Phillip McCutchen appeals from a judgment in a criminal case in which he was convicted of possession with intent to distribute nine grams of cocaine base (crack), in violation of 21 U.S.C. § 841(a)(1), and of distributing nine grams of crack within one thousand feet of an elementary school, in violation of 21 U.S.C. § 860.1 McCutchen was arrested with 119 vials of a chunky white powder in his possession. Fifteen of these vials were later analyzed by a police chemist, who determined that they contained 1381 milligrams of a substance or mixture containing cocaine base.2 From the weight of the substance in these 15 vials, the chemist projected that the total weight of the substance in all 119 vials was 9.866 grams. The district court, relying on the chemist's quantity determination, imposed concurrent sentences of 78 months imprisonment on each of the two drug counts.

The defendant argues that the court erred in including the weight of the substance in the 104 untested vials in the total quantity of drugs to be used to calculate his base offense level. The government responds that extrapolating from the random sample was sufficient to meet its burden of establishing the quantity of drugs by a preponderance of the evidence. We will affirm. In doing so, however, we make clear that when a defendant contests a quantity determination that is based on extrapolation from a test sample, the district court must make a finding that there is an adequate factual basis for the extrapolation and that the quantity was determined in a manner consistent with accepted standards of reliability.

The background facts are essentially uncontested. The defendant was arrested by undercover Philadelphia police officers who, during a routine patrol, observed the defendant give another man several vials in return for cash. The officers also saw what appeared to be the butt of a firearm stuck in the defendant's front waistband. When the officers approached and identified themselves, the defendant attempted to run but was stopped by one of the officers. During the pat down, the defendant dropped a clear plastic bag containing 119 vials, each of which had a clear cap and was filled with a chunky white powder.

At the sentencing hearing, a Philadelphia police chemist testified that the net weight of the substance in the 119 vials was 9.866 grams, a quantity he arrived at by extrapolating from the weight of the substance in the 15 vials that had been tested. While the chemist could not quantify what portion of the substance in each of the 15 vials was cocaine, the tests revealed that all 15 contained cocaine base. As to the process for selecting the 15 vials, the chemist's testimony was as follows:

THE COURT: How did you select these vials, sir? How did you decide which 15 to select?

ANSWER: Well the original four, I did select two which had been field tested. And I tested two others at random. The 11 that I did today, I just picked the first 11 that I got out of the bag.

THE COURT: You're the one that selected them?

ANSWER: That's correct.

THE COURT: And no one suggested to you which ones you selected?

ANSWER: No.

The chemist also testified that it was accepted practice within the Philadelphia police department to test a representative sample, rather than all of the seized drugs:

QUESTION: In your field, is it common to test each and every vial that's contained in a package such as this?

ANSWER: No, it's not common. It would be a very rare instance when we tested every vial in a submission this large. It would be way too time consuming. We're very busy with a lot of cases. And what we do is we take a representative sample and then we use that to project what is probably in the rest of them.

The chemist's testimony together with the lab reports reveal that the first four vials tested contained a substance weighing 97, 102, 99, and 84 milligrams, respectively, for a total weight of 382 milligrams. The substance in the remaining 11 vials was calculated to weigh 1,001 milligrams, or just over one gram. In total, the tested vials produced 1383 milligrams of a substance containing cocaine base. Based on this quantity, the chemist calculated (by extrapolation) that the net weight of the substance from all 119 vials was 9.866 grams.

The presentence report, utilizing the chemist's estimate of quantity, calculated a base offense level of 28 based on the presence of between 5 and 20 grams of cocaine base. See U.S.S.G. §§ 2D1.1(a)(3), (c)(9) & 2D1.2(a)(1).3 The report further determined that the defendant had a Criminal History Category of III, given a prior juvenile commitment and escape.4 See U.S.S.G. §§ 4A1.1(d), (e) & 4A1.2(d)(2)(A). Factoring in an adjustment for acceptance of responsibility, U.S.S.G. § 3E1.1, the court reduced the total offense level to 26. Based on the total adjusted offense level of 26 and a Criminal History Category of III, the court sentenced the defendant to concurrent 78 month terms of incarceration on the two drug counts.5

At the sentencing hearing, in response to the defendant's challenge to the use of the extrapolated quantity, the district court first noted that the chemist had randomly chosen the 15 vials tested. The Court then found:

[W]hen someone is arrested and found to be in possession of a large number of identical vials containing powder and a random sample confirms that each sampled vial contains crack, it is a very reasonable inference to make ... that the other vials also contained crack.

As the foregoing makes clear, to arrive at the defendant's sentence, the court had to rely on the chemist's extrapolation from the 15 tested vials to the 104 untested vials. In order to do so, the court had to assume that the untested vials, like the tested ones, contained cocaine base. Since no test was performed to determine the amount of pure cocaine base in each vial or to determine what other substances were present, and since the milligram net weight per tested vial varied (from a low of 69 to a high of 153 milligrams), the defendant argues that it is neither fair nor reasonable to assume that all of the untested vials also contained cocaine. The defendant contends that while the court may consider the net weight of the mixed substance, in contrast to only considering the quantity of the pure cocaine base, to calculate his offense level, see United States v. Gurgiolo, 894 F.2d 56, 60-61 (3d Cir.1990), the government's preponderance burden6 is not met by assuming that since 15 vials tested positive for cocaine base, the remaining 104 also contain cocaine base.

In response, the government argues that the case law has uniformly upheld quantity estimates based on extrapolation from tested samples. In particular, the government relies on United States v. Pirre, 927 F.2d 694 (2d Cir.1991). Pirre pled guilty to importing 15 bricks of cocaine and was sentenced by the district court to 78 months in prison. He appealed, contending that the estimated weight of 15.09 kilograms was so close to the 15 kilogram threshold of the base offense level that all 15 of the cocaine bricks should have been weighed. As part of the sentencing process, the district court convened a hearing on the weight of the cocaine. At that hearing, the government's chemist testified that all 15 packages appeared to be the same size, contained the same logo, were wrapped in the same type of electrical tape, and each appeared to be a standard one-kilogram brick. Additionally, the chemist testified that he had tested 8 of the 15 bricks, and that each of the tested bricks contained cocaine. The district court found that the government had proved, by a preponderance of the evidence, that the weight of the cocaine was 15.09 kilograms. The court of appeals affirmed, holding that "[i]t is sufficient for the government to show that its method of estimating the total is grounded in fact and is carried out in a manner consistent with accepted standards of reliability." Pirre, 927 F.2d at 697.7

We endorse the analysis and holding of Pirre. If a defendant challenges a drug quantity estimate based on an extrapolation from a test sample, the government must show, and the court must find, that there is an adequate basis in fact for the extrapolation and that the quantity was determined in a manner consistent with accepted standards of reliability.

That does not mean, however, that the government must produce expert statistical evidence on sampling techniques. Rather, reasonable reliability is the touchstone of the determination. See U.S.S.G. § 6A1.3 ("In resolving any reasonable dispute concerning a factor important to the sentencing determination, the court may consider relevant information without regard to its admissibility under the rules of evidence applicable at trial, provided that the information has sufficient indicia of reliability to support its probable accuracy." (emphasis added)); Uwaeme, 975 F.2d at 1021 ("A district court's finding of quantity is not erroneous if it is based on evidence possessing sufficient indicia of reliability to support its probable accuracy."); see also United States v. Miele, 989 F.2d 659, 660 (3d Cir.1993) ("Under the Guidelines regime, however, sentencing determinations are driven by the facts found at sentencing that relate to the defendant's conduct and background.... Thus, under the Guidelines, the reliability of the evidentiary basis for facts found at sentencing is clearly of the utmost importance.").

In the present case, we believe that there was sufficient evidence and testimony adduced by the government to establish that the method used to determine quantity was grounded in fact and met accepted standards of reliability, as well as the government's general preponderance burden, see Uwaeme, 975 F.2d at 1018. The government elicited specific testimony from the chemist on how the analysis was conducted, including testimony establishing that the particular samples tested were chosen at random. Moreover, the chemist's uncontradicted testimony established that the procedure he followed of testing a random sample was the accepted methodology among the chemists working in the Philadelphia Police Department.8

Additionally, other evidence in the case supports the reasonableness of inferring that the untested vials contained cocaine base. While varying somewhat in weight, the vials and their contents were all similar in appearance, each containing a chunky white powder. The context in which the vials were seized also adds considerable support to the reasonableness of the inference. McCutchen was arrested in the midst of selling a chunky white substance. When a chemist tests 12% of the seized vials, all of which contained crack cocaine, a fairly strong inference arises that the remaining vials, which were seized in the same location during the same act of distribution, also contained crack cocaine. It is also relevant that there was a significant margin for error in the present case. McCutchen's sentence was based on a finding of 5 or more grams of cocaine base. Since the total quantity estimated by the chemist was 9.866 grams, a number of the vials could have contained no cocaine base and the total would still have been well over the required 5 grams.

While in the future the district court should address more explicitly whether or not there is an adequate basis in fact for the extrapolation and whether the method of estimating the total quantity was consistent with accepted standards, the court did address the method of determining quantity in light of the supporting facts, and explicitly concluded that:

when someone is arrested and found to be in possession of a large number of identical vials containing powder and a random sample confirms that each sampled vial contains crack, it is a very reasonable inference to make ... that the other vials also contained crack.

Under the circumstances of this case, we find this sufficient. Accordingly, the judgment of the district court will be affirmed.