Roy Kennedy, an engine wiper employed on the S.S. Coastal Rambler of the Alaska Steamship Company on September 11, 1964, stuck a fire hose and its stream of water into the starboard bilge sounding pipe on the weather deck of the ship near the No. 1 hatch. He should have put it into the nearby starboard sounding pipe connected to the double bottom. The water ran for about four or five hours before the mistake was discovered. As a result, canned salmon in the cargo consigned by Hydaburg was seriously damaged.

The ship was berthed at Pier 24 in Seattle. The stream of water intended for the double bottoms would have helped trim the ship. Prior removal of some of the cargo had made this desirable.

Over on the port side there was a pair of sounding pipes, again one for the bilge and one for the double bottom. Apparently one Bergman, first assistant engineer, assisted Kennedy in locating the correct sounding pipe for the water on the port side. Then Soyktis, second assistant engineer, assisted him in finding the sounding pipe (the wrong one) on the starboard side.

Each sounding pipe had a cap and a collar. On the starboard side the cap and collar of the pipe to the double bot-ton seem to have been at the weather deck level. The collar and cap of the bilge sounding pipe seem to have been slightly raised above the weather deck. On the port side the bilge sounding pipe cap was at deck level and the double bottom sounding pipe’s cap was protruding above deck. On the starboard side a square indentation in the cap required a square-headed wrench to open the double bottom pipe. A clover leaf wrench was required to open the bilge pipe. On the port side, it was vice versa, a square wrench for the double bottom pipe cap and a clover leaf for the bilge.

On either side of the ship there had been painted on the deck a white circle around the bilge pipe collar and cap. But Kennedy insists that the circle had become indistinct on the port side. The circle seems to have been clearly apparent on the starboard side. However, there is no indication in the record that such circles have universal meaning on a ship or even on Alaska’s ships. Further, it is not shown that the crew had any instruction as to what the circle meant on the Coastal Rambler, or that there were any written instructions or illustrated charts. If only the starboard bilge cap was circled with white paint on September 11, one might have concluded that the white circle was a warning that the cap and collar were protruding above the floor of the deck — a caution against stumbling.

At the trial Bergman was not called. It was he who helped Kennedy to make the correct water connection to the double bottom pipe on the port side. Soyktis, who supervised the starboard connection, couldn’t be called to testify. He was dead.

The flooding of the bilge which flooded the hold that damaged the salmon went on for hours. If all the water had gone to the double tanks, no damage would have been done and the ship would have been trimmed.

At a time subsequent to the day of the error, Hunter, a marine surveyor, examined the caps and collars of the bilge pipes. On the starboard double bottom pipe he found stamped into the collar, “Sounding Res. F O I I B.” This, he explained, meant “sounding pipe, reserve fuel oil number one inner bottom.” (On the Knot type ships, such as Coastal Rambler, inner bottom meant the same as double bottom. The inner bottoms sometimes were used for fuel oil, other times for water ballast.) Then he reported the bilge sounding pipe (inboard) had a cap and collar with the legend, “Hold number one starboard sounding bilge.” Apparently all the letters were indented by stamping. It was the surveyor’s recollection that the same lettering was on the caps or collars, and visible the night of September 11-12 after the flooding, as he found later on his specific study. The first mate’s testimony is consistent as to the collar-cap markings with that of the surveyor as to what was readable on September 11. Kennedy, the wiper, insists the markings on the caps-collars were indistinct on the port side. We think his testimony points to the same lettering condition on the starboard side (this might be disputed). Kennedy says that on September 11 the painted circle on the port side bilge pipe was obliterated but distinct on the starboard side.

Obviously, we have related a splendid case of carelessness, putting the hose down the bilge pipe, which was bound, in time, to flood the hold. And then there is the wonder that the flooding went on for over four hours.

Alaska Steam says the bill of lading under the Carriage of Goods by Sea Act, 46 U.S.C. § 1300 et seq., protected it. This is right if the event resulted solely from an error in the management of the ship, 46 U.S.C. § 1304 (2) (a). Hydaburg says the accident was caused by unseaworthiness of the ship and that it was Alaska Steam’s burden to prove seaworthiness.

The district court found unseaworthiness in lack of adequate identification on and around the pipe caps and collars and the confusing circumstances created by the reversing of the style of pipes from port to starboard. The court drew some inferences against Alaska by the failure to bring Bergman, who satisfactorily supervised the port connection, to the trial as a witness.

After reviewing the record, we cannot say the finding of the court was clearly erroneous. Therefore, we do not reach Hydaburg’s contention that In re Pacific Mail Steamship Co., 9 Cir., 130 F. 76 , 69 L.R.A. 71 (where an incompetent crew was held to constitute unseaworthiness) applies. The argument has some persuasiveness.

Also, we wonder about this possibility: Suppose the pipes are superbly marked. Suppose a highly trained man, through human error, puts the water into the bilge sounding pipe. Unless we adopt a theory of “instantaneous unseaworthiness,” the act of placing of the hose could hardly qualify as unseaworthiness without more. But what about the second, the third and the fourth hours with the water running ?

Another factor which may have entered into the trial court’s determination is the complete failure of Alaska Steam to produce any photographs of the purported stampings of raised letters in the caps or collars which it claimed were adequate to tell a workman which pipe was which. Unexplained to a trier of fact, this point is quite serious.

In our view, there may have been error in the management of the ship, but in a series of events one may find unseaworthiness and negligence in “management of the ship” causing damage. The presence of the latter does not vitiate the former.

The decree for the damage to Hyda-burg’s canned salmon is affirmed.

. A sounding of the bilge would be used ordinarily to test for the presence of ■water that should not be in the bilge below the hatch. . As part of the pre-trial order, Alaska Steam rightly admitted having the burden of exoneration. . Cf. W. T. Lockett Co. v. Cunard S. S. Co., E.D.N.Y., 21 F.2d 191 . . Union Carbide & Carbon Corp. v. The Walter Raleigh, S.D.N.Y., 109 F.Supp. 781 , aff’d sub nom. Union Carbide & Carbon Corp. v. United States, 2 Cir., 200 F.2d 908 .