United States v. Michael Gary Whitmire and Donald John Williams , 49 A.L.R. Fed. 738 ( 1979 )

GEE, Circuit Judge:

A district court, sitting without a jury, convicted appellants Michael Whitmire and Donald Williams of possession of marijuana with intent to distribute, a violation of 21 U.S.C. § 841(a)(1). Whitmire was also convicted of importation of marijuana in violation of 21 U.S.C. §§ 952(a), 960(a)(1); Williams was acquitted of that charge. On appeal they argue that due process was violated by the trial court’s referral of their motion to suppress evidence to a magistrate under 28 U.S.C. § 636(b)(1)(B). They also challenge the warrantless search that produced the main evidence used against them. Finally, Williams challenges the sufficiency of the evidence to support his conviction of possession with intent to distribute. Disagreeing with each of appellants’ contentions, we affirm their convictions.

I. Referral to Magistrate of Suppression Motion.

The Ninth Circuit has held that a district court may not enter an order contrary to a magistrate’s recommendation on a referred evidentiary matter without itself holding an evidentiary hearing. United States v. Bergera, 512 F.2d 391 (9th Cir. 1975). Appellants would have us not only espouse that position but extend it to hold that a district court cannot even adopt a magistrate’s recommendation without holding a duplicate hearing. They argue that due process of law requires the trier of fact actually to hear the relevant testimony and assess witness credibility. Whatever the merits of such a rule in instances where a district court chooses to ignore a magistrate’s recommendations, a question we need not reach, it is not applicable here. The motion to suppress was referred initially to a magistrate, who recommended that it be denied. After giving counsel an opportunity to object to the magistrate’s conclusions and reviewing the record, the district judge adopted those findings and recommendations. The magistrate had a firsthand look at the witnesses and appraised their credibility. The trial judge retained the power to hear additional testimony or the same testimony all over again if he decided that would be beneficial in determining the motion. As the Supreme Court has said in a related but nonconstitutional context:

The magistrate may do no more than propose a recommendation, and neither § 636(b) [of the United States Magistrates Act, 28 U.S.C. §§ 631-639] nor the General Order gives such recommendation presumptive weight. The district judge is free to follow it or wholly to ignore it, or, if he is not satisfied, he may conduct the review in whole or in part anew. The authority — and the responsibility — to make an informed, final determination, we emphasize, remains with the judge.

Mathews v. Weber, 423 U.S. 261, 270-71, 96 S.Ct. 549, 554, 46 L.Ed.2d 483 (1976).

In the 1976 amendments of the Magistrates Act, Congress enumerated some additional duties that may be assigned to magistrates and clarified, along the lines suggested in Weber, the weight a district court may give to a magistrate’s various findings and determinations. Subsections 636(b)(1)(A) and (B) read jointly now specifically provide that a judge may designate a magistrate to conduct an evidentiary hearing on a motion to suppress evidence in a criminal case and to recommend a disposition of the motion. Subsection 636(b)(1) further provides:

Within ten days after being served with a copy, any party may serve and file written objections to such proposed findings and recommendations as provided by rules of court. A judge of the court shall make a de novo determination of those portions of the report or specified proposed findings or recommendations to which objection is made. A judge of the court may accept, reject, or modify, in whole or in part, the findings or recommendations made by the magistrate. The judge may also receive further evidence or recommit the matter to the magistrate with instructions.

These procedures provide sufficient safeguards both to ensure the integrity of the factfinding process and retention by the judge of final responsibility for ruling on the motion. Indeed, the factfinding process may be improved by the referral practice. In making a final determination, the district court has the benefit of a carefully developed record, a magistrate’s thoughtful consideration of the issues, and argument of counsel regarding specifics not agreeable to the parties. See Weber, 423 U.S. at 271, 96 S.Ct. 549. We conclude that appellants’ rights to procedural due process have not been violated.¹

II. Warrantless Search of Boat.

Around 8:00 a. m. on December 23, 1976, two customs officers were on marine patrol in a Dade County, Florida, intercoastal waterway. They observed a 25-foot Nova boat speeding at an estimated 40-45 miles per hour through Baker’s Haulover Cut, an inlet connecting the ocean to the inland waterway. The boat was coming from the ocean side but was already inside the shoreline when the officers first sighted it. The day was unpleasant for boating: it was overcast, cold, windy, and the water in the cut was choppy. The speeding boat was producing a heavy wake. Each time it came down it threw up a large bow spray, higher than the boat top. To these experienced officers this indicated that there was something heavy in the boat’s bow. They later testified that in the past year there had been about 25 similar cases of boats that size, “riding heavy in the bow, throwing excessive bow wake,” found loaded with marijuana. They observed two males aboard, and no fishing gear was visible. Considering all these facts in the cloudy light of the inclement day, the officers decided to stop the boat to investigate further.

Their boat, however, was too slow to catch the Nova, which turned north into the inland waterway and continued on at high speeds past a customs inspection station and through two “no wake” areas, a most egregious nautical sin. The officers followed and finally closed with the boat as it was being docked on a canal behind Whitmire’s house. At this closer range the officers saw that the boat was encrusted with salt crystals such as might have formed during an extended ocean voyage. They also noticed that both Whitmire and Williams were wearing brand new orange sweatshirts with “BIMINI” printed across the chest.

Appellants had left their boat and were walking toward Whitmire’s back door when the officers called to them, requesting to see their identification and registration papers. Whitmire helped the officers dock and then produced his identification and an unsigned boat registration made out to the Excellent Car Company; Williams could produce no identification. Thinking the boat perhaps stolen and still suspecting that contraband was aboard, one officer boarded the Nova to investigate further while the other watched the two men. As soon as the officer stepped down into the cockpit area in the center of the boat he smelled the odor characteristically referred to in these prosecutions, the “overpowering aroma of marijuana.” Opening the hatch, he saw over a thousand pounds of baled marijuana, the evidence appellants seek to suppress on fourth amendment grounds. The officers arrested the men and thereafter found an American Express receipt in Whitmire’s wallet indicating that he had bought a large quantity of fuel in Bimini the previous day.

Several exceptions to the fourth amendment’s warrant requirement are potentially applicable here: (1) border search, with its extended-border and functional-equivalent-of-the-border glosses; (2) a limited investigatory stop upon presence of reasonable suspicion of law violation under United States v. Brignoni-Ponce, 422 U.S. 873, 95 S.Ct. 2574, 45 L.Ed.2d 607 (1975); (3) pure statutory authority — 19 U.S.C. § 1581(a)— found independently reasonable at least as to some searches on water under fourth amendment analysis because of the unique character and history of law enforcement on our country’s seaways. See United States v. Freeman, 579 F.2d 942, 946-48 (5th Cir. 1978). There are problems with each of these analyses, however.

The instant facts do not fit a border search analysis neatly, since we have consistently required some degree of probability that the vehicle/vessel has crossed the border.² In water cases, the true border is an imaginary line three miles offshore. Though we have not required the coast guard or customs officers to observe the vessel as it crosses this imaginary line,³ our precedent generally has required that the officials have articulable facts from which they may reasonably infer that the boat has come from international waters.⁴ On our facts, the boat first was sighted inside the coastline and, for all the customs people actually knew, might have been on an early morning cruise up and down the intercoastal canal or outer coastline. It is true that a boat’s space at a dock is often considered the functional equivalent of the border⁵ —but this is true only of boats thought with some degree of probability to have just entered the country. Thus, traditional search concepts do not justify this particular search since, though the boat had actually come from Bimini as evidenced by the American Express receipt, the officers initially had no grounds for believing it had.

There are also problems with justifying all the officers’ conduct on BrignoniPonce grounds. There the Supreme Court was faced with a simple investigatory stop of a vehicle and not a vehicle search such as had been condemned, absent probable cause or warrant, in Almeida-Sanchez v. United States, 413 U.S. 266, 93 S.Ct. 2535, 37 L.Ed.2d 596 (1973). The officers’ observations of the boat produced articulable facts and inferences amounting to reasonable suspicion that a load of contraband, carried over seas, was aboard. Under the BrignoniPonce test, therefore, they had sufficient grounds for a limited investigatory intrusion but not for a search or further detention. The officers were thus well within Brignoni-Ponce authority in pursuing the vessel and in approaching Whitmire and Williams for identification and registration documents. Their suspicions justifiably increased when the two were unable to produce proper documents, but, suspicious as these facts may have been, they did not amount to probable cause necessary to justify a search of the vessel. Probable cause did not accrue until after the officer boarded to search and smelled the marijuana. If the boarding for purposes of inspection or search is to be upheld, then, it must be on some other ground, despite the unfortunate circumstance that, had their boat been fast enough to have overhauled the Nova before it reached home and appellants had disembarked, the officers could have boarded the vessel — and thus smelled the marijuana — in the proper course of their Brignoni-Ponce investigation.

We must therefore explore the fourth amendment principles applicable when customs officers’ right to board and inspect or search a vessel rests solely on 19 U.S.C. § 1581(a). Read literally, that statute grants extremely broad authority, the full reaches of which some courts have thought incompatible with fourth amendment protections. It reads as follows:

Any officer of the customs may at any time go on board of any vessel or vehicle at any place in the United States or within the customs waters . . . and examine the manifest or other documents and papers and examine, inspect, and search the vessel or vehicle and every part thereof and any person, trunk, package, or cargo on board, and to this end may hail and stop such vessel or vehicle, and use all necessary force to compel compliance.

We held in United States v. Freeman, supra, that section 1581 provided authority, reasonable under fourth amendment standards, for discretionary customs boardings for routine safety and document checks of any vessel within customs waters, a term defined by 19 U.S.C. § 1401(j), as to an American vessel, as “the waters within four leagues [12 nautical miles] of the coast of the United States.” Recently, in United States v. Whitaker, 592 F.2d 826 (5th Cir. 1979), we found the same section 1581 boarding rights — the right to board even absent reasonable suspicion — as to any vessel sighted in customs waters and followed into intercoastal waterways prior to boarding. The instant facts require us to go further and explore the constitutional limits of customs officers’ section 1581 powers over a boat initially sighted in intercoastal waters. Because we feel it appropriate to explore the application to customs searches of the Supreme Court’s decision in Almeida-Sanchez v. United States, 413 U.S. 266, 93 S.Ct. 2535, 37 L.Ed.2d 596 (1973), in which the Court outlawed roving searches by the Border Patrol, it is best to begin by placing boat searches in their historical context in fourth amendment analysis.

A. Early Doctrine Regarding Boat Searches.

The earliest Supreme Court boat cases did not deal with fourth amendment issues explicitly. The Court seemed to assume that United States agents might approach, stop and board American flag and other vessels on the high seas. The debate, often in a context of who got the spoils, focused instead on whether there had been probable cause to seize the vessel and force it into an American port. See, e. g., Murray v. Schooner Charming Betsy, 2 Cranch 64, 6 U.S. 64, 2 L.Ed. 208 (1804); The Apollon, 9 Wheat. 362, 22 U.S. 362, 6 L.Ed. 111 (1824).⁶ Perhaps the explanation for this approach lies in the Court’s early focus on property rights as a principle of fourth amendment exegesis. In Boyd v. United States, 116 U.S. 816, 6 S.Ct. 524, 29 L.Ed. 746 (1886), for instance, the Court enforced the fourth amendment by disapproving a court order directing a man to produce his books and papers to be used as evidence against him. The majority considered that order the equivalent of an unreasonable search for those personal effects. Contrasting such personal items with other objects of search, the Court stated:

The search for and seizure of stolen or forfeited goods, or goods liable to duties and concealed to avoid the payment thereof, are totally different things from a search for and seizure of a man’s private books and papers for the purpose of obtaining information therein contained, or of using them as evidence against him. The two things differ toto coelo. In the one case, the government is entitled to the possession of the property; in the other it is not. The seizure of stolen goods is authorized by the common law; and the seizure of goods forfeited for a breach of the revenue laws, or concealed to avoid the duties payable on them has been authorized by English statutes for at least two centuries past; and the like seizures have been authorized by our own revenue acts from the commencement of the government.

The first statute passed by congress to regulate the collection of duties, the act of July 31,1789, (1 St. 43), contains provisions to this effect. As this act was passed by the same congress which proposed for adoption the original amendments to the constitution, it is clear that the members of that body did not regard searches and seizures of this kind as “unreasonable,” and they are not embraced within the prohibition of the amendment. So, also, the supervision authorized to be exercised by officers of the revenue over the manufacture or custody of excisable articles, and the entries thereof in books required by law to be kept for their inspection, are necessarily excepted out of the category of unreasonable searches and seizures. . . . But, when examined with care, it is manifest that there is a total unlikeness of these official acts and proceedings to that which is now under consideration. In the case of stolen goods, the owner from whom they were stolen is entitled to their possession, and in the case of excisable or dutiable articles, the government has an interest in them for the payment of the duties thereon, and until such duties are paid has a right to keep them under observation, or to pursue and drag them from concealment; and in the case of goods seized on attachment or execution, the creditor is entitled to their seizure in satisfaction of his debt; and the examination of a defendant under oath to obtain a discovery of concealed property or credits is a proceeding merely civil to effect the ends of justice, and is no more than what the court of chancery would direct on a bill for discovery. Whereas, by the proceeding now under consideration, the court attempts to extort from the party his private books and papers to make him liable for a penalty or to forfeit Ms property.

116 U.S. at 623-24, 6 S.Ct. at 528-529.⁷

Enacted by the very Congress that proposed the fourth amendment, the revenue statutes mentioned by the Court are the direct predecessors of 19 U.S.C. § 1581. The initial statute passed in 1789 provided customs officers with authority to enter any vessel in which they had “reason to suspect” dutiable goods were concealed and therein to search for and seize such goods. Upon “cause to suspect” concealment of goods in a “dwelling house, store, building,” a warrant was to be obtained before search.⁸ By 1815, the reasonable suspicion requirement had been deleted for a customs boarding and search of “any ship, vessel, boat, or raft,” though Congress retained that requirement for searches of carriages, vehicles, persons traveling on foot, and beasts of burden.⁹ None of these provisions contained a requirement that customs officers know or be fairly certain that the searched vessel had just come from foreign waters.

In 1924 in Carroll v. United States, 267 U.S. 132, 45 S.Ct. 280, 69 L.Ed. 543, the Supreme Court reviewed these statutes, noting that none had ever been attacked as unconstitutional and stating that the Court had treated a subsequent version as operative in an earlier decision. Boat searches were not before the Court in Carroll, however. Nodding in the direction of the property concepts expressed in Boyd, the Court there validated the warrantless search of a car for contraband liquor that was subject to forfeiture or destruction. The customs statutes were cited to support the longstanding distinction drawn for fourth amendment purposes between dwellings and movable vehicles, since the latter might flee the jurisdiction before a warrant could be obtained. Though there was support for an additional distinction between boats and cars in the various statutes, the Court did not allude to one. Instead, the distinction between searches of those crossing an international boundary, where no probable cause was required, and searches of those “lawfully within the country, entitled to use the public highways” where interruption or search was forbidden absent probable cause, was seemingly extended to vessels as well. Because, however, the pronouncement as to boats was not necessary to the decision and because the facts that were deemed to constitute probable cause in that case would probably not support today’s more lenient reasonable suspicion standard,¹⁰ Carroll should not preclude our fresh examination and balancing of the interests involved in customs stops, inspections and searches of vessels in our intercoastal and coastal waters.¹¹

B. Toward a Balancing of Interests to Determine Reasonableness in the Boat Search Context.

Those who challenge customs or coast guard searches of boats regularly raise the analogy of Almeida-Sanchez v. United States in an effort to have us invalidate the use of evidence obtained in those searches. In Almeida-Sanchez the Supreme Court held that since the intrusion was not at the border or its functional equivalent, probable cause was required for searches of automobiles by roving Border Patrol agents, despite broader statutory authority. The argued analogy possesses considerable surface appeal: a seemingly-legal vessel is hailed by a roving customs or coast guard boat and boarded and searched for drugs on the “pretext” of the need to check documents or safety equipment.¹² We think it proper, however, to resist the urge to impose uncritically on boat searches the set of standards governing auto searches on our internal highways. In the wake of Almeida-Sanchez the Court surveyed a wide range of interests and circumstances before striking various fourth amendment balances in the context of policing the nation’s land boundaries.¹³ The importance of the nation’s interests in regulating its commerce with other nations, policing its borders, and regulating the conduct of the boats protected under its flag makes such a detailed survey imperative here. We offer these reflections as much to aid in developing a mode of analysis that is cognizant of the unique maritime context as to determine the fourth amendment status of the search here challenged.

We commence by sounding the varying degrees of privacy one may reasonably expect aboard vessels.¹⁴ It is true that certain boats may be “home” to certain sailors,¹⁵ but that is not uniformly or invariably the case. The type of vessel and, as suggested in Rakas v. Illinois,¹⁶ the particular area of it that is searched may be crucial in determining whether a sailor’s expectations are reasonable. Relatively high levels of privacy might be accorded, for instance, to those aboard a houseboat¹⁷ or to the crew’s living quarters on a tanker that travels for months at sea.¹⁸ By contrast it is difficult to see that a crew member might legitimately claim privacy on the open deck of a fishing smack or in the hold of a cargo vessel available for hire.¹⁹ A harder case to assess, reserved for another day, is the enclosed area of a yacht or large sailboat which might be in use as a living area during an extended cruise or might also serve as a mere cargo container on an illicit drug run.

The next factor to be examined is the degree of intrusion on protected privacy that is caused by the challenged governmental action. Pursuant to the generous section-1581 authority Congress has attempted to accord customs officers, the intrusion might range from a completely random stop routinely to check documents or safety equipment, through a limited inspection of potentially cargo-bearing cavities to match their contents against the manifest or to determine whether there are contents that should have been reported, to a no-holds-barred search of every nook, cranny, mattress and trunk aboard. Unless the vessel voluntarily reports to a customhouse, however, there is probably little regularity, in the times and places of these encounters.²⁰ Many intrusions might therefore be deemed arbitrary, with officers claiming “unbridled discretion” to stop boats on the least whim. The presence of such wide discretion requires caution in fourth amendment analysis,²¹ but it is not invariably fatal to the constitutionality of an intrusion.²² Moreover, the potential for provoking fear in the stopped citizen that worried the Supreme Court in recent automobile search cases ²³ seems less likely to attend maritime intrusions. The heavy overlay of maritime law and the long practice of regulatory stops, inspections and searches by these officers, while not dispositive of the constitutional question, must, in addition to coloring our constitutional balancing,²⁴ lead us to suspect that the approach of a governmental vessel to a private boat on the waters may not be as startling as that of a roving Border Patrol car to an unsuspecting citizen cruising our national highways. On the contrary, boaters customarily expect the presence of and rely on aid from coast guard or customs vessels when facing the perils of a maritime environment. It is quite possible, therefore, that smugglers form the main class of sailors in whom the approach of these officers provokes anxiety.

As a final observation in assessing the private interests to be balanced against the public interests discussed below, we note that though sailors on occasion live on their boats, no one lives in a fixed location out in the water. In contrast to the situation involved in extended border searches on land, recognizing relatively generous customs powers over boats does not create a permanent class of citizens who, solely because of where they live, are subject to greater intrusions than are other citizens. While those who venture out into customs waters and interconnecting waterways for business or pleasure may certainly lay claim to the right to proceed unhindered by unreasonable government intrusions, their claims are perhaps less weighty in gauging reasonableness than are those of citizens living near our land borders, with no choice but to use roads in making their daily rounds. In this connection, however, the locale of the detention and thus the frequency with which purely domestic traffic is halted may well bear on the scope of intrusion that is reasonable or on the degree of suspicion, if any, required to validate certain intrusions.

Turning to the governmental interests implicated in intrusions on vessels, it is evident that the major interest lies in executing congressional determinations under its “plenary powers” to regulate commerce with foreign nations.²⁵ But the revenue and regulatory concerns of Congress in this area are not exhausted by application to vessels known for a certainty to have come directly from foreign ports. There is a valid governmental interest in determining whether a boat has come from abroad or whether it contains secreted dutiable goods.²⁶ Moreover, the interest in monitoring exports also may necessitate inspections of vessels in the absence of border crossing facts.²⁷

In addition to customs concerns, some interference with boats is also warranted by the national interests in regulating vessels that fly the American flag²⁸ and in ensuring orderly travel under the navigation laws. Unlike border patrol agents, customs officers in patrolling our intercoastal and customs waters customarily enforce such laws and coast guard regulations under authority shared with other law enforcement agencies.²⁹ Effective enforcement of these long-standing maritime regulatory schemes requires according officials fairly generous powers to act when an occasion presents itself — when a vessel comes into view. Oceanic frontiers of their nature are more difficult to police than are land boundaries. It is impractical to picket the actual maritime border three miles out at sea.³⁰ It is also unreasonable to expect customs agents to observe the major portion of vessels crossing that imaginary line or to pair up sightings before and after most crossings. There may be no effective alternative to random stopping³¹ and some degree of imspection or search because of the sea’s vastness and the difficulty of locating objects traveling on it. Moreover, there are no roads in customs waters to channel the flow of incoming persons. Thus, customs agents have fewer opportunities than their land-based counterparts to establish “traffic checkpoints” and thus monitor entrants who fail to stop voluntarily at a customhouse.³²

We have already judged it reasonable under the fourth amendment for customs officials to stop any vessel found in customs waters and board it for a document check and safety inspection.³³ We have also found that, pursuant to their virtually identical statutory authority, coast guard officers may board American vessels on the high seas to do a document and safety check and look for obvious customs violations.³⁴ On the other hand, we have re-quired at least reasonable suspicion that a houseboat is within the area of customs concern before it may be boarded while docked in an inland marina.³⁵ Considerations such as those outlined above might in the proper case support a blanket authorization of customs personnel to briefly halt vessels sighted in intereoastal waterways for random document and safety boardings. They might also support a rule that, upon reasonable suspicion, customs officers may board boats found there or in customs waters and do a limited inspection for obvious customs violations by viewing the interior of large, potentially cargo-bearing cavities of cargo or fishing vessels and pleasure boats whose construction reveals a capacity to conceal significant amounts of cargo.³⁶ We do not presently face these situations. We need only decide whether the fourth amendment allows the boarding of a pleasure craft, sighted initially in intercoastal waters, as to which officers have a reasonable suspicion of a customs violation — a boarding that occurred after an unsatisfactory document check on shore. Applying the mode of analysis suggested above, we note that the area initially invaded — the cockpit of a 25-foot pleasure craft — is generally visible to those passing nearby. The privacy expected in such an area would be minimal. In addition, the two appellants were no longer aboard; thus the boarding infringed their privacy interests, as contrasted to their property interests, even less severely. The government interests being vindicated, on the other hand, were extremely important. The officers observed appellants flout traffic rules and possessed other facts leading them reasonably to suspect a customs violation and perhaps a theft. Though there was a chance, especially given the place of initial sighting in intercoastal waters, that appellants’ outing would turn out to be only a domestic journey, it was reasonable in light of those strong suspicions to detain the men pending a further brief inspection aboard, even absent known border crossing facts. We need not speculate whether their suspicion would also have justified a brief inspection of the closed hull cavity since, as soon as the boat was boarded, probable cause arose, fully justifying the more detailed search. If exigent circumstances be required, they were present as well. As the search was within statutory authority and was constitutional, its fruits were admissible in evidence.

III. Sufficiency of Evidence on Williams’ Possession Charge.

Having been acquitted on the importation charge, Williams argues that there is insufficient evidence to sustain his conviction of possession with intent to distribute. The evidence against him is not overwhelming, but we believe it sufficient when assessed under the relevant standard of review. We must view the evidence and all reasonable inferences arising therefrom in the light most favorable to the government, the prevailing party. Glasser v. United States, 315 U.S. 60, 80, 62 S.Ct. 457, 86 L.Ed. 680 (1942). We may not substitute our view of the evidence for that taken by the trier of fact since the test on appeal is whether the trier of fact might reasonably conclude that the circumstantial evidence relied on excluded every reasonable hypothesis of innocence. United States v. Sidan-Azzam, 457 F.2d 1309 (5th Cir. 1972).

The evidence indicates that Williams was observed in a 25-foot Nova, speeding into an inland waterway from the “ocean side” of a cut at 8 a. m. on a cold, wet day. He was positioned in the cockpit of the boat, a point at which the odor of the 1500 pounds of marijuana aboard was “overpowering.” He was wearing a soaking wet “Bimini” sweatshirt identical to that of Whitmire whose presence in Bimini just the day before was confirmed by a fuel receipt. He argues that his conviction should be overturned because “mere presence” or merely being a passenger in proximity to contraband is not enough to constitute constructive possession. See United States v. Ferg, 504 F.2d 914, 916-17 (5th Cir. 1974); Williams v. United States, 361 F.2d 280, 281 (5th Cir. 1966). We have also held, however, that where other circumstantial evidence, such as one’s attitude, conduct or relationship to the driver, is sufficiently probative, proximity to contraband coupled with inferred knowledge of its presence will support a finding of guilt on such charges. United States v. Christian, 505 F.2d 94, 96 (5th Cir. 1974). In United States v. Canada, 459 F.2d 687 (5th Cir. 1972), we sustained the conviction of a passenger in a car found to have four large sacks of marijuana in its trunk during a border search. In addition to this proximity to the contraband and the defendant’s “relationship” to the person driving the car, the sole additional evidence supporting an inference of conspiracy and marijuana transportation was the testimony of a border patrol agent, inconsistent with his own earlier testimony, that both defendants had told him that the marijuana had been crossed upriver from the port of entry. In United States v. Christian, supra, we concluded that the evidence that a woman was freely present on a boat for an eight-day sailing trip, during which time bags of marijuana were boarded and placed in the cabin in plain view, was sufficient to support a jury’s finding of possession and importation. We held that in addition to her presence on the boat, the length of the voyage, the relationship between the defendant and a co-defendant whom she had married by the time of trial, and the large quantity of marijuana involved which made it indisputable that she had knowledge of the contraband were factors the jury was entitled to consider.

In the case at bar the trial judge could reasonably have concluded that the presence of such a large amount of marijuana in a boat the size of the Nova can hardly have escaped the attention of Williams. Moreover, the hour of their apprehension, their probable point of departure, their great speed to reach home port, and their flouting of boating regulations all contribute to the inference that Williams must have realized and shared the furtive object of Whitmire’s enterprise. We find the evidence of his guilt sufficient.

AFFIRMED.

1 . Despite some favorable language therein, Wingo v. Wedding, 418 U.S. 461, 94 S.Ct. 2842, 41 L.Ed.2d 879 (1974), does not control this case. In that opinion the Supreme Court decided that, purely as a matter of construing the habeas corpus provisions, referral to a magistrate, coupled with later judicial review of the record, would not satisfy the statutory requirement that federal judges conduct habeas evidentiary hearings.

Neither does our opinion in McKinney v. Parsons, 488 F.2d 452 (5th Cir. 1974), control. That case is distinguishable because the trial judge failed to personally to inspect the allegedly obscene material — a task uniquely his under first amendment doctrine. See Blount v. Rizzi, 400 U.S. 410, 91 S.Ct. 423, 27 L.Ed.2d 498 (1971); Freedman v. Maryland, 380 U.S. 51, 85 S.Ct. 734, 13 L.Ed.2d 649 (1965).

We are aware that our holding differs with that of the Seventh Circuit in United States v. Raddatz, 592 F.2d 976 (7th Cir. 1979). The court there held that due process requires a district judge not merely to review a written record but to rehear all testimonial evidence in any criminal case where “credibility evidence is central to the determination of a material issue of fact.” We think this conclusion an overly restrictive view of the magistrate’s role. In our view, the discretionary power of the district court to adopt or to reject the magistrate’s recommendation and personally to hear any evidence it finds necessary to deciding the issue adequately protects a defendant’s due process rights.

2 . The standard has been variously stated from time to time in this circuit. Before, and to some extent after, Almeida-Sanchez v. United States, 413 U.S. 266, 93 S.Ct. 2535, 37 L.Ed.2d 596 (1973), we required reasonable suspicion of a customs violation and/or a “nexus” to the border for extended border searches, those occurring other than immediately at the border. The required nexus was present when officers formed a reasonable suspicion that the person or vehicle searched had itself crossed a border or had been in contact with some other person or thing that had crossed over. United States v. Hill, 430 F.2d 129 (5th Cir. 1970); United States v. Lonabaugh, 494 F.2d 1257 (5th Cir. 1973); United States v. Bowman, 502 F.2d 1215 (5th Cir. 1974).

In United States v. Brennan, 538 F.2d 711 (5th Cir. 1976), cert, denied, 429 U.S. 1092, 97 S.Ct. 1104, 51 L.Ed.2d 538 (1977), the first exhaustive application of Almeida-Sanchez to customs searches, a panel stated that the border nexus requirement was no longer sufficient justification for a customs search made on the basis of reasonable suspicion instead of full probable cause. 538 F.2d at 719 n.9. The panel went on to conclude that neither border patrol nor customs agents may search on less than probable cause at a point other than the border or its functional equivalent. And to establish as a functional equivalent of the border a location such as the Melbourne, Florida, airport, the panel required (1) a “high degree of probability that a border crossing took place” and (2) a regularity to the intrusion, as where, for example, a flight lands at an international terminal and all passengers know in advance that they will be subject to an inspection that proceeds for each person in substantially the same manner. 538 F.2d at 715-16. Each factor was found lacking in the plane search there reviewed.

Though Brennan was couched in broad language, subsequent panels have declined to follow its full implications. In United States v. Ivey, 546 F.2d 139 (5th Cir.), cert, denied, 431 U.S. 943, 97 S.Ct. 2662, 53 L.Ed.2d 263 (1977), for instance, a nonregular search of an airplane was upheld as a border search because a nexus between it and the border was established with “reasonable certainty” or “a high degree of probability.” See also United States v. Adams, 569 F.2d 924 (5th Cir. 1978), which applied the latter formulations to validate a nonregular search of a mud-spattered van stopped on a road closely paralleling the Rio Grande River. The most recent panel to consider the problem has derived the helpful distinction between particular places and particular searches in applying the doctrine of the functional equivalent of the border. The panel also departed from past articulations regarding individual searches to hold that such a search is not a valid border search unless it appears by a preponderance of the evidence, presumably at the time of the after-the-fact suppression hearing, that a border crossing has occurred. United States v. Johnson, 588 F.2d 147, 154 (5th Cir. 1979).

After Brennan, the reasonable suspicion test continued unabated in the boat search context, either without citation of Brennan or after distinguishing it as an airplane case. See, e. g., United States v. Williams, 544 F.2d 807 (5th Cir. 1977) (no reasonable belief that houseboat fell within customs’ area of concern because no evidence it had entered international waters or contacted other vessels in such waters). And see United States v. Fogelman, 586 F.2d 337, 343 (5th Cir. 1978), in which the reasonable suspicion test was enunciated in upholding the search of a truck that had been in contact with a vessel known with certainty to have crossed the border.

Because the principles from automobile, aircraft and boat cases do not necessarily translate literally from their discrete factual settings to the others, we also will apply the reasonable suspicion test, leaving for another day whether, as questioned in the text at subsequent points, the requirement of a border crossing nexus is sensible in the boat search context.

3 . United States v. Ingham, 502 F.2d 1287 (5th Cir. 1974), cert. denied, 421 U.S. 911, 95 S.Ct. 1566, 43 L.Ed.2d 777 (1975).

4 . See cases in note 2, supra.

5 . The Ninth Circuit in particular has developed the application to harbors of this concept ratified by the Supreme Court in Almeida-Sanchez. See, e. g., United States v. Solmes, 527 F.2d 1370 (9th Cir. 1975); United States v. Tilton, 534 F.2d 1363 (9th Cir. 1976).

6 . Carmichael, At Sea With the Fourth Amendment, 32 U.Miami L.Rev. 51 (1977), surveys additional Supreme Court case law regarding early boat searches. The focus of the article, as with the early cases, is primarily on boat searches away from the coastline.

7 . In Warden v. Hayden, 387 U.S. 294, 87 S.Ct. 1642, 18 L.Ed.2d 782 (1967), the Court finally repudiated the property notions transmitted in Boyd, recognizing that privacy, the fundamental interest sought to be protected by the fourth amendment, may be infringed even during searches for items to which the searched person has no legal claim, for instance, contraband.

8 . Customs Act, Ch. 5, § 24, 1 Stat. 43 (1789). Similar provisions were carried forward in the Act of August 4, 1790, Ch. 35, § 48, 1 Stat. 145, 170; and in the Act of March 2, 1799, Ch. 22, § 68, 1 Stat. 627, 677-78.

9 . Act of March 3, 1815, Ch. 94, §§ 1, 2, 3 Stat. 231-32. An earlier statute regarding the enrollment and licensing of ships to be employed in fishing and “the coasting trade” (maritime commerce between American port cities) had also authorized collections agents without a requirement of particularized suspicion to board vessels within or without their districts and “[there] to inspect, search and examine, and if it shall appear, that any breach of the laws of the United States has been committed” for which the vessel or goods aboard would be liable to forfeiture, to seize the same. Act of Feb. 18, 1793, Ch. 8, § 27, 1 Stat. 305, 315. The Act of March 2, 1799, supra note 8, in addition to § 68, which authorized boarding and search upon reason to suspect concealment of dutiable goods, also provided for boarding and search without suspicion of ships in any United States port, or within four leagues of the coast “if bound to the United States.” Officers were explicitly authorized to have “free access to the cabin and every other part” of the vessel. If boxes or trunks were found in the cabin or in other places away from the cargo, the officers were to make note of it and had power to seal them and report the matter to inspectors at the port to which the ship was bound. Ch. 22, § 54, 1 Stat. 668. See Carmichael, supra note 6, for review of later statutes.

10 . The facts in Carroll were that undercover agents had attempted to buy liquor from the two defendants in September 1921, but the sale' fell through because the defendants’ source was away from his Grand Rapids location. In October, the agents were patrolling the highway between Grand Rapids and Detroit when they sighted the defendants’ car heading toward Detroit. The agents attempted to tail, but lost, the car. Two months later, the agents again saw the defendants headed in their car toward Grand Rapids. The agents stopped them and in the subsequent search found liquor stashed behind the car seat upholstery.

The agents certainly had a hunch that proved correct, but it is doubtful that even the initial stop would be valid today under BrignoniPonce standards because of the passage in time since defendants’ last known illegal attempt and because the outward appearance of the car and its occupants seems to have borne no “clues” as to its illegal contents. Compare the characteristics rejected as insufficient for a valid auto stop by this court in, e. g., United States v. Escamilla, 560 F.2d 1229 (5th Cir. 1977); United States v. Lopez, 564 F.2d 710 (5th Cir. 1977).

11 . We find support for this reexamination in our review of the boat cases from the years immediately following Carroll. In Maul v. United States, 274 U.S. 501, 47 S.Ct. 735, 71 L.Ed. 1171 (1927), the parties conceded and the Court seemed to assume that the Coast Guard would have plenary authority to board, search and seize boats found anywhere on the landward side of the 12-mile line. The majority did caution, however, that the case did not involve “an exercise of asserted authority to board and search a vessel, domestic or foreign, for the purpose of detecting and thwarting intended smuggling.” 274 U.S. at 503-04, 47 S.Ct. at 736. A special concurrence by Mr. Justice Brandéis, joined by Mr. Justice Holmes, includes the telling remark that “there is no limitation upon the right of a sovereign to seize without a warrant vessels registered under its laws, similar to that imposed by the common law and the constitution upon the arrest of persons and upon the seizure of ‘papers and effects.’ ” 274 U.S. at 524, 47 S.Ct. at 744. He cites Carroll for this proposition, but it seems more a revival of the Boyd idea that certain seizures are per se reasonable by their very natures. This categorical view is ambiguously muted in United States v. Lee, 274 U.S. 559, 47 S.Ct. 746, 71 L.Ed. 1202 (1927), in which Mr. Justice Brandéis writes for the whole Court. In

Lee the government argued that, upon probable cause, the Coast Guard had authority to visit, search and seize an American vessel beyond the 12-mile zone. It alternatively argued that the fourth amendment did not apply to small motor boats not apparently used as a place of residence or that the search had been delayed until within “the territorial limits” of the United States where search was “clearly valid.” Mr. Justice Brandéis replied that the government contentions “in the main” were “well founded” and proceeded to offer so many alternative grounds for affirming the search (there was probable cause, there was no search but plain view instead, search within U.S. territory was in any event authorized) that it is impossible to glean any potential reservations regarding plenary search powers closer to shore.

The lower courts were more explicit in this period. District courts tended to be restrictive as to customs or Coast Guard searches. See, e. g., United States v. Coppolo, 2 F.Supp. 115 (D.N.J.1932); United States v. Powers, 1 F.Supp. 458 (E.D.N.Y.1932). In a widely cited opinion, a New York district court questioned the applicability of § 1581 to pleasure boats and intimated that search of such boats was not authorized thereunder in the absence of facts leading agents to believe the boat carried cargo from a foreign port. Fish v. Brophy, 52 F.2d 198 (S.D.N.Y.1931). That court also read Carroll as requiring probable cause for boarding and searching any boat not known to have come from foreign waters. 52 F.2d at 201. The Second Circuit repudiated both of these positions, however. In United States v. Wischerth, 68 F.2d 161 (1933), on facts closely resembling ours, the court ruled that probable cause was not necessary for a search of a vessel, low in the water and apparently carrying cargo, which was first sighted entering New York harbor through “the Narrows.” See also The Atlantic, 68 F.2d 8, 9-10 (2d Cir. 1933) (§ 1581 right to stop and search is not confined to commercial vessels but also applies to “private” or “pleasure” vessels); The Pueblos, 77 F.2d 618 (2d Cir. 1935) (probable cause not required for search of vessel running without lights up the coastline); Await v. United States, 47 F.2d 477 (3d Cir. 1931); Alksne v. United States, 39 F.2d 62 (1st Cir.), cert, denied, 281 U.S. 768, 50 S.Ct. 467, 74 L.Ed. 1175 (1930); Arch v. United States, 13 F.2d 382 (5th Cir. 1926).

12 . Carmichael, supra, note 6 at 54-55.

13 . As mentioned above, in United States v. Brignoni-Ponce, the Court validated limited investigatory stops by a roving boarder patrol upon the presence of reasonable suspicion. In United States v. Martinez-Fuerte, 428 U.S. 543, 96 S.Ct. 3074, 49 L.Ed.2d 1116 (1976), the Court ruled that, even absent suspicion, such a limited stop might be made at a traffic checkpoint not the functional equivalent of the border. But searches may not be performed during either type stop except upon probable cause or consent. Brignoni-Ponce, 422 U.S. at 882, 95 S.Ct. 2574; United States v. Ortiz, 422 U.S. 891, 95 S.Ct. 2585, 45 L.Ed.2d 623 (1975).

14 . Under Katz v. United States, 389 U.S. 347, 88 S.Ct. 507, 19 L.Ed.2d 576 (1967), one’s reasonable expectation of privacy is the touchstone of fourth amendment analysis.

15 . United States v. Cadena, 588 F.2d 100, 101 (5th Cir. 1979), denying petition for rehearing.

16 . 435 U.S. 922, 99 S.Ct. 421, 58 L.Ed.2d 387 (1978).

17 . Though this court’s opinion in United States v. Williams, 544 F.2d 807 (5th Cir. 1977), does not explicitly allude to the degree of privacy expected aboard the gerrybuilt houseboat, the fact that the appellants were residing aboard the boat may have influenced the decision.

18 . United States v. Whitaker, 592 F.2d 826 (5th Cir. 1979); United States v. Cadena, 585 F.2d 1252, 1264 n.27 (5th Cir. 1978), rehearing denied, 588 F.2d 100 (1979).

19 . United States v. Williams, 589 F.2d 210 at 214 (5th Cir. 1979); Whitaker, supra at 829-830.

20 . Regularity in the sense discussed in the Brennan opinion, supra note 2, played a significant role in the Court’s decision in United States v. Martinez-Fuerte, supra.

21 . See United States v. Brignoni-Ponce, 422 U.S. at 882-83, 95 S.Ct. 2574.

22 . In United States v. Biswell, 406 U.S. 311, 92 S.Ct. 1593, 32 L.Ed.2d 87 (1972), and Colonnade Catering Corp. v. United States, 397 U.S. 72, 90 S.Ct. 774, 25 L.Ed.2d 60 (1970), the Court upheld administrative inspection schemes that apparently allowed enforcement officers broad discretion to choose the target and time of search. In Marshall v. Barlow's, Inc., 436 U.S. 307, 98 S.Ct. 1816, 56 L.Ed.2d 305 (1978), its most recent treatment of the administrative search exception to the warrant requirement, the Court refused to allow discretionary nonprobable cause inspections by OSHA absent an administrative warrant. But the Court explicitly left open the status of warrantless searches made under other regulatory statutes, noting that the reasonableness of such schemes will depend upon “the specific enforcement needs and privacy guarantees of each statute.” 436 U.S. at 321, 98 S.Ct. at 1825, 56 L.Ed.2d at 317.

23 . United States v. Ortiz, 422 U.S. at 894-95, 95 S.Ct. 2585; United States v. MartinezFuerte, 428 U.S. at 558-59, 96 S.Ct. 3074.

24 . United States v. Ramsey, 431 U.S. 606, 619 n.14, 97 S.Ct. 1972, 52 L.Ed.2d 617 (1977), citing Carroll v. United States, 267 U.S. at 149, 45 S.Ct. 280.

25 . United States v. Ramsey, supra, 97 S.Ct. at 1979. In Ramsey, the Court validated customs searches of letters arriving from overseas under a statute allowing search upon “reasonable cause to suspect” that they contained merchandise subject to duty or contraband. In the course of developing its border-search rationale, the majority read the early customs statutes as “border search” enactments. None of the statutes nor the Boyd case, supra, required knowledge of a border crossing as a prerequisite for use of the plenary customs search powers, however. Since the basic notion of customs entails the regulation of ingress and egress of goods from the country, the border search principle undoubtedly forms the underpinning for those statutes granting authority to search. This does not necessarily conclude the inquiry whether an official must possess objective information about a boat’s recent whereabouts before Congress may constitutionally authorize exercise of its plenary customs powers.

26 . Certain pleasure craft are subject to special regulation along with larger fishing or cargo vessels. See, e. g., the statutes pertaining to yachts at 46 U.S.C. § 107 et seq.

27 . In Companía Naviera Bascongada v. United States, 354 F.2d 935 (5th Cir. 1966), for instance, this court examined customs statutes regarding vessels “bound for” the United States as applied to a foreign vessel that made a port call in New York City, there fulfilling all customs regulations, and then made additional stops in other American port cities. Though not dealing with fourth amendment issues, the court upheld the right of customs agents to board, check the manifest, inspect the vessel, and inventory its contents at each successive United States port. The latter intrusions would probably not qualify as extended border searches under current tests, which seek to minimize the elapsed time and distance from the known border crossing and generally require some surveillance to validate an intrusion once a vehicle has left the border. See, e. g., United States v. Fogelman, 586 F.2d 337 (5th Cir. 1978). Yet the nation had a valid customs interest to protect in the Compañía situation: goods had been onloaded in New York that remained duty free only if resold or used aboard, and the crew had been reselling them illegally at subsequent stops along the coast.

28 . This interest would not be present in approaches to foreign vessels, though many of the other considerations expressed herein would apply in that context, which is additionally complicated by treaties and international law. See, e. g., Cadena, supra; United States v. Cortes, 588 F.2d 106 [No. 78-5413 (5th Cir. Feb. 6, 1979)]; United States v. Postal, 589 F.2d 862 (5th Cir. 1979).

29 . Under a “two-hat” system of delegation, coast guard officials function as customs agents, and customs officials may be authorized to assist the coast guard. See, e. g., United States v. Warren, 578 F.2d 1058, 1066-67 (5th Cir.), en banc granted on other grounds, 586 F.2d 608 (1978); United States v. Byrd, 483 F.2d 1196, 1198 (5th Cir. 1973); United States v. Thompson, 475 F.2d 1359, 1362-63 (5th Cir. 1973).

30 . United States v. Freeman, 579 F.2d 942, 946 (5th Cir. 1978).

31 . Because of the physical characteristics of especially large boats, it may be difficult for observers to form particularized suspicions based on sluggish handling and other factors like those present in this case.

32 . Recognizing the difficulties of stemming the flow of illegal entrants along our land border with Mexico, courts have approved the use of permanent checkpoints along roads leading from the border. At these places federal officers may briefly detain and question all passers. Martinez-Fuerte; United States v. Santibanez, 517 F.2d 922 (5th Cir. 1975). If the checkpoint fulfills certain criteria, it is considered a functional equivalent of the border, and routine but limited searches of a vehicle’s large cavities may be conducted. United States v. Alvarez-Gonzaiez, 561 F.2d 620 (5th Cir. 1977). At such locations, significant amounts of purely domestic traffic may routinely be interdicted as a result of striking the necessary balance between a citizen’s reasonable expectations of privacy and the pressing law enforcement need to monitor positions of international traffic not otherwise controllable.

The difficulties of similar routine monitoring on many of our waterways and the other factors discussed herein may serve to distinguish maritime law enforcement practices from, e. g., the random stops of automobiles recently prohibited in Delaware v. Prouse, -U.S.-, 99 S.Ct. 1391, 59 L.Ed.2d 660 (1979).

33 . United States v. Freeman, 579 F.2d 942 (5th Cir. 1978).

34 . United States v. Warren, 578 F.2d 1058 (5th Cir.), rehearing en banc granted on other grounds, 586 F.2d 608 (1978).

35 . United States v. Williams, 544 F.2d 807 (5th Cir. 1977). This opinion is somewhat ambiguous as to the minimum degree of suspicion required to validate a boarding. It may be read as allowing boarding if officers have a reasonable suspicion either that a customs violation is being committed aboard or that the boat has been in, or in contact with other boats that have been in, international waters.

36 . In United States v. Odom, 526 F.2d 339 (5th Cir. 1976), and Warren, supra, we allowed coast guard officers to enter the hold of American vessels to check the main beam number against the number listed on the vessel’s documents. These inspections were not analyzed as searches to be justified but rather as routine parts of legitimate document checks, themselves the occasions for observing suspicious “cargo” that formed probable cause for the further search. This practice perhaps offers a fertile ground for elaborating the Supreme Court’s observation that “[n]ot every aspect of a routine automobile ‘inspection’ necessarily constitutes a ‘search’ for purposes of the Fourth Amendment.” United States v. Ortiz, 422 U.S. at 897 n.3, 95 S.Ct. at 2589 n.3. In exploring a possible distinction between inspections and searches, we should be mindful that, in the maritime context, the policies behind several traditionally discrete exceptions to the warrant requirement — the border search doctrine, the administrative inspection exception for certain regulated industries, and the automobile-exigent circumstances doctrine— uniquely converge. See Carmichael, supra, n.6 at 104.