On August 4, 1992, appellant William Jay Cole was convicted of possession with intent to distribute crack cocaine in violation of 21 U.S.C. Sec. 841(a)(1). The district court found that Cole's total relevant conduct was 6.0 grams of cocaine base; therefore, his offense level was 26 and his criminal history level was 1, yielding a sentencing range of 63-78 months. The district court sentenced him to the lower end of the range, 63 months. In United States v. Cole, 27 F.3d 996 (4th Cir.1994), this court vacated that sentence because Cole was not adequately afforded his right to allocution at the sentencing hearing. On remand, the district court again sentenced Cole to 63 months. Cole now appeals that sentence, alleging that the district court improperly estimated his total relevant conduct. Finding no error by the district court, we affirm.

The trial court determined that the government had proven 6.0 grams of crack cocaine by a preponderance of the evidence. We review that determination only to determine if it was clearly erroneous. That it was not. Cole sold 1.2 grams of crack to Arthur Hairston, a government informant, in the controlled buy that led to Cole's arrest. Hairston testified that he had previously purchased crack from Cole "maybe five times or so." J.A. at 27. Moreover, Hairston distinguished between "purchasing" crack and being given crack to pay off the $1200 debt Cole owed him from previous marijuana purchases, suggesting that the total number of cocaine transactions could have been as high as seven or eight.

The district court relied on the presentence report which computed the 6.0 grams by adding the 1.2 grams from the controlled buy to an estimated 4.8 grams in previous crack sales. The presentence report apparently calculated the 4.8 grams by taking the controlled buy (1.2 grams for $300) and extrapolating it out over the $1200 debt that Cole paid Hairston off with cocaine, and so estimated that his previous sales totaled 4.8 grams. Agents from the narcotics task force testified that this was consistent with typical crack street prices. J.A. at 138.

Not only is this conclusion supported by a preponderance of the evidence, it seems a very conservative estimate, since if the report had multiplied his controlled buy (1.2) by five, it would have arrived at 6.0 (which, added to the controlled buy amount of 1.2 would have yielded a total of 7.2). And if the report had multiplied 1.2 grams by the seven or eight transactions that Hairston's testimony suggested (distinguishing between "purchases" and "paying off debt"), the result would have been even higher (8.4 or 9.6 grams, which, added to the 1.2 grams would have yielded a total of 9.6 or 10.8 grams).

The district court had clear authority to approximate the quantity of a drug in its determination of relevant conduct and to rely upon circumstantial evidence and statistical methods in making that determination. See United States v. Uwaeme, 975 F.2d 1016, 1021 (4th Cir.1992) (upholding the use of statistical methods to calculate the total drug weight possessed by a defendant); cf. United States v. Hicks, 948 F.2d 877, 881 (4th Cir.1991) (using street values to compute the drug equivalent of a large amount of cash for sentencing purposes).

Moreover, U.S.S.G. Sec. 2D1.1 Application Note 12 (emphasis added) provides as follows:

Types and quantities of drugs not specified in the count of conviction may be considered in determining the offense level. See Sec. 1B1.3(a)(2) (Relevant Conduct). Where there is no drug seizure or the amount seized does not reflect the scale of the offense, the court shall approximate the quantity of the controlled substance. In making this determination, the court may consider, for example, the price generally obtained for the controlled substance, financial or other records, similar transactions in controlled substances by the defendant, and the size or capability of any laboratory involved. That is exactly what the district court did. We find that that determination was not clearly erroneous.

AFFIRMED.