United States v. David Johnson, United States of America v. Bryant Legree, United States of America v. Rebecca Legree, United States of America v. Bruce Douglas, United States of America v. Kelvin Legree, United States of America v. Sandra Legree Jones

47 F.3d 1166 NOTICE: Fourth Circuit I.O.P. 36.6 states that citation of unpublished dispositions is disfavored except for establishing res judicata, estoppel, or the law of the case and requires service of copies of cited unpublished dispositions of the Fourth Circuit. UNITED STATES of America, Plaintiff-Appellee, v. David JOHNSON, Defendant-Appellant. UNITED STATES of America, Plaintiff-Appellee, v. Bryant LEGREE, Defendant-Appellant. UNITED STATES of America, Plaintiff-Appellee, v. Rebecca LEGREE, Defendant-Appellant. UNITED STATES of America, Plaintiff-Appellee, v. Bruce DOUGLAS, Defendant-Appellant. UNITED STATES of America, Plaintiff-Appellee, v. Kelvin LEGREE, Defendant-Appellant. UNITED STATES of America, Plaintiff-Appellee, v. Sandra Legree JONES, Defendant-Appellant. Nos. 93-5845, 93-5873, 93-5886, 93-5887, 93-5904, 93-5909. United States Court of Appeals, Fourth Circuit. Argued: July 15, 1994 Decided Feb. 17, 1995. ARGUED: Robert Lee Hallman, Jr., Columbia, SC; John Dewey Elliott, Columbia, SC, for Appellants. Cameron Anne Glenn, Assistant United States Attorney, Columbia, SC, for Appellee. ON BRIEF: John G. Felder, St. Matthews, SC, for Appellant Bryant Legree; Jan Strifling, Columbia, SC, for Appellant Rebecca Legree; Douglas W. Truslow, Columbia, SC, for Appellant Johnson; J. Dennis Bolt, Columbia, SC, for Appellant Jones. J. Preston Strom, Jr., United States Attorney, Robert C. Jendron, Chief of Criminal, Columbia, SC, for Appellee. Before RUSSELL, WIDENER, and HALL, Circuit Judges. OPINION PER CURIAM: — 1 — Appellants Bryant Legree, Kelvin Legree, Rebecca Legree, Sandra Legree Jones, David Johnson, and Bruce Douglas appeal various elements of their sentencing under the Federal Sentencing Guidelines for conspiracy, possession with intent to distribute, and distribution of cocaine base in violation of 21 U.S.C. Secs. 841(a)(1), 841(b)(1)(A), and 846. Individual defendants also raise several issues relating to their prosecutions. We affirm as to all issues raised on this appeal. — 2 — The facts of this case essentially are not in dispute, except insofar as defendants object to the district court's calculations of drug amounts used to determine base offense levels under U.S.S.G. Sec. 1B1.3. For convenience, since they are critical to the sentencing objections, we will discuss the facts as they relate to the Probation Office's presentence report and the district court's findings at sentencing. — 3 — In January 1991, the Orangeburg County sheriff's office began investigating the Legrees on suspicion of illegal drug trafficking. In March 1991, Bryant and Kelvin Legree and Bruce Douglas were arrested and pleaded guilty to state charges related to possession of 41.6 grams of crack cocaine. The Legrees were sentenced to three years in prison. The 41.6 grams of crack were attributed to the Legrees and Douglas as relevant conduct under U.S.S.G. Sec. 1B1.3 in this prosecution in establishing their base offense levels at sentencing. — 4 — After the arrest on the state charges, the Legrees' half-brother, Stanley Dash, went to the house of his mother, Rebecca Legree. There, he observed her carrying a full gallon-sized bag of crack cocaine.1 The Probation Office calculated the amount of crack that would fill a gallon-sized bag to be at least 1000 grams, and the district court attributed this amount to the conspiracy as relevant conduct.2 — 5 — Following the release of the Legrees from prison in April 1992, Dash observed the Legrees and Bruce Douglas microwaving powdered cocaine into crack. Dash saw three packages, each approximately the size of a shoe box, containing cocaine.3 The Probation Office estimated that a shoe box would hold 1000 grams of crack cocaine, and the district court accordingly attributed 3000 grams of crack to the Legree conspiracy as relevant conduct. — 6 — In April 1991, a DEA informant, Rickenbacker, accompanied one E. Gaffney to the house of Rebecca Legree.4 Gaffney purchased 55.8 grams of crack. Gaffney named Douglas as the person who had sold him the crack, and stated that he had seen Mrs. Legree in the house at the time of the sale. — 7 — In October 1992, one Brabham was questioned regarding his association with the Legrees. Brabham described a series of purchases of crack that he had made from the Legrees, Douglas, and Mrs. Jones beginning in April 1992. He claimed to have purchased at least a quarter-ounce of crack from the Legrees at least four times a week for a period of approximately six months. The presentence report estimated, and the district court found, that Brabham had purchased 812 grams of crack from the Legrees during this time, and attributed that amount to the conspiracy as relevant conduct. — 8 — At the instance of law enforcement, Brabham purchased crack from Kelvin Legree and Douglas in two transactions during the month of October 1992. In December 1992, at the instance of law enforcement, Brabham purchased crack from Mrs. Jones and Douglas. During this purchase, Brabham observed Mrs. Jones with a gallon-sized baggie that he described as at least two-thirds full of crack. The presentence report estimated that this amounted to at least 500 grams of crack, and the district court attributed this amount to the conspiracy as relevant conduct.5 — 9 — During this time, an informant, Dierdre Irick, related to law enforcement that she had on at least 18 occasions driven associates to the Legrees' home to purchase at least one gram of crack each time. The presentence report estimated that Miss Irick thus had knowledge of at least 18 grams of crack sold by the Legrees, and the district court attributed that amount to the conspiracy as relevant conduct. Miss Irick then made several controlled purchases of crack from members of the Legree organization from December 1992 to February 1993. On one occasion during this period, Miss Irick was informed that the Legrees were out of crack and that she might try the Johnson house around the corner. — 10 — On two occasions in February 1993, at the instance of law enforcement, Brabham went to the Johnson house and purchased crack. — 11 — In October 1992, one Preston informed police that he had purchased crack from the Legrees and Mrs. Jones for approximately four months and that he had purchased approximately 150 grams of crack from them. This amount was included by the court as relevant conduct. — 12 — On March 1, 1993, all of the defendants were arrested. Weapons were recovered from the bedrooms of Johnson, Bryant, Kelvin, and Mrs. Legree. — 13 — In May 1993, one G. Gaffney informed authorities that he had purchased at least one gram of crack cocaine from the Legrees on each of at least 15-20 occasions beginning in 1991 and continuing after the Legrees' parole in April 1992. Fifteen grams were included as relevant conduct without objection. — 14 — Prior to trial, Mrs. Rebecca Legree moved for severance to obtain the favorable testimony of her daughter, Mrs. Sandra Jones, who refused to testify at the joint trial. Rebecca wished Sandra to be tried first so Sandra could have been compelled to testify in Rebecca's case. The district court denied the motion. — 15 — Johnson moved for a directed verdict on the charge of involvement in a single conspiracy, on the ground that his association with the Legrees was insufficient. The court found sufficient relationship and denied the motion. — 16 — All of the defendants were convicted of conspiracy to possess with intent to distribute and to distribute cocaine base, as well as various other offenses. On the basis of the above facts, the presentence report calculated that at least the amount of 5557.5 grams of crack cocaine was to be considered relevant conduct in establishing the base offense levels of the defendants with regard to the conspiracy count, with the exception of Johnson whose knowledge of the conspiracy was limited. — 17 — Defendants objected to the presentence report calculations at a number of points, and a hearing was held to address these objections. During the hearing, the district court adopted the presentence report findings regarding drug amounts with minor modifications. Under U.S.S.G. Sec. 2D1.1(c)(2), at least five kilograms but less than 15 kilograms of cocaine base yields a base offense level of 40. The district court assigned all defendants save Johnson a base offense level of 40. — 18 — The district court adopted a two-point enhancement of the offense levels of Kelvin Legree, Bryant Legree, and Rebecca Legree for possession of a firearm under U.S.S.G. Sec. 2D1.1(b)(1). The district court also credited a two-point enhancement against Kelvin and Bryant Legree under U.S.S.G. Sec. 3B1.1(c) for their roles as "organizer, leader, manager, or supervisor in the drug distribution conspiracy." The district court accepted a two-point reduction in the offense levels of Mrs. Legree and Mrs. Jones for acceptance of responsibility under U.S.S.G. Sec. 3E1.1. — 19 — Kelvin Legree moved the district court for a continuance in order to prepare a proportionality review of his impending life sentence, and the district court denied the motion. — 20 — In accordance with the Sentencing Table, U.S.S.G. pt. 5A, Bryant Legree and Kelvin Legree were sentenced to life in prison. The other conspirators were sentenced to terms ranging from 190 to 300 months. I. — 21 — Appellants raise a number of issues on this appeal. Of these, we find only the issue of drug-calculation error to warrant more than brief mention. Defendants claim that the district court erred in calculating drug amounts attributable to the conspiracy as relevant conduct under U.S.S.G. Sec. 1B1.3(a)(1). We review these findings at sentencing for clear error. United States v. Adams, 988 F.2d 493, 495 (4th Cir.1993); United States v. Uwaeme, 975 F.2d 1016, 1018 (4th Cir.1992). — 22 — A convicted conspirator may be sentenced on the basis of all conduct committed in furtherance of the conspiracy and known or reason ably foreseeable to the defendant. See U.S.S.G. Sec. 1B1.3(a)(1)(A)(B). The government has the burden of proving conduct attributable to a convicted conspirator by a preponderance of the evidence. See United States v. Gilliam, 987 F.2d at 1009, 1013 (4th Cir.1993); Uwaeme, 975 F.2d at 1018. — 23 — The district court may, in calculating the amount of drugs attributable to each defendant as relevant conduct, "rely on the information presented in the presentence ... report so long as the information has 'some minimum indicium of reliability.' " United States v. Vela, 927 F.2d 197, 201 (5th Cir.) (quotation omitted), cert. denied, 60 U.S.L.W. 3263 (U.S. Oct. 7, 1991). If the district court concludes that the presentence report is sufficiently reliable, then "[t]he burden is on the defendant to show [its] inaccuracy or unreliability...." United States v. Terry, 916 F.2d 157, 162 (4th Cir.1990). "Without an affirmative showing the information is inaccurate, the court is 'free to adopt the findings of the [presentence report] without more specific inquiry or explanation.' " Terry, 916 F.2d at 162 (quotation omitted). — 24 — Defendants object to the district court's drug-calculation findings on two separate grounds. They argue first that the district court improperly credited several estimates made by the Probation Office of quantities of drugs seen by witnesses, and second that the district court refused to account for improper double counting of drugs where sales of drugs were made after witnesses reported seeing large quantities of drugs. We address these two issues in turn. A. — 25 — As to the estimation of drug amounts, this court has recognized that " '[w]here 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.' " United States v. D'Anjou, 16 F.3d 604, 614 (4th Cir.), cert. denied, 62 U.S.L.W. 3861 (U.S. June 27, 1994) (quoting U.S.S.G. Sec. 2D1.1, application note 12). Such "very rough estimates of quantity," Uwaeme, 975 F.2d at 1019, "inherently possess a degree of uncertainty," D'Anjou, 16 F.3d at 614. However, such methods "constitute ... valid bas[es] for sentencing purposes." Uwaeme, 975 F.2d at 1019. — 26 — Defendants have specifically objected to the district court's quantification of (1) 812 grams of crack purchased by Brabham following the release of the Legrees from prison;6 (2) 500 grams of crack in a two-thirds full gallon-sized baggie seen by Brabham; (3) 1000 grams of crack in a full gallon-sized bag seen by Dash;7 and (4) 3000 grams of crack in three shoe-box-sized packages of crack seen by Dash.8 After reviewing the evidence, we cannot say that the district court clearly erred in crediting these estimations. — 27 — Defendants offer evidence of disputed testimony and inconsistencies, but no evidence that the district court clearly erred in these calculations. Instead, they emphasize two points at which the Probation Office admittedly erred in calculating drug amounts, from which we are asked to infer that the Probation Office estimates are generally unreliable. First, the presentence report found that E. Gaffney had purchased 81.1 grams of crack from the Legrees, when in fact this was a gross weight not adjusted for approximately 25 grams of packaging.9 Second, the presentence report calculated the series of purchases made by Brabham as occurring over a period of seven months, when in fact it was proved at sentencing that these purchases could only have occurred over a period of six months nine days.10 Although such errors were made, this is precisely the sort of human error which the sentencing hearing is intended, in part, to disclose. It thus appears that the hearing was quite successful in this case. — 28 — Moreover, there was no evidence of bias or misrepresentation by design on the part of the Probation Office. In fact, there is ample evidence in the record that most of the Probation Office's estimations favored the defendants. For example, the Probation Office chose not to quantify what appeared to be several large quantities of crack seen by Dash at Mrs. Legree's house. Also, at almost every point where a range of quantities was described, the Probation Office chose to rely on a minimum or median number in calculating the amount of crack attributable to the defendants. — 29 — Third, and most important, defendants were given ample opportunity to demonstrate that any or all of the disputed estimates were materially inaccurate or unreliable. Defendants were unable to show any significant error beyond these two miscalculations, both of which were adequately accounted for by the district court at sentencing. We therefore decline to infer that the Probation Officer's estimates were generally unreliable. — 30 — We recognize that most of the defendants are less than 750 grams over the five-kilogram threshold, but it is clear to us that the difference of only about 750 grams is the result of favorable treatment by the Probation Office. We find no clear error in the district court's acceptance of the Probation Office estimations of these amounts of crack. B. — 31 — Defendants next contend that the district court double-counted some of the drugs possessed and distributed by various members of the conspiracy. Essentially, they argue that where a court is roughly estimating amounts of drugs possessed and distributed, that court should assume that any amounts distributed within some unspecified time after an observed possession of a large quantity of drugs should not be counted. The theory underlying this argument is that drug sales made within a short time after a quantity of drugs has been observed should be presumed to have originated in that quantity of drugs, and thus already to have been counted. They cite no direct authority for this proposition, relying on the Supreme Court's decision in Townsend v. Burke, 334 U.S. 736 (1948), for the proposition that unreliable evidence used at sentencing may violate Due Process. — 32 — We decline the invitation to assume or create a presumption that any drugs sold within some temporal proximity to possession of a quantity of the same drug originated in that quantity. Instead, we reiterate that the government has the burden of proving at sentencing by a preponderance of the evidence that certain quantities of drugs are attributable to a defendant, upon which showing the defendant must affirmatively show material inaccuracies in such quantities. We are of opinion that the mere allegation that an amount of drugs was sold within weeks or months after a larger quantity of drugs was possessed does not satisfy the defendant's burden. — 33 — Specifically, in this case, defendants argue that on three occasions, the government has failed to demonstrate that drugs sold did not originate in drug quantities earlier held in their possession. First, they claim that "[i]t is clearly inferable" that the 55.8 grams of crack sold to E. Gaffney on April 23, 1991 derived from the 1000 grams of crack seen by Dash on March 29, 1991. Next, they claim that the 812 grams of crack attributed to purchases by Brabham from April to October 1992 are likely to have originated in the 3000 grams of crack seen by Dash after the Legrees were released from prison.11 Finally, Appellants argue that the 150 grams of crack purchased by Preston over approximately four months ending in October 1992 may have been part of the same 3000 grams seen by Dash. — 34 — The government has clearly established by abundant evidence a large and active crack distribution ring, with seemingly continuous inflow and sales of drugs. The local pushers literally, not figuratively, had to "stand in line" to get the drugs. It would not even have been clearly erroneous for the district court to conclude in this context that there was a large volume of steady inflow of drugs and that each of the sales described above came from a new source or supply of crack. Defendants would have been free to disprove this conclusion by showing, for example, that their source of supply had been cut off, or by offering evidence of how much crack they customarily purchased for resale in a given period. Short of this, we find no clear error in the district court's conclusions that each of these sales of crack came from a different infusion of the drug, particularly where the sales occurred over periods of one, four, and six months after a large quantity of inventory was seen. — 35 — We recognize the defendants' plight here: a discrepancy in drug calculations could translate into years of confinement. However, we find that the district court and the Probation Office in this case exercised appropriate caution in estimating and calculating drug amounts attributable to this conspiracy, and that the defendants failed to show any material inaccuracies in such determinations beyond unsubstantiated inferences that are more than likely not well taken. We therefore find no error in the district court's determinations of drug amounts for sentencing. II. A. — 36 — Defendants allege a number of additional errors on this appeal which can be briefly disposed of. First, Kelvin Legree appeals the district court's attribution to him of a two-level sentencing enhancement for his role as supervisor, leader, organizer or manager of the conspiracy under U.S.S.G. Sec. 3B1.1(c). We review this factual determination for clear error, see, e.g., United States v. Melton, 970 F.2d 1328, 1334 (4th Cir.1992), and the government has the burden of proving his role by a preponderance of the evidence. The district court stated that Kelvin and his brother were the two primary managers of the operation and that they were the ones that got their mother and sister involved. There was evidence that Sandra advised a drug customer that she dealt with customers until her brothers (Bryant and Kelvin) got to know the customer; that on one occasion when there was a question about a customer, Kelvin directed Bryant and Bruce Douglas to stop holding up his "home girl;" that Kelvin was present on two separate occasions when Bryant was cooking cocaine powder in the microwave; and that Kelvin was actively involved in four of the controlled purchases. After reviewing the record, we are convinced that there was no error in the district court's finding that Kelvin Legree was a supervisor, leader, manager or organizer of the conspiracy. B. — 37 — Kelvin Legree next asserts error in the district court's refusal to grant him a continuance to prepare a proportionality review of his life sentence under the Eighth Amendment as established in Solem v. Helm, 463 U.S. 277, 290-91 (1983), whether or not overruled by Harmelin v. Michigan, 501 U.S. 957 (1991). We review the denial of a motion for proportionality review with "substantial deference ... to the discretion" of the district court. United States v. Rhodes, 779 F.2d 1019, 1028 (4th Cir.1985), cert. denied, 476 U.S. 1182 (1986). We think this issue is controlled both by our decision in United States v. D'Anjou, 16 F.3d 604, 613-14 (4th Cir.), cert. denied, 62 U.S.L.W. 3861 (U.S. June 27, 1994), which found a life sentence for a highly placed dealer of a large quantity of drugs proportionate in accordance with Solem, and by Harmelin, which held a sentence of life in prison without parole to be within constitutional bounds for the possession of 672 grams of cocaine, not nearly the quantity of drugs involved here. We are of opinion the district court did not abuse its discretion. C. — 38 — Mrs. Legree challenges the district court's denial of her motion to sever her trial to afford her the opportunity to present the favorable testimony of her daughter, Mrs. Jones. Mrs. Jones invoked her Fifth Amendment right not to testify at the joint trial. We review this denial for clear abuse of discretion. United States v. Rusher, 966 F.2d 868, 877-78 (4th Cir.), cert. denied, 61 U.S.L.W. 3285 (U.S. Oct. 13, 1992). It is clear to us that the district court properly applied our precedent in United States v. Shuford, 454 F.2d 772, 778 (4th Cir.1971), and Rusher, 966 F.2d at 878, which hold that it is not abuse of discretion to refuse to grant severance where the co-defendant would testify at the severed trial only if the co-defendant's trial comes first. Moreover, and in all events, because Mrs. Legree was accused of no partic ular misconduct but merely involvement in the conspiracy, because the potential testimony of Mrs. Jones would have gone to an item of particular conduct, and because the record reveals ample other evidence of Mrs. Legree's involvement in the conspiracy, we think that the district court did not abuse its discretion in refusing to sever Mrs. Legree's trial. D. — 39 — Mrs. Legree further alleges error in the district court's attribution to her of over 5000 grams of crack, claiming that that amount was not foreseeable to her. We review this factual finding for clear error. We find in the record ample evidence of Mrs. Legree's significant involvement in the conspiracy and therefore we see no error in the district court's calculations of amounts foreseeable to Mrs. Legree. E. — 40 — Mrs. Legree alleges error in the enhancement of her offense level for possession of a weapon under U.S.S.G. Sec. 2D1.1(b)(1). We review this finding for clear error and find none. Mrs. Legree was an active member of the conspiracy and her house was a significant center for drug transactions. It appears highly probable that she possessed the weapon for just those reasons, and it was not error for the district court to so conclude. F. — 41 — Finally, Johnson challenges the district court's denial of his motion for judgment of acquittal on the count of conspiracy, on the grounds of insufficient evidence that he was a member of the Legree conspiracy. Taking all the evidence in the light most favorable to the government, see Glasser v. United States, 315 U.S. 60, 80 (1942), we are convinced that a jury could rationally have attributed to Johnson sufficient involvement in the Legree conspiracy. The evidence shows at least that there was cocaine and a pistol found in a man's bedroom in Johnson's mother's house; Johnson had possession of a pistol during a sale of drugs; Johnson participated in more than one drug sale; and a drug customer was sent from the Legrees' house to the John sons'. See United States v. Giunta, 925 F.2d 758, 764 (4th Cir.1991). We therefore affirm the district court's denial of Johnson's motion for judgment of acquittal. III. — 42 — The convictions and sentences are accordingly — 43 — AFFIRMED. — 44 — HALL, Circuit Judge, concurring in part and dissenting in part: — 45 — I concur in all of the majority's opinion except with respect to Part II.F. In appeal No. 93-5845, I would reverse David Johnson's conspiracy conviction (Count I) and remand for resentencing. — 46 — The sum of the evidence tying Johnson to the Legree conspiracy1 is this: He lived with his mother, Willie Ann Johnson, in a house some blocks from Rebecca Legree's residence. On two consecutive days in February, 1993, informant Brabham made controlled buys of $100 pieces of crack at Mrs. Johnson's house from David Johnson while co-defendant Eric Benjamin was present.2 When Rebecca Legree suggested on one occasion that Deidre Irick and Kenny Brown, a regular customer of the Legrees, go to Mrs. Johnson's house to buy drugs, Brown told Legree, "I don't know these people, you have to call them and tell them I'm coming." On another occasion, Legree was unable to locate the amount of drugs requested by Irick, so she suggested that Irick go to Mrs. Johnson's house and, "if there wasn't any there, to check back with them [the Legrees] ... either the next day or that Monday." Finally, the March 1, 1993, search of Mrs. Johnson's home turned up 2.6 grams of crack and a pistol in her bedroom. A pistol was also found in David's bedroom. The government points to nothing else in the record that would link the Johnsons and Benjamin to the Legrees. — 47 — This evidence shows only that Rebecca Legree was aware that her neighbors sold drugs and, over the two-year life of a conspiracy that was found to have sold more than 5,500 grams of crack cocaine, she referred customers to the Johnsons on two occasions.3 Based on what little evidence there is linking Johnson to the Legrees, I do not believe that any rational trier of fact could have found beyond a reasonable doubt that he (or Mrs. Johnson or Benjamin, for that matter) had entered into an agreement with any of the undisputed members of the "Legree conspiracy"--Rebecca Legree, her three children, and Bruce Douglas--to act in concert to distribute drugs. See United States v. Giunta, 925 F.2d 758, 765 (4th Cir.1991) ("[T]he danger of guilt being found on the basis of speculation from mere association between criminally disposed people, related criminal behavior, and like considerations is acute"). The most that can be said from this record is that the two groups operated in the same market. Cf. United States v. Smith, 13 F.3d 860, 869 (5th Cir.1994) ("[B]y definition, competitors are not engaged in a jointly undertaken activity") (DeMoss, J., dissenting in part). David Johnson's motion for acquittal on the conspiracy count should have been granted. Footnotes [1] This evidence was disputed at length, including whether or not Dash had actually seen a gallon-sized bag, whether or not he had claimed it was full, and whether or not a full gallon-sized bag would contain 1000 grams. The court concluded that in any event the presentence report calculation on this point was conservative and thus adopted it despite the statements of Dash [2] On another occasion following the 1991 arrest of the Legrees and Douglas, Dash saw what he described as an ounce of crack (approximately 28 grams) at the house of his mother, Rebecca Legree. The Probation Office did not include this amount in its calculation of the base offense level for sentencing. Dash also related, on a different occasion, seeing three Pyrex dishes containing white powder and baking soda which was microwaved. None of this amount was estimated or included in the relevant conduct of the conspiracy for sentencing purposes [3] This testimony was disputed by the parties. See infra notes 7 and 8 [4] Rickenbacker was wearing an electronic monitoring device through which investigating officers heard Gaffney describe his purchases of crack cocaine from the Legrees in recent months. No amounts were attributed to the conspiracy as the result of this conversation [5] There was a taped debriefing which indicated that the bag was three-fourths full. The probation office, however, attributed only two-thirds of a bag to defendants [6] In estimating these purchases, the Probation Office relied on a period of seven months of purchases, which the district court found to be six months and nine days. In addition, the report attributed $350 worth of crack per purchase whereas Brabham had described purchases ranging from $200-$400 each. In addition, Brabham had stated that he had purchased crack 4-6 times a week during this period, and the Probation Office, to be conservative, calculated the purchases based on a frequency of four times per week, the range minimum. Further, the government presented evidence that Brabham had related purchasing at least 1/4 ounce each time, which apparently costs approximately $350 rather than $300, the range median. In any event, we cannot say that the district court's acceptance of this figure was clearly erroneous in the light of all the evidence regarding this series of purchases [7] We note the obvious discrepancy between an estimated 500 grams of crack in two-thirds of a gallon baggie and an estimated 1000 grams of crack in a full gallon bag. But we also note that there was evidence that both estimates were very conservative. A witness at the sentencing hearing, a United States Probation Officer named Avenger, testified that in his estimation at least 1800 to 2000 grams of crack cocaine would fit in a gallon-sized bag. Another witness, Detective Heaton, testified that in his opinion, at least 2000 grams of cocaine hydrochloride would fit in a gallon-sized baggie. He also demonstrated this fact by showing the district court a gallon-sized bag in which were 1000 grams of cocaine, which he said would "easily hold another kilogram." Detective Heaton further testified that the 1000 grams of crack was "a very good representation" of 1000 grams of powdered sugar used by the Probation Office to estimate the mass of crack. Although the density of crack or cocaine hydrochloride is not a measurement presently available to us, the density of sugar is approximately 1.5 times that of water, and a gallon of water weighs approximately 3764 grams (because one liter of water weighs one kilogram, and there are 1.057 quarts in a liter). Thus, a gallon of sugar would weigh approximately 5646 grams Although none of this is precise due to the unavailability of a measurement for the density of crack or cocaine hydrochloride, we are certain that the weight estimations used by the Probation Office and by the district court for a full gallon-sized bag and two-thirds of a gallon-sized bag are quite underestimated. Moreover, while we do not know the size of the shoe boxes referred to in the record, see infra note 8, it seems evident that they would not hold less than one-half gallon, and thus that more than 2000 grams would fit into each of them. Accordingly, those estimates, as well, were slanted favorably toward the defendants. In view of the above calculations, we are convinced that any errors in the calculations of the weights of drug amounts involved in this case quite heavily favored the defendants. [8] Defendants argue on this appeal only that the district court erred in misunderstanding Dash's testimony. They contend that Dash testified only to two shoe-box sized packages of crack, whereas the district court found that he testified to three. At the very least, after reviewing the record, we believe that reasonable minds could differ as to the conclusions derived from Dash's testimony, and we will not attribute clear error to the district court's finding on this point [9] The district court, however, noted this error and accordingly adjusted the presentence-report estimate downward to reflect it [10] The court noted this error, but determined that the estimation by the Probation Office was so conservative that the error was effectively harmless, and credited the report without modification [11] Defendants did not raise this specific issue at sentencing, although it was properly objected to by way of addendum to the presentence report, and thus the district court directly considered it at sentencing only in one instance. However, because this objection was presented as an assertion that the 812 grams should be presumed to have originated in the 3000 grams seen by Dash, we will consider the merits of the contention [1-1] 1 There is no question but that the theory of the prosecution was that this case is about the "Legree conspiracy" and that at most, the Johnsons and Benjamin were bit players [2-1] 2 Johnson does not contest his convictions on the substantive counts arising from these controlled buys [3-1] 3 David was sentenced on the basis of only 6.1 grams of crack, this being the total of (1) the amount found during the search of his mother's house and (2) the amounts sold to informant Brabham in February, 1993