concurring in part and dissenting in part.
I join in the majority’s opinion except for its discussion of the canine sniffing evidence and the portions of Part III and Part IV pertaining to the sufficiency of the evidence supporting Carr’s conviction on Count 21. I write separately not only to explain why I believe the majority has construed 18 U.S.C.A. § 1956(a)(2)(B)(i) too broadly, but also to express my views on the general inadmissibility of canine sniffing evidence.
*1213I. CONSTRUCTION OF 18 U.S.C.A. § 1956(A)(2)(B)©
A. The Plain Meaning of the Statute
Section 1956(a)(2)(B)(i) makes it a crime to transport money out of the United States “knowing that the ... funds ... represent the proceeds of some form of unlawful activity and knounng that such transportation ... is designed ... (i) to conceal or disguise the nature, the location, the source, the ownership, or the control of the proceeds of specified unlawful activity_” 18 U.S.C.A. § 1956(a)(2)(B)® (emphasis supplied). Unfortunately, the language does not clearly convey whether the mens rea requirement of “knowing” applies solely to the fact that the transportation was “designed” to “conceal or disguise” attributes of what really is proceeds of specified unlawful activity — rendering the “proceeds of specified unlawful activity” language a mere passing referent to which no mens rea requirement attaches — or whether the scienter requirement extends also to the fact that the funds represented “the proceeds of specified unlawful activity.” Although the drafting of the statute is opaque, and the question of the mens rea requirement a close one, I respectfully disagree with the majority’s construction, see Maj. Op. at 1206.
After careful meditation on the structure and wording of § 1956(a)(2)(B)®, I am led to conclude in the end that the mens rea of “knowing” also applies to the fact that the funds represent the proceeds of “specified unlawful activity” and that the majority’s reading cabining the mens rea requirement to the “conceal or disguise” element is less persuasive. I find guidance in the Model Penal Code’s learned approach toward culpability. The Model Penal Code’s rule of statutory construction provides that “[w]hen the law defining an offense prescribes the kind of culpability that is sufficient for the commission of an offense, without distinguishing among the material elements thereof, such provision shall apply to all the material elements of the offenses, unless a contrary purpose plainly appears.” Model Penal Code § 2.02(4) (1965). .“Knowing” applies to all material elements of subsection (i) under that rule, and because it identifies the essence of the offense the “proceeds of specified unlawful activity” (as defined by § 1956(c)(7)) element is a material term.1
This construction does not read the phrase “some form of unlawful activity” in § 1956(a)(2)(B) out of the statute: that broader language applies to § 1956(c)(2)(B)(ii), which does not contain the language “proceeds' of specified unlawful activity.”2 Had Congress meant to have the scienter requirement in § 1956(a)(2)(B)® apply only to proceeds of “some form of unlawful activity” and not to “proceeds of specified unlawful activity,” it could (and should) have used the language “some form of unlawful activity” in § 1956(a)(2)(B)® instead of the language “proceeds of specified unlawful activity,” or otherwise conveyed that meaning.
For example, in § 1956(a)(1), Congress partially drafted the statute the way the majority • interprets § 1956(a)(2) to read. That section begins, “Whoever, knowing that the property involved in a financial transaction represents the proceeds of some form of unlawful activity, conducts or attempts to conduct such a financial transaction which in fact involves the proceeds of specified unlawful activity....” 18 U.S.C.A. § 1956(a)(1) (Supp.1994) (emphasis supplied). The majority reads § 1956(a)(2)(B)® as if it similarly read “knowing that such transportation ... is designed ... to conceal or disguise the nature, the location, the source, the owner*1214ship, or the control of what in fact represents proceeds of specified unlawful activity,” whereas the statute as written leaves out the emphasized words.
The case the majority relies upon to support its construction, United States v. Massac, 867 F.2d 174, 177-78 (3d Cir.1989), did not decide the issue the majority cites it for. In that case there was no question that the defendant knew the proceeds derived from drug trafficking; the only question we confronted was whether the transaction was designed to conceal or disguise that fact.
B. Application of the Rule of Lenity
Even if ultimately I am wrong about the difficult question concerning the mens rea standard for the element “proceeds of specified unlawful activity” in § 1956(a)(2)(B)(i), it seems to me that at the very least the statute’s mens rea requirement is ambiguous. Therefore the rule of lenity should come into play and require the same outcome I reach above.
The Supreme Court has held that the rule of lenity should be employed to “resolv[e] any ambiguity in the ambit of [a criminal] statute’s coverage.” Crandon v. United States, 494 U.S. 152, 158, 110 S.Ct. 997, 1001, 108 L.Ed.2d 132 (1990) (emphasis supplied).3 An ambiguity exists for purposes of the rule in “those situations in which a reasonable doubt persists about a statute’s intended seope even after resort to ‘the language and structure, legislative history, and motivating policies’ of the statute.” Moskal v. United States, 498 U.S. 103, 108, 111 S.Ct. 461, 465, 112 L.Ed.2d 449 (1990) (quoting Bifulco v. United States, 447 U.S. 381, 387, 100 S.Ct. 2247, 2252, 65 L.Ed.2d 205 (1980)); see Crandon, 494 U.S. at 158, 110 S.Ct. at 1001-02.
Since there is no legislative history, and since neither the language, structure, nor the animating purposes of the statute provide sure direction, see supra Part I.A, the statute is at best ambiguous. Therefore the rule of lenity mandates that the heightened scienter requirement, rather than no scienter requirement (as the majority would have it), applies to that element of the offense.
II. The Relevanoe of Canine-Alert EvIDENCE
Although I believe that the introduction of the dog-sniffing evidence was error, it does not rise to the level of plain error, inasmuch as I also believe there was sufficient evidence to prove beyond a reasonable doubt that the cash involved represented the proceeds of drug trafficking even absent the canine-sniffing evidence, and hence that the evidence was not prejudicial.4 Moreover, the question whether the currency in fact represented the proceeds of drug trafficking is not relevant to the question of whether Carr subjectively believed he was transporting currency de*1215rived from drug trafficking in order to launder it. Nevertheless, I feel obliged to comment on the nature of the evidence in light of what I believe to be its highly prejudicial potential in future proceedings.5
My concern stems from numerous studies and other evidence cited in United States v. Fifty-Three Thousand Eighty-Two Dollars in United States Currency, 985 F.2d 245, 250 n. 5 (6th Cir.1993), United States v. Six Hundred Thirty-Nine Thousand Five Hundred & Fifty-Eight Dollars ($639,558) in United States Currency, 955 F.2d 712, 714 n. 2 (D.C.Cir.1992), and elsewhere, which collectively persuade me that a substantial portion of United States currency now in circulation is tainted with sufficient traces of controlled substances to cause a trained canine to alert to their presence. Although the mounting evidence and studies were not made part of the record at trial and thus are not directly before us on review, I set them forth in the margin.6
*1216If any of the many studies cited above is valid, then the fact that a dog alerted to a large number of bills in United States currency which has circulated in a major metropolitan center (at which the studies are directed) is meaningless and likely quite unfairly prejudicial, see Fed.R.Evid. 403, and evidence thereof should have been excluded. Although having been directed to many of the studies I have cited by Carr’s brief, the government in its brief has not disputed the validity of any of the studies mentioned above. It has not pointed to any countervailing studies — whether of record or otherwise — -which contradict them, or mustered any arguments pointing to the relevancy of its canine-alert evidence, despite shouldering the burden of establishing its relevancy.
It is thus my considered opinion that the fact that numerous studies by governmental and private agencies, studies which stand unrefuted, strongly suggest that a trained canine will alert to all bundles of used currency does not permit the jury to draw a reasonable inference that the person in prior possession of such currency was a drug trafficker or associated with one.7 Indeed, I am *1217inclined to the view that the information now available establishes a strong presumption against the admissibility of evidence of a canine’s alert to currency, and that the government can rebut that presumption only if it first clearly and convincingly establishes, outside the presence of the jury, the relevance and non-prejudicial character of the offered evidence. Since the government has failed its burden of. showing the relevance of its evidence, in order to protect the “public reputation of judicial proceedings,” United States v. Young, 470 U.S. 1, 15-16, 105 S.Ct. 1038, 1046-47, 84 L.Ed.2d 1 (1985), I would exclude that prejudicial and irrelevant evidence from our consideration of this case.8
However, I think that the other evidence and the nature of Carr’s conviction renders the admission of that evidence harmless er*1218ror in this case. In particular, the canine-sniffing evidence appears to have been relevant only to the issue of whether the currency Gonzalez exchanged for the government’s fresh $100 bills in fact represented the proceeds of drug trafficking.9 I believe that the other evidence proffered by the government, including the fact that the money originated from Gonzalez, who was convicted of drug trafficking, sufficed to establish that the large amount of cash in fact represented the proceeds of drug trafficking.10 In sum, although I think; admission of the canine alert evidence was not plain error in this case under Federal Rule of Criminal Procedure 52(b), I. think for purposes of appellants’ sufficiency of the evidence challenge we should only give the evidence the weight due it — absolutely none.
III. GaRR’s Guilt UndeR the PROPER Reading op § 1956(a)(2)(B)(i)
Under my reading of § 1956(a)(2)(B)(i), I think that even after drawing all reasonable inferences in favor of the government, the jury could not properly have found Carr guilty beyond a reasonable doubt on the charges contained in Count 21 of the indictment. I base my conclusion on the relevant facts that are set forth in the majority’s opinion at pages 1198-1200, 1204-07 (that is, all the facts except the ones pertaining to canine alerts), which I will briefly repeat here.
There is evidence in the record of Carr’s repeated other short trips to the Cayman Islands and Colombia, trips which he took in close proximity to the government’s exchange of $100 bills for cash in smaller denominations and only after making a stop at the travel agency, and of the passport incident occurring on May 2, 1990. From this amalgamation of evidence, the jury could legitimately have inferred that on several other occasions Carr carried cash from the United States to a foreign country, including Colombia. The evidence also established that Carr was knowingly transporting a large amount of cash ($186,000), that Carr lied to the customs agent about the amount and source of the cash he was transporting, and that he phoned Mrs. Gonzalez after the government confiscated the money to report the loss. These facts allowed the jury to conclude, as it did, that Carr illegally, in. violation of the currency reporting laws, knowingly transported cash out of the country.
But I think that all these facts in combination fall shy of establishing that Carr was knowingly a money launderer, as there is no evidence indicating that Carr knew the cash represented the proceeds of drug trafficking.11 Although Carr obtained the cash from *1219(subsequently convicted) drug trafficker Gonzalez, there is no evidence Carr was anything but a business acquaintance of Gonzalez’s12 or that he knew of Gonzalez’s illicit activities. Specifically, the government introduced no evidence that Carr was aware that the money represented the proceeds of drug trafficking, as opposed to, for example, money illegally skimmed from Gonzalez’s two legitimate businesses, or money that Gonzalez was attempting to smuggle to Colombia on behalf of his Colombian friends who were avoiding United States or Colombian taxes or on behalf of illegal immigrants sending their wages to their families in Colombia. None of the probative circumstantial evidence the majority cites — the frequent trips, the passport incident, the phone conversations — alone or in combination has the “logical and convincing connection,” United States v. Casper, 956 F.2d 416, 422 (3d Cir.1992), to drug trafficking needed to elevate the drug trafficking theory of what Carr knew over any other theory.
The majority today allows a jury to speculate on the mere basis of a short trip to Colombia with a large amount of cash not only that the carrier is a money launderer of drug trafficking proceeds, but that he knows he is. Especially in a criminal case, however, it is improper to speculate about what the defendant knew. The Constitution requires the government to prove each element of the offense beyond a reasonable doubt, not merely to a reasonable suspicion. In short, the evidence did not establish beyond a reasonable doubt that Carr knew the funds represented proceeds from drug trafficking.
The district judge did give the jury a charge on willful blindness. See United States v. Caminos, 770 F.2d 361, 365 (3d Cir.1985) (holding that willful blindness occurs when “the defendant himself was subjectively aware of the high probability of the fact in question”). But Caminos, in which we concluded that the willingness of two individuals to pay over $1,000 to have the defendant deliver a $60 Brazilian wood carving was sufficient to sustain a finding of defendant’s willful blindness to the fact that there was something illicit hidden in the carving, is distinguishable. There the charge on willful blindness was quite broad, applying to “willful blindness to the existence of facts which indicated that there is a high probability that some forbidden or illegal substance may be contained therein,” id. at 366 (emphasis supplied), whereas here I believe the government had to prove that Carr knew the money represented the proceeds of “specified unlawful activity,” see supra Part I. Carr could easily have believed merely that he was transporting the cash in violation of the currency reporting laws instead of laundering it, whereas (given the broad wording of the willful blindness charge) no such alternative explanation existed in Caminos.13 Thus it has not been established beyond a reasonable doubt that Carr was subjectively aware of a high probability that the money he was transporting represented the proceeds of drug trafficking.
For the foregoing reasons, I respectfully dissent from the judgment insofar as it affirms Carr’s conviction on Count 21.
. I do not mean to suggest that the government is obliged to prove knowledge of unlawfulness or that the government must prove that the defendant knew that the statute defines as "specified unlawful activities” what the defendant knew the proceeds derived from. I only mean that the government must prove that the defendant knew that the money in fact (but not in law) represented the proceeds of one of the great many forms of "specified unlawful activities” enumerated in • § 1956(c)(7).
. That subsection makes it a crime to transport funds in to or out of the United States "(B) knowing that the ... funds involved in the transportation ... represent the proceeds of some form of unlawful activity and knowing that such transportation is designed in whole or in part ... (ii) to avoid a transaction reporting requirement under State or Federal law”. See 18 U.S.C.A. § 1956(a)(2)(B)(ii) (Supp.1994).
. This Court has recurrently echoed the same view. See Government of V.I. v. Knight, 989 F.2d 619, 633 (3d Cir.1993) ("The rale of lenity requires that any ambiguity concerning the meaning of a criminal statute be resolved in favor of the criminal defendant.” (emphasis supplied)), cert. denied, - U.S. -, 114 S.Ct. 556, 126 L.Ed.2d 457 (1993); United States v. Tabaka, 982 F.2d 100, 103 (3d Cir.1992) (same); United States v. Knox, 977 F.2d 815, 819 (3d Cir.1992) (same), vacated on other grounds, - U.S. -, 114 S.Ct. 375, 126 L.Ed.2d 325 (1993); United States v. Mobley, 956 F.2d 450, 452 (3d Cir.1992) (same); cf. 3 Norman J. Singer, Statutes and Statutory Construction § 59.04, at 118 (5th ed. 1992 rev.) ("The words of a criminal statute must leave no reasonable doubt as to its meaning or the intention of the legislature, and where such doubt exists the liberty of the citizen is favored." (footnotes omitted)).
. See Fed.R.Crim.P. 52(b); e.g., United States v. Young, 470 U.S. 1, 15-16, 105 S.Ct. 1038, 1046-47, 84 L.Ed.2d 1 (1985) (stating that plain error applies only to "particularly egregious errors” which "seriously affect the fairness, integrity or public reputation of judicial proceedings,” and that plain error must be evaluated "by viewing such a claim against the entire record” (internal quotations omitted)); Government of V.I. v. Parrilla, 7 F.3d 1097, 1100 (3d Cir.1993) (‘“Plain error’ analysis requires a case-by-cáse determination that includes examining factors such as 'the obviousness of the error, the significance of the interest protected by the rule that was violated, the seriousness of the error in the particular case, and the reputation of judicial proceedings if the error stands uncorrected — all with an eye toward avoiding manifest injustice.’ ” (quoting United States v. Thame, 846 F.2d 200, 205 (3d Cir.1988))); Government of V.I. v. Smith, 949 F.2d 677, 681 (3d Cir.1991) (“Plain errors are those that undermine the fundamental fairness of the trial and contribute to a miscarriage of justice,” and courts are to find them "sparingly.” (internal quotation omitted)).
. The prejudicial nature of canine-alert evidence can manifest itself in various proceedings in addition to trials on the merits — for example, at suppression hearings investigating the existence vel non of probable cause to search'based on a canine’s reaction to currency or in grand jury proceedings.
. The Sixth Circuit wrote:
[TJhe evidentiary value of a narcotics detection dog’s alert has recently been called into question. As noted by one district court, "[tjhere is some indication that residue from narcotics contaminates as much as 96% of the currency currently in circulation.” United States v. $80,760.00 in United States Currency, 781 F.Supp. 462, 475 & n. 32 (N.D.Tex.1991) [aff'd, 978 F.2d 709 (5th Cir.1992)]. See also Use of Drug-Sniffing Dogs Challenged: ACLU Backs Complaint by Men Whose Pocket Cash Was Seized, Wash. Post, May 6, 1990, at DI ("I would not want to walk into court and rely exclusively on a dog sniff for a forfeiture of money,” said Charles S. Saphos, Chief of the U.S. Justice Department's Narcotic and Dangerous Drugs Section. "There are a lot of guys out there that have shown that there is a trace [of] dope on a lot of money out there. And for that reason alone, I’d want more than just the dog.”); Dirty Money, United States Banker, October 1989, at 10 (discussing study by Lee Hearn, Chief Toxicologist for Florida's Dade County Medical Examiner's Office that 97% of bills from around the country tested positive for cocaine; noting also that banks play a role in spreading the cocaine traces when tellers count and recount money, rubbing one bill against another). Thus, a court should "seriously question[] the value of a dog's alert without other persuasive evidence_” $80,-760.00, 781 F.Supp. at 476 (and cases cited therein).
985 F.2d at 250 n. 5. The District of Columbia Circuit surveyed similar evidence:
In order to blunt the implications of [drugs being found on the money, the defendant] called an expert, Dr. James Woodford, who testified that 90 percent of all cash in the United States contains sufficient quantities of cocaine to alert a trained dog. Officer Beard, the dog handler, suggested on the basis of hearsay that the number was lower, near 70 percent. (There is at least one study indicating that up to 97 percent of all bills in circulation in the country are contaminated by cocaine, with an average of 7.3 micrograms of cocaine per bill. Crime and Chemical Analysis, 243 Science 1554, 1555 (1989).)
... It has been estimated that one out of every three circulating bills has been involved in a cocaine transaction. R. Siegel, Intoxication 293 (1989). Cocaine attaches — in a variety of ways — to the bills, which in turn contaminate others as they pass through cash registers, cash drawers, and counting machines at banks and commercial establishments, id.; Crime and Chemical Analysis, supra, at 1555. Dr. Woodford testified that, as a result, bills may contain as little as a millionth of a gram of cocaine, but that is many times more cocaine than is needed for a dog to alert.... See generally Taslitz, Does the Cold Nose Know? The Unscientific Myth of the Dog Scent Lineup, 42 Hastings L.J. 15, 29 & n. 71 (1990).
955 F.2d at 714 n. 2 (some citations omitted). Many other sources have, apparently without exception, reached the same conclusion. See United States v. Mendez, 827 F.Supp. 1280, 1283 (S.D.Tex.1993) ("Dogs alert to the odor from trace amounts in bags and money of unsuspecting, innocent persons from drugs long gone.”); Jones v. United States Drug Enforcement Admin., 819 F.Supp. 698, 719-21 (M.D.Tenn.1993) (recognizing that "a growing chorus of courts [is] finding[ ] the evidence of the narcotic-trained dog's 'alert' to the currency is of extremely little probative weight” and concluding that “the continued reliance of courts and law enforcement officers on dog sniffs to separate 'legitimate' currency from 'drug-connected' currency is logically indefensible”) (citing a 1987 memorandum by a DEA scientist, admitted into evidence, revealing that “one-third of the bills in a randomly selected sample were contaminated with ... cocaine”); United States v. $87,375 in United States Currency, 727 F.Supp. 155, 160 (D.N.J.1989) (stating that the defendant’s expert witness concluded "[a]fter analyzing random samples of currency in varying denominations from banks throughout the Northeast ... that all the bills contained cocaine residue"); Mark Curriden, Courts Reject Drug-Tainted Evidence, 79 A.B.A.J. 22 (Aug. 1993) (quoting forensic chemist Dr. James Wood-*1216ford of Atlanta as stating that "[t]he probability that every single person in the United States is carrying drug-tainted money is almost certain" and reporting that a 1987 study by a Drug Enforcement Agency scientist found that one-third of all money at the federal Reserve Building in Chicago was tainted with cocaine); John Dillin, Law Would Reign in Agents from Seizing Money, Christian Sci. Monitor, June 17, 1993, at 2 (reporting that Rep. Hyde stated that "97 percent of all currency now carries at least a trace of narcotics”); Howard R. Reeves, Dusty Money, Orlando Sentinel Trib., July 1, 1992, at A14 ("According to a May 1992 newsletter circulated by the Hartz Group Inc. and prepared by the International Health Awareness Center of Kalamazoo, Mich., 97 percent of all paper money in the United States contains traces of cocaine."); Jeff Brazil & Steve Beriy, You May Be Drug Free, But Is Your Money?, Orlando Sentinel Trib., June 15, 1992, at A6 (reporting that all used bills taken from prominent communily leaders — -including a circuit judge, the police chief, a state senator, a mayor, a college president, and others — tested positive for cocaine, and that toxicologist Wayne Morris, a former crime-laboratory specialist for the Florida Department of Law Enforcement, has testified in hundreds of criminal cases that as much as 90 percent of currency in some cities tests positive for cocaine); Virtually All U.S. Paper Money Is Contaminated with Cocaine, Albany Times Union, Aug. 12, 1991, at A-8 (discussing studies by two private institutions indicating over 80% of U.S. currency is tainted with cocaine); Nation's Money Supply Dusted with Cocaine, United Press Int’l, Dec. 13, 1989 (available in LEXIS' News Library) (reporting that tests by Toxicology Testing Service Inc. of Miami indicated 10 of 11 bills nationwide and all 24 randomly selected bills from Orange County, Cal. tested positive for cocaine, and that Steven Fike of the U.S. Customs Laboratory in San Francisco stated that 70% of the bills he tests contain detectable traces of cocaine residue); cf. In re Daniels, 478 N.W.2d 622, 623 (Ia.1991) (holding that a "dog's detection of the odor of a controlled substance [on currency] did not provide substantial evidence connecting the money with drug dealing”); Commonwealth v. Giffin, 407 Pa.Super. 15, 25, 595 A.2d 101, 106 (1991) (holding that a dog's alert to seized funds did not establish that the currency was "more likely than not derived from or used to facilitate a violation of the Controlled Substances Act" by the owner).
. The majority opines, without any support whatsoever, that "positive alerts by trained drug sniffing dogs indicate that much of the money likely was used in drug transactions.” Maj. Op. at 1206. It may be true, although given the aforementioned studies which detail how traces of drugs may be transferred between bills it is doubtful, that "much of the money likely was used in drug transactions.” But what the evidence obviously fails to describe is who was involved in those drug transactions and who knew about them. If the person standing before me in the grocexy line pays for his or her goods with cash earned in a drug sale, and I receive that cash as change, then I am clearly in possession of "money ... used in drug transactions,” but I have not thereby become a drug trafficker (or a money launderer for that matter). Drug-tainted money is passed around as quickly and as effortlessly as money not tainted with illegal drugs, and given the vast amounts of cash apparently consumed by the black market in drugs it is easily conceivable that, as the studies indicate, between 70%-97% of all used bills come tainted with traces of illegal drugs.
It would seem that the odds that out of the $180,000 in cash exchanged for the government's $100 bills, assuming it was composed of an average of $10 bills (i.e., assuming about 18,000 bills), none was ever (i.e., since being printed) exposed to drugs (whether directly or by virtue of contact with tainted money or sorting equipment) are exceedingly low. Even if I discount as grossly excessive the statistics cited in the studies, which indicate that between 70%-97% of U.S. currency is contaminated with sufficient traces of drugs for a trained canine to alert to, and assume that only 1% of currency meets that test, the odds that none out of a random collection of 18,000 bills in U.S. currency contains traces of illicit drugs are so small (less than 1 in 1078 — by comparison, there are only about 1057 atoms in the sun) as to be equivalent to zero.
*1217The government's dog handlers testified in unison at trial that they could not tell how many in a bag of bills were tainted, that dogs would alert if only one out of a thousand bills was tainted, and that there was no way of knowing when a particular bill became tainted. Eichmann at 95-97; Hamlers at 118-19, 121; Kinsky at 97-98. Against this background, the grossly prejudicial potential of canine-alert evidence is readily apparent. I also note that the government witnesses' concession that a dog could alert to one tainted bill in a bundle of one thousand untainted bills completely discredits the majority's speculation that the canines' positive alerts indicated that "much of the money likely was used in drug transactions.” Rather, it would appear from the government's own witnesses that even if just a few of the thousands of bills sniffed were "used in drug transactions” that the dogs would have alerted to the whole bundle.
The facts would have been different if the government had randomly selected out several hundred of the bills and had subjected each one individually to a canine. This could establish within a degree of probabilistic certainty what percentage of the bills in the whole bundle contained traces of drugs. If as a statistical matter that percentage departed significantly from the percentage of used bills in general circulation to which the canines respond, the evidence might bear some relevancy and with proper instructions might not be prejudicial. Alternatively, the government could have washed the bills and obtained exact measurements of the amount of drug traces the bills contained, and compared this quantity statistically to that found to taint currency in general circulation. But as the test was conducted, the government’s proof is fatally and prejudicially flawed.
. The majority acknowledges the studies, but declines to consider them as evidence because they do not satisfy the standard set by Federal Rule of Evidence 201 for judicial notice of adjudicative facts. See Maj. Op. at 1202-1203 n. 3. I see the studies from a different perspective: I am inclined to the view that they are adequate for legislative factfinding, which is not governed by Rule 201. That is, I do not see them as evidence of whether the canine alert evidence was relevant in this case, but as evidence of whether evidence of a canine's alert to currency is generally admissible or probative. For this purpose, the studies make sufficiently convincing reading for me to require the government to come forth with affirmative evidence of relevance in every case. See In re Asbestos Litig., 829 F.2d 1233, 1247 (3d Cir.1987) (Becker, J., concurring) ("As Justice Holmes recognized long ago, 'the court may ascertain as it sees fit any fact that is merely a ground for laying down a rule of law.' To forbid such recognition would force courts to fashion laws without reference to reality." (citation omitted)), cert. denied, 485 U.S. 1029, 108 S.Ct. 1586, 99 L.Ed.2d 901 (1988); Bulova Watch Co. v. K. Hattori & Co., 508 F.Supp. 1322, 1328 (E.D.N.Y.1981) (Weinstein, C.J.) ("A court’s power to resort to less well known and accepted sources of data to fill in the gaps of its knowledge for legislative and general evidential hypothesis purposes must be accepted because it is essential to the judicial process.”); Fed.R.Evid. 201 — advisory committee note. See generally Michael J. Saks, Judicial Attention to the Way the World Works, 75 IowaL. Rev. 1011, 1017-18 (1990); Robert E. Keeton, Legislative Facts and Similar Things: Deciding Disputed Premise Facts, 73 Minn.L.Rev. 1, 8-10, 14-28 (1988); Ann Woll-handler, Rethinking the Judicial Reception of Legislative Facts, 41 Vand.L.Rev. Ill, 112-14 (1988); Kenneth C. Davis, An Approach to Problems of Evidence in the Administrative Process, 55 Harv. L.Rev. 364, 402-10 (1942).
Although I am driven to conclude that there is no reason to assign any weight to that evidence, the majority apparently labors under the false assumption, for which there is also no support in the record, that only currency which has been used by the immediately preceding possessor in a drug transaction is likely to contain traces of drugs. In other words, the majority is sub silen-to engaged in similar legislative factfinding, although it has nothing to support its facts other than unfounded and baseless assumptions. In particular, the majority impliedly assumes that only a vety small portion of currency in circulation is tainted with drugs — since even if only 1 in 10,000 bills in general circulation contains enough drug traces for a trained canine to alert to, the probability that in a random collection of 18,000 bills there would be at least one such bill is approximately 84%. Beyond that, the majority also assumes that only drug traffickers and their co-conspirators are in possession of such tainted money. As far as I can tell, this assumption is based on a belief, unsupported by rhyme or reason, in the chemical purity of the money supply.
. The majority, without providing any support for or explanation of its reasoning, seems to conclude that the mere fact that the money in fact represented the proceeds of drug trafficking — a conclusion it draws, I believe quite erroneously, from the canine-sniffing evidence — is relevant evidence to prove that Carr knew that it did. Maj. Op. at 1205-1206. Even if I could sign on to the majority's assumption about the probativeness of canine-sniffing evidence, the logical connection between this circumstantial evidence and Carr's knowledge escapes me. The majority cannot be suggesting that because the dogs smelled the drugs on the money so Carr could have too, as Carr was not present when Gonzalez exchanged his tainted cash with the government informant’s cash, and the cash Carr actually carried had no drug traces on it. The fact that some money which Carr never had contact with was once used in drug trafficking, a fact which only a canine could detect (assuming for the moment the evidence of a canine alert is probative on that point), has no bearing on what CaiT believed about the origin of the (different) money which he later was caught transporting. The majority's reasoning is tantamount to using the fact that a radio was stolen circumstantially to prove that the defendant knew he or she was purchasing stolen goods.
. I agree with the majority that there was ample evidence of drug trafficking adduced with respect to Cardona. See Maj. Op. at 1201-03.
. The majority posits that the jury may have convicted Carr of money laundering on the alternative ground that Carr believed that the money represented the proceeds of illegal money laundering. See Maj. Op. at 1205. The problem with this theory is that the government did not charge it in the indictment, did not adduce any evidence of it, did not argue it to the jury, and did not raise it as a basis for affirmance on appeal. Cf. United States v. Campbell, 977 F.2d 854, 857 (4th Cir.1992) ("All of the Government's evidence was designed to show that [the defendant] knew that Lawing was a drug dealer. There is no indication that the jury could have believed that Law-ing was involved in some form of criminal activity other than drug dealing.”), cert. denied, - U.S. -, 113 S.Ct. 1331, 122 L.Ed.2d 716 (1993). In fact, the government specifically *1219charged in the indictment that Carr knew that the cash represented the proceeds of drug trafficking, an indictment which was never amended to rid it of that "surplusage” (if indeed that is what it was). The majority seems to cull its theoiy from Carr's brief on appeal, which mentions it only to observe that the government had not pursued it.
Insofar as the majority fashions new grounds of criminal liability on appeal, it departs from established and conventional norms of notice and due process apropos criminal convictions. See U.S. Const, amend. VI. Issues of notice and due process aside, I do not see how the jury could have convicted Carr of "knowing” the money represented the proceeds of money laundering when the jury was not instructed that money laundering is a "felony under state, federal, or foreign law,” Maj. Op. at 1204, for purposes of § 1956(a)(2)(B), and when the government did not press that argument upon it.
. The majority's suggestion about an "intimate relationship,” Maj. Op. at 1205 & n. 6, is not supported by the record. While there was evidence that Carr went on numerous business trips with Gonzalez, there was no evidence that the two were close friends or that they confided in each other. In fact, on the July 11 trip, the only one about which there was any evidence relating to their precise travel arrangements, the two did not even sit in the same section of the airplane.
. I note also that the arguments raised here were not discussed in Caminos.