's mere failure to continue previously active participation in a conspiracy ... is not enough to constitute withdrawal; there must be an affirmative and bona fide rejection or repudiation of the conspiracy, communicated to the coconspirators." ( Id. )

As an illustrative example of the problems engendered by the implicit-explicit approach, if we assume a conspiracy in which proof of an explicit agreement to conceal is shown, a coconspirator can withdraw from acting in furtherance of the concealment objective only by confessing his crime to the authorities. If a conspirator in such a case communicates his "withdrawal" to the other conspirators after the substantive offense is completed and if he is thereafter approached by someone investigating the crime, he can fail to act in furtherance of the conspiracy to conceal only by revealing his own complicity. The "withdrawn" conspirator must necessarily answer investigative inquiries so as to conceal the involvement of his former associates in order to conceal his own involvement, thereby demonstrating his continued commitment to the concealment "objective" of the conspiracy. [15] Indeed, the very failure to affirmatively confess his crime, even in the absence of an investigation, can be viewed as "playing the conspirators' role to the hilt." (Note, supra, 70 Yale L.J. 1311 , 1342.) [16] A requirement that a conspirator confess his crime before the statute of limitations will begin to run is certainly not calculated to encourage the abandonment of criminal associations or goals, nor is the elimination of the withdrawal defense which would follow from the sanctioning of a theory of extended "explicit" conspiracies to conceal. (See Note, Application of Federal Statute of Limitations (1954) 29 N.Y.U.L.Rev. 1470, 1473-1475; Note, supra, 70 Yale L.J. 1311 , 1340-1343; Developments, supra, 72 Harv.L.Rev. 920, 957-960.) [17]

Finally, we see no really significant benefit gained by characterizing some criminal schemes as "explicit" conspiracies to conceal while classifying others as merely "implicit" conspiracies to conceal. In reality, the distinction suggested by Mr. Justice Jackson in Krulewitch does little more than frustrate on technical grounds the prosecutor's attempt to negate the statute of limitations as a defense to conspiracy charges. [18] There is simply no meaningful difference between one group of persons conspiring to commit a crime whose members firmly understand that such crime will require acts of concealment but who do not "explicitly" agree to commit such acts, and another group of conspirators who "explicitly" decide before the crime, as an example, which of them will hide a weapon to be used during the crime after the actual commission thereof. Certainly in terms of the "group danger" rationale underlying the crime of conspiracy there is no significantly greater degree of danger posed to society by the "explicit" conspiracy than by the "implicit" conspiracy to conceal. [19] Yet this implicit-explicit distinction could indefinitely extend the duration of one conspiracy while terminating the duration of a virtually identical conspiracy.

Finally, an arbitrary requirement of evidence to show an "explicit" agreement to conceal is contrary to the well established rule that the unlawful design of a conspiracy may be proved by circumstantial evidence without the necessity of showing that the conspirators met and actually agreed to commit the offense which was the object of the conspiracy. (E.g., People v. Steccone (1950) 36 Cal.2d 234, 238 [ 223 P.2d 17 ]; see also, 1 Witkin, Cal. Crimes (1963) § 108, p. 103; Fricke, Cal. Criminal Law (1970) p. 126.) Were we to adopt the implicit-explicit distinction which Grunewald suggests we would create a rather confusing situation. On the one hand the jury could rely on all the evidence, both circumstantial and direct, in determining the scope and design of a conspiracy for the purpose of imposing substantive liability on the conspirators. At the same time, the jury would be limited in assessing the time span of the conspiracy for statute of limitations purposes by reference to direct evidence alone. Although such a result is not necessarily unworkable, we believe that there is a better solution to the problem.

For the foregoing reasons, we perceive no justification for the adoption of an implicit-explicit conspiracy to conceal test to determine the commencement of the running of the period of the statute of limitations in conspiracy cases. (3) We conclude accordingly that acts committed by conspirators subsequent to the completion of the crime which is the primary object of a conspiracy cannot be deemed to be overt acts in furtherance of that conspiracy. Consequently, upon successful attainment of the substantive offense which is the primary object of the conspiracy, the period of the statute of limitations for the conspiracy begins to run at the same time as for the substantive offense itself. [20] Our holding leaves unaffected the basic proposition that the limitation period begins to run upon completion of the last overt act in furtherance of the conspiracy ( People v. Crosby, supra, 58 Cal.2d 713, 727-729 ); it simply means that for purposes of the statute of limitations an overt act in furtherance of the conspiracy cannot be committed subsequent to the completion of the object which made the conspiracy unlawful in the first instance. [21] ( Lonabaugh v. United States, supra, 179 F. 476, 481 .)

Application of the foregoing principles to the instant case makes it clear that the instant conspiracy charges are barred by the statute of limitations. The conspiracies to commit arson and to burn insured property were completed on April 10, 1968, when the fire was ignited in the Garcia Road building. The conspiracy to commit grand theft was complete with receipt of the last insurance payment on September 16, 1968. Since those dates are more than three years prior to June 22, 1972, when the indictment was returned and filed, the charges contained in counts I, II, and III were barred by the three-year statute of limitations (§ 800), and all the conspiracy convictions must be set aside.

We now turn our attention to the grand theft convictions and the interpretation of the "discovery" provision of section 800.

Since the amendment of section 800 in 1969, grand theft has fallen within a very special class of felonies subject to a three-year period of limitation. Prior to that time the limitation period for grand theft began to run from the commission of the theft even in cases where the victim was unaware of the loss of his property or did not know that the transfer of his property had been obtained by false pretenses. (See People v. Swinney (1975) 46 Cal. App.3d 332, 341 [ 120 Cal. Rptr. 148 ]; cf. People v. Darling (1964) 230 Cal. App.2d 615, 618-621 [ 41 Cal. Rptr. 219 ].) In 1969 the Legislature specifically addressed the latter two situations by amending section 800 to provide that the three-year limitation period did not begin to run until "discovery" of the theft. [22] Although the statutory period for some grand thefts still commences to run at the time the offense is committed, as in the case of a robbery when the commission and discovery of the crime coincide, it is deferred by operation of section 800 in cases when the offense is concealed at the time of its commission such as in a theft by false pretense.

On its face the 1969 amendment appears to set the limitation period running from the actual "discovery" of a theft. The new provision has nevertheless been interpreted to include the same requirement of "reasonable diligence" in discovering the facts of a theft that the courts have read into the "discovery" provision of the statute of limitations for tort actions based on fraud as set forth in Code of Civil Procedure section 338, subdivision 4. ( People v. Swinney, supra, 46 Cal. App.3d 332, 340-343 .) [23] In that context we have held that the word "discovery" is not synonymous with actual knowledge. ( Bainbridge v. Stoner (1940) 16 Cal.2d 423, 430 [ 106 P.2d 423 ].) "The statute commences to run ... after one has knowledge of facts sufficient to make a reasonably prudent person suspicious of fraud, thus putting him on inquiry. Section 19 of the Civil Code provides: `Every person who has actual notice of circumstances sufficient to put a prudent man upon inquiry as to a particular fact, has constructive notice of the fact itself in all cases in which, by prosecuting such inquiry, he might have learned such fact.'" ( Hobart v. Hobart Estate Co. (1945) 26 Cal.2d 412, 437 [ 159 P.2d 958 ] [last italics added].) Based on the foregoing rule we have required the plaintiff, in cases where an action is brought more than three years after the commission of the fraud, to plead and prove: (1) when and how the facts concerning the fraud became known to him; (2) lack of knowledge prior to that time; (3) that he had no means of knowledge or notice which followed by inquiry would have shown at an earlier date the circumstances upon which the cause of action is founded. ( Bainbridge v. Stoner, supra, 16 Cal.2d 423, 430 .) The reason for such specificity is quite simple by looking to the manner in which the "discovery" is alleged to have been made, the trial court can more easily ascertain if the plaintiff has been negligent in seeking out the fraud. Once the court determines that the facts stated in the pleadings are sufficient and do not show, as a matter of law, that in the exercise of reasonable diligence the plaintiff could have discovered the fraud at an earlier time then the reasonable diligence question becomes an issue for the trier of fact. ( Hobart v. Hobart Estate Co., supra, 26 Cal.2d 412, 440 ; see also Helfer v. Hubert (1962) 208 Cal. App.2d 22, 26-27 [ 24 Cal. Rptr. 900 ].)

In the present case the People sought to bring the grand theft prosecution within the "discovery" provision of section 800 by alleging in both counts IV and V that the acts of grand theft were not discovered until February 22, 1972. No attempt was made to allege all three of the factors which we have required a plaintiff to aver in order to sufficiently plead the civil fraud "discovery" provision of section 338 of the Code of Civil Procedure. [24] After the court had already ruled against defendants' demurrers and motions to set aside the indictment (§ 995), defendants made a nonstatutory motion to dismiss the two grand theft counts for failure to plead facts showing that reasonable diligence had been exercised to discover the thefts and that earlier discovery was not possible. At the time the motion was made no appellate opinion had construed the 1969 amendment to section 800. Defense counsel failed to uncover the legislative history of the amendment which gave guidance for its application. Defendants accordingly supported their motion by citation to People v. Doctor (1967) 257 Cal. App.2d 105 [ 64 Cal. Rptr. 608 ], a case in which the court construed a Labor Code section directing that the period of limitation for certain violations of that code ran from "discovery" of the violation. Doctor not only imputed to the statute a reasonable diligence requirement but also held that the accusatory pleading must specify the same factors necessary to sufficiently plead the "discovery" provision of section 338 of the Code of Civil Procedure. ( Id., at p. 112.) Defendants in the instant case argued that the same analysis should be applied to section 800, and requested that the court hold an evidentiary hearing to resolve the reasonable diligence issue. The People strenuously objected to both the request for a hearing and defendants' construction of the statute. (4) (See fn. 25.) After considerable discussion of the matter with counsel, the court granted the requested hearing and allowed defendants to call as witnesses a number of law enforcement personnel who took part in the investigations in either 1968 or 1972. [25]

At the conclusion of the hearing the court ruled that People v. Doctor, supra, 257 Cal. App.2d 105 , was not apposite to the instant case in the pleading context because the Labor Code section there involved and section 338 of the Code of Civil Procedure both related to fraud whereas concealed or "undiscovered" thefts are not necessarily the result of fraud. (5) (See fn. 26.) The court held that the pleading of the "discovery" provision of section 800 in counts IV and V of the indictment was thus sufficient. [26] The court also found no lack of reasonable diligence in the investigation of the thefts by the fire and police department investigators. As a result of the ruling the limitation issue was again litigated at trial.

Prior to their deliberations the jurors were instructed on the various sub-issues of the limitation "discovery" question including the reasonable diligence requirement. [27] Therefore the jury's verdicts of guilty on counts IV and V contain implicit findings that the facts constituting the acts of grand theft were not discovered more than three years before the indictment was filed nor could they have been discovered before that time with the exercise of reasonable diligence. The jury's findings on the "discovery" issue were on questions of fact and on appeal they are tested by the substantial evidence standard. (6a) We have examined the record and are unable to conclude that there is substantial evidence to support those findings. Furthermore, judged by the standards which have developed through application of the analogous civil statute (Code Civ. Proc., § 338), we hold that as a matter of law the uncontradicted evidence produced at trial shows that with the exercise of reasonable diligence the facts constituting the acts of grand theft could have been discovered at an earlier time. (See Helfer v. Hubert, supra, 208 Cal. App.2d 22, 26-27 .) Because the reasonable diligence findings of the jury ultimately rest on proof of a negative proposition, i.e., that earlier "discovery" was not possible, we must add to the basic summary of the facts already recited in order to show that due diligence was not exercised to discover the thefts.

Fire department arson investigators were on the scene only minutes after the blaze on Garcia Road was extinguished. During the initial inquiry by Inspectors Bishop and Bailey it was ascertained that the natural gas supply to the house had been previously disconnected and was therefore not a cause of the fire. Residue apparently left by a flammable liquid was discovered in several different burn patterns and Bishop was informed by the firefighters that the blaze had had all the characteristics of a fire fueled by such a liquid. Inspector Bishop also noticed that the house had been recently painted and that a strong odor of paint, paint thinner, and lacquer thinner was present in the air.

The following morning a pane of glass from a kitchen window was found 62 feet away from the house. The fire had started near the window and Inspector Bishop concluded from the condition of the glass and its location that it had been blown out by an explosion. Blood stains were found on the kitchen sink and leading away from the house. A flashlight, a smoking pipe, and a pair of eyeglasses thought to belong to the arsonist were found in the kitchen. Finally close examination of wood paneling confirmed the initial conclusion that a flammable liquid had been poured over it.

Fred Cobler, the painter who had been employed by Zamora, arrived and led Bishop on a tour of the house identifying and locating his various paint cans. An empty can labeled lacquer thinner was disavowed by Cobler and he also indicated that some of his paint thinner was missing. Cobler told Bishop that some unknown person had been entering the house each night after he locked it and left for his home, leading him to conclude that the visitor must have had a key. Finally, Cobler informed Bishop that he was in the process of purchasing an apartment building from Oakdale Manor, owner of the Garcia Road building. Cobler did not disclose the strange painting instructions he had been given by Zamora, nor did he state his suspicions that the Garcia Road fire had been planned because he only suspected arson and because he "wasn't asked the right question."

Daryl Skare, Zamora's brother-in-law, appeared, identified himself as a Santa Barbara police officer, and informed Bishop that he was to be the next tenant in the house. He indicated that Saling Realty was managing the property and had had some trouble evicting the previous tenants who had been "hippies." Skare suggested that Cobler's mysterious nocturnal visitor might have been Guido Hanak. Skare also stated that he had an interest in Oakdale Manor, and identified Zamora as a principal in the corporation.

Fire inspectors Bailey and Bishop followed the bloody trail leading from the house for some distance until it disappeared on a nearby road. Because they believed that someone in the house had been cut and burned in the fire they decided that the local hospitals should be contacted to ascertain whether anyone had been admitted with such injuries. Bishop also learned that Bailey had already responded to a fire call about 7 o'clock that morning at 1623 Los Canoas Road. Bailey had discovered a small fire burning in a backyard barbeque pit and was told by a man calling himself Szymanski that an explosion had occurred when he came out and lit a fire in the pit. Szymanski suggested that the explosion might have resulted from someone pouring gasoline in the fire pit the night before. This, of course, was an attempt by defendants to corroborate Hanak's alibi. In spite of Szymanski's explanation, Bailey did not believe that an explosion had taken place because he examined Szymanski and saw no evidence of singed hair or flash burns which would have been present if the explosion story and gasoline hypothesis were true.

Detective Trainee Lenz of the Santa Barbara Police Department was also at the fire scene the morning after the blaze. He talked to Officer Skare who stated that the details of ownership and past tenants could be obtained from Saling Realty. Upon proceeding to the realty office Lenz met Saling who informed him who the owners, prior tenants, mortgage holder, and insurer were. During the conversation Zamora entered the office and Saling introduced him to Lenz explaining his relationship to the Garcia Road property. Zamora volunteered that he had inspected the property on the previous afternoon and had seen nothing out of the ordinary. Lenz concluded that it was unnecessary to talk with Zamora further because he felt that Saling had given him all available information.

On the evening of April 11th, Inspector Nielsen telephoned Inspector Bishop at his home and told him that Guido Hanak had been admitted to St. Francis Hospital that morning with severe burns, singed hair and a cut. Nielsen also indicated that Hanak had been brought to the hospital by Saling who resided at 1623 Los Canoas Road. Hanak was the only person who had been admitted to a local hospital with injuries of the type the inspectors believed had been suffered by the Garcia Road arsonist. Bishop and Nielsen then discussed the 1967 residence fire on Clifford Street which Hanak had been suspected of setting. Inspector Bailey's early morning fire call and his experience with Szymanski was noted [28] as was the involvement of Saling and Szymanski with the Clifford fire.

Detective Trainee Lenz met with Inspectors Bishop and Bailey on April 12th. He had already been informed about Hanak's condition and the connections with Saling and Szymanski. They agreed that the fire department would be allocated the task of ascertaining the immediate facts of the arson, i.e., such aspects as determination of the flammable liquid used to set the fire and observations of the neighbors when the fire began. The police department and Lenz in particular were to have the responsibility for investigating Guido Hanak and reconciling all of the evidence. Bishop indicated to Lenz that the fire might have been set for insurance purposes or in order to correct the lot lines problem, an apparent reference to the fact that the Garcia Road building sat across a lot line and blocked access to a rear canyon lot.

By April 15 the fire inspectors were convinced that the fire had been the result of arson. Bishop had discovered another broken window in the house with hair, pieces of skin and dried blood on it. [29] From all the foregoing evidence Bishop concluded that Hanak had set the fire using lacquer thinner, had been trapped in the ensuing explosion, and had tried to escape by throwing himself out the window, suffering lacerations in the process. Saling and Szymanski were implicated because it was clear that the barbeque-pit story was a fabrication and because they had been involved with the Clifford Street fire investigation.

The next day, April 16, Lenz sought to connect Hanak with the eyeglasses found at the scene of the fire. At some point in his investigation Lenz located a photo of Hanak in which he wore glasses similar to those recovered in the house. Although he subsequently checked a number of optical outlets Lenz was never successful in supplying the missing link because Hanak had purchased the eyeglasses from an optician with offices in a department store where no inquiries were made. On April 16 Lenz also went to St. Francis Hospital and learned that the barbeque-pit alibi could not have been true because no debris from a wood fire had been found in the area of extensive burns on Hanak's body. The burns, furthermore, were of a "flash" type and would have been so painful that anyone in normal circumstances would have been forced to seek immediate medical attention. Based on this information Lenz returned to Saling's office on April 17 and requested a written statement from Saling relative to the barbeque-pit accident. Although he initially agreed to supply such a statement Saling later called Lenz and informed him that he would not comply with the request on the advice of counsel. Lenz never attempted to substantiate or disprove Saling's claim that Hanak had been at a party at Saling's home on the night of April 10 or even if there had been a party that night.

An officer not involved in the investigation informed Lenz that a local realtor, Jim Baker, had told him that Hanak's former girlfriend, Marie Handelman, could supply information about the fire on Garcia Road. Lenz met with her on April 17 and although she promised to convey any information she subsequently learned, she claimed to have no knowledge of the fire. She indicated however that Hanak had left her apartment on the night of the fire about 8 p.m. stating that he was leaving to make a painting estimate. [30]

Near the end of April Lenz spoke with a deputy in the district attorney's office and discussed the Garcia Road case along with a number of others. He mentioned that Hanak had no apparent motive and questioned whether he should investigate the possibility of an insurance fraud. When he received no encouragement he terminated this line of investigation.

As time progressed the fire department investigators became increasingly concerned over the lack of progress in the police investigation, particularly because Lenz seemed to have relegated the case to his spare time. The fire department inspectors volunteered to investigate the optical outlets in San Jose where Hanak had previously lived but their efforts were, of course, unsuccessful since the glasses had been purchased in Santa Barbara. [31]

Lenz talked with James Brewer, an insurance claims adjuster, on April 26 and told Brewer that he was investigating because the property had been insured. When Lenz asked if any money could have been made by burning the property Brewer was noncommittal but indicated that if any "foul play" was involved the insurance company would want to hire its own investigators to conduct an independent inquiry. [32]

In May during the course of the fire department's continuing investigation Inspector Bishop learned that Zamora was the son-in-law of Mr. and Mrs. Wood, two of the principals in Oakdale Manor and the original purchasers of the Garcia Road property. At the same time he was informed that either the Woods or Zamora had been experiencing financial difficulties, including the possibility of foreclosures on some real properties they held. Finally Fire Marshall Bennett told him that Karen Jorgenson, a secretary associated with Oakdale Manor, had indicated that Hanak had set the fire and had been transported from the immediate area by Saling. No further inquiries were made to check these leads.

The last police activity in the case during 1968 occurred on June 17 when Lenz was again told that Jim Baker, the realtor, believed that Marie Handelman could confirm that Szymanski had been to the Garcia Road building on the afternoon of the fire and had spread a flammable liquid in the house, that Szymanski and Hanak had driven to the property on the night of the fire, that Hanak had gone inside to light the fire, and that Hanak had been burned in the explosion which resulted from Hanak's lack of information that Szymanski had already spread other flammables. Lenz met with Handelman but she once more claimed to have no knowledge of the fire. Despite the fact that Lenz had received the tips about Handelman's purported knowledge of the fire from another individual, Lenz made no effort to investigate that source or the origin of that information.

Lenz testified at trial that while the case was not officially closed no other investigation took place in 1968. He was busy with other investigations since he managed a workload of about 50 cases at any one time, and had been told that his inability to supply a motive for the arson or to connect Hanak to the eyeglasses meant that the case could not be taken to the district attorney. In November 1968, during a routine clearance of the police evidence room, Lenz told other officers that a door from the house could be disposed of since it had been analyzed. The other physical evidence including the eyeglasses, the flashlight, and the smoking pipe were mistakenly destroyed. Lenz had nothing further to do with the case until 1972 when he was approached by investigators from the district attorney's office who were following the leads given by Fred Cobler in his interview with Investigator Arca on February 22, 1972.

Our review of the record persuades us that neither the victim of the thefts, the fire department investigators, nor the police department investigators actually knew that an insurance fraud had been perpetrated in 1968. However, we have already indicated that lack of actual knowledge is not required to bring the "discovery" provision of section 800 into play. The crucial determination is whether law enforcement authorities or the victim had actual notice of circumstances sufficient to make them suspicious of fraud thereby leading them to make inquiries which might have revealed the fraud. Judged by that standard of reasonable diligence the uncontradicted evidence which we have set forth compels the conclusion that a prudent man apprised of the information known in 1968 would have pursued a more vigorous inquiry. We are, of course, aided by the benefit of hindsight. Yet we do not believe that our conclusion is any less compelling because it can be argued that the investigators were confronted with a mass of unrelated and unorganized information as they pursued their inquiries. (7) The purpose behind the "discovery" provision of section 800 is not to directly penalize criminals who conceal their unlawful conduct. Rather it is simply to avoid the possibility that they may completely escape punishment for their acts because they have left no hint that a crime has even been committed. When the efforts at concealment of an offense are unsuccessful to the extent that law enforcement authorities become aware of the facts warranting an investigation, the situation becomes no different from any other crime. It does not follow that such offenders forever lose the benefit of the statute of limitations which the Legislature has seen fit to enact on the theory that they cannot benefit from their wrong in concealing the crime. [33] As we noted in the conspiracy context, anyone who contemplates the commission of a crime realizes that he may be called upon to conceal evidence of his offense. In spite of that almost universal proposition, the Legislature has nevertheless been motivated to set a limitation period for the prosecution of most offenses. We have furthermore indicated that one of the purposes behind the concept of a statute of limitations is the encouragement of the swift and effective enforcement of society's laws. The present case is one in which that purpose would be disserved by a conclusion that reasonable diligence had been exercised in this case.

We are cognizant of the ever increasing burdens placed on our law enforcement personnel to investigate crimes against society. The caseload carried by Detective Trainee Lenz in the instant case is ample testimony to that problem. Yet the Legislature did not incorporate an "over-burdened investigator" exception into the "discovery" provision of section 800. Nor can the failure to follow through on leads discovered in the present case be excused because Lenz was only a detective trainee. This is not a case in which a connection between a fire and possible insurance fraud was clearly lacking. The uncontradicted evidence demonstrates that not only was such a possibility discussed by the investigators but that the task of investigating such a possibility was specifically allocated. The fact that that portion of the investigation was not properly handled is not a burden which falls on the defendants, unsavory and offensive as they are. (6b) Trainee or not, Lenz failed to investigate obvious discrepancies in the statements of Saling. He never talked to Zamora after meeting him in Saling's office, despite discovering evidence that the property had been intentionally burned by persons to whom Zamora had entrusted its management. No attempt was made to substantiate the claim that a party had taken place at Saling's residence on the night of the fire. Nor was Hanak ever confronted with the overwhelming evidence which implicated him. No attempt was made to interview Jim Baker, the realtor, who had supplied the information that Marie Handelman could confirm facts about the fire. It is manifest that if Baker knew about the fire he must have had a source. Further, although Bishop learned that an employee of Oakdale Manor had claimed knowledge about details of the fire, nothing was done to contact her for an interview. Despite the fact that an arson had clearly been committed by known persons, the investigation was simply allowed to die in the summer of 1968. We have noted only some of the lapses in investigation which suggest a lack of reasonable diligence. Our recitation of the facts has necessarily been limited to those lapses and perhaps does not adequately reflect the overall efforts of the investigators. We are nevertheless compelled to conclude that there is not substantial evidence to support the jury's implied findings that reasonable diligence was exercised to discover the fraud in 1968. As a matter of law that lack of diligence means that the "discovery" provision of section 800 is inapplicable to defendants' case. The charges of grand theft were barred by the three-year statute of limitations.

In our decision today we set aside convictions for a number of offenses which the record amply demonstrates were actually committed by defendants. It bears repeating that we have not done so out of any sympathy for these defendants. Our action has been mandated by adherence to the rule that statutes of limitation are to be strictly construed in favor of the accused. The Legislature has determined that the limitation period for certain offenses, including those in the instant case, should be three years. Our decision effectuates that legislative determination.

The judgments are reversed.

Tobriner, J., Mosk, J., and Sullivan, J., concurred.

RICHARDSON, J.

I concur in the majority opinion to the extent that it reverses the judgment as to the first and second conspiracy counts (conspiracy to commit arson, and conspiracy to burn insured property with intent to defraud the insurer). As to these counts, the applicable three-year statute of limitations (Pen. Code, § 800), in my view, commenced to run when the substantive offense was committed, and expired prior to the time the indictment was filed.

I respectfully dissent, however, from the majority's reversal of count III (conspiracy to commit grand theft) and counts IV and V (two separate acts of grand theft). In my opinion, the statutory period with respect to the conspiracy count was extended by virtue of the express agreement between the conspirators, made prior to the receipt of the insurance proceeds, to pay off one of the conspirators and to conceal evidence of the arson. This agreement was coupled with two overt acts committed in furtherance thereof, all occurring within the three-year limitations period. Further, with respect to the grand theft counts, the statutory limitations period did not commence to run until the investigating officers, acting with due diligence, discovered the acts of grand theft in February 1972, once again well within the three-year limitations period.

As the majority recite the facts at considerable length, I will summarize only the essential points. The prosecution established that on April 10, 1968, the night of the Garcia Road fire, the conspirators met and expressly agreed to pay the medical expenses of arsonist Hanak, and to conceal the source of his burns, the cause of the fire, and Zamora's connection with Hanak. The prosecution further established the existence of two overt acts specifically committed in furtherance of the foregoing agreement, which acts occurred within the three-year limitations period, namely, (1) Saling's offer on June 30, 1970, to obtain trust deeds from Zamora in payment for Hanak's unlawful services, and (2) Saling's attempt, in December 1970 or January 1971, to obtain money from Zamora to pay to Hanak for his participation in the arson.

The majority hold that since the last insurance payment was made on September 16, 1968, more than three years prior to June 22, 1972, when the indictment was returned, the grand theft conspiracy charge was not timely filed. I respectfully suggest, however, that the relevant cases of this court, and of the United States Supreme Court, support the contrary view.

First, in People v. Leach (1975) 15 Cal.3d 419 [ 124 Cal. Rptr. 752 , 541 P.2d 296 ], we recently reaffirmed our holding in People v. Saling (1972) 7 Cal.3d 844, 851-853 [ 103 Cal. Rptr. 698 , 500 P.2d 610 ], to the effect that although a conspiracy usually terminates when the underlying substantive offense is completed, "`[p]articular circumstances may well disclose a situation where the conspiracy will be deemed to have extended beyond the substantive crime to activities contemplated and undertaken by the conspirators in pursuance of the objectives of the conspiracy,'" and that it is a question for the trier of fact to determine when the conspiracy ended. ( 15 Cal.3d at p. 431 , quoting from 7 Cal.3d at p. 852 .)

In Saling , for example, we upheld the jury's implied finding that a murder conspiracy did not terminate until one of the conspirators had been paid for his services rendered in furtherance of the conspiracy. Likewise, in the present case the jury, as trier of fact, might reasonably have concluded that the grand theft conspiracy had not yet terminated when Saling offered, and later attempted, to arrange payment to Hanak. The majority, however, refuse to reach the foregoing conclusion, reasoning that since Saling and Leach involved an evidentiary question, rather than application of the statute of limitations, those cases are not applicable.

However no such distinction was recognized in Saling and Leach , which applied general principles of conspiracy law for purposes of deciding when a criminal conspiracy terminates. Both cases involved the application of the rule that a conspirator's statements are admissible against his coconspirators only when made during the conspiracy. Yet these general principles invoked by us in Saling and Leach certainly have equal force and application in the present case, for once again, this time in ascertaining the commencement of the statute of limitations, we are faced with the question when has a criminal conspiracy terminated? While consistency is not everything it is something.

I think it significant that the United States Supreme Court has declined to follow the majority's course: As the majority concede, the Warren Court in Grunewald v. United States (1957) 353 U.S. 391 [ 1 L.Ed.2d 931 , 77 S.Ct. 963 , 62 A.L.R.2d 1344 ], a statute of limitations case, applied without hesitation the general conspiracy principles enunciated in two prior cases, Krulewitch v. United States (1949) 336 U.S. 440 [ 93 L.Ed. 790 , 69 S.Ct. 716 ] and Lutwak v. United States (1953) 344 U.S. 604 [ 97 L.Ed. 593 , 73 S.Ct. 726 ], both coconspirator-hearsay rule cases.

The majority not only decline to consider the "extended conspiracy" doctrine announced by our own Saling and Leach cases, but they likewise fail to follow the principles set forth by the United States Supreme Court in Grunewald, supra, to the effect that an express agreement between conspirators to cover up their crime may extend the applicable statute of limitations. In this connection the majority insist that Grunewald "never indicated what quantum of evidence would be sufficient direct proof of an agreement to conceal which would extend the duration of the conspiracy...." ( Ante, p. 554.) To the contrary, with due deference, I must note that the high court carefully explained that the requisite agreement cannot be implied from mere acts of concealment, but must consist of an express agreement between the conspirators. Commenting on the nature of the evidence in Grunewald, the high court explained: "There is not a shred of direct evidence in this record to show anything like an express original agreement among the conspirators to continue to act in concert in order to cover up, for their own self-protection, traces of the crime after its commission. " (353 U.S. at p. 404 [1 L.Ed.2d at p. 943], italics added.)

In the present case, in contrast, there was direct evidence, in the form of testimony from coconspirator Hanak himself, that the conspirators met on the day of the Garcia Road fire and expressly and specifically agreed to cover up the crime, and their connection with it, and to pay for Hanak's medical expenses, thus meeting any reasonable standards as to the quantum of evidence required. As I interpret the Grunewald principles, applicable to the facts in the case before us, this express agreement would have the effect of extending the grand theft conspiracy beyond the date of final payment of the insurance proceeds, and the statutory limitations period would not commence to run until the last overt act committed in furtherance of that express agreement, namely, Saling's efforts to provide payment for Hanak.

The majority suggest that, for purposes of applying the statute of limitations, conspirators who expressly agree to conceal their crime must be treated no differently than any other offenders who take the usual steps to avoid being caught. It seems obvious to me that an express conspiracy of concealment poses a far greater hindrance to effective police investigation than that associated with the usual escape following commission of a routine criminal offense. In the matter before us the participants actively and by prearrangement and design aided and abetted each other by supplying alibis, cover stories, and subsequent financial and other assistance. It seems to me that, under such circumstances, the conspiracy should be deemed to continue until completion of the last of the overt acts done in furtherance of the prearranged plan and scheme.

The majority acknowledge that their extended decision herein sets aside multiple convictions for offenses which, in the words of the majority, "were actually committed by defendants." ( Ante, p. 574.) I respectfully suggest that the majority's conclusion need not follow and that the Leach and Saling cases of this court, and the Grunewald decision of the Supreme Court, furnish sufficient legal precedent for affirmance of the grand theft-conspiracy conviction.

In addition to the conspiracy counts, defendants were also convicted of two counts of grand theft, based upon their unlawful receipt of insurance proceeds following the fire. The applicable statute of limitations (Pen. Code, § 800) requires that the indictment for grand theft must be found "within three years after its discovery." Although the People alleged and proved that they discovered the insurance fraud on February 22, 1972 (only four months before the indictment was returned), the majority reverse the conviction on the ground that the People failed to proceed with "due diligence" in investigating the fraud. As I explain, and again, respectfully, the majority clearly err in so holding.

The majority concede that, although the language of section 800 refers to "discovery," California case law holds that the People must plead and prove that the theft could not have been discovered earlier in the exercise of due diligence. Further, the existence of such exercise is a factual issue for the trier of fact to decide. ( People v. Swinney (1975) 46 Cal. App.3d 332, 344-345 [ 120 Cal. Rptr. 148 ]; see Hobart v. Hobart Estate Co. (1945) 26 Cal.2d 412, 440 [ 159 P.2d 958 ].)

In the present case, both the trial court and jury made separate determinations on the issue of the People's diligence: (1) At a preliminary hearing, the trial court denied a motion to dismiss, expressly finding that the officers proceeded with due diligence in investigating the insurance fraud, and (2) at the trial, the jury was expressly instructed on the subject of due diligence and, as the majority concede, the verdicts of guilty "contain implicit findings that the facts constituting the acts of grand theft were not discovered more than three years before the indictment was filed nor could they have been discovered before that time with the exercise of reasonable diligence." ( Ante, p. 565.)

Despite the contrary findings of both trial judge and jury, the majority assert that there is no substantial evidence to support these findings of due diligence. My examination of the record, however, indicates what I think is overwhelming support for these findings of both the trial court and jury.

As the majority observe, section 800 and its "due diligence" requirement were based upon the discovery provision of the statute of limitations for tort actions based on fraud (Code Civ. Proc., § 338). Yet were the present case an ordinary tort action brought by a private plaintiff, it is highly doubtful that this court would, for lack of due diligence, reverse a damage award for plaintiff under the circumstances in this case.

The extensive evidence disclosing the diligent efforts of detective-trainee Lenz and other investigating officers are summarized at length in the briefs and majority opinion herein. Rather than repeat the substance of this evidence, suffice it to say that Lenz and the other officers conducted an exhaustive investigation (including photographs, witness interviews, forensic analysis, and the like) which confirmed the officers' suspicion that arson had occurred, but which afforded insufficient evidence of an insurance fraud. In fact, in 1968 Lenz had discussed the matter with both the district attorney's office and his own superiors, and was advised that the case against defendants was not strong enough to justify prosecution at that time. It seemed quite unlikely to the officers that insurance fraud was the motive for the arson, since the value of the residence was substantially equal to the amount of insurance.

As was well stated by Justice Loring in the vacated opinion prepared by him for the Court of Appeal, Second Appellate District, in this case, "Merely collecting insurance in an amount equal to the value of a residence is not a profitable operation and, therefore, normally does not constitute an adequate motive. It is only when additional unique factors were discovered that motivation became understandable. Those factors were that Oakdale (Zamora and Saling) wanted to remove the Garcia Building which was astride the property line without loss of money so that Oakdale could build two houses on and develop property in the rear, which development was prevented by the location and existence of the Garcia Building. Under these unique facts, arson of a residence insured only for its actual value became a profitable operation and therefore constituted motivation for the arson. The motivation was not to collect insurance at a profit but to destroy the Garcia Building without financial loss. Those unique facts were not discovered by the People until 1972, when Cobler, Torresani and Hanak finally told the truth and the whole truth. Prior to the discovery of such unique facts, the People were not chargeable with knowledge that the mere collection of the insurance constituted grand theft. When those additional unique facts were added to the known facts, the People were chargeable with knowledge that the collection of the insurance constituted grand theft." (Italics in original.)

It seems to me that there was substantial evidence in the record to support the finding, of both the jury and trial court, that the People acted diligently under the unusual circumstances in this case. No public policy is served by the imposition of unrealistic standards of diligence upon already overburdened investigative officers. (The record discloses that in the present case, for example, during the period from April through June 1968 Detective Trainee Lenz was working on approximately 100 cases, with approximately 50 cases pending at any one time.) The majority, admittedly assisted by the inevitable advantage of hindsight, note several "leads" which assertedly Lenz failed to follow. Yet the majority acknowledge that their recitation of the facts "has necessarily been limited to those lapses [in investigation] and perhaps does not adequately reflect the overall efforts of the investigators." ( Ante, p. 573.) The majority appear to marshall only those facts favorable to the defendants while ignoring our long recognized appellate function of determining whether any substantial evidence supports the jury's implied finding of due diligence. (See Hobart v. Hobart Estate Co., supra, 26 Cal.2d 412, 440-441 ; Schaefer v. California-Western States Life Ins. Co. (1968) 262 Cal. App.2d 840, 845-846 [ 69 Cal. Rptr. 183 ].)

I fear that the unwitting effect of the majority holding is to reward the devious. It may also induce prosecutors in future cases of this type to file premature or unsubstantiated criminal charges before fully completing the course of the investigation, in order to avoid dismissals based upon the exhaustion of statutes of limitations. No discernible public policy is served by encouraging such precipitous prosecutorial action in the course of insisting on the exercise of "due diligence."

For all of the foregoing reasons I would affirm the grand theft-conspiracy and grand theft convictions.

CLARK, J.

I concur in reversing counts I and II, but, for the reasons so well expressed by Justice Richardson, dissent from reversing counts III-V.

Regarding counts IV and V, the grand theft counts, the standard of diligence required by the majority is altogether unreasonable. That they make such unrealistic demands upon our hard pressed law enforcement officials suggests that the majority subscribe to what a Rand Corporation study refers to as the "media image of the working detective," i.e., "that of a clever, imaginative, perseverant, streetwise cop who ... roam[s] the entire city for days or weeks trying to break a single case, which is ultimately solved by means of the investigator's deductive powers." (1 Greenwood & Petersilia, The Criminal Investigation Process, Summary and Policy Implications (Rand Corp. Report prepared under a grant from the Nat. Institute of Law Enforcement and Criminal Justice, L.E.A.A., Dept. of Justice, 1975) p. 5 [hereinafter referred to as "Rand study"].)

The reality of criminal investigation, the Rand study reveals, is very different. To determine what factors contribute to case solution the Rand researchers analyzed a large sample of cleared crimes from a variety of crime types. In more than half of the cleared cases, they found, the identification of the offender was available at the time of the initial report because (1) the offender was arrested at the scene; (2) the victim or witness identified the suspect by name and address; or (3) some evidence available at the crime scene, such as a license plate or employee badge number, uniquely determined the identity of the suspect. Most of the remaining cases that were eventually cleared were solved through routine administrative actions: fingerprint search, informant tips, reviewing of mug shots, or arrests in connection with the recovery of stolen property. On the basis of these findings, the Rand study concluded that "with the possible exception of homicide, if investigators performed only the obvious and routine tasks needed to clear the `easy' cases, they would solve the vast majority (97 percent) of crimes that now get cleared. All their efforts in relation to other cases have a very marginal effect on the number of crimes cleared. " (Rand study, pp. 13-14, italics added.)

One of the policy implications of these findings, according to Rand, is that police departments should reduce follow-up investigation on all cases except those involving the most serious offenses. The rationale of this proposal: "Our data consistently reveal that a regular investigator's time is preponderantly consumed in reviewing reports, documenting files, and attempting to locate and interview victims and witnesses on cases that experience shows will not be solved. Our data show, moreover, that most cases that are solved are solved by means of information spontaneously provided by a source other than those developed by the investigator. It follows that a significant reduction in follow-up investigative efforts would be appropriate for all but the most serious offenses in which public confidence demands some type of response." (Rand study, p. 27.) That Rand made this recommendation that the present level of follow-up investigative effort be reduced is all the more significant in light of its finding that, under present practice, investigative efforts in over 86 percent of unsolved cases are suspended by the end of the first week. (Rand study, p. 19.)

In light of the Rand study, we can see that more, not less, than reasonable diligence was exercised in the investigation of this case. Despite the length of the majority opinion, this was, after all, not the crime of the century. It was a $15,000 arson and insurance fraud that involved no injury to innocent persons. Nevertheless, several investigators worked on it intensively for over two months before suspending their efforts because the suspect had an alibi and no motive was apparent. To demand a higher standard of diligence, given society's limited resources and apparently unlimited propensity to crime, is completely unrealistic.

McComb, J., concurred.

Respondent's petition for a rehearing was denied January 13, 1977. Clark, J., and Richardson, J., were of the opinion that the petition should be granted.

[*] Reporter's Note: This case was previously entitled "People v. Saling."

[1] All statutory references herein are to sections of the Penal Code unless otherwise specified.

[2] Two distinct statute of limitations issues were raised by all defendants. As to the conspiracy counts, they argued that the general provisions of section 800 barred the indictment. ( People v. Crosby (1962) 58 Cal.2d 713 [ 25 Cal. Rptr. 847 , 375 P.2d 839 ].) As to the grand theft counts, they contended that the "discovery" provision of section 800 which tolls the statute of limitations for certain crimes was inapplicable in this case.

[3] Defendant Zamora has raised a number of other contentions but our disposition of the statute of limitations question eliminates the need to address those additional issues.

[4] Crosby holds, inter alia, that the limitation period applicable to a conspiracy violation commences to run from the date of the commission of the last overt act alleged. ( People v. Crosby, supra, 58 Cal.2d 713, 728 .)

[5] The defendant in Saling is not the same individual involved in the present case.

[6] It should be noted that the rule is to the contrary in a number of other jurisdictions. In some states the statute of limitations may not be raised in a collateral attack and can be waived if not asserted sometime before a verdict is rendered. ( Commonwealth v. Ashe (1944) 154 Pa.Super. 397 [ 36 A.2d 249 ]; State v. Brinkley (1927) 193 N.C. 747 [ 138 S.E. 138 ]; United States v. Taylor (2d Cir.1953) 207 F.2d 437 ; see Askins v. United States (D.C. Cir.1958) 251 F.2d 909, 913 [ 102 App.D.C. 198 ].)

[7] The origins of this rule and the meaning of the term "overt act" are both crucial to the issues of this case. The inception of the Crosby rule can be traced through People v. Hess (1951) 104 Cal. App.2d 642, 678 [ 234 P.2d 65 ], and People v. Ware (1924) 67 Cal. App. 81, 85 [ 226 P. 956 ], to Hyde v. United States (1912) 225 U.S. 347 [ 56 L.Ed. 1114 , 32 S.Ct. 793 ]. In Hyde the defendants conspired to fraudulently secure title to numerous tracts of federal land in Oregon and California by way of a scheme involving separate transactions over a period of years. The court held that in such a case the conspiracy "continued" beyond any particular transaction or act because the conspirators contemplated a continuous course of conduct similar to that presented in United States v. Kissel (1910) 218 U.S. 601 [ 54 L.Ed. 1168 , 31 S.Ct. 124 ]. Kissel had held that conspiracies which required cooperation of the conspirators and the commission of various acts such as a conspiracy in restraint of trade should be deemed to continue until abandonment or success. Support for the Kissel and Hyde rule was found in a number of lower federal court decisions at the turn of the 19th century.

The first federal courts to confront the question of the commencement of the running of the statute of limitations in conspiracy cases took the view that conspiracy was an "instantaneous" offense, because the crime became complete and subject to prosecution at the instant the first overt act in furtherance of the conspiracy was committed (of course necessarily after an agreement had been reached by the conspirators). Under this theory, the limitation period ran from the first overt act and subsequent overt acts did not affect the expiration of the statutory period. (See, e.g., United States v. Owen (D.Ore. 1887) 32 F. 534, 537-538 ; United States v. McCord (W.D.Wis. 1895) 72 F. 159, 161-166 ; Ex parte Black (E.D.Wis. 1906) 147 F. 832, 840 , affd. sub. nom., United States v. Black (7th Cir.1908) 160 F. 431 .) The "first overt act" rule gradually fell from favor apparently beginning with United States v. Greene (E.D.Ga. 1902) 115 F. 343, 349-350 , affd. (5th Cir.1907) 154 F. 401 , cert. den. (1907) 207 U.S. 596 [ 52 L.Ed. 357 , 28 S.Ct. 261 ], in which the court applied a "continuing conspiracy" theory to a situation where the defendants conspired to defraud the government and win certain construction contracts over a period of years from 1891 through 1897. In Greene the government charged two conspiracies in one count of the indictment, the first occurring in 1891 when the scheme was first hatched and the second occurring in 1897 when the conspirators attempted to win a new government contract. Citing no authority, the court held that such pleading was not improper because proof on the 1897 conspiracy opened the door to proof of the earlier acts of the conspirators which would have otherwise been barred by the statute of limitations. ( Id., at p. 350.) Greene was subsequently distinguished in Ex parte Black, supra, 147 F. 832 , as a case in which the "continuing conspiracy" theory was appropriate because the scheme involved the commission of a series of acts each of which would be sufficient to prove a separate and new conspiracy. However, the court held that in the ordinary conspiracy in which all the acts necessary to achieve the substantive crime intended by the conspirators had been committed, in other words successful completion of the crime (in Black the transfer of title of certain lands to the conspirators), the statute ran from the first overt act. ( Id., at pp. 840-842.) A careful reading of Black gives a clue to the correct resolution of the instant case particularly where the court stated, "I believe no case can be found where the overt act postdated the consummation of the conspiracy. Where would be the locus penitentiae in such a case?" ( Id. at p. 840.)

In Ware v. United States (8th Cir.1907) 154 F. 577 , the court attempted to answer the query posed by the court in Black . The conspiracy in Ware was one of the many Homestead Act frauds in which the conspirators hired individuals to file false land claims which were then transferred to the conspirators. As in Greene, the scheme contemplated many such transfers over a period of years. Consequently, the court was able to hold that there was a locus penitentiae after the performance of each new overt act, since any conspirator could withdraw from the conspiracy before new frauds were completed thereby escaping liability for subsequent acts of the remaining conspirators. ( Id., at p. 579.) Since the crime charged was not limited to formation of a conspiracy, the court found no difficulty in holding that in a continuing conspiracy the statute of limitations ran from the last overt act. (See also Jones v. United States (9th Cir.1908) 162 F. 417, 426-427 , cert. den. (1908) 212 U.S. 576 [ 53 L.Ed. 657 , 29 S.Ct. 684 ].) It was this theory which was adopted by the Supreme Court in Kissel and Hyde and which also underlies the Crosby rule in California. However, adoption of a "last overt act" rule does not answer the questions of what constitutes an "overt act" and when an "overt act" must occur. It is the answers to these questions which ultimately determine the result in the instant case.

[8] Courts have struggled through the years to formulate a definition for the term "overt act," and it has been said that no single definition can be adequate for all conspiracy cases. For our purposes, it is sufficient to say that "an overt act is an outward act done in pursuance of the crime and in manifestation of an intent or design, looking toward the accomplishment of the crime." ( Chavez v. United States (9th Cir.1960) 275 F.2d 813, 817 ; see also, § 184; People v. Olson (1965) 232 Cal. App.2d 480, 490 [ 42 Cal. Rptr. 760 ]; 1 Witkin, Cal. Crimes (1963) § 118, at p. 111.) The requirement of an overt act before conspirators can be prosecuted and punished exists, as previously mentioned, to provide a locus penitentiae an opportunity for the conspirators to reconsider, terminate the agreement, and thereby avoid punishment for the conspiracy. ( Id. ; United States v. Britton (1883) 108 U.S. 199, 204-205 [ 27 L.Ed. 698, 700 , 2 S.Ct. 531 ]; People v. Olson, supra, at p. 490 .)

Unfortunately, the term "overt act" has often been used in a loose fashion without regard to the particular issue under consideration. As a result the significance of the timing of "overt acts" has been considerably blurred. In the specific context of the statute of limitations we think the rule was correctly stated in Lonabaugh v. United States (8th Cir.1910) 179 F. 476 . "It is not enough that the conspiracy be directed to the attainment of some unlawful object, or to the attainment of some lawful object by unlawful means; it must be directed to the attainment of one of the objects specified. Nor is it enough that the overt act be directed to the attainment of some other object; it must be directed to the attainment of the object which brings the conspiracy within the interdiction; and when that object is attained `the object of the conspiracy,' in the sense of the statute, is effected." ( Id., at p. 481.) A careful reading of Penal Code section 184 leads to the same conclusion, i.e., that an overt act cannot succeed the completion of the substantive offense which is the object of the conspiracy. The same point was made in Ex parte Black where the court said, "It is a contradiction of terms to speak of an [overt] act done to effect the purpose of the conspiracy after the conspiracy has been accomplished." ( Ex parte Black, supra, 147 F. 832, 840 .) Finally, the rule is the same in California: "`The overt act... cannot succeed the completion of the contemplated crime.'" ( People v. Olson, supra, 232 Cal. App.2d 480, at p. 491 , and cases cited therein; see also, 15 C.J.S., Conspiracy (1967) § 43(2) at p. 748, fn. 77.)

With the reasons for the Crosby rule and the definition of the term "overt act" clarified by the foregoing historical analysis, the remaining question for us to determine as to the conspiracy counts in the instant case is what constitutes "successful completion of the contemplated crime" for purposes of the statute of limitations ? And, as we have already pointed out in our discussion of the inapplicability of People v. Leach, supra, 15 Cal.3d 419 and People v. Saling, supra, 7 Cal.3d 844 to the present case, the result reached for purposes of the statute of limitations may vary from the result in the context of hearsay questions or conspirator liability for the acts of fellow conspirators.

[9] Overt act No. 16, is charged in the following language: "That thereafter, on or about June 30, 1970, in pursuance of said conspiracies, and to effect the object thereof, the said defendant MILO P. SALING offered to attempt to obtain for Guido Hanak a quantity of second trust deeds from PAUL ZAMORA in compliance with his promise to pay Guido Hanak for his participation in the burning of 1010 Garcia Road, said Guido Hanak at this time having never been paid anything by defendant PAUL ZAMORA."

[10] Overt act No. 17 is charged in the following language: "That thereafter, in late December of 1970 or early January of 1971, in pursuance of said conspiracies, and to effect the object thereof, the said defendant MILO P. SALING, in compliance with the promise of defendant PAUL ZAMORA to pay Guido Hanak for his participation in the crime, did attempt to obtain money from defendant PAUL ZAMORA to be given to Guido Hanak in payment for the participation of Guido Hanak, who at that time had never been paid for his participation in the arson promised by defendant PAUL ZAMORA."

[11] This conclusion flows from the fact that the court dismissed the following evidence as inadequate to prove an explicit agreement to conceal the conspiracy: (1) the construction of a hiding place for secret papers by one of the conspirators; (2) the misleading of revenue agents as to expenses listed in the records of the conspirators; (3) communications between the involved parties instructing silence before the grand jury; and (4) the warning of a secretary by a conspirator to tell the investigators that she "didn't remember." (Note, supra, 70 Yale L.J. 1311 , 1325-1326.)

[12] We note that this court held in People v. Trim (1870) 39 Cal. 75, 77-79 , that declarations made after the commission of arson with intent to defraud an insurer but before receipt of the insurance proceeds were in furtherance of a conspiracy to violate the predecessor to Penal Code section 548. Whether Trim accords with our decisions in People v. Leach, supra, 15 Cal.3d 419 , and People v. Saling, supra, 7 Cal.3d 844 , is a question that we need not answer in the present context. In any event, we hold for purposes of the statute of limitations issue that such a conspiracy ends at the time that the arson is committed with the requisite intent.

[13] As an example of the lengths to which this doctrine has been extended see Pinkerton v. United States (1946) 328 U.S. 640 [ 90 L.Ed. 1489 , 66 S.Ct. 1180 ].

[14] These developments have not gone without considerable criticism. One commentator has rather concisely summarized the opposing viewpoint by stating that "This fear of conspiracy, frequently expressed in terms of sinister characters meeting in dark corners and secretly plotting organized criminality, depicts the climate of thinking which has nourished the inordinate growth of the conspiracy doctrine and the development of an intricate network of subtle rules, loose practices and nebulous criteria of proof." (Klein, Conspiracy The Prosecutor's Darling (1957) 24 Brooklyn L.Rev. 1, 4-5.)

[15] For an illuminating example of just such a situation, see the jury instruction given as to one of the defendants in Hyde v. United States, supra, 225 U.S. 347, 371-372 [ 56 L.Ed. 1114, 1127-1128 ].

[16] We need not deal with hypotheticals for such an example. In Eldredge v. United States (10th Cir.1932) 62 F.2d 449 , a participant in a bank embezzlement conspiracy performed one last act of concealment and then told his coconspirators that he "`was absolutely through and would have nothing further to do with the shortage....'" ( Id., at p. 450.) Nevertheless, his conviction for conspiracy was upheld despite the fact that the indictment against him had not been filed until four years later when the statute of limitations arguably had run. The court stated that "an essential part of the conspiracy planned was to conceal the shortage...." ( id., at p. 451), and therefore an expressed intent to withdraw was not sufficient for a defense of withdrawal because it would not result in termination of the entire conspiracy nor in revelation of the crime.

The result reached in Eldredge is only one short step from withdrawal cases in California. In People v. Jones (1961) 197 Cal. App.2d 503, 508 [ 17 Cal. Rptr. 252 ], the court held that the defendants had failed to show withdrawal from a conspiracy because they had done nothing to prevent commission of the offense. (See also People v. King (1938) 30 Cal. App.2d 185, 204 [ 85 P.2d 928 ].) If the only portion of the conspiracy which is still being committed is the concealment of the crime, then Jones would logically require a conspirator to confess his crime to the authorities before he could withdraw from further criminal conduct.

[17] It should also be noted that the difficulties in withdrawing from a conspiracy which such an approach would create would serve to subject the conspirators to a greater threat of blackmail, the very opposite result from one of the rationales underlying the statute of limitations.

[18] Language in Grunewald itself supports this conclusion, for in holding the substantial evidence of a concealment objective to be insufficient to extend the duration of the conspiracy the court said, "The crucial teaching of Krulewitch and Lutwak is that after the central criminal purposes of a conspiracy to conceal 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. As was there stated, allowing such a conspiracy to conceal to be inferred or implied from mere overt acts of concealment would result in a great widening of the scope of conspiracy prosecutions, because it would extend the life of a conspiracy indefinitely. Acts of covering up, even though done in the context of a mutually understood need for secrecy, cannot themselves constitute proof that concealment of crime after its commission was part of an initial agreement among the conspirators. For every conspiracy is by its very nature secret; a case can hardly be supposed where men concert together for crime and advertise their purpose to the world. And again, every conspiracy will inevitably be followed by actions 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, as well as extend indefinitely the time within which hearsay declarations will bind co-conspirators." ( Grunewald v. United States, supra, 353 U.S. 382, 401-402 [ 1 L.Ed.2d 931, 941-942 ] [italics added].)

Although we agree that the sanctioning of such a theory could increase the scope of conspiracy prosecutions beyond reasonable limits, elimination of this threat does not require the implicit-explicit distinction.

[19] Further, common sense indicates that almost everyone who contemplates the successful perpetration of a crime realizes beforehand that acts of concealment will probably be required after the commission of the offense. Indeed, the act of escape from the scene of a crime is manifestly a form of concealment. In addition, our analysis has indicated that a greater number of participants in an unlawful act is not a guarantee of success in concealing the crime in fact, the likelihood of concealment would seem to be an inverse ratio to the number of persons involved. Yet the danger of more effective concealment by a single perpetrator has not resulted in any special tolling provision for the statute of limitations in such cases. Thus not only does there appear to be no greater danger presented by an "explicit" rather than an "implicit" conspiracy to conceal but there does not seem to be any significantly different danger to society between either kind of conspiracy to conceal and the sole offender's intention to conceal his crime. If the dangers posed by the concealment of conspiracies call for particularized provisions in the statute of limitations, then such action is within the realm of the Legislature, not this court, and the Legislature has by its past action in providing for a "discovery" section applicable to only certain crimes demonstrated its concern for such problem areas.

[20] In the case of murder and other offenses which are excluded from the statute of limitations, the statute for conspiracy purposes, of course, runs from the completion of the substantive offense itself.

[21] However, when the conspiracy is truly a "continuing conspiracy," such as an auto theft ring or a conspiracy in restraint of trade ( United States v. Kissel, supra, 218 U.S. 601 [ 54 L.Ed. 1168 ]), the statute of limitations will not bar a conviction for conspiracy if an overt act occurs in the three years preceding the filing of an indictment or information because the conspiracy has not been completed or abandoned. (See Ware v. United States, supra, 154 F. 577 ; Jones v. United States, supra, 162 F. 417, 425-427 , giving a "series of robberies" example; cf. Nye & Nissen v. United States (1949) 336 U.S. 613, 616-617 [ 93 L.Ed. 919 , 923-924, 69 S.Ct. 766 ], affg. 168 F.2d 846 (9th Cir.1948); Braverman v. U.S. (1942) 317 U.S. 49, 50-54 [ 87 L.Ed. 23, 26-29 , 63 S.Ct. 99 ].) It is important to note that in such circumstances only one charge of conspiracy for the entire course of conduct will lie as to each substantive offense ( Braverman v. U.S., supra ); it would be improper to charge a separate conspiracy as to each auto theft, for example, and then extend the duration of each of those "conspiracies" by resort to the final or most recent theft within the preceding three years.

[22] Grand theft was explicitly excluded from the general provision of the three-year period and a final sentence was added to section 800 which read: "An indictment for grand theft ... shall be found, an information filed, or case certified to the superior court within three years after its discovery."

Section 800 has been subsequently amended to include other felonies within the "discovery" provision but the modifications do not affect the issues in this case.

[23] The imputation of a reasonable diligence requirement by the court in Swinney was based on a legislative statement of intent set out in a report which accompanied the bill enacting the "discovery" amendment to section 800. In relevant part, the report stated: "It is the intent of the Assembly Criminal Procedure Committee that the word `discovery' as it is used in that bill shall be interpreted in the same manner as the word `discovery' has been interpreted in section 338 of the Code of Civil Procedure." (Assem. J., July 14, 1969, p. 6402.)

[24] In each of the two separate counts the allegations of grand theft are set out in a routine pleading format. At the end of those allegations the following statement appeared: "It is further alleged that the above mentioned crime was not discovered until on or about February 22, 1972."

[25] In light of the disagreement over the propriety of holding such a hearing and the trial court's justifiable uncertainty on the issue because of the paucity of authority, we believe that we should comment on the subject even though none of the parties have questioned the trial court's ruling.

In our earlier discussion of the rationales underlying the statute of limitations, we referred to the rule that the statute of limitations may properly be raised at any time before or after judgment. ( In re Demillo, supra, 14 Cal.3d 598, 601 ; People v. McGee, supra, 1 Cal.2d 611, 613 .) However, McGee and its progeny have dealt only with cases in which the accusatory pleading is deficient on its face, i.e., where it fails to plead facts showing that the prosecution is not barred by the statute of limitations. The decision of the trial court in the instant case rested in large part on McGee and the belief that since the statute of limitations is jurisdictional a defendant should not be forced to undergo a trial on many other issues when an early resolution of the limitation question might obviate the necessity of undertaking a more rigorous trial on the merits. Our own research indicates that the propriety of such a hearing is a question of first impression in California.

The meager authority from other jurisdictions on the issue is split. In the federal courts the rule is that it is within the discretion of the trial court to determine whether or not to hold a hearing on the limitation issue before proceeding to the trial of the general issues. ( United States v. Haramic (W.D. Pa. 1954) 125 F. Supp. 128, 129 ; United States v. J.R. Watkins Company (D.Minn. 1954) 16 F.R.D. 229, 232-233 ; cf. United States v. Mathues (E.D. Pa. 1928) 27 F.2d 137 .) However, the rule is based on specific authority, rule 12(b)(4) of the Federal Rules of Criminal Procedure. In contrast, the court in State v. Snyder (1904) 182 Mo. 462, 503-505 [ 82 S.W. 12 ], held that the defendant had no right to a special preliminary trial on the limitation issue.

We believe on balance that the federal rule is the preferable practice. We are aware of our past statements that the limitation issue is, upon proof of probable cause to believe that the statute is not a bar, a question for the trier of fact. But the limitation question is a basic jurisdictional issue and the bar thereof is aimed as much at the prevention of untimely prosecutions as it is at the prevention of untimely convictions. If it appears possible that the evidence will establish as a matter of law that the period of limitation has run, then judicial economy may be far better served if the issue is resolved at the earliest possible stage of the proceedings rather than waiting until an entire trial on multiple issues is completed. Moreover, the determination of the trial court after such a hearing will be similar to that on a motion for a directed verdict (§§ 1118, 1118.1) without the necessity of first litigating the merits of the case. We conclude therefore that a trial court has within its discretion the power to hold an evidentiary hearing for purposes of determining whether as a matter of law the statute of limitations bars the prosecution. At such a hearing, it may properly be considered whether the reasonable diligence requirement of section 800 has been complied with. We emphasize that there is no right to such a preliminary determination of the limitation issue. In each case the court should, before granting a hearing on the issue, consider such factors as the likelihood that the People will be unable to meet their burden of proof on the question (a preponderance of the evidence, see People v. McGill (1935) 10 Cal. App.2d 155, 159-160 [ 51 P.2d 433 ]) and the potential length of both the hearing and a full trial on the merits. If the People prevail after such a hearing, then the limitation issue must still be resolved by the jury if it remains disputed by the defendant.

[26] The court's ruling on the sufficiency of the pleading raises a second issue which we must discuss although neither of the parties has urged the point on appeal. The trial court's ruling on this issue came, of course, before any appellate decision construing section 800's "discovery" provision. When the court in People v. Swinney, supra, 46 Cal. App.3d 332 , interpreted section 800 to include a reasonable diligence requirement it disagreed with the decision in People v. Doctor, supra, 257 Cal. App.2d 105 , insofar as Doctor required pleading of the same specificity as for section 338 of the Code of Civil Procedure. The Swinney court contended that such a standard would mandate technical and detailed pleadings inconsistent with the concepts of modern criminal pleading, holding that "To satisfy the demands of Penal Code section 800, it is enough to allege that the theft was not discovered until a date within the three-year period; relative to the element of reasonable diligence, it is enough to allege in general terms that in the exercise of reasonable diligence the theft could not have been discovered at an earlier date. Contrary to the Doctor case, the `facts' which support the general allegation are a matter of proof, not pleading." ( Id., at p. 344.)

We think that the better view lies somewhere between the standards set forth in Doctor and Swinney . While we agree that an accusatory pleading should not stand or fall solely on the facts averred on the face of the pleading, Swinney goes too far in allowing a generalized pleading. The general rule is that the accusatory pleading must allege facts showing that the prosecution is not barred by the statute of limitations. ( People v. Crosby, supra, 58 Cal.2d 713, 724 .) We do not believe that technical rules of pleading in criminal cases will be resurrected by requiring that the following facts be alleged in an accusatory pleading which seeks to avoid the bar of the statute of limitations by pleading the "discovery" provision of section 800: (1) the date on which the offense was "discovered"; (2) how and by whom the offense was "discovered"; (3) lack of knowledge, both actual or constructive, prior to the date of "discovery"; (4) the reason why the offense was not "discovered" earlier. A requirement that such facts be pleaded will serve several important functions. First, in the case of an indictment such pleading will focus the attention of the grand jury on the crucial factors which should be considered in assessing the evidence before it. If as in the instant case the grand jury is not made aware of the reasonable diligence requirement or of the information which was available to the victim or investigators at an earlier time, then it cannot assess whether a prosecution is barred by the statute of limitations. (Cf. Johnson v. Superior Court (1975) 15 Cal.3d 248, 255 [ 124 Cal. Rptr. 32 , 539 P.2d 792 ].) Second, in order to hold a defendant over for trial the People bear the burden of producing evidence (either before the grand jury or at the preliminary hearing) which demonstrates that there is probable cause to believe that the prosecution is not barred by the statute of limitations. ( People v. Crosby, supra, 58 Cal.2d 713, 725 .) We believe that the committing magistrate or the trial court will be aided in making the probable cause determination by specific pleading on the limitation issue. Such specificity will certainly serve to clarify the task of determining whether any diligence has been shown, particularly because the People bear the burden of proving a negative proposition, i.e., that knowledge of the offense could not have been obtained at an earlier time. With the basic facts on the "discovery" issue set forth in the pleading it will be much easier to then consider the People's additional proof on the issue which will be presented either to the grand jury or before the magistrate. Finally, specific pleading of the facts will serve to focus the attention of the trier of fact on the critical questions to be resolved preliminary to the ultimate decision of the limitation issue. To the extent that People v. Swinney, supra, 46 Cal. App.3d 332 is inconsistent with the foregoing standard of pleading it is disapproved.

[27] There is no record indication that the jury was ever instructed as to the lesser standard of proof that the People bear on the statute of limitations issue. Nor did the court indicate what standard it applied in making the preliminary determination on the issue. The proper burden is a preponderance of the evidence and thus when a limitation issue goes to the jury the preferable practice should be to carefully instruct the jury as to that burden making it clear the lesser burden applies solely to the limitation issue. (See People v. McGill, supra, 10 Cal. App.2d 155, 159-160 .)

[28] The record does not indicate if the inspectors discussed Szymanski's failure to mention Hanak's alleged fall into the fire pit the night before when he talked with Bailey. Bailey, however, had noted that there had been no evidence of any earlier fire in the barbeque pit.

[29] The blood on the glass was too small a quantity to type, but blood on the kitchen sink was found to be type "O," the same type as Hanak's blood.

[30] Saling had told Lenz that Hanak had been at his house on the night of the fire from at least 6:30 p.m.

[31] A letter was sent from the Chief of the Santa Barbara Fire Department to the San Jose Fire Department requesting that they check all the local optical outlets. The letter described Hanak and the basic facts of the case. Significantly, the letter began, "We have an arson case here that we have worked and are about ready to go to grand jury. However, we have one more loose end to tie up."

[32] Brewer later indicated that he was not told that the investigators thought the fire was an arson. He did not learn until 1972 that Hanak, Saling and Szymanski were under suspicion when he first talked to Lenz. He proceeded with settlement of the claim but promised to keep Lenz apprised of the settlement arrangements. Lenz apparently never inquired as to the specifics of the insurance policy, for if he had he would have learned that Bank of America had requested coverage of only $15,000. However, when the policy was issued in February 1968, the policy amount was $25,000. And, only 12 days before the fire on April 10th, the policy was amended to add an additional $10,000 of coverage and to reflect that a W.D. Taylor, an associate of Zamora, was a second trust deed holder.

On April 11, 1968, Brewer met with the independent insurance agent who had written the policy for Zamora to examine the burned structure. Zamora and Taylor were present and both of them expressed an opinion that the house was a total loss ($35,000). Taylor was asked to submit a bid since he was a contractor and about three days later Brewer received three different repair estimates, one for $15,000, one for $17,000, and Taylor's continued assertion that the structure was a total loss. Brewer's opinion was that Zamora was displeased with his appraisal that the property could be repaired and he heard Zamora comment, "I wonder how much it would cost to have this building demolished."

About May 13 Brewer telephoned Lenz and told him about Taylor's claim that the building was a total loss. Although Lenz knew that Taylor held a second trust deed on the property he apparently did not investigate further. During the call Brewer made it clear that Zamora had a financial interest in the property.

Eventually Brewer accepted the lowest repair bid and the property was rehabilitted for about $15,000. None of the money was ever received by Zamora, Saling, Szymanski or Oakdale Manor, but the charge of grand theft was proper because the value of the repairs to the structure had been secured by a false representation, i.e., that the insured had had nothing to do with the fire.

[33] In the trial court the People argued that the limitation period for concealed thefts should not begin to run until law enforcement authorities have sufficient evidence of the crime to prove their case beyond a reasonable doubt. We reiterate that such a theory misconstrues the purpose of the "discovery" provision of section 800. The statutory period is not tolled in such cases because it is more difficult to gather evidence of a concealed crime it is tolled because no one other than the criminals themselves even knows that a crime has been committed.

If a bank is robbed in broad daylight and the criminals successfully escape apprehension and prosecution because they are able to conceal their identities, the limitation period is not tolled on the ground that the authorities are unable to bring a case for prosecution. If an indictment or information is not filed within the ensuing three years the offenders may never be convicted of bank robbery because the limitation period will have run. The situation is no different in the case of a concealed theft to which the "discovery" provision of section 800 applies. Once there is sufficient knowledge (judged by the standards we have set forth herein) that a theft has been committed the limitation period will begin to run. If authorities do not file an indictment or information in the ensuing three years the perpetrators of the once concealed theft will, like the bank robbers, be immune from conviction for that crime.