SOBELOFF, Senior Circuit Judge:

Appellant Louis Tucker, a longshoreman, sought damages from the Calmar Steamship Corporation, owner of the S.S. PENNMAR, for injuries he sustained while assisting in transferring a load of steel pipe to the vessel's cargo hold from a railroad gondola car stationed on an adjacent pier. Tucker alleged that the accident was caused by the ship's unseaworthiness and the Calmar Corporation's negligence. The District Judge found for the defendant on both grounds.

The only question before us is the correctness of the District Judge's ultimate decision that Tucker failed to prove unseaworthiness or negligence. We conclude that the District Court's finding that the ship was seaworthy is clearly erroneous and we therefore reverse.

I

At approximately 9 p. m. on the evening of March 5, 1966, appellant was working on a railroad gondola car loaded with steel pipe. His duty was to secure a draft of pipe for transfer to the ship's cargo hold by fastening a metal strap around one end of the draft. A co-worker would similarly secure the other end of the draft. A draft ordinarily contains about 15 to 20 steel pipes. The pipe involved in the accident was 4 to 6 inches in diameter and about 30 feet long. To get the straps around both ends, one end of the draft would be "broken-out" — lifted by a sling — so that the two longshoremen could reach under the pipe and fasten the straps at each end. Tucker, working on the end not raised in the breaking-out operation, was seriously injured when a pipe slipped out of the sling and rolled down on him, throwing him from the railroad car onto the pier.

At the time of the accident, the breaking-out was accomplished by the use of a house fall. The house fall, pictured in Figure 1, is a cable running from the

ship's port winch to a pulley attached to a post on the roof of the pier's warehouse and then down to a hook to which the sling is attached. A load on the sling at the end of the fall is raised by winding the cable onto the winch. On the occasion with which we are concerned, the pipe slipped out of the sling because the house fall had lifted the sling holding one end of the draft too high. Whereas the sling is normally lifted only 3 or 4 feet in the breaking-out procedure, in this instance it was lifted nearly 6 feet. At that elevation and angle, the sling could not hold the draft and one pipe slipped out and descended on Tucker.

Joseph Haskins, a man experienced in breaking-out pipe where a house fall is in use, was the winch operator at the time of the accident. Called as a witness by the shipping company, he explained how the sling came to be lifted too high. He testified that the tension on the house fall caused by the weight of the load on the sling tended to pull the ship towards the pier. When the load was released, the tension on the cable slackened and the ship returned to its original position vis-a-vis the dock. On the night of the accident, as often occurs, both the fore and aft winches of the ship were loading cargo. Haskins was operating the forward winch next to the #1 hold; another operator was working a winch to the rear of the ship by the #3 hold. When both winches were lifting a load and one released the tension on its house fall before the other, the ship would start to return to its original position and this movement away from the pier would pull the other, taut house fall, causing the load on the end of the still taut line to be lifted higher than anticipated. Haskins explained that the winch at the #3 hold had just released its load and the ship had begun to move away from the dock. Consequently, Haskins' house fall raised the draft at the end of his line higher than desired. Haskins' explanation of the cause of the accident was not disputed and was accepted by the District Judge.

Because of the recognized proclivity of a ship to move away from the pier when one of two operating house falls is relaxed, resulting in a variation in the height of the load on the other, still taut fall, the winch operator is required to make quick compensatory maneuvers to avoid an accident of the kind which injured plaintiff. To correct for the ship's motion and keep the load at the proper height, the winch operator must, with exact timing, relax the tension of his fall as the ship moves away from the dock. Haskins testified that in the darkness of night a winch operator could not see the movement of the sling so as to make the necessary fine correction, but had to relay on a deckman standing on the ship to relay signals from the longshoremen breaking-out the pipe. Haskins said on cross-examination that, although a deckman was present during daylight, from his experience he did not rely on the deckman when he could see the breaking-out operation for himself; but he always needed a deckman at night. The lighting on the night of the accident was described as "very poor" by witnesses for both parties, and at the time of the accident the longshoremen were searching for additional lights. While Haskins said the lighting that night was adequate for him to see the deckman, it is noteworthy that no one contradicted Haskins' further assertion that at night, even after the lighting was somewhat improved, he could not see the movement of the sling.

Haskins in his testimony took pains to distinguish between day and night operations with a house fall. He emphasized that in daylight, by looking at the gondola car, rather than at the deckman, he could better and more speedily compensate for the motion of the ship.

[I]f I'm looking at the load, at the men working at the time [the ship moves] I can just slack off a little bit and give it a little more slack to extend, but if I'm watching the deckman, I can't do this. (Transcript p. 136).

In the daytime mostly, we can usually see those things happening that way and you have more chance of slacking away and extending, giving it more fall so that it won't pull the load any at all. (Transcript p. 138).

[You have less accidents because] when you take a signal from the deckman, you're usually a few seconds late but if you're getting it direct — if you're looking directly at a situation, you are more accurate. (Transcript p. 159).

By ignoring the deckman in daylight and relying on his own observation Haskins said he had avoided several accidents by reason of his quicker reaction time in response to the ship's motion. Several times earlier on the day of the accident, during the daylight hours, Haskins said he was readily able to correct for the ship's movement.

There was uncontroverted testimony that other methods of breaking-out the pipe were available, namely the use of the ship's boom or a land-based heister. Both alternative methods were described by Haskins as safer and steadier than the house fall method because only the use of the house fall cause the ship to move towards the pier when the fall is carrying a heavy load. In Haskins' view, if one of the other means had been used that night, rather than the house fall, the accident would not have occurred.

Reviewing these facts, the District Judge nevertheless found the house fall method of breaking-out to be reasonably safe and its use neither unseaworthy nor negligent. Plaintiff contended that the ship was unseaworthy in failing to provide lighting adequate to enable the winch operator to see the movement of the sling so as to quickly make the necessary corrections in the height of the sling thereby preventing the accident. In response, the District Judge reasoned that the particularly poor lighting on the night of the accident was not the proximate cause of the injury since, even when better lighting was supplied after the accident, the winch operator still required the assistance of a deckman at night. He found the lighting at all times to be adequate to allow Haskins to see the deckman, and impliedly found that lighting sufficient to enable the winch operator to see the sling, not merely the deckman, was not required to make the use of the house fall that night reasonably safe.

Regarding the availability of other, safer equipment, the Judge noted the well established admiralty maxim that a ship need not provide the best or safest gear or method of operation, only gear or a method which is reasonably safe and fit for the purposes for which it is intended. Because he found the house fall to be reasonably safe to use that night, the Judge concluded that, even though other method of breaking-out might have been safer, the ship was under no duty to provide them. While acknowledging that practices customary in an industry do not establish the reasonable safety of those practices, the Judge also observed that the house fall was a recognized method of breaking-out pipe, used in both day and night operations and, in 1966, was preferred over the use of heisters because the house fall could break-out larger drafts of pipe than the heisters then available. The Judge concluded that the accident was simply an "unfortunate" water-front mishap, but one for which the ship and its owner were not liable.

II

A district court's findings regarding unseaworthiness and negligence are generally treated as findings of fact reviewable under the "clearly erroneous" standard of F.R.C.P. Rule 52(a). Texas Menhaden Co. v. Palermo, 329 F.2d 579 (5th Cir. 1964); Morales v. City of Galveston, 275 F.2d 191 (5th Cir. 1960).

In explaining this rule, it has been said:

In a non-jury case, [the court of appeals] may not set aside a finding of fact of a trial court unless there is no substantial evidence to sustain it, unless it is against the clear weight of the evidence, or unless it was induced by an erroneous view of the law.

Cleo Syrup Corp. v. Coca-Cola Co., 139 F.2d 416, 418 (8th Cir. 1944). With due regard for the limitations on our reviewing role under Rule 52(a), we conclude that the Judge's finding of reasonableness in the chosen method of operation is not supported by substantial evidence and is clearly erroneous.

The reasonableness of a ship's method of loading is determined by "matching operating proficiency against anticipated operating conditions." Mills v. Mitsubishi Shipping Co., 358 F.2d 609, 613 (5th Cir. 1966); Walker v. Harris, 335 F.2d 185, 191 (5th Cir.), cert. denied, 379 U.S. 930, 85 S.Ct. 326, 13 L.Ed.2d 342 (1964) Of course these factors cannot be stated with mathematic precision, but a thorough assessment of the evidence leads to the firm conclusion that it was not reasonably safe to use the two house falls simultaneously under the conditions prevailing on the night of the accident, without lighting sufficient to allow the winch operator to take quick corrective action by looking directly at the movement of the sling.

All of the evidence in the record commands this conclusion. As we have seen, the ship had a recognized tendency to move under a heavy load when a house fall was used; when two falls were in simultaneous operation, the movement of the ship had a significant potential to cause accidents unless compensated for by the winch operator. We have the undisputed testimony of Haskins, the winch operator, that accidents were more likely when darkness requires reliance on a deckman due to the time lag in relaying the signal. He specifically testified that "you have less accidents" when the gondola car can be viewed directly from the winchman's position.

In view of this increased risk of extremely serious harm occurring at night, the existence of other, safer methods of breaking-out the pipe becomes significant. The rule that the vessel need not supply the safest or best equipment, but only equipment which is reasonably safe does not provide guidance in determining what is reasonably safe equipment. The availability of safer methods of operation must be taken into account in making the initial determination whether the equipment used was in fact reasonably safe.

The use of the ship's boom or land based heisters to break-out drafts of pipe does not entail the hazard that the ship will move during the operation. Although the ship's boom could not reach the track where the gondola car was stationed, the boom could have been employed if the gondola car being unloaded had been moved to the open track closest to the ship. Land-based heisters were also available. The District Judge found that they were not customarily used to break-out heavy drafts because the heisters available in 1966 could not handle as much weight as the house fall. But, that heisters were not at that time customarily used to break-out drafts due to their smaller capacity is not proof that the method of operation chosen instead was reasonable. As this circuit has had occasion to note, an industry's customary practices are not necessarily determinative of reasonableness.

[The shipowner's obligation to provide a seaworthy ship] is not satisfied by a showing that he * * * has complied with industry practice which may be guided by economy in disregard of the men's safety.

Venable v. A/S Det Forenede Dampskibsselskab, 399 F.2d 347, 353 (4th Cir. 1968) (emphasis added). If, independent of industry custom, it was not reasonable under the circumstances to use the house fall, a heister or boom would have to be used, notwithstanding that they may lift smaller drafts.

In sum, under the uncontroverted testimony, we think the conclusion inescapable that, because of the adverse conditions prevailing that night and the availability of safer equipment, the risk of serious injury was too great to permit the use of the house fall to break-out the pipe without lighting sufficient to enable the winch operator to observe the sling independently and to correct for the anticipated movement of the ship. Binding the winch operator to reliance on a deckman and interposing an inevitable delay while the deckman transmitted signals to the winchman aggravated the risk to an unreasonable degree when matched against the known hazard of the ship's motion. The unsafe method of loading present here made the ship unseaworthy. Blassingill v. Waterman Steamship Corp., 336 F.2d 367 (9th Cir. 1964); Ballwanz v. Isthmian Lines, Inc., 319 F.2d 457 (4th Cir. 1963). To be seaworthy the ship and its appurtenances must be reasonably safe to use and fit to perform their assigned tasks. Gutierrez v. Waterman Steamship Corp., 373 U.S. 206, 213, 83 S.Ct. 1185, 10 L.Ed.2d 297 (1963).

In the words of Mr. Justice Reed, after reviewing the entire record, we are "left with the definite and firm conviction that a mistake has been committed," United States v. United States Gypsum Co., 333 U.S. 364, 395, 68 S.Ct. 525, 542, 92 L.Ed. 746 (1948), and hold that the judgment against plaintiff must be reversed.

It should be made clear that our present holding does not mean that the house fall system under other conditions is an unreasonable method of operation. There is no evidence that using just one house fall at night, with the aid of a deckman, is as dangerous as using two house falls simultaneously. In addition, the house fall was also used on the night of the accident to transport strapped steel pipe from the gondola car to the hold after the breaking-out process was completed. The winch operator testified that there was no risk of injury caused by the motion of the ship in that transit operation. Likewise we do not hold that the use of the house fall in daylight is unseaworthy, even though steadier alternatives are available.

Since the District Judge stated hypothetically that he would find the use of the house fall to be the proximate cause of the accident if he were to determine that the house fall was not reasonably safe, no further finding of fact on the question of liability need be made. The case is remanded to the District Court with directions to enter judgment for the plaintiff and to assess damages. On remand, the District Court will determine the merits of the shipowner's claim to indemnity from the stevedoring company.

Reversed and remanded.