concurring in part and dissenting in part.
I agree that the convictions of these seven defendants should be affirmed, and I join in Part II of the court’s opinion. I dissent in part from Part III of the court’s opinion because the district court improperly enhanced the sentences of Payne, Ransom, and Alexander by basing their offense level on amounts of crack cocaine that were not charged in the indictment and were not proven at trial.
There was sufficient evidence to support the charge in the indictment that the defendants conspired to distribute powder cocaine. The indictment, however, neither specified the amounts of powder cocaine involved in each count nor alleged anywhere that the conspiracy involved crack. This practice should not be permitted because it gives the prosecution the opportunity to “indict for less serious offenses which are easy to prove and then expand them in the probation office.” United States v. Ebbole, 917 F.2d 1495, 1501 (7th Cir.1990) (quoting United States v. Miller, 910 F.2d 1321, 1332 (6th Cir.1990) (Merritt, C.J., dissenting)). See also United States v. Kikumura, 918 F.2d 1084, 1121 (3d Cir.1990) (Rosenn, J., concurring). Unfortunately, that is exactly what happened here.
The district court found that Alexander and Ransom were responsible for distributing 65 ounces or 1,843 grams of powder cocaine. The court reached this conclusion by reviewing evidence presented at trial. *294That evidence indicated that members of the conspiracy bought cocaine by transferring money from Nebraska to Florida, that Ransom and Alexander were responsible for some $156,875 of those money transfers, and that the $156,875 would buy approximately 1,843 grams of powder cocaine. There is adequate support in the record for these findings, and the district court’s conclusion that Ransom and Alexander were responsible for distributing 1,843 grams of powder cocaine is not clearly erroneous.
Had these two defendants been sentenced on the basis of the 1,843 grams of powder cocaine, I would have no difficulty affirming both their convictions and their sentences. The district court, however, also found that about approximately 191 grams of the 1,843 grams of powder cocaine had been converted into 40 grams of crack cocaine by Ransom and Alexander. These 40 grams of crack are, for sentencing purposes, the equivalent of 4,000 grams of powder cocaine. The district court added these 4,000 grams to the remaining 1,652 grams of powder cocaine for a total of 5,652 grams. The resulting base offense level was 32, six levels higher than that for the original 1,843 grams of powder cocaine. The effect of this “conversion to crack” was to more than double Ransom’s sentence and to nearly double Alexander’s:
3r Ransom
Final offense level 32/III 30/1
& guidelines range based on 1,843 g. powder cocaine (151-188 months) (97-121 months)
Final offense level 38/III 36/1
& guidelines range based on 1,652 g. powder cocaine + 40 g, crack cocaine (292-365 months) (188-235 months)
Sentence imposed 292 months 225 months
Likewise, the district court sentenced Payne on the basis of 1.97 grams of crack cocaine, which under the guidelines equals 197 grams of powder cocaine. The conversion of powder cocaine into crack cocaine at least doubled Payne’s sentence:
Payne
Final offense level 10/11
& guidelines range based on 1.97 g. powder cocaine (8-14 months)
Final offense level 16/11
& guidelines range based on 1.97 g. crack cocaine (24-30 months)
Sentence imposed 30 months
These three sentences were increased because the guidelines provide a much stiffer penalty for the distribution of crack than for powder cocaine. U.S.S.G. § 2D1.1, comment (Drug Equivalency Tables). No record has been made in this case to challenge the Sentencing Commission’s action in making this 100 to 1 equivalence, even though the standard may well impact more severely on black defendants than on white defendants. This, however, does not excuse the government’s failure to charge that these defendants were members of a conspiracy whose purpose was to distribute *295powder cocaine and crack cocaine.2 Indeed, some United States Attorneys make it a practice to charge crack cocaine, and there .is no good reason why crack cocaine could not have been charged here, as law enforcement officials were aware of the nature of the drugs involved in the conspiracy before the indictment was returned.
The government’s failure to charge crack meant that the defendants did not have notice of the charges and likely penalties against them. It is true that the ultimate sentences here were within statutory limits, but that fact means little. Under the guidelines regime, it is the guidelines range and not the statutory maximum that determines the punishment received by the vast majority of defendants. In any event, Congress set maximum sentences at a time when prisoners were eligible for parole, see United States v. Brady, 928 F.2d 844, 852 (9th Cir.1991), and prisoners served, on the average, less than half the sentence imposed. U.S. Dep’t of Justice, Federal Bureau of Prisons, Statistical Report, Fiscal Year 1986 at 12 (1987). As a result, an offender sentenced under the guidelines who receives a sentence that is below the statutory maximum may well serve far more time in prison than an individual sentenced to the statutory maximum in the preguidelines era.
The failure to charge crack also allowed the government to use the sentencing proceeding, with its lower burden of proof and relaxed evidentiary safeguards, to dramatically enhance the sentence of these defendants. I do not believe the existence of crack to be merely a “sentencing fact” when such a factual finding can mandate a doubling or tripling of a defendant’s sentence. Uncharged conduct, as in this case, can become “the tail that wags the dog” of the substantive offense, and failure to prove such conduct beyond a reasonable doubt violates the due process clause of the Constitution. Cf. McMillan v. Pennsylvania, 477 U.S. 79, 88, 106 S.Ct. 2411, 2417, 91 L.Ed.2d 67 (1986).
For this reason alone, I believe the cases of all three should be remanded for resen-fencing based on powder cocaine only. If this were done, the sentencing ranges would be 151-188 months for Alexander, 97-121 months for Ransom, and 8-14 months for Payne.
There is an additional reason why the matter should be remanded as to Payne and Ransom: no evidence at trial showed these two had participated in the distribution of crack cocaine. At sentencing the only evidence linking Ransom with the distribution of crack was a paragraph in an affidavit by Detective Michael Garnett that contained the following hearsay and double hearsay:
I spoke with Greg Rohleder who told me that he saw Ray Ransom with packages of powdered cocaine and “crack” strapped to Ransom’s body during the summer of 1987, which Ransom had carried to Nebraska form [sic] Florida; Stan Stoy of the FBI told me that Amos James, who had entered into a plea agreement with the government, told Stoy that in October or November of 1987, Ransom told James that the cocaine he shipped to Nebraska, was being made into “crack” by Ransom’s customers;
This “evidence” is just too flimsy to serve as a basis for doubling Ransom’s sentence.
Likewise, there was no evidence presented at trial to indicate that Payne ever distributed crack or knew that anyone else involved in the cocaine conspiracy was doing so. While Payne was convicted of four substantive counts that apparently involved 1.97 grams of crack, his conviction was on a Pinkerton theory. To secure a conviction, the government did not need to prove that crack cocaine was involved. It needed to show only that cocaine was involved. Thus, Payne’s conviction on the four substantive counts, without more, is inadequate proof that Payne knew that crack or anything other than powder cocaine was involved during those four offenses. To sentence Payne on the basis of crack cocaine is an unwarranted expansion of Pinkerton.
*296The majority opinion relies on United States v. Lawrence, 915 F.2d 402 (8th Cir.1990), to support the proposition that defendants may be sentenced on the basis of types of drugs not charged in the indictment. I do not agree with this proposition, but even if I did, Lawrence does not apply to this case. In Lawrence, the defendant was charged with conspiracy to distribute marijuana. At sentencing, the government presented evidence showing that the defendant personally sold cocaine as part of the same conspiracy, and the district court accordingly based Lawrence’s sentence in part upon the cocaine that Lawrence himself had sold.
In Payne’s case, however, the government presented absolutely no evidence at sentencing showing that Payne personally sold crack cocaine as part of a pattern of uncharged conduct. The government points to sentencing testimony presented by a crack cooker named Eric Beckwith, but that testimony establishes only that Payne went to Alexander’s house in August 1987 and purchased and smoked some crack there. This certainly does not establish that Payne distributed crack.
For the foregoing reasons, I dissent as to the sentences imposed on Alexander, Ransom, and Payne. I concur in all other respects.
. Although the government should have charged all defendants as to crack cocaine, the district court’s decision to sentence on the basis of crack as well as powder cocaine affected only the sentences of Payne, Alexander, and Ransom.