OPINION

WIDENER, Circuit Judge:

This case requires us to interpret the regulations that govern fishing for the Atlantic Sea Scallop. Between October 9, 1989 and October 12, 1989, agents of the National Marine Fisheries Service thawed, sampled and counted ten pounds of Atlantic sea scallops from the catch of the Alice Amanda. Based on the count, the agents declared the entire catch illegal and demanded the proceeds from the sale of the catch. The owner, on the advice of counsel, refused to deliver the proceeds.

A month later, the Alice Amanda was boarded and seized by the Coast Guard. See 16 U.S.C. § 1861. The United States then initiated this forfeiture proceeding against the Alice Amanda in the district court. See 16 U.S.C. §§ 1860, 1861. The ship was released on the posting of a $300,000 bond. See 16 U.S.C. § 1860(d)(1)(B). After a bench trial, the district court ordered that the United States recover $25,000 in partial forfeiture of the Alice Amanda.

The Alice Amanda appeals the order of forfeiture. The United States cross-appeals, arguing that the district court abused its discretion by setting the amount of forfeiture too low. We now reverse the order of forfeiture. Accordingly, we decide the government's cross-appeal is without merit.

I

A. Statutory and Regulatory Background

The Magnuson Fishery Conservation and Management Act (the Act), 16 U.S.C. § 1801 et seq., was enacted in 1976 to preserve the nation's fishery resources. The Act establishes an Exclusive Economic Zone in the waters off the coast of the United States, extending from the seaward boundary of each of the coastal States to 200 nautical miles offshore. 16 U.S.C. § 1811. Within this zone, all fish are subject to the exclusive fishery management authority of the United States. The Act is administered by the Department of Commerce, National Oceanic and Atmospheric Administration, National Marine Fisheries Service (the Agency). See 16 U.S.C. § 1802(20). The Act divides the Exclusive Economic Zone into eight regions and creates a Regional Fishery Management Council for each region. 16 U.S.C. § 1852.

A Fishery Management Council's primary responsibility is the preparation and amendment of Fishery Management Plans for various fisheries. The fishery involved in this case is the Atlantic Sea Scallop Fishery, which covers the species Placopecten magellanicus, the Atlantic sea scallop, in the entire Atlantic coast Exclusive Economic Zone. See 50 C.F.R. §§ 650.1 et seq. The New England Fishery Management Council has primary responsibility for the Atlantic Sea Scallop Plan.

After a Council has adopted a Fishery Management Plan, or an amendment to an existing Plan, it forwards the Plan and suggested regulations to the Agency. The Agency then undertakes notice and comment rulemaking procedures established by the Magnuson Act if it proposes to give effect to the Council's action. 16 U.S.C. § 1854(a) and (b). The Magnuson Act also allows the Agency itself to prepare a Fishery Management Plan, or an amendment to a Plan, under certain circumstances. 16 U.S.C. § 1854(c). Regulations to implement such a Plan also must go through notice and comment rulemaking procedures. 16 U.S.C. § 1854(c)(2).

As a final method of controlling fishing practices, the Act provides for the promulgation of emergency regulations by the Secretary without regard to the existence of a Fishery Management Plan. 16 U.S.C. § 1855(c). The Act requires that "[a]ny emergency regulation promulgated ... shall be published in the Federal Register together with the reasons therefor." 16 U.S.C. § 1855(c)(3)(A). Emergency regulations are limited in duration to a maximum of 180 days. 16 U.S.C. § 1855(c)(3)(B).

The regulations governing the Atlantic Sea Scallop Fishery are codified at 50 C.F.R. Part 650. The regulations require vessel permits, control certain aspects of possession and landing and control the minimum size of scallops permitted in possession. See 50 C.F.R. §§ 650.4, 650.7, 650.20, and 650.21. Only the minimum size requirements are at issue here.

The regulations use two size measures. For scallops in the shell, they specify a minimum shell height. 50 C.F.R. § 650.20(a). For shucked scallop meats, the regulations specify a maximum number of meats per pound. Id. This latter measure is known as the "meat count." All other things being the same, a higher meat count indicates smaller scallops. Because the catch of the Alice Amanda had been shucked at sea, meat count, rather than shell height, is the relevant measure in this case.

At the time that the Alice Amanda landed her catch, the maximum allowable meat count was 33 meats per pound. 50 C.F.R. § 620.20(a), (c)(1). The regulations make it unlawful to "[p]ossess, at or prior to the first transaction in the United States, any nonconforming Atlantic sea scallops," 50 C.F.R. § 650.7(a), including those that exceed the maximum allowable meat count.

The procedure for measuring the meat count is governed by 50 C.F.R. § 650.21. This section provides in relevant part as follows:

Compliance with the specified meat count and shell-height standards will be determined by inspection and enforcement up to and including the first transaction in the United States as follows:

(a) Shucked meats. The Authorized Officer will take one-pound samples at random from the total amount of scallops in possession. The person in possession of the scallops may request that as many as ten one-pound samples be examined as a sample group. A sample group fails to comply with the standard if the averaged meat count for the entire sample group exceeds the standard. The total amount of scallops in possession will be presumed in violation of this regulation if the sample group fails to comply with the standard.

As indicated, a catch is presumed to be in violation when the average meat count from samples exceeds the maximum allowable meat count.

The Magnuson Act makes it "unlawful . . . to violate . . . any regulation . . . issued pursuant to" the Act. 16 U.S.C. § 1857(1)(A). The Act provides that violations may be punished by a civil penalty up to $25,000 per violation, 16 U.S.C. § 1858(a), or by forfeiture of the catch, its fair market value, or the fishing vessel. 16 U.S.C. § 1860(a). In this case, the government sought forfeiture of the fishing vessel, the Alice Amanda, under section 1860.

Regulations implementing the Fishery Management Plan for Atlantic Sea Scallops were first enacted in 1982. See 47 Fed. Reg. 20,776 (May 14, 1982). From the outset, the principal management measure for shucked scallops was a maximum meat count. With minor exceptions, the sections defining the manner in which the meat count was to be determined have not changed.

In 1982, when the regulations were adopted, Atlantic sea scallops were landed in one of two ways, in the shell, or shucked on ice. During the fishing voyage, shucked scallops were placed in cheesecloth bags and packed on ice. Because scallops could not be kept fresh on ice for longer periods, the average voyage lasted 10 days.

Around 1988, a new technology, the freezer vessel, began to be used in scallop fishing. The Alice Amanda is a freezer vessel. By 1989, when this case began, the Alice Amanda still was one of only a handful of freezer vessels in the Atlantic sea scallop fleet.

The Alice Amanda is equipped with freezer bins below decks and a plate freezer on deck. Instead of storing scallops on ice, they are quick frozen in the plate freezer then stored below decks in the freezer bins. On the Alice Amanda, scallops are shucked soon after being collected in the ship's dredges. The shucked scallops then are rinsed in a bucket and allowed to drain in a washer, a container equivalent to a large colander. The drained scallops then are packed in cheesecloth bags. Each bag holds sixteen to eighteen pounds of scallops. The bags are placed in the plate freezer, where they are kept at twenty below zero for four hours. The bags of frozen scallops then are transferred to the freezer bins below decks. Because it was able to store the scallops frozen, instead of packed in ice, the Alice Amanda's fishing trips averaged thirty days, twenty days longer than the average trip for a boat storing shucked scallops on ice. The longer trip allows a larger catch, the catch at issue here being 19,668 pounds.

Upon being landed frozen, a catch from the Alice Amanda routinely was unloaded, weighed and then thawed for processing. The scallops were thawed by immersing the frozen mass in a mixture of water and tripolyphosphate. The scallops then were weighed a second time, sorted by size and packaged for sale.

B. Frozen Scallop Sampling Procedure

Beginning in early 1988, Agency personnel began to consider the issues involved in sampling scallops frozen at sea on freezer vessels. In response to an inquiry from the owner of the freezer vessel Mr. Big, the Agency's Regional Director, Richard Roe, advised that frozen shucked scallops would be subject to the same sampling procedures as shucked scallops on ice, but that he expected the New England Council soon would address the issue. Roe then wrote the Chairman of the New England Council to urge that the frozen scallop sampling issue be placed on the Council's agenda. Roe wrote: "The scallop regulations currently in force do not address such handling. Enforcing the meat count average against these vessels will be difficult, if not impossible." The New England Council Chairman disagreed in his interpretation of the regulations, responding as follows: "I would recommend that the NMFS immediately implement an interim policy to sample frozen scallops to determine compliance with the meat count standard.... There is nothing here that precludes sampling frozen scallops."

On December 1, 1988, however, the Scallop Oversight Committee of the Council voted unanimously that two studies be undertaken by the Regional Director's office "before enforcement efforts on frozen product." The first of the two recommended studies would compare scallop meat weight before freezing and after thawing, the purpose being to obtain a formula that would convert the weight of thawed samples to the equivalent weight of a fresh sample, in order to compensate for dehydration during freezing and thawing. The second recommended study would be a statistical analysis of the problems associated with sampling sorted scallops. Obviously this latter study was to be undertaken to establish the statistical validity of the sampling process. On December 15, 1988, the New England Council voted unanimously to undertake these two studies. This resolution is discussed in more detail later in this opinion.

On April 20, 1989, the National Marine Fisheries Service General Counsel in conjunction with the Chief Enforcement Officer issued a memorandum to all law enforcement agents. That memorandum set out a procedure to be used in sampling frozen shucked scallops to determine compliance with the meat count limit. Agents were instructed to examine at least ten one-pound samples, chosen "from various parts of the load in a patternless manner, without definite aim and without examining the size of individual scallops." The scallops then were to be thawed "at room temperature," the thawed scallops mixed in each of the ten containers, and a sample of at least one pound scooped from each container onto a scale and weighed. The number of scallop meats in each of the ten samples then was to be counted and divided by the exact weight of the sample to obtain the number of meats per pound. The meats per pound figures for each weight category were arranged and then all the samples are analyzed to obtain an average meats per pound figure for the entire sample group.

We have noted that at the time the April 20th Enforcement Memorandum was issued, the Fisheries Management Council had adopted formal resolutions that prior to enforcement a factor had to be arrived at to equate the meat count of the scallops after freezing with that of the scallops before freezing, and indeed we suggest that any such factor which did not equate the meat count of the scallops landed after freezing to that of scallops landed after being stored on ice, would be so subject to universal criticism as to be practically unattainable, even if not outright illegal, because of an unabashed discrimination in favor of the old technology and against the new.

The reasons for the council resolutions are apparent from the record. There was at hand study after study which had been made through the years with respect to weight change by scallops after they had been shucked at sea. As we have recited, those shucked at sea were stored packed on ice until they were landed. While the precise amount of weight gain is not formally found as a fact by the agency and is not shown with precision by the record, the only reasonable conclusion which can be drawn from the studies at hand is that scallops shucked at sea and kept packed on ice for the duration of the ordinary voyage of about ten days gained some significant weight by the absorption of water. So a meat count of 40, given as one example, at sea, might well be 34 upon landing. And we note that the meat count upon landing was that which was enforced by the Agency, for under the regulations, the critical time is at or prior to the first transaction in the United States. 50 C.F.R. § 650.7(a). Similar but related studies had also been done by the Fisheries Service, or were in hand by the Fisheries Service, with respect to any weight change of scallops which had been shucked at sea but which had been frozen, as were those on board the Alice Amanda, and later thawed. Again, while the precise amount of any weight loss was not formally ascertained by the Agency or is not shown with precision in the record, the only reasonable conclusion which can be reached is that when scallops are shucked at sea, quick plate frozen, stored frozen in cheese cloth bags for several weeks, and then thawed in air, there is a significant weight loss. So the meat count of such scallops will be higher after thawing than when caught.

Thus, any meat count of thawed in air but previously frozen scallops will be higher than that which existed when the scallops were caught, and even higher than a meat count for the same scallops which had been caught and stored on ice until landing.

C. Enforcement Against the Alice Amanda

Upon completion of her maiden voyage on April 4, 1989 at Hampton, Virginia, enforcement agents attempted to sample the Alice Amanda's catch. Because the catch was frozen and the Agency had not yet issued the Enforcement Memorandum, the agents left without taking samples. The memorandum was issued before the end of the Alice Amanda's next voyage, and agents sampled the Alice Amanda's catch at the end of each of her next four voyages. On each occasion they found no violation.

Finally, at the end of her sixth voyage, the agents obtained a meat count in excess of the legal limit. By the time the agents arrived at the pier, unloading already had begun and 400 of a total of 1100 bags of scallops had been unloaded and taken to cold storage at another location. The entire catch, having been kept in the freezer bins below decks, consisted of frozen shucked scallops in cheesecloth bags weighing fifteen to seventeen pounds each. As they were being unloaded onto the pier, an enforcement agent, Passer, selected seven bags from among the 700 bags that had not yet been taken to cold storage. Meanwhile, a second agent, Hermann, went to the cold storage warehouse to select three bags from among the 400 that were there. The agents had decided to select one sample bag per hundred bags, which amounted to one bag per pallet. Agent Passer attempted to choose randomly from various points on the pallets.

Agent Passer took all ten samples to a storage area, where he placed the ten bags into two 96-quart Igloo coolers. The storage area was an unheated, locked room within a storage facility. The scallops remained there from the evening of October 9th until the morning of October 11th. On the morning of October 11th, Passer took the partially thawed scallops to S S Seafood, a scallop processing plant operated by Liston Shackelford, the owner of the Alice Amanda. At S S Seafood, Passer placed each of the ten samples into a large plastic container. Passer then left the ten large containers in a hallway area to finish thawing.

Throughout the day, enforcement agents kept an eye on the scallops. During the entire thawing process, Passer periodically stirred the contents of each container, endeavoring, however, not to break the individual scallop meats. Around 3:30 p.m., Passer moved the scallops into a small break room at the S S building. In contrast to the rest of the building, which was kept cool for scallop processing, the break room was heated by an electric heater.

When the scallops had thawed enough to be counted, about an hour after moving them to the heated break room, Passer began the actual sampling procedure. He first called the Alice Amanda's captain, Captain Belvin, to witness the sampling. Using a sixteen-ounce cup, Passer scooped scallops from the center of each large plastic container and placed them into a separate container. After removing all bits and pieces, Passer placed the whole scallops on the scale. Another agent recorded the number of scallop meats and the exact weight of each sample. This process was performed once for each of the ten large containers of scallops, a total of ten samples of about sixteen ounces each. Five samples were below 36.3, and five above. The resulting average meat count for all ten samples was 40.6 meats per pound which exceeded the maximum allowable meat count of 36.3.

Following the agency's penalty guidelines, the agents told Liston Shackelford and Captain Belvin that the severity of the violation would require a seizure of the entire value of the catch. Shackelford asked to consult with his lawyer, and the agents agreed to return the next morning. The next morning, Shackelford refused to relinquish the proceeds of the catch.

A month later, on November 16, 1989, the Coast Guard seized the Alice Amanda. The agency then initiated this proceeding, seeking forfeiture of the Alice Amanda under 16 U.S.C. §§ 1860, 1861. Alice Amanda, Inc., a corporation controlled by Liston Shackelford, filed a claim to the Alice Amanda. Pending trial of the forfeiture action, the vessel was released on the personal recognizance bond of Liston Shackelford. The bond was for $300,000 and was conditioned on redelivering the ship should the court so order. See 16 U.S.C. § 1860(d)(1)(B). After a bench trial, the district court found for the government, ordering that the United States recover $25,000 in partial forfeiture of the Alice Amanda. This appeal followed.

II

A. Effect of the Enforcement Memorandum

As a preliminary matter, we must determine what effect, if any, the Enforcement Memorandum should have on our application of the regulations to the facts of this case. We find that the memorandum is, at best, only persuasive authority and we are not persuaded.

It is undisputed that the sampling procedures recited in the memorandum were not adopted after notice and comment as the Magnuson Act requires for the adoption of a regulation. See 16 U.S.C. § 1854(a)(b) and (c). Neither the memorandum nor the substance of its provisions was published in the Federal Register. Even the emergency procedures of § 1855(c) were not used so it, therefore, cannot even qualify as a valid emergency regulation adopted under 16 U.S.C. § 1855(c)(3).

The government argues that the memorandum did not constitute rule making subject to the notice and comment requirement. Rather, the argument goes that the memorandum "embodied internal agency policies and procedures which are exempt from publication or notice and comment." This argument begs the critical question. Even assuming that the memorandum was not subject to the requirements of notice, comment and publication, we still face the question whether the agency's opinion of what constitutes a violation of the regulations requires any deference from a court when the court is required to interpret those regulations in the course of a forfeiture action initiated in court by the government.

At the outset, we note that, in this particular case, the agency's failure to publish the memorandum in the Federal Register is not itself dispositive. Although the Freedom of Information Act, 5 U.S.C. § 552(a)(1), provides that "a person may not in any manner be required to resort to, or be adversely affected by, a matter to be published in the Federal Register and not so published," this statute by its terms does not apply "to the extent that a person has actual and timely notice of the terms" of the unpublished matter. 5 U.S.C. § 552(a)(1). In this case, the evidence shows that the Vice President of the claimant, Alice Amanda, Inc., had received a copy of the memorandum prior to the October 9 sampling. Thus the statutory defense of 5 U.S.C. § 552(a)(1) is not available.

It is idle to contend, however, that the actions taken under the Enforcement Memorandum were not agency actions, and we hold that they were. A ship was seized and forfeiture was sought.

B. Review Under the Administrative Procedure Act

The Agency argues that it should be able to seize and forfeit the Alice Amanda or the proceeds of her whole scallop catch because its sampling procedure revealed a violation of the meat count standard. This is so, contends the Agency, even though the Alice Amanda is a freezer ship, with a scallop load of over 19,000 pounds, a type of vessel and size of catch not contemplated by the 1982 Fishery Management Plan's body of regulations. The Alice Amanda maintains instead that scallop fishing technology has changed, that freezer ships should be sampled differently, that the present sampling technique is inaccurate, and that the Agency was aware of all of these facts yet still decided to enforce new and admittedly unproven methods on Alice Amanda's catch. Alice Amanda argues that the Agency's enforcement and the resulting seizure of the Alice Amanda was arbitrary and capricious because the decision to enforce was made with full knowledge of the distinction between freezer vessel scallop fishing and the methods in vogue previously, illustrated by the fact that the enforcement methods used are in contradiction to the New England Fishery Management Council's own formal resolutions.

We will "hold unlawful and set aside agency action, findings, and conclusions found to be ... arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law." 5 U.S.C. § 706(2)(A). Agency action is arbitrary and capricious if the agency relies on factors that Congress did not intend for it to consider, entirely ignores important aspects of the problem, explains its decision in a manner contrary to the evidence before it, or reaches a decision that is so implausible that it cannot be ascribed to a difference in view. Bedford County Mem. Hosp. v. Health Hum. Services, 769 F.2d 1017, 1022 (4th Cir. 1985) (quoting Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Aut. Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 2866, 77 L.Ed.2d 443 (1983)). Although we should not venture into the area of agency expertise, we need not "accept without question administrative pronouncements clearly at variance with established facts." National Labor Rel. Bd. v. Morganton Full Fash. Hos. Co., 241 F.2d 913, 915-16 (4th Cir. 1957).

There is no substantial evidence in the record except that which reflects that the Agency was aware of the substantial difference between the new technology of freezer ship scallop fishing and the previous methods of scallop fishing contemplated by the regulations promulgated under the 1982 Fishery Management Plan. It is also beyond doubt that the Agency had actual knowledge of the problems with applying the old regulations to these new freezer ship catches. The Regional Director of the Fisheries Service, Richard Roe, repeatedly voiced his concern over freezer ship enforcement. In a letter dated September 26, 1988 to David Borden, Chairman of the New England Fishery Council, he stated: "Enforcing the meat count average against these vessels will be difficult, if not impossible." He urged that Borden place the issue on the agenda of the next New England Fishery Council Meeting. At the next Council meeting on October 25, 1988, Roe advised the Council that "he did not know how to enforce [freezer ship scallop fishing] at the moment because the current regulations do not take into account freezing at sea." The issue was then placed on the agenda of the next meeting of the Scallop Oversight Committee.

At the Scallop Oversight Committee's meeting the Committee received a report made by an ad hoc group of government and state staff workers recommending that two studies be completed by the Regional Director of freezer ship scallop fishing before implementing an enforcement policy. The Scallop Oversight Committee unanimously voted to submit these recommendations to the New England Fishery Management Council. The New England Fishery Management Council took the matter up at its December 15, 1988 meeting and unanimously adopted the recommendations made by the Scallop Oversight Committee.

The resolution adopted was (adopting the Scallop Oversight Committee report):

1. The committee voted unanimously that the following two studies should be completed by the Regional Director prior to enforcement efforts on frozen product:

1) A technical study must be done to compare scallop meat weight before freezing and after thawing. The results of the study will provide a factor to convert the meat count from thawed samples to fresh meat count equivalents.

2) A statistical analysis of the sampling frame for sorted (stratified) scallops must be done. The purpose of sampling sorted sea scallops is to a) determine that meat count categories are valid, b) measure the meat count for the entire catch. The results of the study will determine the number of samples necessary from each strata, and whether secondary sampling will be necessary to measure the meat count for the entire catch.

At that meeting, Roe advised the Council that there was no enforcement for scallops frozen at sea because there was no procedure yet established. The minutes show, however, an interjection by a certain Geidt implying that there had been some enforcement and that the frozen scallops had to be defrosted. This interjection is not explained in the record.

While Part 2) of the resolution does not affect this case because the catch of the Alice Amanda was not sorted, Part 1) of the resolution directly affects the catch, as is at once apparent. The resolution may only be said to acknowledge the facts shown by the various studies that the meat count before freezing and after thawing is different, for it required that "a factor to convert the meat count from thawed samples to fresh meat count equivalent" be provided "prior to enforcement efforts."

Notwithstanding Roe's previously professed concerns for enforcement without factual studies upon which appropriate procedures might be based, on April 14, 1989, Roe sent a memorandum to one MacDonald, the general counsel of the Fisheries Service, urging him to begin enforcement of the scallop meat count for scallops frozen at sea and appeared before a meeting of the Scallop Oversight Committee meeting on March 23, 1989, at which he assured the Committee that the Fisheries Service could start to enforce the regulations on scallops frozen at sea pending completion of the two studies being done to facilitate enforcement. How this would be accomplished Roe did not explain, and there is no explanation in the record for his about-face.

Then, without regard to any studies or discussion of policy, the Fisheries Service General Counsel, in conjunction with the Chief of Law Enforcement, one McCarthy, issued an Enforcement Memo on April 20, 1989. The memo set out a procedure for sampling and stated that freezer ship catches would be required to comply with the normal scallop meat size requirements. The Alice Amanda's catch at issue here was sampled pursuant to this procedure and found to be in violation of the Magnuson Act on October 11, 1989. The Enforcement Memo was contrary to the formal resolutions of the New England Fishery Management Council, which had voted unanimously to stay enforcement until the recommended studies were made to determine an accurate method to be used by the Fisheries Service field agents for sampling freezer ship catches. The Agency gave no explanation for this memo at the time, nor does it now.

When reviewed under the standards of Bedford County Memorial Hospital, Motor Vehicle Manufacturers Association, and Morganton Full Fashion Hosiery Company, we are of opinion the actions of the Agency in this case are arbitrary and capricious under 5 U.S.C. § 706(2)(A). Because the Agency failed to consider the relevant factors that possibly distinguish between frozen scallops and iced scallops, there is an absence of substantial evidence to support its action.

While there is no contention that the Agency relied on any factor that Congress did not intend for it to consider, the Enforcement Memorandum ignored the important aspect of the problem that the meat count after thawing in air of scallops frozen at sea was different from the meat count before freezing. There is no evidence in the record that this is not so, and there is also substantial evidence in the record that the meat count of scallops, frozen at sea, which are thawed at room temperature in air, and not in water, had their meat count considerably raised above those scallops which had been counted after having been packed on ice as was the older practice. The decision reached here by the general counsel and the chief enforcement agent, was so implausible that it could not be ascribed to a difference in view. We do not invade the realm of the agency expertise when we refuse, as we do here, to accept without question the administrative pronouncement, implicit in the enforcement of the same meat count, that the meat count of scallops frozen at sea and thawed at room temperature would be the same as scallops not frozen at sea but packed on ice. Such an administrative pronouncement is clearly at variance with established facts which were even acknowledged by formal resolution of the New England Fisheries Council, the principal rule-making originator of the agency involved.

Accordingly, the judgment of the district court is reversed and the government's cross-appeal is held to be without merit.

No. 91-7555 — REVERSED.

No. 91-7575 — APPEAL DENIED.