UNITED STATES of America, Plaintiff-Appellee, v. FUNDS IN THE AMOUNT OF THIRTY THOUSAND SIX HUNDRED SEVENTY DOLLARS ($30,670.00), Defendant, Antonio Calhoun, Claimant-Appellant, 403 F.3d 448

KANNE, Circuit Judge. Two Drug Enforcement Administration (“DEA”) agents encountered Antonio Calhoun at Chicago’s Midway Airport as he attempted to board a flight destined for Phoenix. The agents confiscated $30,670 in cash that Calhoun was carrying, and a drug detection dog later alerted to the currency, indicating the presence of narcotics. The United States filed a civil forfeiture action with regard to the cash, and the district court granted

summary judgment in the government’s favor. Calhoun appeals the judgment and, moreover, asks that we sua sponte grant

summary judgment in his favor. We decline Calhoun’s request and affirm the judgment of the district court. I. Background A. Factual History Around 9:00 A.M. on September 5, 2000, Antonio Calhoun paid cash at Chicago’s Midway Airport for a one-way ticket to Phoenix, Arizona. He checked no luggage, but carried with him a gym bag. As Calhoun attempted to pass through security on the way to his departure gate, he was approached by DEA agents. One agent informed Calhoun that they had heard that a person matching Calhoun’s description would be coming through the airport carrying a large amount of cash.

The agent also explained to Calhoun that he was not under arrest and could leave at any time. After identifying himself to the agents with an Illinois state identification card, Calhoun explained that he had lost his job and was traveling to Phoenix in order to find work. He also indicated that he had no luggage other than his carry-on gym bag. He confirmed that the bag belonged to him and that he had packed the bag himself.

The agents asked Calhoun if he was carrying any narcotics or large amounts of currency, and Calhoun responded that he was carrying “about $1000.” Calhoun then consented to the agents’ request to search his bag. The search revealed a few changes of clothes, assorted toiletries, and two separate bundles of cash wrapped with rubber bands. The agents again asked Calhoun how much cash he was carrying, and this time Calhoun answered that he was carrying $1700. The agents also noticed that Calhoun was stuttering his answers and appeared increasingly nervous.

The agents asked Calhoun if he was certain that he was carrying no other money, and Calhoun replied, “No.” One of the agents noticed that Calhoun had several bulges in the front and sides of his clothing, and he asked Calhoun for permission to search his person. Calhoun backed away, asking, “What do you want to search me for?” The agent brushed Calhoun’s body with the back of his hand and felt an odd protrusion under Calhoun’s clothes.

At that point, both agents escorted Calhoun to a vacant DEA office at the airport, where the agents made an interesting discovery regarding the mysterious bulges in Calhoun’s clothing — under his clothes, Calhoun wore a woman’s girdle stuffed with 27 additional bundles of cash totaling $28,970. Calhoun denied that he was the owner of the money but refused to say who the actual owner was. He also signed a “Voluntary Disclaimer of Interest and Ownership” form. The agents gave Calhoun a receipt for $30,670 (the sum of the cash in Calhoun’s gym bag and girdle) and sent him on his way.

Calhoun never traveled to Phoenix that day, nor did he ever claim a credit for his unused ticket. After Calhoun left the DEA office, the agents placed all of Calhoun’s bundles of cash into a plastic bag and enclosed the bag inside one of many empty suitcases in the room. The agents then enlisted the services of “Bax,” a Cook County Sheriffs Police narcotic detector dog, to conduct a sweep of the room containing the suitcase.

The agents informed Officer Robert Arrigo, Bax’s handler, only that confiscated currency had been hidden in the room, but not where (the room contained the suitcases as well as two desks and two filing cabinets). Officer Arrigo brought Bax into the room and commanded him to search for drugs. Bax sniffed around the room before fixing on the suitcase containing the bundles of cash. At that point, Bax alerted by scratching at the suitcase’s surface.

When the agents removed from the suitcase the plastic bag containing the cash, Bax continued to scratch and bite at the plastic bag, indicating the presence of narcotics. B.

Calhoun’s Story On February 27, 2001, the government filed a verified complaint of forfeiture pursuant to 21 U.S.C. § 881(a)(6), alleging that the seized cash “constitute[s] proceeds of narcotics trafficking and [was] intended to facilitate a narcotics transaction, or [was] to be furnished in exchange for narcotics .... ” Shortly thereafter, Calhoun, despite his earlier disclaimer of ownership of the cash, filed an answer and a verified claim for the $30,670. Discovery ensued, and additional facts emerged regarding Calhoun’s Phoenix trips and the sources of his cash hoard.

Regarding his aborted trip on September 5, 2000, Calhoun testified in his deposition that he recently had lost his job in Chicago, so he was traveling to Phoenix to “start over” — in other words, to move there permanently. He planned to find work there and to continue a budding relationship with a young woman identified as “Rochelle,” whom he had met at a Phoenix barber shop. Calhoun could not, however, recall Rochelle’s last name, stating that it was “either Brown or Burns.” Calhoun was similarly unable to provide Rochelle’s address or phone number.

Calhoun further testified that he had made a total of three previous trips to Phoenix in July and August 2000. He always purchased one-way tickets under his own name and flew American Trans Air (“ATA”). He claimed to have stayed at the same hotel in Phoenix on each trip. He could not, however, remember which hotel (“I stayed at a hotel on 55th and the [I — 10] expressway. I think it’s called Holiday Inn or the Hampton Inn or something.”). Calhoun testified that he did not know anyone in Phoenix and did not visit anyone while there.

He also claimed that he filled out “a couple of [job] applications” but could not provide details of these prospective employers other than his mention of one unnamed gas station, which never responded to his application. He indicated that each of his stays in Phoenix lasted less than a week. He also testified that while in Phoenix on his first trip, he spent more than $300, and he spent about $700 on each of his subsequent trips.

As for the sizable hoard of cash stashed in his girdle and gym bag, Calhoun indicated that he had accumulated the cash from various sources, including savings and gambling earnings that he started to accumulate in 1994 or 1995. He provided no evidence, such as receipts or bank statements, to substantiate these claims. C. The Government’s Motion for

Summary Judgment Following discovery, the government moved for

summary judgment. The government pointed out a number of problems with Calhoun’s story, including various contradictions in Calhoun’s explanations of his trips to Phoenix. For example, the government presented records subpoenaed from ATA. These records indicated that Calhoun did not, as he testified in his deposition, make three prior trips to Phoenix. Instead, from July 1 to August 27, 2000, he made no fewer than seven round trips. For each of these flights (both to and from Phoenix), Calhoun purchased a one-way ticket and made his flight reservation either the day of, or the day before, his departure.

The government also subpoenaed records showing that Calhoun never stayed at the Holiday Inn, Hampton Inn, or Super 8 hotel, all of which are located at 55th Street and the expressway in Phoenix. Instead, records indicated that Calhoun stayed at a Traveler’s Inn, located at least four blocks from the location Calhoun indicated in his deposition. In sum, based on his flight schedule, Calhoun stayed a total of 27 nights in Phoenix over the course of two months.

The government contended that all the evidence regarding Calhoun’s prior travels — along with evidence that Calhoun left his car in the Midway Airport parking garage on September 5, 2000, did not bring his driver’s license with him, made no arrangements for his two daughters to join him in Phoenix, and did not continue on to Phoenix (nor exercise a credit on his unused flight) that day or later — indicates that Calhoun had no plan to move permanently to Phoenix as he had testified.

In addition, the government pointed out obvious contradictions in Calhoun’s income (as indicated in various financial records) and Calhoun’s stated sources for his cash hoard. For example, in November 1998, Calhoun filed for

Chapter 7 bankruptcy. In his bankruptcy petition- — signed under penalty of perjury — Calhoun declared that he had no cash on hand. On his 1998-2000 tax returns, Calhoun filed as head of household, took the Earned Income Tax Credit, and claimed no interest income from savings. Calhoun’s tax returns indicated no gambling earnings. Moreover, Calhoun supported both of his daughters, who lived with him in 1999 and 2000. Calhoun did not have any bank accounts, savings or otherwise.

From November 1998, when Calhoun filed for bankruptcy, until the day the cash was seized, Calhoun’s combined reported income totaled $47,098. Calhoun’s reported expenditures in the same period (including rent, food, car payments, clothing, and recreation), plus the money Calhoun spent in connection with his travels to and from Phoenix, totaled at least $73,412. Thus, Calhoun’s expenditures exceeded his stated income in the relevant period by at least $26,394.

In any event, the government argued, Calhoun’s financial papers did not support his assertions regarding the alleged sources — gambling or savings — of the $30,670. The government contended that all of these facts, coupled with the circumstances leading to the seizure of Calhoun’s cash hoard — in particular, Bax’s positive alert to the cash — proved by a preponderance of the evidence that the cash represented the proceeds of (or was intended to be furnished in exchange for) an illegal drug transaction and thus subject to forfeiture.

Calhoun, who was represented by counsel, did not deny any of the material facts set forth in the government’s statement of material facts, nor did he offer any evidence of his own in opposition to the government’s motion. Instead, Calhoun offered his unsupported and conclusory contention that genuine issues of material fact precluded

summary judgment because the government’s evidence was “nothing more than extrapolated speculations stemming from innocuous facts.” In considering the government’s motion, the district court found that Calhoun’s actions “show[ed] a pattern that is consistent with illegal narcotics transactions[.]” Specifically, the court considered evidence of Calhoun’s earnings and expenses and concluded that Calhoun could not possibly have accumulated $30,670 in any aboveboard manner consistent with his explanations. The court also found Calhoun’s statements regarding his travels to Phoenix to be inconsistent and implausible.

Finally, the court determined Bax to be a reliable drug detection dog based on evidence the government offered to that effect, and thus Bax’s alert to Calhoun’s cash, in addition to the aforementioned factors, demonstrated a link between the cash and illegal drug activity.sufficient for the government to prevail in its forfeiture action. The court therefore granted

summary judgment in the government’s favor, finding that an examination of all the evidence indicated “a pattern that is consistent with illegal narcotics transactions.” Calhoun, acting pro se, appealed the judgment. After the government filed its brief, we located appellate counsel to act on Calhoun’s behalf and requested additional briefing from both parties to illuminate several narrow questions related to Calhoun’s case, specifically:

(1) How frequently do drug detection dogs falsely alert to currency that is not demonstrably related to the drug trade, but has been contaminated by prior owners?

(2) What steps were taken to ensure that the suitcase into which officers placed the currency was not a possible source of drug odors (or to determine, after the fact, whether the suitcase rather than the currency may have been the source)?

(3) What inference, if any, may be drawn with respect to the nature of Calhoun’s activity from the fact that he falsely asserted that the currency came from personal savings? and, relatedly,

(4) Under what circumstances, if any, does a false explanation establish as a matter of law that the adversary’s version of the facts is true, as opposed to establishing a material issue for trial? We requested additional briefing in part because Calhoun’s pro se appellate brief was not a model of clarity. In addition, as the government concedes, the propriety of this forfeiture case turns on the dog alert evidence — if dog alerts to currency have no probative value, then the government cannot possibly link Calhoun’s cash to illegal narcotics activity.

Evidence regarding the probity of dog alerts to currency is therefore crucial to resolving this appeal, and the record unfortunately does nothing to illuminate the issue. There appears to be some scientific and legal debate over this very issue, which is why we invited the parties to supplement their arguments with “publicly available empirical information” addressing the probative value of dog alerts to currency. We address the parties’ answers to these questions and their other arguments below. II. Discussion We review de novo the district court’s grant of

summary judgment. See Celotex Corp. v. Catrett, 477 U.S. 317, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986). A party is entitled to

summary judgment when “there is no genuine issue as to any material fact and ... the moving party is entitled to a judgment as a matter of law.” Fed.R.Civ.P. 56(c); Celotex, 477 U.S. at 322, 106 S.Ct. 2548. We must construe all facts in the light most favorable to Calhoun, but “we are not required to draw every conceivable inference from the record.” Bell v. Duperrault, 367 F.3d 703, 707 (7th Cir.2004) (citation and internal quotation marks omitted). The government filed its action pursuant to the civil forfeiture provision of the Controlled Substances Act, 21 U.S.C. § 881(a)(6).

This statutory provision makes subject to forfeiture “[a]ll moneys ... furnished or intended to be furnished by any person in exchange for a controlled substance ..., all proceeds traceable to such an exchange, and all moneys ... used or intended to be used to facilitate [such an exchange] .... ” Id. Thus, Calhoun’s cash hoard may be subject to forfeiture if the currency at issue represents the proceeds of an illegal drug transaction or was intended to facilitate such a transaction. Civil forfeiture standards are now subject to the Civil Asset Forfeiture Reform Act of 2000 (“CAFRA”), 18 U.S.C. § 983(c)(1).

CAFRA heightens the government’s evidentiary burden in civil forfeitures — the government must demonstrate by a preponderance of the evidence that the property sought is subject to forfeiture. See id. (“In a suit or action brought under any civil forfeiture statute for the civil forfeiture of any property ... the burden of proof is on the Government to establish, by a preponderance of the evidence, that property is subject to forfeiture^]”).

Furthermore, § 983(c)(3) provides that “if the Government’s theory of forfeiture is that the property was used to commit or facilitate the commission of a criminal offense, or was involved in the commission of a criminal offense, the Government shall establish that there was a substantial connection between the property and the offense.” Id. Turning to the matter at hand, we repeat that Calhoun did not dispute any of the material facts asserted by the government in support of

summary judgment, so Calhoun has adopted them as true. N.D. Ill. L.R. 56.1(b)(3)(B) (“All material facts set forth in the statement required of the moving party will be deemed to be admitted unless controverted by the statement of the opposing party.”); see also Smith v. Lamz, 321 F.3d 680, 683 (7th Cir.2003). Thus, we must decide what to make of the undisputed facts in the record along with the additional briefing and information we requested from the parties. Calhoun argues on appeal that the district court not only should have denied the government’s motion for

summary judgment, it should have granted

summary judgment in his favor. Calhoun also argues that, in any event, no reasonable juror could find a substantial money-drugs connection by a preponderance of the evidence in the record, and thus the district court erroneously granted the government’s motion. Calhoun presents a “contaminated currency” argument — the notion that most circulated currency in the U.S. is tainted by cocaine, so any dog alert to currency (Bax’s included) should not be entitled to probative weight.

Calhoun also claims that the conduct of the sniff search failed to safeguard against possible cross-contamination from other sources of drugs, thus undercutting any probative value that Bax’s positive alert may otherwise have had. Finally, Calhoun argues that the remaining factors in the case, such as his inconsistent or implausible testimony, do not suffice to meet the government’s preponderance of the evidence burden. We disagree with Calhoun’s contentions. As detailed below, the publicly available empirical information presented in this case cuts against Calhoun’s currency contamination argument.

The totality of the circumstances, especially Bax’s alert to Calhoun’s cash, leads us to conclude that reasonable jurors would agree that it is more likely than not that Calhoun’s cash is subject to forfeiture pursuant to 18 U.S.C. § 881(a)(6) and thus that the government is entitled to

summary judgment. A. The Dog Alert As mentioned earlier, the crucial threshold issue is whether Bax’s alert, which linked Calhoun’s cash hoard to illegal drug activity, is entitled to any probative weight. Calhoun argues that it is not, citing what he describes as “the consensus that most U.S. currency in circulation in the United States (and specifically the Chicago area) is tainted with cocaine” in levels detectable by drug dogs.

Thus, assuming a large proportion of the currency in general circulation is tainted with enough cocaine to trigger a drug dog alert, such an alert would be virtually useless in establishing the necessary link to illegal drug activity justifying forfeiture. But it is a matter of some scientific debate whether dogs alert only to cocaine itself or rather to the odor of a cocaine byproduct, such as methyl benzoate. Thus the critical question is not whether most currency in general circulation is tainted with cocaine, but whether the cocaine itself is what triggers dog alerts to currency.

If properly trained dogs alert to cocaine and not another substance or a byproduct of cocaine, then Calhoun’s argument is stronger — dog alerts to currency would have little or no probative weight if it can be shown that most currency is “innocently” tainted with detectable levels of cocaine.

On the other hand, if dogs alert to methyl benzoate as opposed to cocaine per se, and the byproduct is volatile enough to evaporate from the currency within a short period, then a dog alert likely would be more probative because even assuming that most currency is tainted with particles of cocaine, dogs will not alert to it unless it contains the odor of methyl benzoate.

Calhoun argues that even if dogs alert to methyl benzoate and not cocaine, the government must prove that the dog in question has been properly trained to alert only to the methyl benzoate, as opposed to something else, such as the currency itself. Thus, Calhoun challenges the evidentiary worth of any dog alert to currency and Bax’s reliability in particular. 1. Calhoun’s Empirical Evidence Calhoun relies principally on the work of Dr. Charles Mesloh. See Charles Mesloh et al., Sniff Test: Utilization of the Law Enforcement Canine in the Seizure of Paper Currency, 52 J. Forensic Ident. 704 (2002) (“Sniff Test ”).

Dr. Mesloh’s

article at first glance seems to support Calhoun’s currency contamination argument because it cites several studies indicating that 70-90% of circulated paper currency in major U.S. cities is contaminated with trace amounts of cocaine. See id. at 721-22; see also Adam Negrusz et al., Detection of Cocaine on Various Denominations of United States Currency, 43 J. Forensic Sci. 626, 626-29 (1998) (finding cocaine in amounts up to 10 micrograms per bill on randomly selected general circulation currency).

Calhoun also cites caselaw, including from this circuit, which seems to adopt the currency contamination theory and concludes that dog sniffs are entitled to little or no probative weight. See United States v. $506,231 in U.S. Currency, 125 F.3d 442, 453 (7th Cir.1997) (“[W]e are unwilling to take seriously the evidence of the post-seizure dog sniff.... Even the government admits that no one can place much stock in the results of dog sniffs because at least one-third of the currency in the United States is contaminated with cocaine in any event.”); Muhammed v.

Drug Enforcement Agency, 92 F.3d 648, 653 (8th Cir.1996) (holding that a dog alert is “virtually meaningless” because an “extremely high percentage of all cash in circulation in America today is contaminated with drug residue”); United States v. $5000 in U.S. Currency, 40 F.3d 846, 849 (6th Cir.1994) (same); United States v. U.S. Currency, $30,060.00, 39 F.3d 1039, 1042 (9th Cir.1994) (same). A closer read of Dr.

Mesloh’s article, however, reveals that it does not support Calhoun’s contention that “widespread contamination of currency at levels detectable by a drug-detection dog renders insupportable any inference that the most recent holder of the currency was involved with illegal narcotics activity.” Rather, Dr. Mesloh’s

article merely provides practical and helpful recommendations to law enforcement agencies in the wake of apparent judicial adoption of the currency contamination theory. See, e.g., Mesloh et al., Sniff Test at 708 (questioning Ninth Circuit conclusion that, given 75-90% contamination rate of currency, “it is extremely likely a narcotics detection dog will positively alert when presented with a large sum of currency”) (citing $30,060, 39 F.3d at 1043). Indeed, Dr. Mesloh himself suggests that dog sniffs are reliable even if a large proportion of circulated currency is tainted by narcotics.

See Mesloh et al., Sniff Test at 708 (noting that “recent research has shown that dogs will not alert on large amounts of currency!,]” and that “the drug odor on which the dogs are trained to alert dissipates over time, leaving only trace amounts on currency!.]”). Dr. Mesloh also cites scientific studies concluding that dogs likely do not alert to cocaine itself but rather to methyl ben-zoate. See id. at 708-09 (citing the research of Dr. Kenneth Furton). Dr.

Mes-loh acknowledges scientific findings that methyl benzoate “was found to evaporate rapidly from the surface of paper currency at a rate of approximately 90% in 120 minutes” and allows that “research does suggest that methyl benzoate may be one of the components of the odor that dogs are trained to detect.” Id. at 709. Thus, Dr. Mesloh concludes, these scientific findings “restore the credibility of the canine sniff[,]” id., contrary to the view expressed by the Ninth Circuit in $30,060 and by other courts. See, e.g., $30,060, 39 F.3d at 1042-43.

Therefore, we do not read Calhoun’s own proffered empirical evidence as bolstering his currency contamination theory. The cited work of Dr. Mesloh proposes practical solutions to reduce the likelihood that a court will overturn dog alerts on the basis of the currency contamination theory. See Mesloh et al., Sniff Test at 715 (“[A]ny agency that is intent on conducting drug money forfeiture must be aware of its obligations to scientific rigor....

If the [recommended] policies are implemented and maintained, a department will ultimately stand to benefit from the value of the property that has been seized and from more effective and cost efficient defenses of the seizures during judicial review.”). 2. The Government’s Empirical Evidence Assuming as true that most currency is contaminated by trace amounts of cocaine, the studies cited by the government (and in part by Dr. Mesloh) directly address whether dogs alert to cocaine itself or to methyl benzoate.

The authors of these studies hold advanced scientific degrees and have extensive experience in chemistry and forensic toxicology — fields that seem especially relevant to our analysis. Stefan Rose, for example, is an M.D. with years of research and training in the field of forensic toxicology and is a member of the faculty at Florida International University. Kenneth Furton holds a Ph.D. in analytical chemistry, has years of experience in the study of odor signatures, and is a former chair of the chemistry department at Florida International University.

Since 1993, both men have conducted extensive research on the scientific aspects of dog sniffs and have been retained to offer expert opinions on the subject in both state and federal courts. See, e.g., United States v. $242,484.00, 389 F.3d 1149, 1165 n. 9 (11th Cir.2004) (en banc); United States v. $22,474.00 in U.S. Currency, 246 F.3d 1212 (9th Cir.2001) (see Rose Aff., Gov’t App. at 21); United States v. One Lot of U.S. Currency Totalling $14,665, 33 F.Supp.2d 47, 58 n. 9 (D.Mass.1998); People v. Puertas, 2002 WL 31160304, *8 (Mich.Ct.App. Sept.27, 2002) (unpublished) (per curiam).

Both scientists have criticized the currency contamination theory and its uncritical adoption by courts. See, e.g., Kenneth G.

Furton et al., Odor Signature of Cocaine Analyzed by GC/MS and Threshold Levels of Detection for Drug Detection Canines, 14 Current Topics in Forensic Sci. 329, 329 (1997) (“Odor Signature ”) (“These conclusions are alarming in light of the fact that there are no definitive peer-reviewed studies determining the ‘range of a drug dog’s detection capability' to cocaine nor are there any statistically significant determinations of the degree and extent of cocaine contamination on U.S. currency.”) (critiquing the Ninth Circuit’s analysis in $30,060 and the sources cited therein). The research of Dr. Furton and Dr.

Rose established “to a reasonable scientific certainty that a narcotics detection dog alerts to the odor of methyl benzoate as the dominant odor of illicit cocaine, and not to cocaine itself.” (Aff. of Dr. Rose, Gov’t App. at 21; see also Furton et al., Odor Signature at 331.) In addition, the research indicates that dogs do not alert to byproducts other than methyl benzoate and would not alert to synthetic “pure” cocaine unless methyl benzoate was added. See Kenneth G.

Furton et al., Field and Laboratory Comparison of the Sensitivity and Reliability of Cocaine Detection on Currency Using Chemical Sensors, Humans, K-9s and SPME/GC/MS/MS Analysis, in Investigation and Forensic Science Technologies 41, 42 (Kathleen Higgins ed., 1999) (“Field Comparison”). Indeed, it seems that dogs cannot smell cocaine at all because the narcotic acts as an anesthetic that deadens olfactory senses. (Rose Aff. ¶ 2, Gov’t App. at 22) (“[Cjocaine is a local anesthetic and as such blocks the transmission of nerve impulses.

Therefore, cocaine should block the transmission of olfactory ... nerve fibers resulting in non-detection.”). In addition, methyl benzoate is highly volatile and evaporates at an exponential rate from tainted currency, so currency recently exposed to cocaine and returned to general circulation will quickly lose any detectable odor of methyl benzoate, even if the particles of cocaine remain. Furton et al., Field Comparison at 46; see also Kenneth G.

Furton et al., Novel Sample Preparation Methods and Field Testing Procedures Used to Determine the Chemical Basis of Cocaine Detection by Canines, in Forensic Evidenoe Analysis and Crime Scene Investigation 56, 58 (John Hicks et al. eds., 1997) (“Novel Methods ”) (“Whereas . the parent cocaine molecule is nonvolatile and can remain [on] currency for long periods of time, ... methyl benzoate dissipate[s] quickly .... ”).

A single cocaine-tainted bill will lose 90% of the odor of highly volatile methyl benzoate through evaporation within two hours of its removal from the presence of cocaine, but will lose the odor more slowly if stacked with other bills. Furton et al., Odor Signature at 332 (concluding that thirty stacked tainted bills lose less than 10% of the methyl benzoate odor in the span of two hours); (Rose Aff. ¶ 2, Gov’t App. at 22 (“The more closed the environment, the slower the rate of evaporation and the longer the smell remains.

One would expect that currency involved in the recent transaction of significant amounts of illicit cocaine and bundled would retain the odor of methyl benzoate for days and weeks after the exposure.... ”).) Dr. Furton and Dr. Rose undertook some 200 field and laboratory trials and ultimately concluded that dogs do not alert to innocently tainted currency in general circulation because the amount of cocaine and methyl benzoate is too small for detection. (Cf.

Rose Aff. ¶ 1, Gov’t App. at 21 (“[A] positive alert to U.S. currency by a properly trained ... canine indicates that the currency had recently ... been in close or actual proximity to a significant amount of narcotics, and is not the result of any alleged innocent environmental contamination of circulated U.S. currency by microscopic traces of cocaine.”).) The research indicated that, in contrast to the levels found on general circulation currency, the “threshold level of cocaine and methyl benzoate required for a canine to signal an alert was substantial and reproducible .” Furton et al., Odor Signature at 332.

As Dr. Furton and Dr. Rose conclude: Calculation from the amount of methyl benzoate required for a reliable detector dog alert (>85% [detection success] at 10 [micrograms]), the amount of methyl benzoate observed in street cocaine sampled,] ... and the diffusion rates of methyl benzoate from individual bills (ca. 10% remaining after 2 hours) indicate a required amount of recently contaminated cocaine residue of ca. 500 [milligrams] cocaine (initially). Furton et al., Field Comparison at 46. This required amount is “50,000 higher than the amount reported on circulated currency (ca. 10 micrograms/bill).” Id. at 46.

Therefore, generously assuming that all bills in general circulation are tainted by 10 micrograms of cocaine, it would take at least fifty thousand bills to generate enough methyl benzoate to trigger a dog alert. And, as Dr. Furton and Dr. Rose point out, “the odor produced by contaminated bills stacked together does not increase proportionally to the number of bills, but rather is a function of the available surface area” of the bills. Id.

This indicates that stacked or bundled bills, which obviously have less contaminated surface area exposed to the air, would exude a correspondingly smaller odor signature and the 50,000 figure therefore may be too small by orders of magnitude when tainted bills are bundled together (although stacked bills do retain the methyl benzoate for longer periods). Cf. id. at 45. In sum, the research led Dr. Furton and Dr.

Rose to conclude that “[o]verall, the scientific results indicate that circulated currency, innocently contaminated with [microgram] quantities of cocaine would not cause a properly trained detection canine to signal an alert even if very large numbers of bills are present.” Id. at 46. Given the apparently rigorous empirical testing giving rise to this conclusion, it is likely that trained cocaine detection dogs will alert to currency only if it has been exposed to large amounts of illicit cocaine within the very recent past.

As a result (and contrary to Calhoun’s assertions), it seems that a properly trained dog’s alert to currency should be entitled to probative weight. Likewise, we find the dog sniff caselaw cited by Calhoun either distinguishable or simply unpersuasive with regard to whether dog alerts to currency are entitled to probative weight. The conclusions reached in these cases rest on uncritical adoption of the currency contamination theory. In at least some of these cases, even the government seemed to assume the truth of the currency contamination theory.

See, e.g., $506,231, 125 F.3d at 453 (case in which the government not only did not contest the currency contamination theory, it allowed that “no one can place much stock in the results of dog sniffs because at least one-third of the [U.S.] currency ... is contaminated with cocaine ....”). More recently, however, courts have taken the approach suggested by Dr. Fur-ton and Dr. Rose and moved away from unquestioning acceptance of the currency contamination theory.

As Calhoun himself concedes, “the federal courts have become more open-minded toward dog alert evidence.” See, e.g., $242,484, 389 F.3d at 1165-66 & n. 9 (declining to “write the [currency contamination] theory into the law of [the Eleventh] [C]ircuit” and disagreeing with district court conclusion that probative value of dog alert to currency was weak, because “no one with any expertise testified in support of [the claimant’s] ever-lasting scent, global contamination theory”); United States v. $22,474.00 in U.S.

Currency, 246 F.3d 1212, 1216 (9th Cir.2001) (distinguishing earlier Ninth Circuit decisions that adopted the currency contamination theory because, unlike in the earlier cases, “the government presented evidence that the dog would not alert to cocaine residue found on currency in general circulation. Rather, the dog was trained to, and would only, alert to the odor of ... methyl benzoate[, and] ... unless the currency [the defendant] was carrying had recently been in the proximity of cocaine, the detection dog would not have alerted to it”).

We therefore conclude that the empirical information provided in this case indicates that dog alerts to currency should be entitled to probative weight. Cf. Furton et ah, Odor Signatures at 332 (“[T]here is no valid scientific basis for the theory that innocently contaminated currency contains sufficient quantities of volatile chemicals to signal an alert from a properly trained drug detector dog. Therefore, an alert to a specimen (including money) ... indicates that the item contains a significant quantity of narcotic odor.”). 3.

Bax’s Reliability Calhoun contends that there is no evidence that Bax is a reliable drug detector dog, and thus Bax’s alert to his cash is entitled to no probative weight. On this issue, however, we are satisfied that the record amply demonstrates Bax’s reliability. Bax is a certified narcotic detector dog, having received his certification in November 1997 upon completion of 400 hours of training with his handler, Officer Arrigo. Bax’s diploma indicates that he was qualified to detect marijuana, cocaine, and heroin; in 1999, Bax underwent recertification at which time he was certified to detect methamphetamine.

During the performance of his duties in the years after certification, Bax racked up an impressive record. Drugs or currency were found after 97.6% of his alerts. Drugs were found after 70.1% of his alerts. Only five of Bax’s alerts (the remaining 2.4%) were unambiguous false positives, and none of those took place in the two years prior to the seizure of Calhoun’s cash hoard. These facts are reflected in the record and set forth in the government’s statement of material facts, and Calhoun disputed none of these facts in his opposition to

summary judgment. But Calhoun now attacks Bax’s reliability by pointing to evidence that Bax conducted certification training with both actual cocaine and pseudococaine (which consists primarily of methyl benzoate). Furton et al., Novel Methods at 61 (“The active ingredient in ... pseudococaine is actually methyl benzoate.”).

According to Calhoun, “a dog’s training regimen will assure probativeness of future alerts to currency only if pseudococaine/methyl benzoate is the only material used during training.” Calhoun cites to some caselaw for this proposition and contends that Bax was not properly proofed to alert only to methyl benzoate, or some-how lacked adequate training generally.

See, e.g., $22,474, 246 F.3d at 1216 (holding that a dog alert to currency was probative because the government offered evidence that “the dog was trained to, and would only, alert to the odor of a chemical byproduct of cocaine called methyl ben-zoate”) (emphasis added); United States v. $10,700 in U.S. Currency, 258 F.3d 215, 230 (3d Cir.2001) (declining to determine the evidentiary weight to be accorded dog alerts to currency because the government had not presented any evidence concerning the dog’s training or its degree of accuracy in detecting narcotics on currency).

Thus, Calhoun argues that Bax’s record “does nothing to show that Bax has been trained to distinguish (or has otherwise shown the ability in practice to distinguish) recently tainted currency from currency in general circulation.” This is a distinction without meaning, however. For the reasons exhaustively set forth above, dogs apparently do not alert to cocaine at all, but rather to methyl benzoate.

So it does not matter whether Bax was trained with actual cocaine and pseudococaine, because he was actually trained to detect and alert to the odor of methyl benzoate emanating from the cocaine and pseudococaine, not the odor of cocaine per se — which is impossible to detect in any event, due to its anesthetic qualities. Furton et al., Odor Signature at 331 (“When a dog is trained to alert to ... drugs, ... the canine is actually being trained to alert to a scent associated with the item rather than the item itself....

This is the basis of commercial training aids developed, such as pseudo cocaine, ... which[ ] in fact[ ] contain no controlled substances but, rather, only the controlled substance odor.”); (Rose Aff. ¶2, Gov’t App. at 22.) And, as discussed, currency in general circulation, even if it is tainted, does not exude enough methyl benzoate to trigger a dog alert (unless the cash is very recently tainted and present in massive quantities) due to the substance’s high rate of evaporation.

An alert to currency in most circumstances would indicate that the currency is not innocently tainted, but instead tainted through contact or close proximity to illegal narcotics. Equally unavailing is Calhoun’s suggestion that Bax might alert to general circulation currency itself (the smell of cash) even in the absence of the odor of methyl benzoate. Bax was certified to detect certain narcotics, not currency — and apparently there are dogs trained for that specific purpose. See Normal Lorenzo et al., Laboratory and Field Experiments Used to Identify Canis Lupus Var.

Familiaris Active Odor Signature Chemicals from Drugs, Explosives, and Humans, 376 Analytical & Bioanalytical Chemistry 1212, 1213 (2003) (listing various detector dog types). There is no indication in the record (nor does Calhoun offer any meaningful argument) to suggest that Bax alerts to general circulation currency itself or to some other substance or byproduct that may be on currency by virtue of innocent contamination — indeed, extensive research by Dr. Furton and Dr. Rose indicates that trained dogs do not alert to any byproduct of cocaine other than methyl benzoate.

See Furton et al., Odor Signature at 331 (“None of the dogs tested alerted to byproducts other than methyl ben-zoate .... ”); Kenneth G. Furton et al., Identification of Odor Signature Chemicals in Cocaine Using Solid-Phase Microextraction-Gas Chromatography and Detector-Dog Response to Isolated Compounds Spiked on U.S. Paper Currency, 40 J. Chromatic Sci. 147, 155 (2002) (“Identification ”). In addition, as we previously have recognized, there is no need to “proof’ a dog off currency when there is ample evidence to illustrate the dog’s reliability in the field. See, e.g., United States v.

Limares, 269 F.3d 794, 798 (7th Cir.2001) (“Whether you can confuse a drug-detection dog depends ... on how dogs perform in practice, not, as Limares believes, how they were trained and ‘proofed off currency.”). Thus, we do not accept Calhoun’s contention that Bax is an “unsophisticated” narcotics detector dog, and Calhoun’s cited caselaw does not convince us otherwise. Calhoun cites $22,171 in support of his view that dogs must be shown to alert only to methyl benzoate as opposed to cocaine itself — a view resting on a faulty premise of which we have already disposed (cocaine per se has no odor).

In fact, we read the Ninth Circuit’s reasoning in that case to be consistent with the empirical information offered by the government — dogs alert to the odor of methyl benzoate, not to the minute particles of cocaine present on currency in general circulation. Cf. $22,474, 246 F.3d at 1216 (finding dog alert probative because of government evidence that “the dog would not alert to cocaine residue found on currency in general eirculation[,]” but rather to methyl benzoate). Likewise, Calhoun’s citation to $10,700 does not carry the day.

In that case, the Third Circuit attached no significance to a positive alert to currency because the government failed to present any evidence concerning the dog’s past training or accuracy. $10,700, 258 F.3d at 230. This is quite unlike the situation presented here, in which the government presented uncontroverted evidence demonstrating Bax’s qualifications and reliability in the field. Given these facts, we agree with the district court’s conclusion that Bax was a reliable detector dog at the time of his alert to Calhoun’s cash.

Certainly we may assume that Bax is wrong on rare occasion, as evidenced by his handful of false positives over the years. But Bax’s high rate of success (drugs or currency were found after 97.6% of his alerts, and drugs were found after 70.1% of his alerts), coupled with the additional empirical information before us in this case, is more than adequate to indicate his reliability in this case.

See Limares, 269 F.3d at 798 (“[E]ven if all alerts to currency are treated as false positives, [drug detection dog] Wendy has been right 62% of the time [as to the presence of drugs], enough to prevail on a preponderance of the evidence ....”) (emphasis in original). 4. Dog Alert Evidence — Conclusion In sum, on the evidence contained in the record as supplemented by empirical evidence, we conclude that Bax’s positive alert is entitled to probative weight in this case.

The publicly available empirical information offered in this case supports the conclusion that no properly trained dog could have alerted to Calhoun’s cash hoard, which comprised 1700 individual bills, if it had contained only innocently tainted bills. As indicated earlier, at least 50,000 such bills would be necessary to trigger a dog alert (likely many more, given that Calhoun’s bills were stacked and bound in 29 separate bundles).

In addition, the very ephemeral nature of the methyl benzoate byproduct of illicit cocaine makes it highly likely that Calhoun’s cash hoard was in very close proximity to large amounts of the drug within hours or days of Bax alerting to it. B. The Methodology of Bax’s Sniff Alert Calhoun’s next contention is that the “imprecise testing methodology” undertaken by the DEA agents and Bax’s handler failed to prevent the possibility of cross-contamination.

Thus, even assuming that Bax is reliable, the possibility that a source independent of Calhoun’s cash triggered Bax’s alert might deprive it of any probative weight. Calhoun makes several related arguments in support of this contention. First, Calhoun argues that the government agents did not adhere to the methodologies recommended by Dr. Mesloh, and thus the conduct of the dog sniff was somehow defective. See, e.g., Mesloh et ah, Sniff Test at 713.

Relatedly, Calhoun claims that the government failed to provide a package of “control” currency in order to rule out the possibility that Bax would alert to general circulation currency. Calhoun also contends that the agents failed to safeguard against possible cross-contamination of his cash hoard (perhaps the agents had handled other illicit substances recently but failed properly to sanitize themselves) and that the agents failed to ensure that the suitcase in which they placed his cash was free of narcotics residue or odor.

In support of these arguments, he offers his own conjecture and the methodologies recommended by Dr. Mesloh. It is true that the record does not contain great detail regarding the methodology of Bax’s sniff search. But at the

summary judgment stage, Calhoun failed to challenge the methodology used by the DEA agents and Bax’s handler in conducting the dog sniff, nor did he offer any contrary evidence of his own. The government asserts that it would have provided more detailed evidence addressing Calhoun’s contentions had Calhoun properly raised these arguments in opposition to

summary judgment. It is axiomatic, however, that Calhoun— the party opposing

summary judgment — • had the burden of coming forward with properly supported arguments or specific evidence to show a genuine issue of material fact. Celotex Corp., 477 U.S. at 324, 106 S.Ct. 2548. Instead, Calhoun accepted the government’s facts as true and only now offers arguments challenging the agents’ methodology. For example, Calhoun posits the sniff search methodology must be improper because the government faded to offer evidence that the agents employed a new plastic bag, sealed the plastic bag, or cleaned the suitcase before the sniff search. This turns

summary judgment on its head. The government already came forward with evidence regarding the methodology of the sniff search, and Calhoun never disputed or challenged this evidence before the district court. Calhoun’s speculative musings do not amount to a genuine issue of material fact requiring a remand for trial. See, e.g., Liu v. T & H Mach., Inc., 191 F.3d 790, 796 (7th Cir.1999) (citation omitted).

Our order requesting additional briefing on the methodology the agents employed was intended to focus the arguments of the parties (particularly Calhoun, as he was acting pro se at the time) on issues potentially dispositive in this case. We did not invite the parties to sidestep longstanding rules of

summary judgment and waiver. The time for Calhoun to challenge the methodology and offer evidence was at the

summary judgment stage, not for the first time on appeal. Cf. United States v. $174,206.00 in U.S. Currency, 320 F.3d 658, 662 (6th Cir.2003) (“That the burden [post-CAFRA] is on the government does not change the fact that, if the government meets its burden, it will prevail unless the claimant[] introducéis] evidence to support [his] case.”). Even if Calhoun has not waived a challenge of the methodology, we conclude that the evidence presented in the record is enough to establish that there was nothing untoward about the methodology employed by the DEA agents and Officer Arrigo.

In the first place, we disagree that any failure to apply Dr. Mesloh’s recommended methodology somehow invalidates Bax’s sniff search and alert. While we have no reason to doubt that Dr. Mesloh’s recommended methods may be sound, the failure to apply them does not indicate that the methods undertaken by the agents were improper — especially since Dr. Mesloh’s recommendations were, in part, designed to respond to judicial adoption of the currency contamination theory.

As set forth in detail above, we decline to accept Calhoun’s currency contamination arguments in light of the empirical information showing that trained dogs alert to methyl ben-zoate, not cocaine. And, as we concluded above, Bax is a trained and reliable narcotics (not currency) detector dog, so we do not see how the use of a control package of “clean” currency would have made any difference. See Limares, 269 F.3d at 798. Calhoun’s remaining contentions also lack merit.

In the absence of any contrary evidence offered by Calhoun, we conclude that the methodology used by the DEA agents and Officer Arrigo was not defective. The record discloses that the agents, outside the presence of Officer Arrigo and Bax, put Calhoun’s money into a plastic bag and placed the bag into one of many empty suitcases in the DEA office. Officer Arrigo then commanded Bax to search, and the dog sniffed around the room before alerting to the suitcase. The agents removed the plastic bag from the suitcase and Bax continued to alert to the bag, not the suitcase.

Evidence in the record regarding other sniff searches in the same office underscores the conclusion that the mere possibility of cross-contamination does not deprive Bax’s alert of probative weight. The government cites two incident reports detailing sniff searches by Bax in the same office in the two days immediately following the confiscation of Calhoun’s cash. On September 6, 2000, Bax alerted to a suitcase containing .methamphetamine that had been placed in the DEA office along with three other suitcases.

Likewise, on September 7, Bax alerted to a large sum of currency ($18,345) that had been hidden in the drawer of one of the desks in the same office. (See Gov’t App. at 56-61.) The government also introduced a voluminous account of Bax’s other sniff searches over the years. (R. 15, Ex. O.) This record indicates that Bax had undertaken more than 40 other sniff searches in the same office, and each time Bax alerted only to the location in which the suspect currency or narcotics were hidden (whether a suitcase, file cabinet, or desk drawer). Id.

No evidence suggests that, in any of these instances, Bax falsely alerted to other objects in the DEA office, which presumably he would have if these objects had been sufficiently contaminated or if agents had employed sloppy methodology. The volatility of methyl benzoate and the quantity needed to trigger an alert suggests that whatever minute particles of cocaine that might remain in the office would not lead to a false alert. Furthermore, no evidence suggests that the DEA actually stores narcotics or contraband in the office that might conceivably lead to false alerts.

While not conclusive, these points reinforce our conclusion that Bax is reliable and that the likelihood of cross-contamination leading to a false alert was extremely low in Calhoun’s search. Calhoun has offered nothing other than conjecture to convince us otherwise. See Bell, 367 F.3d at 707 (“Inferences that are supported by only speculation or conjecture will not defeat a

summary judgment motion.”). Contrary to Calhoun’s speculations, it is far more likely that Bax alerted to the only thing in the room exuding sufficient odor to trigger an alert — Calhoun’s cash hoard. Given all of these facts and Calhoun’s failure to raise an effective challenge on this basis before the district court, we conclude that Calhoun’s attack on the methodology of the sniff search fails. C. Calhoun’s Testimony and Other Factors Supporting Forfeiture The final issue upon which we sought additional briefing is the impact of Calhoun’s false explanations and the inferences, if any, to be drawn therefrom.

The government argues that Calhoun’s income and stated expenditures, as documented in his tax records, show that Calhoun could not possibly have earned the cash that he claims to have stockpiled over the years — his expenditures (including his $30,670 cash hoard) exceeded his stated income by at least $26,394 from 1998 to 2000. The government’s evidence and argument are premised on Calhoun’s filing of bankruptcy in 1998, in which he declared that he had no cash on hand (so the government began counting Calhoun’s earnings from 1998 to the date his cash was seized). In his opposition to

summary judgment, Calhoun did not dispute any of the government’s detailed accounting of his income and expenditures. The district court granted

summary judgment in part based on this obvious and substantial disparity between Calhoun’s expenditures and his reported income. Calhoun now claims on appeal that he did not give a false explanation for the source of his cash hoard. Calhoun offers three specific arguments in support of this claim. First, Calhoun artfully parses his response to a government interrogatory asking that Calhoun set forth the facts and documents supporting his position that the seized funds were not linked to illegal drug activities. Calhoun’s response was: See attached documents.

In addition, the claimant’s tax returns support the claimant’s position that the funds were not monies furnished or intended to be furnished in exchange for a controlled substance, or were not monies used or intended to be used to facilitate a narcotics transaction. Calhoun claims that the highlighted “in addition” indicates the existence of documentary proof other than his tax returns, and therefore the context of the answer makes clear that they [the “attached documents”]'are different documents from Calhoun’s 1998-2000 tax returns.

According to Calhoun, these mystery documents show that his explanation for his funds did not rest solely on his tax returns, so any disparity between his earnings and expenditures do not necessarily make his explanations false. Second, Calhoun alleges that he had begun saving cash in the mid-1990s, and thus the government should not have treated 1998 as a zero baseline from which to measure his expenditures against his reported income. He concedes that he failed to report these savings in his 1998 bankruptcy filing, in which he declared under penalty of perjury that he had no cash on hand.

Calhoun also concedes that he may not have reported gambling winnings on his tax returns, but alleges that any failure to report these earnings does not lead to the conclusion that he falsely explained the source of his funds. We find none of these arguments convincing. As Calhoun admits, the “attached documents” were not submitted to the district court, nor are they part of the record on appeal. If these documents had existed, and Calhoun had ensured that they were part of the

summary judgment record, matters might be different. But these documents are not before us in the record now, and their unexplained absence does not raise a disputed issue of material fact, especially when Calhoun failed to raise the issue before the district court. Likewise, Calhoun’s belated admission that he failed to account for his purported savings and gambling earnings in his bankruptcy filing and tax returns does not create a triable issue of fact that entitles him to a remand.

It is true, as Calhoun concedes, that the false statements on his bankruptcy and tax filings may expose him to fines or sanctions unrelated to civil forfeiture. But it is also true that the court properly could draw inferences and grant

summary judgment on the basis of the substantial and documented differences between the sources of income properly accounted for in Calhoun’s filings and his claimed sources of income (which he failed to substantiate). Cf. $174,206, 320 F.3d at 662 (“[Ejvidence of legitimate income that is insufficient to explain the large amount of property seized, unrebutted by any evidence pointing to any other source of legitimate income or any evidence indicating innocent ownership, satisfies the burden imposed by [§ 881(a)(6) ].”) (emphasis added).

The district court inferred that Calhoun’s explanations about the sources of his cash hoard were demonstrably false. In the absence of any meaningful argument or evidence to the contrary, it is hard to see how the district court could have come out any other way. Even if Calhoun had admitted his falsehoods in an affidavit submitted in opposition to

summary judgment, Calhoun could not prevail on this basis.

Summary judgment would be meaningless if litigants could manufacture genuine issues of material fact through self-serving and unsupported “admissions” materially different from positions taken in the past. This is why courts do not countenance the use of so-called “sham affidavits,” which contradict prior sworn testimony, to defeat

summary judgment. See, e.g., Bank of Ill. v. Allied Signal Safety Restraint Sys., 75 F.3d 1162, 1168-69 (7th Cir.1996) (“We have long followed the rule that parties cannot thwart the purposes of Rule 56 by creating ‘sham’ issues of fact with affidavits that contradict their prior depositions.... If such contradictions were permitted, ... ‘the very purpose of the

summary judgment motion — to weed out unfounded claims, specious denials, and sham defenses — would be severely undercut.’ ”) (internal citations omitted) (collecting authority). Although Calhoun’s representations on his bankruptcy and tax papers were not sworn testimony given in the course of litigation, Calhoun signed the documents under penalty of perjury.

We do not see why the “sham affidavit” principle would not apply here — at least to the extent that Calhoun’s false statements and failure to account for his cash hoard (plus his failure to challenge the government’s version of the facts) easily support the inference that the government’s version of the facts was true. Calhoun can not conjure a triable factual issue on the basis of his newfound and convenient respect for the truth, particularly when this “truth” is unsupported by any evidence other than Calhoun’s say-so. D.

The Totality of Circumstances Supports Forfeiture We conclude that the government has satisfied its burden to prevail on

summary judgment. The totality of the circumstances in this case, including the issues discussed above, lead to only one reasonable conclusion — Calhoun’s cash hoard was substantially connected to illegal drug trafficking and properly subject to forfeiture. We have discussed at length the most important of these factors: Bax’s positive alert to Calhoun’s cash and Calhoun’s demonstratively false explanations regarding the source of his funds. The remaining factors present in this case also indicate that the government is entitled to

summary judgment. For example, Calhoun’s explanations regarding his travel to Phoenix are suspect. On the day his cash was seized, Calhoun was traveling to Phoenix, a recognized source city for illegal narcotics. See, e.g., $22,474, 246 F.3d at 1216 (“Phoenix[ ][is] a known source city for drugs.”); cf. United States v. Currency, U.S. $42,500.00, 283 F.3d 977, 981 (9th Cir.2002) (giving weight to fact that claimant was “traveling from New York to San Diego, well known source cities for drugs”); United States v. $141,770.00 in U.S.

Currency, 157 F.3d 600, 604 (8th Cir.1998) (giving weight to fact that claimant was traveling from California, “a drug source state”). He had made frequent trips to Phoenix — seven trips within two months, not three as he claimed. Calhoun alleged that he stayed at the same hotel each trip (at “55th and the expressway”) but could not recall the hotel’s name; subpoenaed travel records indicate that Calhoun did not stay at any hotel at “55th and the expressway.” Yet he stayed in Phoenix at least 27 nights during the two months he had been traveling there (making his forgetfulness all the less credible).

All of these inconsistencies are relevant in weighing whether the government has established its burden justifying forfeiture. See $242,484, 389 F.3d at 1164 (finding it proper to consider claimant’s inconsistent statements and changing stories in considering whether the government’s burden is met); $22,474, 246 F.3d at 1217 (“[Claimant’s] inconsistent statements about the money and his reasons for being in Phoenix tended to support an inference that the money was drug-related.”); United States v. $67,220.00 in U.S.

Currency, 957 F.2d 280, 286 (6th Cir.1992) (“[M]isstatements are probative of possible criminal activity.”). Calhoun’s travel

schedule and arrangements were also suspicious. He always paid cash for one-way airplane tickets, which he usually purchased the same day of the flight (and sometimes only minutes before). As any savvy air traveler knows, this practice virtually guarantees higher fares, and someone in Calhoun’s position— unemployed and with no regular income for months — would seem especially unlikely to be a spendthrift under such circumstances. In addition, Calhoun’s travel

schedule was haphazard — in some instances, he remained in Phoenix for over a week, but other times he stayed only overnight before returning to Chicago. The gaps between trips varied as well, with at least one instance in which Calhoun flew back to Phoenix the day after returning to Chicago. These factors are consistent with the travel habits of drug couriers, who apparently must be very flexible in their travel plans.

Cf. $242,484, 389 F.3d at 1163 (holding that it was appropriate for the district court to consider the claimant’s travel arrangements in light of evidence that “drug couriers often travel on tickets purchased with cash, ... and frequently change their return date, as [claimant] did twice in two days.”). Calhoun also suggested that he was moving to Phoenix permanently because he had lost his job in Chicago.

As Calhoun points out, no evidence indicates that he was making that permanent move on the very day that his cash was seized (indeed, Calhoun left his car at the Midway Airport parking garage and did not have his driver’s license with him). But it is also true that Calhoun never explained why he was carrying a substantially large sum of money if his trip (like the others) was to be only temporary (and Calhoun testified that he had only spent $300-$700 on each of his prior trips). Cf. United States v. One Lot of U.S.

Currency ($36,634), 103 F.3d 1048, 1055 (1st Cir.1997) (“Carrying a large sum of cash is ‘strong evidence’ of [a connection to illegal drug activity] even without the presence of drugs or drug paraphernalia.”) (internal quotation marks and citation omitted). Other aspects of his testimony were suspicious as well. For example, one reason Calhoun gave for the move was his desire to pursue a romance with “Rochelle,” whose last name, phone number, and address he was unable to provide.

He also claimed to have been searching for work in Phoenix but could provide no proof — not even the name of any of the prospective employers. Cf. $22,474, 246 F.3d at 1217. Moreover, Calhoun’s actions at the airport were suspicious. When confronted by agents, Calhoun first claimed to be carrying about $1000, then he claimed that he was carrying $1700 dollars. Cf. $67,220, 957 F.2d at 286 (finding probative the fact that claimant “twice understated the amount of money he was carrying”). He appeared increasingly nervous and evasive during his conversation with the DEA agents.

He finally admitted to be carrying an additional $29,000, which the agents verified when they discovered the bundles of cash stuffed into a girdle that Calhoun wore underneath his clothes. Cf. $42,500, 283 F.3d at 981 (concluding that, coupled with other factors, “[possession of a large amount of cash is strong evidence that the money was furnished or intended to be furnished in return for drugs”) (internal quotations and citation omitted). Calhoun denied ownership of the cash, alleging that it belonged to a friend (whom he refused to name), and he signed a form disclaimer.

Cf. $141,770, 157 F.3d at 604 (finding that “initial denial of ownership over the money” cut against innocent explanation). Calhoun provided no receipts or other proof regarding the origins of the cash. Finally, Calhoun’s explanations regarding the sources of his cash did not add up. As explained at length above, an exhaustive accounting of Calhoun’s income and expenditures (reflected largely in Calhoun’s income tax filings and admissions) revealed that Calhoun’s expenditures exceeded his stated income by over $25,000.

Calhoun offered no documentation or other proof to account for this wide discrepancy; similar circumstances have supported forfeiture in other cases. See, e.g., $174,206, 320 F.3d at 662 (fact that claimants’ “legitimate income was insufficient to explain the large amount of currency found in their possession” supported forfeiture); $141,770, 157 F.3d at 604 (“[The claimant’s] statement that this money constituted legitimate business proceeds is undercut ... by his inability to produce any tax records regarding the source of this income .... ”).

Calhoun attempts to characterize this array of factors as nothing more than a series of unrelated factoids adding up to a “drug courier profile” that is not to be accorded evidentiary weight. See United States v. $84,000 U.S. Currency, 717 F.2d 1090, 1099 n. 10 (7th Cir.1983). He further argues that, to the extent any of these factors fall outside the drug courier profile evidence, at most “it shows that Calhoun may have engaged in some unlawful activity,” but not necessarily activity involving illegal narcotics.

Calhoun also highlights certain facts in an effort to show that his cash was not linked to illegal narcotics activity. For example, Calhoun notes that he always traveled under his own name, whereas drug couriers tend to travel incognito. See $67,200, 957 F.2d at 285. Calhoun also makes much of the fact that he was never charged or convicted of a drug offense following the seizure of his cash. These facts, Calhoun claims, show that his cash should not be subject to forfeiture. Again we must disagree with Calhoun’s contentions.

It is true, as Calhoun argues, that many of these factors, taken alone, would not be enough to carry the government’s burden for forfeiture, particularly under CAFRA’s heightened standard. We decline, however, to implement Calhoun’s divide-and-conquer approach with respect to the factors present in this case. Instead, we consider the totality of the evidence as a whole and in the appropriate context.

See $42,500, 283 F.3d at 981 (finding that although some factors taken alone may be innocent, “the aggregate of facts raise more than a mere suspicion of a connection between the seized money and drugs”); see also $242,484, 389 F.3d at 1160 (“In evaluating the evidence of proceeds traceable to drug transactions, we ... eschew[ ] clinical detachment and en-dorsee ] a common sense view to the realities of normal life applied to the totality of the circumstances.”) (internal quotation marks, brackets, and citation omitted); $36,634, 103 F.3d at 1054 (“Even where no particular circumstance is conclusive, it is the aggregate of the facts that is examined.”) (internal quotation marks and citations omitted).

Taken together with Bax’s positive alert, the multitude of factors in the aggregate sufficiently establish a substantial connection between the cash hoard and illegal narcotics activity. Calhoun’s “exculpatory” facts do not convince us otherwise. Bax’s alert outweighs any suggestion that the cash may have been linked to non-drug-related illegal activity. The fact that Calhoun traveled under his own name and was not subsequently charged in a criminal indictment is of little consequence, especially when stacked up against all of the other factors in this case.

See, e.g., $10,700, 258 F.3d at 224 n. 6 (“[FJorfeiture under § 881(

a) is not conditioned upon an arrest or conviction for a drug offense.”). Furthermore, we are unswayed by Calhoun’s efforts to distinguish or deem inap-posite forfeiture caselaw decided before the enactment of CAFRA. Those cases, of course, applied the less-rigorous probable cause standard — a lighter burden for the government to carry in civil forfeiture cases. See United States v. $87,118.00 in U.S. Currency, 95 F.3d 511, 518 (7th Cir.1996) (describing pre-CAFRA forfeiture standard); accord $174,206, 320 F.3d at 661-62 (comparing pre- and post-CAFRA burdens).

But the cases are not inappo-site merely because they were decided pre-CAFRA. The government’s burden may have increased in the wake of CAF-RA, but it did not become insurmountable. Factors that weighed in favor of forfeiture in the past continue to do so now — with the obvious caveat that the government must show more or stronger evidence establishing a link between forfeited property and illegal activity. We have considered all relevant caselaw while remaining cognizant of the increased quantum of proof necessary to support forfeiture in post-CAFRA cases like this one.

Cf. $174,206, 320 F.3d at 662 (concluding that the government met its burden under either pre- or post-CAFRA standards). That quantum is present here. We conclude that the government has carried its burden of showing by a preponderance of the evidence that Calhoun’s cash hoard is subject to forfeiture. Bax’s positive alert to Calhoun’s cash is strong probative evidence of illegal narcotics activity. That alert, coupled with the totality of other circumstances present in this case, indicates a substantial connection between Calhoun’s cash and illegal narcotics activity. The government therefore is entitled to

summary judgment. III. Conclusion For the foregoing reasons, the judgment of the district court is AffiRmed. . The cash was in the following denominations: 10 fifty dollar bills, 1327 twenty dollar bills, and 363 ten dollar bills. . Bax (pronounced "box”), a German Shepherd, is a certified drug detection dog trained to detect several different types of drugs, including cocaine. Bax signals ("alerts”) the presence of narcotics by biting and scratching. . We also indicated in our order that "[i]f the record is silent on this issue, counsel should refer to any publicly available empirical information.” .

Forfeiture proceedings commenced prior to the effective date of CAFRA (August 23, 2000) applied a lesser standard of proof — probable cause. See United States v. 5 S 351 Tuthill Road, Naperville, IL, 233 F.3d 1017, 1023 (7th Cir.2000) (CAFRA "requires the government to prove the connection between the property to be forfeited and the drug activity by a preponderance of the evidence, rather than to prove merely probable cause to believe there is a connection.”). . Dr. Mesloh holds a Ph.D. in Public Affairs and is an assistant professor at Florida Gulf Coast University. Dr.

Mesloh is a former police officer with wide experience and scholarship in the practical uses of canines in support of law enforcement, including dog sniff searches. Dr. Mesloh appears to have coauthored the cited

article while he was a candidate for his Ph.D. . Although his academic background is in public affairs, Dr. Mesloh is an experienced former law enforcement professional and former police dog handler, which lends some weight to his practical recommendations regarding methodology as discussed below. . The research of Dr. Furton and Dr. Rose also indicates that methyl benzoate is always associated to some degree with cocaine. See Furton et al., Odor Signature at 332. Pharmaceutical grade or pure cocaine contains only trace amounts of methyl benzoate, but illicit cocaine contains far more due to impurities.

See Furton et al., Novel Methods at 58, 62; Furton et al., Field Comparison at 46. . This also cuts against Calhoun's assertion that he had been accumulating his cash hoard since the mid-1990s, for any tainted cash he may have been stuffing in his mattress or cookie jar over the years would have long since lost any detectable odor of methyl ben-zoate, even if bundled or kept in closed spaces. See Rose Aff. ¶ 2, Gov't App. at 22. .

Indeed, in the course of his duties at the airport, Bax conducted sniff searches of some 80,000 suitcases (checked bags, not planted in the DEA office), and only detected drugs two times. This underscores the conclusion that Bax has a discriminating nose. (R. 15, Ex. O.) . Calhoun is not helped by caselaw regarding false or pretextual statements made in the employment discrimination context. See, e.g., St. Mary’s Honor Ctr. v. Hicks, 509 U.S. 502, 113 S.Ct. 2742, 125 L.Ed.2d 407 (1993); Anderson v. Baxter Healthcare Corp., 13 F.3d 1120 (7th Cir.1994).

This line of authority teaches that a finding of pretext does not automatically entitle the plaintiff to judgment as a matter of law. So it is in this case — • Calhoun's false statements alone do not trigger

summary judgment, but they certainly raise the inference that the government’s version of the facts is true (particularly when Calhoun offered no evidence to dispute those facts) and may be considered along with the other factors present in this case, particularly Bax's positive alert to his cash.