Paul Y.B. Hahn was convicted after a jury trial on four counts related to the illegal possession of a firearm and slightly less than a gram of methamphetamine. In similar cases involving this amount of methamphetamine, the United States Sentencing Guidelines prescribe a range of ten to sixteen months’ incarceration. Yet on the drug possession offenses, Hahn was sentenced to prison for ninety-seven months — just over eight years — because of “relevant conduct” which occurred more than five months prior to the conduct for which Hahn was convicted. The issue before us concerns whether conduct extraneous to a conviction is nevertheless relevant in passing sentence. We vacate Hahn’s sentence and remand for resentencing.

FACTS AND PROCEEDINGS

On March 18, 1989, Honolulu police approached Hahn’s parked car to investigate a discrepancy between the car’s apparently valid registration sticker and a report that the car’s registration had expired. When Hahn opened one of the car’s tinted windows, the police saw a pistol lodged between the car’s seats. Hahn and Lori Mit-sunaga, the car’s other occupant, were then arrested. Two searches of the car resulted in the seizure of approximately 92/100ths of a gram of methamphetamine contained in a total of nine separate packets.

Hahn was indicted on four counts, all based strictly on the events of March 18, 1989: (1) possession of a firearm in commerce by an unlawful user or addict of a controlled substance, in violation of 18 U.S.C. § 922 (g)(3); (2) possession of a controlled substance, in violation of 21 U.S.C. § 844 ; (3) carrying a firearm in relation to a drug trafficking crime, in violation of 18 U.S.C. § 924 (c)(1); and (4) possession of approximately four grams of a controlled substance with intent to distribute, in violation of 21 U.S.C. § 841 (a)(1).

At trial, the government presented testimony from Mitsunaga, her brother, and others, that Hahn was involved in large-scale methamphetamine dealing and had carried firearms in the year prior to his arrest. Law enforcement officers also testified that Hahn admitted dealing methamphetamine for more than a year prior to his arrest and selling a quarter of a pound of the drug per day in July and August of 1988. Pursuant to Fed.R.Evid. 404(b), the district court instructed the jury to consider this evidence only for the purpose of determining the intent element of the crime concerning the charge of possession with intent to distribute. The jury convicted Hahn on all four counts.

After trial, a presentence report was prepared. The report recounted not only Hahn’s conduct relating to the offenses charged in the indictment, but also his methamphetamine dealing and carrying of firearms in the prior year. The report relied on interviews with witnesses, law enforcement officers, and Hahn, rather than on the trial testimony. The report included Hahn’s admission to a law enforcement officer that he had made daily sales of a quarter pound of methamphetamine in July and August of 1988. It also included Hahn’s admission to the probation officer that he had sold an ounce of the drug every three days between June and September 1988.

In calculating Hahn’s base offense, level for the possession with intent to distribute conviction (Count Four of the indictment), the presentence report did not use the less-than-a-gram quantity actually found in Hahn’s possession when arrested in March 1989. Neither did it use the four-gram quantity mentioned in the indictment. Instead, the report relied on Hahn’s admission that he had sold an ounce every three days between June and September 1988. In his report, the probation officer multiplied days by ounces-sold-per-day to arrive at a total amount of forty ounces sold. When Hahn was sentenced on October 16, 1989, the Sentencing Guidelines prescribed a base offense level of twenty-eight for possession of forty ounces of methamphetamine with intent to distribute. With Hahn’s base offense level at twenty-eight, the Guidelines indicated the range for Hahn’s sentence on Counts Two and Four to be seventy-eight to ninety-seven months. If the report had employed the less-than-a-gram of methamphetamine underlying Hahn’s convictions, however, Hahn’s possible sentence for Counts Two and Four would have ranged only from ten to sixteen months. See United States Sentencing Commission, Guidelines Manual, § 2D1.1 (Oct. 15, 1988).

Hahn objected to the use of the estimated forty ounces of methamphetamine to determine his base offense level. At the sentencing hearing, the government called to the witness stand Wayne Wong, the probation officer who prepared the presen-tence report. Officer Wong testified to Hahn’s admission on which the forty ounce estimate was based. Furthermore, Drug Enforcement Administration Agent Howard testified to Hahn’s admission that, during his “peak times” in the summer of 1988, he was selling a quarter pound of methamphetamine per day. During the sentencing hearing, the government also expressly referred to testimony at Hahn’s trial concerning narcotics activity between September 1988 and Hahn’s arrest in March 1989. Further reference was made to “a certain notebook ... and certain papers of the defendant^]” detailing drug sales. At sentencing, however, defense counsel objected to this documentary evidence. From this objection, and the government’s response thereto, it does not appear that this evidence was considered in determining whether the government proved facts sufficient to support Hahn’s sentence.

The district court denied Hahn’s objections and adopted the presentence report’s calculations. The court sentenced Hahn to 157 months in prison, including ninety-seven months for the possession with intent to distribute conviction. Hahn timely appeals and raises the following four contentions concerning his sentence: (1) The Sentencing Guidelines deny due process in failing to provide for an individualized sentence; (2) the district court improperly relied on the June-September 1988 activities in sentencing Hahn for activities occurring in March 1989; (3) if the 1988 events are to form the basis of Hahn’s sentence, they should be proved beyond a reasonable doubt; and (4) in any event, the district court erred in relying solely on Hahn’s own admission of his activities occurring in 1988.

We reject summarily the first contention. See United States v. Brady, 895 F.2d 538, 540 (9th Cir.1990). Our resolution of the second contention obviates any present need to consider either the third or the fourth. We therefore leave these latter issues for future consideration by the district court.

STANDARD OF REVIEW

We review the legality of Hahn’s sentence de novo. United States v. Turner, 898 F.2d 705, 708 (9th Cir.), cert. denied, 495 U.S. 962 , 110 S.Ct. 2574 , 109 L.Ed.2d 756 (1990). Whether conduct extraneous to an offense of conviction is part of the same “course of conduct” or “common scheme or plan” as the offense of conviction so as to be considered “relevant conduct” within the meaning of Guidelines § lB1.3(a)(2), is reviewed for clear error. United States v. Motz, 936 F.2d 1021, 1026 (9th Cir.1991); United States v. Dillard, 929 F.2d 500, 504 (9th Cir.1991); Turner, 898 F.2d at 711 .

■ DISCUSSION

A. Background

As indicated by its accompanying commentary, Guidelines section 2D1.1 governs the determination of a defendant’s base offense level for possession with intent to distribute a controlled substance. Specifically, section 2D1.1 requires that the base offense level be determined in part on the basis of the quantity of illegal drugs attributable to the offense of conviction.

Under the Guidelines, the relevant quantity of controlled substances is not limited to the amount of illegal drugs directly involved in the offense of conviction. United States v. Nakagawa, 924 F.2d 800, 803 (9th Cir.1991). Instead, Guidelines § lB1.2(b) instructs federal sentencing courts to determine the pertinent amount of drugs in accordance with the principles of “relevant conduct” set forth in Guidelines § 1B1.3.

Pursuant to section lB1.3(a)(2), the Guidelines in effect at the time of Hahn’s sentencing advise that, “in a drug distribution case, quantities and types of drugs not specified in the count of conviction are to be included in determining the offense level if they were part of the same course of conduct or part of a common scheme or plan as the count of conviction.” U.S.S.G. § 1B1.3 comment, (backg’d) (Oct. 15, 1988) (emphasis added). The district court applied this provision to sentence Hahn on the basis of the June through September 1988 drug-related activities — activities for which Hahn was never indicted, tried, or convicted.

Under pre-Guidelines sentencing practice, a district court could properly look to almost any activity of the defendant in determining a just punishment. United States v. Restrepo, 946 F.2d 654, 655 (9th Cir.1991); id. at 665-66 & n. 1 (Norris, J., dissenting); United States v. Ebbole, 917 F.2d 1495, 1496 (7th Cir.1990). With the institution of the Sentencing Guidelines, however, consideration of certain conduct at sentencing became mandatory rather than merely discretionary. Restrepo, 946 F.2d at 655 (accepting defendant’s premise that “the Sentencing Guidelines ... mandate the sentencing effect of uncharged crimes”); id. at 664-65, 669 (Norris, J., dissenting) (discussing mandatory effect of relevant conduct under the Guidelines); cf. United States v. Belgard, 894 F.2d 1092, 1100 (9th Cir.) (describing remnants of discretion with which district courts are still endowed in passing sentence), cert. denied, — U.S. -, 111 S.Ct. 164 , 112 L.Ed.2d 129 (1990). With this development, “[t]he Guidelines have made the defendant’s interest in a fair sentence more defined and protectible.” Restrepo, 946 F.2d at 659 . “A convicted defendant has an interest in the accurate application of the Guidelines within statutory limits, nothing more, nothing less.” Id. (footnote omitted) (emphasis added).

In mandating penal consequences for “relevant conduct” in certain cases, the Guidelines implicate the principles enunciated in In re Winship, 397 U.S. 358 , 90 S.Ct. 1068 , 25 L.Ed.2d 368 (1970). See Restrepo, 946 F.2d at 659 -60 & n. 9; id. at 664 (Norris, J., dissenting) (“[i]n allowing ... separate crimes to be used as sentencing factors with mandatory penal consequences, the Guidelines encounter the due process mandate of [ Winship ]”); United States v. Miller, 910 F.2d 1321, 1330-31 (6th Cir.1990) (Merritt, C.J., dissenting), cert. denied, — U.S. -, 111 S.Ct. 980 , 112 L.Ed.2d 1065 (1991). Winship , of course, interprets due process to “protect[ ] the accused against conviction except upon proof beyond a reasonable doubt of every fact necessary to constitute the crime ... charged.” 397 U.S. at 364 , 90 S.Ct. at 1073 . Although facts pertinent to sentencing under the Guidelines normally need only be proved by a preponderance, Restrepo, 946 F.2d at 661 , such facts frequently amount to criminal conduct apart from the offenses of conviction. Thus, without a clear demarcation of facts pertinent to sentencing, a defendant may well be punished solely for a crime of which she has not been convicted. In the present case, for example, we would be unable to uphold the Guidelines’ requirement that Hahn be sentenced in part on the basis of criminal conduct which occurred more than five months prior to the incident from which Hahn’s convictions arises unless there exists a meaningful standard for determining relevant conduct under section lB1.3(a)(2). Thus, we must determine whether the concepts embodied in the Guidelines’ notion of a “course of conduct” and a “common scheme or plan” are sufficiently concrete to permit the “accurate application” of Guidelines section lB1.3(a)(2) contemplated by this Court in Restrepo, id. at 659, and to avoid the constitutional problem to which this ease points.

Our inquiry as to the meaning of “relevant conduct” is further prompted by the Guidelines themselves, which seek “reasonable uniformity in sentencing by narrowing the wide disparity in sentences imposed for similar criminal offenses committed by similar offenders.” U.S.S.G. Ch. 1, Pt. A, in tro. comment.; see also 18 U.S.C. § 3553 (a)(6); 28 U.S.C. § 991 (b)(1)(B). Unless the principles of “relevant conduct” can be applied in a fairly uniform manner, the Guidelines will fail in their stated goal of narrowing sentencing disparity. In this case, for example, Hahn would have received a sentence of between ten and sixteen months’ incarceration on his drug convictions but for relevant conduct that increased his sentence at least sixfold. His actual sentence of ninety-seven months reveals the tremendous potential a haphazard application of the “relevant conduct” provision would have for undermining perhaps the most important objective of the Guidelines.

B. Defining Belevant Conduct Under Section lB1.3(a)(2)

To begin with, it is not clear that the Sentencing Commission intended “relevant conduct” to include all conduct upon which a sentence constitutionally might be based. Rather, in requiring the inclusion of all drug amounts involved in the same “course of conduct” or in a “common scheme or plan,” the Guidelines sought to avoid a problem which the Commission associated with offenses such as those involving drug distribution.

Specifically, the Commission looked beyond the amount of drugs directly involved in the offense of conviction because drug distribution offenses “often involve a pattern of misconduct that cannot readily be broken into discrete, identifiable units that are meaningful for purposes of sentencing.” U.S.S.G. § 1B1.3 comment, (backg’d). Thus, when illegal conduct does exist in “discrete, identifiable units” apart from the offense of conviction, the Guidelines anticipate a separate charge for such conduct. Cf. United States v. Wood, 924 F.2d 399, 403-05 (1st Cir.1991) (section lB1.3(a)(2) is an exception to the Guidelines charge-offense approach to sentencing which should not be interpreted so as to frustrate the charge-offense approach). Certainly a separate conviction for such conduct is the preferable means for taking the conduct into account at sentencing, not only because a separate conviction minimizes constitutional concerns, but also because separate convictions facilitate a more uniform application of the Guidelines. Cf. Ebbole, 917 F.2d at 1501 (noting that Guidelines § 1B1.3 constitutes a “dubious invitation” for prosecutors “ ‘to indict for less serious offenses which are easy to prove and then expand them in the probation office’ ”) (quoting Miller, 910 F.2d at 1332 (Merritt, C.J., dissenting)).

Under a clear error standard of review, however, it is inevitable that conduct which should have been prosecuted separately nevertheless will be included from time to time as “relevant conduct” in sentencing a defendant for a particular offense. See United States v. Lawrence, 915 F.2d 402, 406 (8th Cir.1990) (distinguishing between a narrow “relevant conduct” standard evident in cases where the conduct occurred within a few weeks of the offense of conviction, and a broader approach in which decisions considering more attenuated conduct as “relevant” were upheld as not clearly erroneous). Consequently, in order to avoid constitutional difficulties and to implement the uniform application of the Guidelines, it is necessary to develop a factually-oriented test for determining whether a district court has clearly erred in considering as “part of the same course of conduct or part of a common scheme or plan as the count of conviction” conduct which exists in “discrete, identifiable units” apart from the offense of conviction.

The pertinent factors to be considered are well stated in United States v. Santiago, 906 F.2d 867 (2d Cir.1990). In assessing whether conduct is “relevant” within the meaning of Guidelines § lB1.3(a)(2), “the sentencing court is to consider such factors as the nature of the defendant’s acts, his role, and the number and frequency of repetitions of those acts, in determining whether they indicate a behavior pattern.” Id. at 872. There must be “ ‘sufficient similarity and temporal proximity to reasonably suggest that repeated instances of criminal behavior constitute a pattern of criminal conduct.’ ” Id. (quoting William W. Wilkins, Jr. & John R. Steer, Relevant Conduct: The Cornerstone of the Federal Sentencing Guidelines, 41 S.C.L.Rev. 495, 515-16 (1990)). Thus, the essential components of the section lB1.3(a)(2) analysis are similarity, regularity, and temporal proximity.

Of course, it is for the district court to determine in the first instance whether these components exist in proper amounts and proportions to support a finding that certain extraneous conduct is nevertheless relevant for sentencing purposes. See Motz, 936 F.2d at 1026 (applying clearly erroneous standard of review); Lillard, 929 F.2d at 504 (same). We cannot formulate precise recipes or ratios in which these components must exist in order to find conduct relevant. When one component is absent, however, courts must look for a stronger presence of at least one of the other components. In cases such as the present one, where the conduct alleged to be relevant is relatively remote to the offense of conviction, a stronger showing of similarity or regularity is necessary to compensate for the absence of the third component. Compare Phillippi, 911 F.2d at 151 (holding that the dates and nature of conduct occurring “as remotely as two years before [the defendant’s] arrest” must be “clearly establish[ed]” in order to be considered relevant) with United States v. Cousineau, 929 F.2d 64, 68 (2d Cir.1991) (“Because of the continuous nature of the conduct and the circumstances of this case, we are not reluctant to consider relevant the conduct that occurred during the course of a two year period.”).

Thus, when conduct alleged to be relevant is temporally remote from the conduct underlying the conviction, and the relevance of the extraneous conduct depends primarily on its similarity to the conviction, it is not enough that the extraneous conduct merely amounts to the same offense as the offense for which the defendant was convicted. See Kappes, 936 F.2d at 231 ; United States v. White, 888 F.2d 490 , 500 (7th Cir.1989). Rather, in looking beyond the offense of conviction, a district court must consider whether specific similarities exist between the offense of conviction and the temporally remote conduct alleged to be relevant under Guidelines section lB1.3(a)(2).

When regularity is to provide most of the foundation for temporally remote, relevant conduct, specific repeated events outside the offense of conviction must be identified. Regularity is wanting in the case of a solitary, temporally remote event, and therefore such an event cannot constitute relevant conduct without a strong showing of substantial similarity. See Kappes, 936 F.2d at 230-31 ; cf. Santiago, 906 F.2d at 873 (“The fact that a period of some eight months had elapsed before a controlled buy could be arranged did not preclude a finding that Santiago’s attempt to sell to Shattuck for the 13th time was part of the same course of conduct as his prior 12 sales.”); United States v. Mak, 926 F.2d 112, 114-16 (1st Cir.1991) (finding of relevant conduct upheld where four similar drug deals were separated by six, three, and nine months, respectively).

C. Application and Conclusion

Once Hahn objected to the presentence report’s use of his Summer 1988 activity in calculating the recommended sentence, the burden fell upon the government to prove that Hahn’s 1988 activity was “relevant conduct” under Guidelines § lB1.3(a)(2). See United States v. Howard, 894 F.2d 1085, 1090 (9th Cir.1990). As we have indicated, this requires a showing of similarity, regularity, and temporal proximity in sufficient proportions so that a sentence may fairly take into account conduct extraneous to the events immediately Underlying the conviction. This test is especially important in cases where the extraneous conduct exists in “discrete, identifiable units” apart from the conduct for which the defendant is convicted.

In the present case, evidence of specific similarity and regularity is important in view of the gap of more than five months between Hahn’s June-September 1988 conduct and his March 1989 conduct. We are unsure what evidence the district court considered in concluding that the earlier conduct was relevant to the later conviction or whether the government showed any evidence of a common scheme or plan. We therefore remand for resentencing and findings to answer the question whether Hahn’s 1988 activities can be deemed “relevant conduct” in sentencing Hahn for crimes committed in March 1989.

The sentence is vacated and the cause is remanded for proceedings consistent with this opinion.

VACATED and REMANDED.

. The four-gram amount alleged in the indictment was derived from a rough estimate of the quantity of methamphetamine recovered from Hahn on March 18, 1989, later determined to weigh slightly less than a gram. . Counts Two and Four were grouped for sentencing purposes. A consecutive sentence of sixty months was mandated for Count Three. See 18 U.S.C. § 924 (c)(1). The penalty for Count One was considered to be subsumed within the penalty for Count Three. . Specifically, the version of the Sentencing Guidelines in effect at the time of Hahn’s sentencing indicates that one gram of methamphetamine is equivalent to two grams of cocaine, that the base offense level for someone possessing less than twenty-five grams of cocaine with intent to distribute is twelve, and that a person with Hahn’s criminal history category of "I” would thus be subject under the Guidelines to a period of incarceration ranging from ten to sixteen months. The result would be the same if the presentence report used the four grams alleged in the indictment. . The district court did not explicitly receive or reject this proffer. The relevant exchange between the United States Attorney and the district judge reads as follows: MR. BURKE: Your Honor, I’ll ask the record to reflect that there was live testimony during trial regarding [the amount of drugs Hahn was dealing between September 1988 and March 1989]. THE COURT: How was that reflected here in the presentence report? MR. BURKE: I don’t believe it was, Your Honor.... [Nevertheless, t]here is evidence on record, sworn testimony subject to cross-examination regarding the man's drug dealing after the summer up to the time of his arrest. THE COURT: All right but that wasn’t considered in the presentence report as far as ... relevant offense conduct, was it? MR. BURKE: It was not, Your Honor. Sentencing Transcript at 29-30; see also id. at 31-32 (probation officer indicates that Hahn’s drug activities between September 1988 and March 1989 were not “calculated”). The district court’s comments suggest that, before admitting evidence at sentencing concerning events occurring between September 1988 and March 1989, the government should have requested the probation office to amend the presentence report to reflect such events. .This conclusion is based on the following exchange, which took place at the sentencing hearing: [MR. BURKE]: [W]ould you tell the Court, in summary, what those notes and reports show. MR. YAMAMOTO: Your Honor, I’m going to object to that question. I don't know what it’s relevant to. I don't think there is any objections [sic] on the part of the defendant that certain documents were not recovered from him. MR. BURKE: That’s fine, Your Honor. I can move on, if the Court wishes. THE COURT: Very well. MR. BURKE: If there’s no objection to that portion of the presentence report that refers to Agent Howard’s investigation as to the documents. BY MR. BURKE: Q Then I’ll turn your attention to a question of whether or not you interviewed the defendant. ... Thus, it appears the presentence report contains the only information regarding Hahn’s notes that the district court considered. Because the presentence report says nothing about the dates of the transactions detailed in the notes and papers, this evidence ultimately cannot be considered relevant in determining whether the June-September 1988 activity was "relevant conduct” in sentencing Hahn for the crimes committed in March 1989. See United States v. Phillippi, 911 F.2d 149, 151 (8th Cir.1990) (failure to establish date of alleged "relevant conduct” precluded consideration of conduct at sentencing), cer t. denied, — U.S. -, 111 S.Ct. 702 , 112 L.Ed.2d 691 (1991). We note, however, that the papers recovered from Hahn contained "no records of transactions between January 25th, of 1989, and March 11, 1989." It further appears that the district court did not consider Agent Howard’s testimony concerning a tape recording in which Hahn discussed drug dealing with his wife. Although the presentence report indicates that the tape recording was made in November 1988, no testimony in support of this date was offered at trial or sentencing, even though Hahn questioned the existence of such a tape. . This portion of the background to section IB 1.3 describes the practical effect of subsection (a)(2). Instead of referring to "drug distribution cases,” as the background does, section lB1.3(a)(2) refers to "offenses of a character for which § 3D 1.2(d) would require grouping of multiple counts.” . Our Circuit’s recent en banc decision in Res-trepo did not address this issue. There was no question in Restrepo that the alleged conduct occurring outside the offenses of conviction could be considered “relevant conduct" under Guidelines § 1B1.3. See 946 F.2d at 661 n. 11; see also the second decision of the three-judge panel, 903 F.2d 648, 655 (9th Cir.1990), withdrawn in part, 946 F.2d 654 (9th Cir.1991) (en banc), cert. denied, — U.S. -, 112 S.Ct. 1684 , 118 L.Ed.2d 211 (U.S. Dec. 31, 1991) (No. 91-6859). The only question in Restrepo was whether the government must prove the allegations by more than a preponderance. 946 F.2d at 655 . In the present case, however, it is not clear that Hahn’s Summer 1988 conduct is “relevant conduct,” even if the events can be proven by the appropriate standard. Cf. United States v. Kappes, 936 F.2d 227, 230 (6th Cir.1991) (“In this case, the fact that Kappes concealed information in violation of federal law in 1983 is not in dispute. Instead, the issue is whether the 1983 offense was ‘part of the same course of conduct or common scheme or plan as the offense of conviction!.]’ ”)■ . This can also happen when a defendant actually has been indicted for a separate crime encompassing the “relevant conduct” but, for whatever reason, has not been convicted of this separate crime. However, recent decisions of this Circuit foreclose to some extent the possibility that conduct for which a separate charge has been brought might also be used as "relevant conduct” in sentencing. See United States v. Faulkner, 952 F.2d 1066 (9th Cir.1991) (amended Dec. 24, 1991) (“It would be patently unfair if the court were ... allowed to hold Faulkner to his part of the bargain — his plea of guilty on five counts — while simultaneously denying him the benefits promised him from the bargain by relying on the ... dismissed counts in sentencing him.”); United States v. Fine, 946 F.2d 650, 651-52 (9th Cir.1991); United States v. Brady, 928 F.2d 844, 851 (9th Cir.1991) ("[I]t does not follow that the Guidelines permit a court to reconsider facts during sentencing that have been rejected by a jury’s not guilty verdict.”). . In particular, the focus of our concern is conduct outside the immediate time frame of the offense of conviction. In extreme cases, the span of time between the alleged "relevant conduct” and the offense of conviction may be so great as to foreclose as a matter of law consideration of extraneous events as "relevant conduct.” See Kappes, 936 F.2d at 230-31 (“Although similar, the two offenses were unrelated acts separated by the passage of six years.... [I]t would take an impermissible stretch of the imagination to conclude that the 1983 offense was part of the same ‘course of conduct’ as the 1989 offense.") (citing, inter alia, U.S.S.G. § IB 1.3, comment, (backg’d)). . The Santiago court focused on the meaning of “same course of conduct.” 906 F.2d at 872-73 . Subsequent decisions have thus emphasized that “same course of conduct" and "common scheme or plan” are two different things. See United States v. Azeem, 946 F.2d 13, 16 (2d Cir.1991) (Sneed, J.); United States v. Perdomo, 927 F.2d 111, 114 (2d Cir.1991); Lawrence, 915 F.2d at 407 . The purpose for making this distinction was, in essence, to rule that direct evidence of a common scheme or plan was not a prerequisite to a finding that certain conduct was relevant within the meaning of Guidelines § lB1.3(a)(2). See Azeem, 946 F.2d at 16 ; Perdomo, 927 F.2d at 115 . Having made this ruling, there appears no other reason to distinguish between the two types of conduct made relevant under Guidelines § lB1.3(a)(2). . Direct evidence of a "common scheme or plan," or a conspiracy conviction, may also permit attenuated conduct to be considered "relevant” under Guidelines § lB1.3(a)(2). See, e.g., United States v. Lokey, 945 F.2d 825, 840 (5th Cir.1991) (“In determining Lokey’s sentence, the district court included all of Lokey’s drug transactions from 1984 to 1989 ... because they were part of the same conspiracy for which he was convicted.”) (citing United States v. Woolford, 896 F.2d 99, 104 (5th Cir.1990)).