Appellant Walter E. Trayer challenges an order of the district court denying his motion to suppress evidence (cocaine) that police found in a nonconsensual search of his train roomette. The officers conducted the search because of information gathered from a computerized Amtrak reservation list and the results of a dog sniff of the exterior of Trayer's roomette. On the facts found by the district court, we agree that the officers had probable cause to believe that appellant's roomette contained illicit drugs. We therefore affirm appellant's conviction.

On August 2, 1988, Officer Sauve of the Amtrak Police Department in Washington, D.C. learned, by examination of Amtrak records, that a passenger identified as W. Trayer would be traveling round trip from Delray Beach, Florida to Philadelphia, stopping in Washington's Union Station for 20 minutes the next day. The train manifest recorded that on the evening of August 1, Trayer reserved a first class private roomette for his trip to Philadelphia but only a coach seat for his return on August 25, that he had purchased his ticket approximately 30 minutes before the train's scheduled departure time on August 2, that he had paid the $327 fare in cash, and that he had provided a call-back telephone number that, when called by Officer Sauve, was not in service.

Based on this information, Officer Sauve decided to interview Trayer when his train arrived at Union Station. At Sauve's request, the Washington Metropolitan Police Department sent Detectives Buss, Cassidy, Dione, and a trained narcotics dog, Ben II, to assist him with his investigation. Ben II is a Golden Retriever and 1987 graduate of the Metropolitan Police Department's K-9 Division training school. Officer Buss attended the school with Ben II and became his handler. By August 3, 1988, the two had investigated suspected contraband approximately sixty times.

Amtrak changed the train's engines at Union Station on the morning in question, causing a temporary electric power shut down while the train was in the station. So, after boarding the train, Officer Sauve used a flashlight to illuminate the numbers on the roomette doors until he found Trayer's roomette, # 7, approximately 15 feet down the corridor from the vestibule where the officers had embarked. Sauve then returned to the vestibule and asked Officer Buss to have the dog sniff the rooms in the corridor. After passing several roomettes, Ben II froze at # 7 and pointed at the air vent at the bottom of the door. Officer Buss and Ben II then returned to the vestibule, and Officer Buss told Officer Sauve that the dog had alerted to drugs in roomette # 7. Officer Buss admitted that when he escorted the dog down the corridor he was aware that roomette # 7 was under suspicion.

Officer Sauve knocked on Trayer's door, and when Trayer answered, told him that a narcotics dog had indicated the presence of drugs inside his roomette. When Sauve asked for consent to search the roomette, Trayer offered him his suit bag but did not consent to a search of the room. Sauve nevertheless searched the room, finding a leather briefcase secreted behind a collapsible seat. Trayer also refused to consent to a search of the briefcase, saying that he did not have the combination to its lock. While Sauve continued to search the roomette, another agent took the briefcase to the platform and asked Officer Buss to have Ben II sniff it. The dog responded by prancing around the briefcase and biting at its corners. The officers then arrested Trayer and took him and his luggage to the police office at the train station. After they obtained a warrant to search the briefcase, the officers opened it and found one kilogram of cocaine.

At the suppression hearing, Trayer called Thomas A. Knott, who, from 1970 to 1986, had been a trainer for the Baltimore City K-9 Corps in charge of training dogs to detect narcotics. Knott testified that, in order to prevent a handler from conveying knowledge to his dog--even unconsciously--the handler should not know where the target of the search is. He also opined that after the dog alerted outside the roomette, the proper procedure would have been to bring the dog inside the roomette to define the target in order to show whether the dog alerted to a residue of drugs that might have been in the roomette. On cross-examination, however, Knott testified that he had not heard anything to indicate that Officer Buss suggested to Ben II that there was cocaine in roomette # 7.

The district court denied Trayer's motion to suppress. United States v. Trayer, 701 F. Supp. 250 (D.D.C.1988). The court held that, considering the reduced expectation of privacy that a passenger has in a train roomette, the exigencies of the situation,1 and the government's interest in thwarting drug traffic, a search was permissible if the officers had reasonable suspicion to believe the roomette contained drugs. The court also held, however, that the officers had probable cause to search the room. The district court rejected appellant's challenges to the manner in which the search was conducted. Addressing appellant's claim that the dog should have been brought into the compartment to define the scope of the search, the court held that since the facts before it made clear that the officers had probable cause to search the entire compartment, the failure to have the dog sniff within the compartment was irrelevant.. As for the claim that it was inappropriate for Buss to know which roomette was under suspicion, the court noted that appellant's expert witness had conceded that there was no evidence that Buss had actually conveyed any inappropriate cues to Ben II and refused to adopt the expert's opinion as to the possibility that the dog was unconsciously "cued" on the ground that it was "at best speculative and conjectural." Trayer, 701 F. Supp. at 255.

Appellant begins his challenge to the search of his roomette by disputing the reasonableness of the police's targeting him for investigation. But, of course, there is no constitutional right to be free of investigation. See United States v. Lloyd, 868 F.2d 447, 451 (D.C.Cir.1989). And we have held that a dog sniff of a train compartment is not a fourth amendment search. See United States v. Colyer, 878 F.2d 469, 477 (D.C.Cir.1989). So it is not apparent that any defined level of suspicion is necessary before a dog may be escorted along the corridor of a train, past a given compartment.

Officer Sauve went on, however, to search Trayer's roomette without his consent once Ben II had frozen and pointed to the vent of Trayer's compartment door. Appellant disputes the district court's conclusion that the search was supported by probable cause to believe that appellant's roomette contained illicit drugs.2 We believe, however, that the district court was justified in determining that Ben II's behavior at appellant's door together with the information about Trayer that Officer Sauve obtained from Amtrak's reservation records established probable cause.

The district court's conclusion was based on factual circumstances nearly identical to those present in other decisions in which we determined that probable cause existed to search train roomettes. See Colyer, 878 F.2d at 471; United States v. Tartaglia, 864 F.2d 837, 839 (D.C.Cir.1989). Here, as in those cases, appellant's travel arrangements were those characteristically used by drug couriers. Florida is known to law enforcement authorities to be a source area for cocaine distributed along the East Coast. Drug couriers are likely to desire privacy while transporting drugs but are indifferent to it after having delivered them. And travelers paying cash and offering non-working call-back telephone numbers typically want to ensure that their identity or the location of their residence cannot be verified. We do not, therefore, hold that Ben II's alert, on its own, provided the district court with sufficient evidence to support a probable cause determination. For as appellant points out, a dog may alert to traces of drugs left in a roomette by previous inhabitants. But here, the dog alert in combination with the information obtained from Amtrak's train manifest supported the search.

That of course assumes that the dog alert was itself a good indicator of the presence of narcotics within the roomette--in other words, that the alert was a reliable one. It is on this point that appellant makes his strongest challenge. It is argued that Ben II, who had been trained to exhibit aggressive behavior on detecting drugs, instead froze and pointed at the compartment as would a bird dog. Officer Buss explained, however, that Ben II alerted in that fashion "on a majority of occasions" and had alerted correctly (finding drugs) in 58 out of 60 alerts.3 That Ben II's behavior is idiosyncratic (perhaps a triumph of genetics for a Golden Retriever) does not diminish its reliability so long as the dog's peculiar brand of alert is a trustworthy method of communicating the presence of drugs.

Appellant also argues that Ben II's alert was tainted by unprofessional handling. Thomas Knott, the retired Baltimore police dog trainer, testified that it is possible for a handler through voice or physical cues to compromise a dog's objectivity. He criticized the officers' behavior in this case as not up to the best standards. Officer Sauve while in the dog's view walked along the corridor with his flashlight until he reached roomette # 7 and then returned to the vestibule. And Officer Buss took Ben II sniffing down the corridor only as far as that roomette. Given the reliance that this and other courts have placed on the results of dog sniffs in reviewing probable cause determinations, we find Knott's testimony quite troubling. Still, Knott was unwilling to say, from the evidence, that the handler had actually cued the dog. Officer Sauve illuminated several roomettes before he reached # 7, and Officer Buss did not, according to the evidence, behave any differently at roomette # 7 than he did as the dog passed other compartments along the corridor. We think, then, that the district court was entitled to reject the notion that the dog was cued as to the target roomette. But we are mindful that less than scrupulously neutral procedures, which create at least the possibility of unconscious "cuing", may well jeopardize the reliability of dog sniffs.

Appellant's last argument, that the entire roomette should not have been searched--that is, that the dog should instead have been brought inside to sniff more directly, in order to localize the search--is foreclosed by our precedent. In United States v. Tartaglia, 864 F.2d 837, 843 (D.C.Cir.1989), we held that a search of an entire train compartment was justified under similar circumstances.

* * * * * *

Accordingly, we affirm the district court's judgment.