This action is the result of a collision that occurred early on the morning of January 31, 1975 on the Delaware River in Marcus Hook, Pennsylvania. The collision took place when the chemical carrier S.S. EDGAR M. QUEENY (QUEENY), an American steamship, struck the crude oil tanker S.T. CORINTHOS (CORINTHOS), a Liberian steam tanker (owned and operated by Villaneuva Compania Naviera S.A. (Villaneuva), as the CORINTHOS discharged a cargo of crude oil at the B.P. Oil Inc./Sohio Petroleum Company Terminal (BP/Sohio).1 It resulted in the loss of 26 lives as well as other personal injuries, extensive damage to the terminal, the destruction of the CORINTHOS, oil pollution in the Delaware River, minor damage to the QUEENY, and damage to neighboring properties. The owners and operators of each vessel subsequently brought suit in admiralty, claiming that the other vessel was exclusively at fault and that they should be exonerated from liability or, in the alternative, that their liability should be limited.

The issues on appeal concern whether the district court erred with respect to the following: first, its denial of the petitions for limitation of liability filed by the owners and operators of the QUEENY (Bankers Trust Company (Bankers Trust), Monsanto Company (Monsanto) and Keystone Shipping Co. (Keystone Shipping ) (collectively the QUEENY INTERESTS or Keystone)); and, secondly, its ruling that under the preemption doctrine, federal legislation precluded it from holding the owners and operators of either the CORINTHOS or BP/Sohio liable. We will reverse each of those holdings, 503 F. Supp. 337. We find that Keystone's management was not privy to the cause of the accident and that the district court was not preempted from considering the liability of the CORINTHOS and BP/Sohio. Moreover, we rule that the alternative finding of the trial court that the CORINTHOS was unseaworthy was in error.2

The parties have saturated us with thousands of pages of transcript, hundreds of exhibits, a plethora of expert testimony and extensive briefs. This inundation of evidence may have been necessary for the trial court to make proper findings, but a landlubber's analogy might be more appropriate to put this case in context.

Every novice automobile driver knows that, by the laws of physics, when one seeks to make a U-turn in a small area where the space is less than the circumference required for the full turning arc of the car, the turn can be successfully maneuvered only with a series of forward and backward motions. On a river, where the area is not wide enough for a turn in one maneuver, a 180 degree turning process is an even more complex maneuver. The "roadbed" of the ship moves with the momentum of the tide. Ships have no brakes which can keep an unanchored vessel at a permanent standstill so as to avoid drifting and possible collisions. Thus, even an experienced captain of a ship must rely on professional pilots whose function on the ship is to aid the captain in making intricate turns, maneuvers and docking. In nautical terminology, we say that the pilot conns3 the vessel. Tugs are also often used to assist him in this process.

The instant case arises because in the process of making a 180 degree turn on the Delaware River, the purportedly experienced pilot of the QUEENY who conned the vessel made tragic errors of judgment. He failed to back and fill4 enough and to use efficiently the tug that was assisting him in the turn.

Clearly the collision was attributable primarily to the pilot's and the captain's negligence in executing the 180o turn. On this appeal, the parties have not questioned the trial court's finding that the collision was also caused in part by the failure of the owners to correct a purportedly defective astern guardian valve a vital component of the ship's engine mechanism. The question we must consider is whether the owners have privity or knowledge of this deficiency in the valve.

The QUEENY is a single screw, steam powered tank vessel of 19,046 gross tons. It is 660.2 feet in length, 90 feet in breadth, and powered by a 15,000 horsepower steam turbine. It was designed and built during the late 1960's as a multiproduct chemical tanker by the Bethlehem Steel Corporation (Bethlehem) for Monsanto and it was delivered to Keystone as an operator in September of 1970. Pursuant to a subcontract with Bethlehem, its turbine set and controls were supplied by the General Electric Company (GE). GE subcontracted with the William Powell Company (Powell) for the manufacture of a marine astern guardian valve a valve to be used in the turbine. As will be noted later, the functioning of this valve is of great importance in this litigation.

The CORINTHOS was a single screw, steam powered tank vessel of 30,705 gross tons. It was 723.7 feet in length, 106 feet in breadth and powered by a 17,000 horsepower steam turbine. On January 30, 1975, the CORINTHOS was moored at the BP/Sohio refinery dock on the Delaware River where it was discharging a cargo of crude oil. Across the river, the QUEENY was moored in Bridgeport, New Jersey in order to discharge part of its cargo.

On January 31, 1975, the QUEENY left port headed for the next discharge facility in Paulsboro, New Jersey. She was under the command of Captain Fay Kellog. Pilot Sverre Sorenson was at the conn of the ship. Both Captain Kellog and Pilot Sorenson had performed docking and undocking maneuvers on many occasions aboard the QUEENY. To proceed upstream to the next facility, they knew she had to make a 180o starboard turn. Visability was clear from eight to ten miles. The eastern half of the Marcus Hook channel was closed for dredging operations reducing the channel width from its standard 800 feet to 400 feet and, therefore, making turning in the river considerably more difficult.

The captain and pilot were on the bridge at the time of the undocking maneuver. To undock, the QUEENY used its main engine and bow thruster, and its rudder was hard left. The Tug Tanda 12, which had been engaged to help with the undocking, was located on the port bow. Pilot Sorenson was at the conn and in radio communication with the tug.

When the vessel had cleared the dock, the QUEENY began backing and filling. To do this, it used a series of astern and ahead maneuvers in conjunction with the vessel's rudder, the bow thruster, and the tugboat. This facilitated the starboard turn of the vessel and positioned her to go upstream. Both the captain and pilot knew the eastern half of the channel was closed.

To assist in the turning maneuver, at the early stages, the Tug Tanda 12 pushed at the QUEENY's port bow. Before the attempt to turn was completed, the QUEENY backed and filled at least twice to prepare for it. Tug Tanda 12 was then released from the port bow and instructed to stand by in Paulsboro, New Jersey. Captain Kellog was on the starboard wing of the bridge and Pilot Sorenson was on the port wing.

When the tug was released the vessel's capacity to turn dramatically decreased, but Captain Kellog was not alarmed. He expected Pilot Sorenson to make additional back and fill maneuvers as he had done on many other occasions in similar situations. The QUEENY's forward acceleration, however, actually increased and, consequently, the captain grew apprehensive, feeling that the QUEENY might not clear the CORINTHOS due to the pilot's failure to continue to back and fill. He voiced his concern about the "closeness" of the situation, but the pilot replied, "Captain, she should make that o.k." Later, Captain Kellog again voiced his concern stating that they should go astern because they would be very close on the maneuver. When Pilot Sorenson did not respond, Captain Kellog became quite alarmed about the proximity of the QUEENY to the well lighted CORINTHOS and he gave a "full astern" order. The third mate rung up the order on the engine telegraph. It was promptly acknowledged and then executed by the first assistant engineer. Then, Pilot Sorenson ordered a "double jingle." This order was received and promptly acknowledged by the engine room, but it had no effect on the operation of the engines because the first assistant engineer had already opened the throttles as far as he felt they safely could be opened. Accordingly, the rate of the QUEENY's turn to the right was reduced as the propellor responded to the full astern order.

The bow of the QUEENY continued to ease closer to the CORINTHOS, and the pilot, as a result, recommended that the starboard anchor be dropped. It seems that Captain Kellog acquiesced to this recommendation, because he ordered it via his walkie-talkie. No response to Captain Kellog's order was received, however, because the bow lookout, Arvie Harris, stationed at the windlass anchor watch position, had fled the extreme bow in the face of the impending collision.

Shortly thereafter the QUEENY crashed into the CORINTHOS. The initial contact produced sparks as the two ships scraped one another. The shock of this contact caused both ships to roll and pitch, and the vessels collided intermittently. The port fluke anchor of the QUEENY punctured the port shell plating of the CORINTHOS's No. 4 and/or No. 5 cargo tanks, resulting in a series of explosions and then a fire.

The fires and explosions aboard the CORINTHOS were out of control throughout the early morning hours of January 31, 1975, and caused the ship to break in half and sink, which resulted in extensive damage to the dock, oil pollution in the Delaware River, damage to neighboring properties, and the loss of 26 lives as well as other personal injuries.5

Keystone, the owners and operators of the QUEENY, and Villaneuva filed petitions for exoneration from or limitation of liability, Civil Action Nos. 75-364 and 75-2110, pursuant to the Limitations of Liability Act, 46 U.S.C. § 183 et seq. This act, in pertinent part, provides as follows:

The liability of the owner of any vessel ... for any ... loss, or destruction by any person of any property, goods, or merchandise shipped or put on board of such vessel, or for any loss, damage, or injury by collision, or for any act, matter, or thing, loss, damage, or forfeiture, done, occasioned or incurred, without the privity or knowledge of such owner or owners, shall not ... exceed the amount or value of the interest of such owner in such vessel, and her freight then pending.

46 U.S.C. § 183(a). Keystone and Villaneuva were damage claimants in each other's limitation proceeding, along with all those who suffered injury or loss.

On the second day of trial, the attorney for Keystone wisely admitted that his client no longer sought exoneration but merely pressed for the limitation of his liability. The district court, sitting without a jury, examined the issues raised in the limitation of liability petitions and the contrary claims made by the three major parties. The issue of damages was bifurcated with respect to the question of liability. In a memorandum opinion dated February 19, 1980, the trial court ruled that Keystone's petition for limitation of liability would be denied, and that Villaneuva's petition for exoneration from and limitation of liability would be granted.

III. THE QUEENY INTERESTS' PETITIONS TO LIMIT THEIR LIABILITY

Since March 3, 1851, federal legislation, under certain circumstances, has permitted a shipowner to limit his liability to the value of the specific ship involved in an accident.6 Thus, "even where a ship or fleet of ships is owned by a corporation, the privilege of limitation will insulate the remaining corporate assets from claims to which they would otherwise be subject." See G. Gilmore & C. Black, The Law of Admiralty, 818 (2d ed. 1975) (Gilmore & Black). Many assert, as the QUEENY INTERESTS have, that the purpose of the limitation act is to promote competitive United States shipping by reasonably fixing investors' liability to the value of the vessel placed in trade and that "the Act was passed to encourage investment at a time when American shipping was struggling to become commercially competitive and that (that) condition is just as valid in the United States today as it was ... when the Act became law." Bankers Trust brief at 12.7 Others assert that "(s)ince approximately 1930 the early enthusiasm, both legislative and judicial, for the limitation principle has cooled." Gilmore & Black at 821.8

Whatever may be the long term trend, in this case, as in most major calamities, high stakes are involved in the granting or rejecting of Keystone's petitions for limitation of liability. Keystone placed a fund of $11,169,501 in the court representing the limitation value of the QUEENY while claims at that time were well above $40 million.

In support of its limitation of liability argument, Keystone asserts that the accident was caused solely by the negligent navigation of the QUEENY by Captain Kellog and Pilot Sorenson; that Keystone was without privity or knowledge of these negligent acts; and that, as a consequence, Keystone is entitled to limit its liability to the value of the vessel and freight on the ship at the time of the accident.

Villaneuva and BP/Sohio contend that Keystone is not entitled to have its liability limited because the QUEENY was unseaworthy and that its unseaworthy condition was a contributing cause of the collision. They maintain that the astern turbine on board the QUEENY was defective, and that it prevented the QUEENY from obtaining her full rated backing power prior to the accident. They submit that had the astern turbine functioned properly the QUEENY, notwithstanding the negligent acts of the captain and pilot, would have been able to avoid the collision with the CORINTHOS.

The district court found that the QUEENY's astern guardian valve and astern turbine were defective on the night of the accident and that these defective conditions were a contributing cause of the tragic collision. It ruled that Keystone should have known of this defective condition and held them in privity. Hence, it denied Keystone's petition for limitation of liability.

IV. THE RELATIONSHIP OF THE ASTERN GUARDIAN VALVE TO THE

For the want of a nail the shoe was lost,

For the want of a shoe the horse was lost,For the want of a horse the rider was lost,

For the want of a rider the battle was lost,

For the want of a battle the kingdom was lost

And all for the want of a horseshoe nail.

Similarly many of the parties assert that for the want of a proper functioning astern guardian valve the collision was caused.

Physically, the astern guardian valve is located on the main steam line just outside the steam inlet to the astern element of a low pressure turbine. It is the last valve the steam passes through before it enters the astern turbine. Its purpose is to protect the astern turbine from an accidental opening of the astern throttle valve when the turbine is turning forward RPM.9 When the valve is open steam passes into the astern turbine and reverses the engine. If steam were accidentally released into the astern turbine while the turbine operated at full speed ahead, it would be seriously damaged.

The valve itself is a disc 5 in diameter and 6 high attached to the end of a stem. As the stem is raised or lowered (by a motor or handwheel), it moves up and down inside the valve body. The disc's bearing surfaces, which touch the valve body, are called guide surfaces. At normal room temperature, the clearances between the guide surfaces and valve body are 10-12 /1,000ths of an inch. In this case, the disc and valve body were manufactured with the same material (ASTM-2176 grade WC6) (cast steel). Neither was treated to obtain differing degrees of hardness.

The valve operates when steam enters an inlet on its side and exits through a hole in its bottom. As the disc moves up and down, it is kept in position by the guide surfaces which come into contact with the valve's body. The stem obviously does not prevent contact between the valve's disc and body. It simply moves the disc up and down.

The valve's defect was caused by the galling or gouging of its surface due to the rubbing of two metal castings manufactured from the same grade of steel. This rubbing caused the surfaces to weld together, and, consequently, broke off pieces of the metal surfaces. This 'chiseling effect' allowed the steam in the valve to project pieces of metal into the astern turbine where they damaged the turbine's blades. This damage reduced the QUEENY's astern power. Thus, when Captain Kellog ordered full astern, the QUEENY's engines were unable to respond with its former full power.

The QUEENY's turbines were designed for normal sea use. They were guaranteed to produce an astern RPM equivalent to 70% of 109. If undamaged, therefore, they should have responded to the full astern command by producing 76.3 RPM. In fact, on the QUEENY's sea trial run, they reached 76 RPM. On the day of the accident, however, they were unable to produce the guaranteed level of astern RPM. Hence, the court below found the QUEENY unseaworthy, and ruled that had the astern turbine functioned properly the collision between the QUEENY and the CORINTHOS would not have occurred.10

V. THE VALVE INCIDENT AND ITS LEGAL CONSEQUENCES

In reviewing the district court's finding of liability, the critical question we must address is whether the QUEENY INTERESTS had knowledge and privity of any defect in the astern guardian valve. Even if it is conceded that the valve was defective and that the collision was caused by the valve's defect, this alone does not constitute a basis to deny the QUEENY INTERESTS' petition to limit their liability. To constitute a per se basis to deny their petition, they must have had knowledge and privity of the valve's defect.

The statute provides that the owners' liability shall be limited to the "value of the interest" of the "vessel and her freight" for any act which occurs "without the privity or knowledge" of the owner. 46 U.S.C. § 183(a). In this case, therefore, the magic words are "privity or knowledge," for a shipowner is responsible for unseaworthiness of his vessel only if he has knowledge, or, in the use of ordinary care, should have knowledge of that condition. Spencer Kellog & Sons, Inc. v. Hicks, 285 U.S. 502, 52 S. Ct. 450, 76 L. Ed. 903 (1932); Gibboney v. Wright, 517 F.2d 1054 (5th Cir. 1975); Waterman Steamship Corp. v. Gay Cottons, 414 F.2d 724 (9th Cir. 1969); China Union Lines, Ltd. v. A. O. Andersen & Co., 364 F.2d 769 (5th Cir. 1966), cert. denied, 386 U.S. 933, 87 S. Ct. 955, 17 L. Ed. 2d 805 (1967); The Cleveco, 154 F.2d 605 (6th Cir. 1946).

In this respect, corporate shipowners, are held liable for the acts or knowledge of managing officers or supervisory employees. Coryell v. Phipps, 317 U.S. 406, 410-412, 63 S. Ct. 291, 293-294, 87 L. Ed. 363 (1943); Avera v. Florida Towing Corporation, 322 F.2d 155 (5th Cir. 1963); Complaint of Allied Towing Corp., 409 F. Supp. 180, 188 (E.D.Va.1976), aff'd, 580 F.2d 792 (4th Cir. 1978). Usually, however, they will not be held liable for the actions of crew members in the navigation or operation of the ship at sea. In re Black Sea Steamship Company, 382 F. Supp. 907, 911 (D.C. Canal Zone 1974). To hold Keystone privy, therefore, the evidence must show that some individual at the management level of Keystone had, or should have had, knowledge of the QUEENY's damaged astern guardian valve. BP/Sohio and the CORINTHOS bear the burden of proving negligence or unseaworthiness. Keystone, on the other hand, must demonstrate a lack of knowledge or privity to show that it is entitled to have its liability limited. Farrell Lines, Inc. v. Jones, 530 F.2d 7, 10 (5th Cir. 1976).11

Under these precepts of law, what were the facts in this case which would justify the trial court's finding of privity or knowledge?

The district court found that the QUEENY's astern guardian valve and astern turbine were defective on the night of the accident and that this defective condition was a contributing cause of the tragic collision. It ruled that Keystone should have known of this defective condition and held them in privity. Thus, it denied Keystone's petition for limitation of liability.

The court below based its conclusion of privity and knowledge on the following findings. First, on or about September 14, 1970, more than four years before this accident and not long after the QUEENY had been delivered by Bethlehem, problems developed with the QUEENY's astern guardian valve. The court held that Keystone had notice of these problems. Second, the court ruled that Keystone would have discovered the defect in the astern turbine had it examined the RPM logger tapes. Third, it found that, in 1973, Keystone was negligent because it failed to determine the cause of the turbine damage on a sister ship of the QUEENY, and relate this information to the earlier problem with the QUEENY's valve. Finally, the court held that the QUEENY INTERESTS had a duty to follow-up on Captain Kellog's March 12, 1972 letter which purportedly revealed concerns about defects to the turbine. These findings of the trial court, of course, must be measured by the traditional standard in admiralty that they will not be set aside unless clearly erroneous. Fed.R.Civ.P. 52(a); see McAllister v. United States, 348 U.S. 19, 20, 75 S. Ct. 6, 7, 99 L. Ed. 20 (1954); SS Amazonia v. New Jersey Export Marine Carpenters, Inc., 564 F.2d 5 (2d Cir. 1977).

There is no vision which is as clear as hindsight. Images come into almost perfect focus through the evidentiary microscope of a lawyer's hindsight. After thousands of daily and weekly occurrences of maritime life are discarded, the lawyer can isolate a few incidents for examination with such a microscope to document things such as privity or knowledge. This process proceeds as if a reasonable person had no other responsibilities or phenomena or events to observe other than those incidents now carefully isolated and focused on through this hindsight perspective. If life were so simple, foreseeability under the law would be the equivalent of omniscience.

We feel that the appellees in this case have built such a hindsight microscope slide one that is patently divorced from reality. Scattered and isolated items taken out of context were presented to the court, and, surprisingly, the trial judge found a nexus of interrelationships to establish privity and knowledge. These findings defy the realities of life even when recognizing a shipowner's undelegable obligation to correct equipment that is known to be defective. The most significant gap which the trial judge failed to appreciate involved a four-and-a-half-year span during which appellants at no time had any realistic notice that the astern guardian valve on the QUEENY was functionally defective.

The QUEENY's maiden voyage was made in September of 1970 when, shortly after it was delivered, it traveled from Baltimore to Texas. On that voyage, the astern guardian valve malfunctioned when it either stuck or jammed in the closed position. When this happened, it became impossible for the steam to pass through to the astern turbine.

The valve usually is automatically operated by the limitorque motor a motor that provides torque to open or close it. This motor could not move the valve on the QUEENY's initial voyage. Each attempt to do so overloaded the electrical circuit and tripped the circuit breaker. The valve, however, was equipped with a handwheel that could be used to operate it manually. The QUEENY's first assistant engineer, Michael Breton, attempted to turn the handwheel after the malfunction. But he was unable to do so. Breton and Bethlehem's operating engineer, John Jensen, later were able to free the handwheel. They used a wrench (called a "cheater") to give them extra leverage so as to apply excessive force to the handwheel to open the valve.

The instruction manual for the astern guardian valve observes that a "cheater" must never be used on the handwheel. Furthermore, it states as follows:

Excessive handwheel effort can indicate the following:

(1) improperly lubricated or damaged valve stem; (2) valve packing gland too tight; (3) improperly lubricated valve; (4) stem nut too tight on valve stem; (5) faulty or damaged valve parts.

Breton and Jensen, however, simply reset the limitorque circuitry to allow the valve to open and shut without further investigating what had caused the initial problem.

The QUEENY INTERESTS were informed of the problems with the valve on September 14, 1970, when the technical assistant to the vice president of Keystone, Samuel Spencer, sent an internal memo to Keystone's president, Adolph B. Kurz. Spencer also notified the Chief Guarantee Engineer, E. L. Insley, at Bethlehem, and Bethlehem listed the valve as an item for repair at the guarantee yard dry docking.

Bethlehem's "guarantee agreement" with Keystone required it to repair or replace any part or system on board the QUEENY that malfunctioned for reasons attributable to the builder. Vessel personnel were to notify both QUEENY management and the builder promptly when guarantee items needed to be repaired. Hence, on September 15, 1970, the day after Keystone learned that the astern guardian valve was sticking, it notified the proper Bethlehem representative of the problem. A Bethlehem Guarantee Engineer, Mr. McKendrick, met the QUEENY at her first northern port of call on September 20, 1970. On September 21, 1970, he reported that the valve was operating properly. At that time, therefore, Keystone had every reason to believe the valve was in working order because Bethlehem reported that it worked properly within only six days of notice of the problem.

The trial court held that the cause of the jamming and sticking of the valve in 1970 was not properly investigated, and that Keystone should have insisted on an inspection of the valve and turbine during the guarantee repair period. Yet, Keystone was told by the experts Bethlehem, the company which built the ship that the valve had been fixed, and for the next four-and-a-half years subsequent to this statement, it seemed to perform properly. Under these circumstances, the shipowner should not have been obligated to insist upon an inspection of the valve and turbine during the guarantee period. It makes no more sense to impose such a duty on a shipowner than it would to require a taxicab company to take a transmission apart on its new cab when the car's manufacturer had advised them that the new cab's transmission had been repaired and it worked properly after the original repair.

It is, of course, well settled that a shipowner has a duty to use reasonable means to acquire knowledge calculated to inform him of conditions likely to produce or contribute to unseaworthiness. He must create reasonable procedures for identifying the need for repairs, and relaying reports of ship conditions to shoreside management. See ARGENT, 1940, A.M.C. 508 (S.D.N.Y.1915). He must also enforce employee compliance with those procedures, The Pennsylvania v. Troop, 86 U.S. (19 Wall.) 125, 22 L. Ed. 148 (1873); Avera v. Florida Towing Corporation, 322 F.2d 155 (5th Cir. 1963). Finally, he must use those procedures to keep abreast of ship conditions. Spencer Kellog & Sons, Inc. v. Hicks, 285 U.S. 502, 52 S. Ct. 450, 76 L. Ed. 903 (1932); Complaint of Allied Towing Corp., 409 F. Supp. 180 (E.D.Va.1976), aff'd, 580 F.2d 792 (4th Cir. 1978).

In this light, in the instant case we hold, as a matter of law that the trial court erred because the record demonstrates that Keystone made every reasonable effort necessary to assure the valve was properly repaired.

Secondly, the court below found that Keystone would have discovered the defect in the astern turbine had it examined the RPM logger tapes which monitor the revolutions per minute of the engine. In so finding, it held that Keystone had notice of a problem with the astern guardian valve, and, therefore, had a duty to monitor the turbine to assure it was unaffected by the earlier difficulties with the valve. The record reveals, however, that the RPM tapes, dating from the sea trials to the night of the collision, were excluded at trial. They were neither marked nor placed into evidence as an exhibit. The trial court admitted into evidence only the sea trial tape, and the tape for the day of the accident. The court did state that it would also consider the opinions testified about in court by Dr. Corlett with respect to the tapes. But, Dr. Corlett's testimony at trial, for the most part, compared the performance of the QUEENY on the night of the collision with its performance during the sea trials. He nowhere expressed an opinion related to the performance of the QUEENY over her sea life from the time of the sea trials until the accident. We hold, therefore, that since the tapes were not in evidence it was error for the trial court to review that data and to attribute to Keystone the duty to have acquired that knowledge and to have acted upon it. We must note that even if the tapes had been admitted, they would have had minimum probative value if any.

Thirdly, the district court found that Keystone was negligent because it did not properly investigate the cause of the turbine damage on a sister ship of the QUEENY in 1973, which purportedly would have alerted it to a defect in the QUEENY's astern guardian valve. In effect, the court held that due to an allegedly improper repair of another ship's, (the LIBERTY), "turbine," the defect in the QUEENY's "valve" was never found. Yet, the record nowhere indicates that the astern guardian valve on the LIBERTY was damaged in 1973. Nor is there any evidence that personnel on board the LIBERTY had ever experienced a problem with the astern guardian valve similar to the one experienced on the QUEENY.

Nothing in the record, therefore, supports the trial court's ruling that the damage discovered on board the LIBERTY should have caused Keystone to be concerned about damage on the QUEENY. Those attending the LIBERTY had determined that the damage to her turbine was the result of small bits of debris left in the steamline which was a plausible explanation.12 And were we to rule that negligence with regard to the repair of one vessel could be imputed to the operation of another ship that is involved in a collision over a year later, we certainly would be extending the notion of foreseeability beyond reasonable limits. Thus, we hold that it was error for the trial court to find Keystone negligent for failure to determine the cause of the turbine damage on the LIBERTY and relate that information to the QUEENY's earlier valve problem on its maiden voyage.

Finally, the trial court found that Keystone should have known of the damaged condition of the QUEENY's turbine because Captain Kellog was reluctant to perform a crash astern test that was not required. The captain did write a letter to Keystone indicating he felt "such a test might damage the intricate and delicate propulsion system" on the QUEENY. But, he had not been ordered to perform the test. Rather, in 1972, Keystone had sent out Master's Circular No. 5565 which stated, in significant part, as follows:

In general, crash stop data was recorded during Builder's trials before delivery of a new vessel or before redelivery after a major conversion. You will appreciate that an emergency crash stop is a severe exercise on both machinery and hull. It is not desirable to undergo such a strain unnecessarily. Therefore, where this information developed by the Builder is available from your records, the emergency crash stop need not be repeated. In this case, make the data readily available for reference.

(emphasis added.) Captain Kellog was on board the QUEENY during its sea trials, so he was well aware that this information was available in the ships records.

In any event, the Kellog incident took place in March 1972, about a year and a half after the valve sticking incident in 1970, and there simply is not enough evidence in the record to lead us to the conclusion that Kellog's letter was related to the valve problem, or that it should have led Keystone to the discovery of the problem with the QUEENY's valve.

In conclusion, isolated events, such as those discussed above, over such an extended period of time, do not place a shipowner on notice with respect to his vessel's alleged seaworthiness. In this regard, the courts must consider how many vessel activities are logged and monitored each day with respect to a vessel, let alone a fleet of vessels. The numbers are enormous. We find, therefore, that the separate and distinct events relied on by the trial court are, as a matter of law, not significant enough to support a ruling that Keystone had knowledge or privity of either the acts of negligence on board the QUEENY, or her purported unseaworthiness.

The QUEENY INTERESTS assert that the owners and operators of the CORINTHOS were at fault because they did not inert its tanks. The trial court held that "the absence of inerting systems was a contributing cause of the damages," but it ruled that it would grant Villaneuva's petition for exoneration of liability because it was preempted by Congress from finding it unseaworthy.

The record indicates that once an uninerted tanker has discharged a cargo of crude oil, it may be likened to a floating time bomb. Inerting is a process that fills the cargo tanks of ships when they are emptied of oil cargo during discharge with inert gas rather than air which contains oxygen. This process reduces the likelihood of explosion or fire on such vessels because it prevents the build-up of fuel vapor which, when combined with oxygen, explodes when there is an ignition source.

Although the concept of inerting is not new,13 the CORINTHOS was not fitted with an inert gas system when it was built in 1963. Villaneuva argues that the trial court was precluded from requiring it to have retrofitted the CORINTHOS with an inert gas system prior to the accident of January 31, 1975 because Congress in October of 1978 enacted a specific schedule for the inerting of vessels like the CORINTHOS that does not require them to be inerted until June 1, 1983. The Port and Tanker Safety Act of 1978, Pub.L. No. 95-474, 92 Stat. 1471, (codified at 33 U.S.C. §§ 1221-1232; 46 U.S.C. §§ 214, 391(a)) (the 1978 Act). Moreover, Villaneuva maintains that the trial court improperly found that the CORINTHOS was unseaworthy since as of January 31, 1975, it had complied with all of the existing international treaties and American legislation prescribing tank vessel equipment for the prevention of explosion and fire. Lastly, it contends that, in any event, an inert gas system would have been unable to prevent the explosions that followed the collision.

In relevant part, the district court held as follows:

Though in 1975, no law required the inerting of cargo tanks on vessels the size of the CORINTHOS, we believe that at that time, the custom of the industry was insufficient to satisfy common law standards of care. However, the CORINTHOS interests are fortunate that Congress has entered this field of law, thereby preempting the Courts from deciding this issue.

It was, however, a misapplication of the preemption doctrine for the district court to hold that it was preempted from considering a conflict between the federal legislature and the federal judiciary. The preemption doctrine, by definition, delimits the relation between the federal legislature and the states. Where the federal legislature regulates a particular area of conduct or activity with an eye toward maintaining uniformity among the states, no state may enact measures which circumvent the federal intent. Gibbons v. Ogden, 22 U.S. (9 Wheat.) 1, 6 L. Ed. 23 (1824).

The federal judiciary, on the other hand, is vested with the authority to interpret the acts of the legislature and executive within the limits prescribed by the Constitution. Within these limits, federal courts are not preempted from interpreting statutes, reviewing executive orders, or establishing common law standards of care merely because another branch of government has entered and regulated a sphere of conduct. And, certainly, in this circuit, industry standards and practices normally establish only minimal requirements, and do not create immunity from liability. Indeed, in Voegele Company, Inc. v. OSHRC, 625 F.2d 1075, 1078 (3d Cir. 1980), we observed that

other courts have evaluated the custom and practice of the industry as one aspect of the reasonable person test. These courts have refused to limit the reasonable person test to the custom and practice of the industry because "(s) uch a standard would allow an entire industry to avoid liability by maintaining inadequate safety ...." General Dynamics v. OSHRC, 599 F.2d 453, 464 (1st Cir. 1979). We find this policy reason for not making industry standards determinative to be quite compelling. See Cape & Vineyard Div. v. OSHRC, 512 F.2d 1148, 1152 (1st Cir. 1975) (in context of § 1910.132(a), "(t)here may ... be instances where industry practice fails to take reasonable precautions against hazards generally known in the industry; in such event it may not be unfair to hold the employer to a standard higher than that of actual practice."); Brennan v. Smoke-Craft, Inc., 530 F.2d 843, 845 (9th Cir. 1976); American Airlines, Inc. v. Sec. of Labor, 578 F.2d at 41.

We hold, therefore, that in precluding itself from considering the legislation in question in this case, the court below was in error. Furthermore, we rule that the district court's alternative finding that the CORINTHOS was unseaworthy was improper for the reasons that follow.

On October 17, 1978, Congress enacted the 1978 Act. This Act makes provisions for the inerting of existing tank vessels of the type, capacity, age and size of the CORINTHOS. In Section 391(a)(7)(f), it states that:

an existing crude oil tanker of 20,000 deadweight tons or above, (should) not later than June 1, 1983, be equipped with an inert gas system.

The statutory history of this piece of legislation reflects an attempt to balance the various interests concerned with the safety of tanker operations. The House Report on the Port and Tanker Safety Act, H.R.Rep. No. 95-1384-Part 1, 95th Cong. 2d Sess., reprinted in (1978) U.S.Code Cong. & Ad.News 3270, 3271, states as follows:

In considering corrective legislative action, the committee has kept itself informed of the ongoing activity in the international arena, particularly in the Inter-Governmental Maritime Consultative Organization (IMCO); and H.R.13311 reflects, in part, certain requirements accepted internationally, as developed in the recent meeting of the International Conference on Tanker Safety and Pollution Prevention, held in London in February, 1978.

The 1978 Act. Further, the Senate Report on a related statute, the Port and Waterways Safety Act, S.Rep.No.724, 92d Cong. 2d Sess. 426, reprinted in (1972) U.S.Code Cong. & Ad.News 2766, 2783, declares that

(t)he committee fully concurs that multilateral action with respect to comprehensive standards for the design, construction, maintenance and operation of tankers for the protection of the marine environment would be far preferable to unilateral imposition of standards. However, standards are slow in coming from the multilateral forums ...

(Therefore) while the committee remains committed to the proposition that multilateral action in this area is preferable, it is not willing to sacrifice the objective of protection of the marine environment on the altar of that principle. Much more rapid and comprehensive action will be required if the United States is to continue to rely on multilateral forums.However, in the light of the preference for multilateral action shared by the administration and by the committee, provisions were adopted I.R. § 40 which attempted to resolve this dilemma, primarily thru the use of a deferral procedure.

In this light, the 1978 Act was a legislative response to a Presidential Message sent to Congress on March 17, 1977 a message with regard to the marine transportation of oil. In this message, President Carter recommended reforms in ship construction and equipment standards (including the requirement of inert gas systems on oil tankers over 20,000 deadweight tons). He hoped that these requirements would be "fully effective within five years." 123 Cong.Rec. 7931 (1977).

Notwithstanding the President's hopes for the future, however, on January 31, 1975, the day of the collision, the CORINTHOS was in compliance with the existing international treaties, and United States legislative and regulatory requirements concerned with fire fighting and explosions. It complied with all the fire fighting and anti-explosion provisions of the Safety of Life at Sea (SOLAS) Conventions of 1948 and 1960.14 Furthermore, in accordance with the SOLAS requirements, her owners had ensured not only that the vessel was equipped for fire fighting, but also that her crew was properly trained to fight fires. Finally, the CORINTHOS was also recognized by the American Bureau of Shipping (ABS)15 to be seaworthy. Thus, we find it was clearly erroneous for the lower court to rule that the CORINTHOS was unseaworthy.

Over the years a series of laws and regulations have been enacted with regard to tanker safety so as to phase in equipment requirements for ships. Congress has determined after considering many factors, including shipyard capacity and the available technology, that five years should be allowed before the retrofitting of tankers over 20,000 deadweight tons becomes mandatory. Were we to impose liability on vessels before this time period is up, the Congressional timetable would be rendered useless. Hence, in this instance, we will be guided by the federal legislation developed by Congress.

For the reasons stated above, we will reverse the district court with respect to its denial of the petition of limitation of liability to Bankers Trust Company, Monsanto Company and Keystone Shipping Co. Moreover, we will hold that the preemption doctrine is not applicable as to Villanueva. To the extent that the court made an alternative finding to deny Villanueva's petition for limitation of liability because it failed to use an inert gas system on the CORINTHOS, we will rule that the alternative finding was error as a matter of law. Of course, in admiralty, appeals from interlocutory decrees are permissible as a matter of statutory right.16 In this case, there was no decree entered by the trial court as to the issues raised in the appeals filed by B.P./Sohio in 80-1450 and 80-1495, and, therefore, the appeals in those cases must be dismissed for lack of appellate jurisdiction. We remand the case to the trial court for further proceedings consistent with this opinion.