The Doris Kellogg, 18 F. Supp. 159 (S.D.N.Y. 1937)

*160 Kirlin, Campbell, Hickox, Keating & McGrann, of New York City (Charles R. Hickox and James H. Herbert, both of New York City, of counsel), for petitioner, Kellogg Steamship Corporation.

Hunt, Hill & Betts, of New York City (John W. Crandall and Frank J. Zito, both of New York City, of counsel), for claimant, Atlantic Refining Co.

GODDARD, District Judge.

Petition for limitation and of exemption from liability under sections 4282-4287 of the Revised Statutes (46 U.S.C.A. §§ 182-187).

The petition was filed by the Kellogg Steamship Corporation as a result of the loss of a cargo of Hobbs crude oil which occurred when the tank steamer Doris Kellogg caught fire from an explosion and burned while on a voyage from Atreco, Tex., to Philadelphia, Pa., on December 29, 1932. She was owned by the New York & Philippine Steamship Company and was under a bare boat charter to the Kellogg Steamship Corporation. The Kellogg Steamship Corporation, by a tank voyage charter party, dated December 5, 1932, had chartered the vessel to the Atlantic Refining Company for voyages from gulf ports to ports of Philadelphia or New York.

On June 6, 1933, a libel in personam was filed by the Atlantic Refining Company against the Kellogg Steamship Corporation to recover the value of the cargo of oil which was lost. It was followed on December 19, 1933, by the filing of this petition for exoneration from and limitation of liability for the loss by reason of the fire happening on board the vessel.

The Doris Kellogg, formerly the Cedarhurst, was built in 1920 or 1921 by the International Shipbuilding Corporation at Hog Island as an ordinary dry cargo carrier. She was a three island well decked vessel with a main deck and underneath that a second deck. Her dimensions were 390 feet in length, 51 feet wide, and 32 feet deep. The space between those decks was known as the 'tween deck space and below the second deck were the holds. There were five hatches and five holds. Within the year after her completion, she was converted into a tanker. In 1929 Spencer Kellogg & Son, Inc., purchased her and at that time and ever since she has been classed by the American Bureau of Shipping as a "Crude oil carrier A-1." Upon her purchase Spencer Kellogg & Sons, Inc., chartered her to the Kellogg Steamship Corporation, the petitioner, on a bare boat basis for five years and this charter provided that "charterer shall at its own expense man, operate, victual, fuel and supply the steamer. * * *" Shortly after acquiring title, Spencer Kellogg & Sons, Inc., transferred it to the New York and Philippine Steamship Company and that company ratified and adopted the charter party previously made to Spencer Kellogg & Sons, Inc., and the Kellogg Steamship Corporation.

On December 29, at about 1:35 p. m. while proceeding off the coast of North Carolina with a cargo of 51,127.88 net barrels of Hobbs crude oil, she was suddenly shaken by a heavy explosion occurring aft of her amidships section; this was followed by fire. A few minutes later there was another explosion and, according to some witnesses, a third explosion occurred after another interval of several minutes. The officers and crew escaped in life boats and were rescued by another tanker, the Delaware Sun which fortunately happened to be in the vicinity. The Doris Kellogg was last seen at 11 o'clock that evening when she was ablaze from amidships aft. Later she sank.

At the time the Doris Kellogg was converted into a tanker it was not considered practical to use the entire space under deck for tanks as that would have given her too much dead weight when loaded. The bulkhead between Nos. 4 and 5 holds was retained and a longitudinal bulkhead was installed amidships from the pump room bulkhead to the after peak bulkhead. This extended from the top of the shaft alley to the under side of the 'tween decks. The lower holds were further subdivided by transverse bulkheads and Nos. 7 and 8 port and starboard oil tanks were thus formed in old No. 5 hold. The Nos. 4 and 5 'tween decks remained as Nos. 4 and 5 dry cargo spaces and were not used. The No. 5 dry cargo space was approximately *161 24 feet wide and 52 feet long. Its height was 8 feet except in the way of the hatch where it was 12 feet. The original hatch which served the No. 5 'tween deck and hold was the same width as the No. 5 dry cargo spaceabout 24 feet. But in December, 1930, the wooden hatch covers which she then had were removed and were replaced by a permanent steel cover and a portable hatch 6 feet by 4 feet was installed in the center. The after end of the original hatch was some 4 feet forward of the poop bulkhead and then ran forward a distance of 27 feet. The surrounding hatch coaming stood about 3 feet above the main deck. The Nos. 4 and 5 dry cargo spaces were separated by a steel bulkhead. Underneath and on all sides of these dry spaces were tanks in which the Hobbs crude oil was stowed. At the forward end of the No. 5 dry cargo space and near the center line of the ship there was a man hole 15 inches by 23 inches, providing access to the No. 5 dry cargo space. This man hole was on a line with the forward bulkhead of this space and an iron ladder was affixed to the bulkhead to enable a man to climb down. In the forward end of the No. 5 dry cargo space was a cowl ventilator about 18 inches in diameter which was permanently closed at the time in question.

The engine room, boiler room, and officers' quarters were located in about the center of the vessel. From the after end of the officers' quarters, the well deck continued back to the poop bulkhead. After the poop bulkhead of the main deck and the deck above were the crew's quarters, steering engine, lockers, etc.

She was equipped with two dynamos built by the General Electric Company, each of 15 kilowatts, 125 amperes, and 110 volts and it was the practice to use them on alternate days as one dynamo was enough to provide the needed electricity. Until the explosion occurred, these dynamos or generators were in good order. The wiring system was what was known as a two-wire system, the current being conducted from the generator by one wire and returned through the other one. A line consisting of a positive and negative wire left each generator and ran to a switchboard located in the engine room. Between the generator and the switchboard each line was protected with two fuses with a capacity or "blowing out" limit of 100 amperes. On the switchboard there was a circuit breaker of 75 amperes. The positive and negative wire to each circuit as it left the switchboard was protected by a fuse of 30 or 60 amperes. The amperage of the fuse depended upon the amperage the line was carrying. From the switchboard 10 different lines or circuits ran to various parts of the vessel. Some to furnish lights; others to operate machinery; one to operate the radio and others for various purposes. These lines are generally referred to as "feeders." At the panel box the electricity supplied by the "feeder" was then broken up into sublights or circuits. In the panel boxes each of these sublights or circuits was equipped with two fuses of 15 or 20 amperes.

From the time the Doris Kellogg was converted into a tanker up to December, 1932, four lines of insulated wire in conduits ran through the No. 5 dry cargo space. These conduits were located on the starboard bulkhead of the No. 5 dry cargo space separating it from the Nos. 7 and 8 tanks and were at about the height of a man's head. They were made fast one above the other and held by clamps at intervals screwed into the bulkhead. One conduit on the after end of, but within the No. 5 dry cargo space, passed up through the deck above and ran to panel box "F", which was located in the starboard alleyway in the poop deck. The three other lines continued through the after bulkhead of the No. 5 dry cargo space on the starboard side; one line ran to the alarm gongs; another to the stern light which was located on top of the booby hatch on the poop deck; and a third to the telephone in the steering engine room. Panel box "F", from which ran the circuits supplying the electricity for all the lights in the poop section, was located on the inboard bulkhead of the starboard poop alleyway about 4 feet aft of the door in the poop bulkhead, which lead to the main deck. The bottom of the panel box, which was about 18 by 20 inches and 4 inches deep was some 4 feet above the deck. One feeder line ran to this panel box and four circuits ran from it. Although there is some evidence indicating to the contrary, the weight of the proof is that this panel box "F" was what is known as the "dead face type" where the bus bars and live parts were inclosed and could only be gotten at by taking out some screws and opening a door or cover, and unless opened only the switches and fuses were visible.

*162 Several days before the Doris Kellogg reached New York after her voyage from San Francisco ending December 2, 1932, three of her navigating lights the starboard light, the masthead light, and the range light went out of commission. Temporary lights were rigged as the weather was too rough for repairs to be made at that time. Upon her arrival in New York an employee of the Robbins Dry Dock Company examined the circuits for "grounds" and found three on these circuits. The ground on the foremast and the one on the main mast circuit was located at the plug box above the junction box at the foot of the mast; the ground on the side light cable was at the plug box. Repairs and replacements were then made by the Robbins Dry Dock Company and the following conduit piping was removed: The piping running up to the cross-arm to the foremast; from the foremast aft through Nos. 1 and 2 dry cargo spaces; and from the engineer's store room aft through No. 4 dry cargo space up to the junction box in the forward end of the No. 5 dry cargo space. The conduit containing the stern light feeder was removed from the No. 5 dry cargo space and the stern light was connected to the steering engine aft. All of these old conduits and wiring were replaced with lead and armored cable. After the work was completed, an employee of the Robbins Dry Dock Company inspected and tested the electrical system on the Doris Kellogg and found it to be in satisfactory condition. The result of these replacements and repairs was that although a large part of the Doris Kellogg's wiring system was replaced with the lead and armored cable, the old conduits and wires were left in No. 5 dry cargo space, except the stern light feeder, which was removed and instead of there being four conduits containing wires running through the No. 5 dry cargo space, there were only three such conduits. The system, except for the parts which were renewed at this time, had been installed not later than 1921, when the Doris Kellogg was converted into a tanker.

In the early part of October, 1932, the Doris Kellogg was drydocked for repairs and she was examined for defective rivets. 245 scattered rivets were found to have started and were renewed. 1,638 rivets, which leaked slightly, were caulked and made tight, together with 354 feet of seam. There are altogether upwards of 1,000,000 rivets on the Doris Kellogg. A visual examination of the rivets of the bulkheads was made and no leaking rivets were found. The bulkheads in the No. 5 dry cargo space were painted. All tanks were examined visually and found to be tight and in good condition. Some of the tanks were subjected to a hydrostatic test, but not Nos. 7 and 8 tanks; these were last hydrostatically tested in December, 1931. The repairs, as well as a survey, were made under the supervision of a surveyor from the American Bureau of Shipping who stated that he himself examined all tanks and found them in good condition. On completion of the survey and when she left the dry dock in the middle of October, the surveyor for the American Bureau of Shipping recommended the Doris Kellogg as being in seaworthy condition and qualified to retain her class in the Bureau. Upon completion of these repairs the Doris Kellogg proceeded to the Pacific coast light and there took on a cargo of sardine and perilla oil for Bayway, N. J. Upon her arrival on December 2, 1932, at Bayway her cargo was discharged without shortage. A chemist, representing the consignee of the cargo, boarded the Doris Kellogg to supervise the discharging of cargo. He made an examination of all the dry cargo spaces and has testified that he examined all dry spaces to see if there had been any loss of cargo and said that he found no leakage nor signs of leaking in the bulkheads. After the sardine and perilla oil had been pumped out, the valves were closed and all cargo tanks steamed out and there was then no sign of leakage.

Upon the completion of loading of oil at Atreco for the voyage upon which this explosion occurred, the ullage in No. 7 tank was: Port side 1 ft. 5¼ in.; starboard 1 ft. 4¼ in.; and in the No. 8 tank: Port 1 ft. 5¾ in.; starboard 1 ft. 5 in. The ullages of the tanks were taken twice a day morning and afternoon.

Shortly after the Doris Kellogg got under way from Atreco, it was noticed that the ullage was decreasing due to the expanding of the oil and a small amount of oil was pumped out of each tank into the No. 1(a) tank to restore the other ullages to the same amounts they had when she sailed. On December 28, between 9 and 11 o'clock in the morning, she was discovered to be down by the stern and some of the oil from the after tanks was pumped *163 into a forward tank. After this was done, the ullage in the Nos. 7 and 8 tanks was about 6 feet.

The ship's logs were lost from a small boat after the Doris Kellogg sank so there is no record of the actual air and water temperatures, but the report of the United States Weather Bureau shows that the air temperatures in that vicinity range from 66 to 72 degrees and that the water from 64 to 78 degrees; the weather was fair.

During the afternoon of December 28, which was the day before the explosion, the riding light was reported to be out of commission and the chief engineer, Doriza, examined it and found that the wire where it came up from the deck and entered the bottom of the outlet box, was broken. As it was late and this light was used only when the vessel was at anchor, repairs were deferred until the following afternoon when Doriza, with the third assistant engineer, Russell, went aft and Doriza told Russell how it should be repaired. Doriza said to disregard the broken wire which was connected to the box and to pull some of the slack wire up through the deck, lead it to the outlet box, and make a new connection. Russell says that he examined the wire and found it in good condition and that the break in it looked like a recent one; that he then went over to the fuse box in panel box "F", threw off the switch and unscrewed the fuses so as to cut off the electric current flowing through this line. Before doing so, he saw that lights were burning brightly in the poop quarters. He then went back to the outlet box on the poop rail and spliced the wire, taped it, and made it water tight. After finishing these repairs, he returned to the fuse box in the starboard after alleyway and soon after heard a muffled explosion. He does not remember whether he had screwed back any of the fuses or touched any of the switches before the explosion, nor what, if anything, he did there.

The witnesses are in agreement that the first explosion occurred about 1:35 p. m., but they do not agree as to just where the first explosion occurred nor as to the condition of the deck in the vicinity of the explosion immediately afterward. The testimony of these several witnesses was all taken by deposition with the exception of one witness, McNulty, who testified at the trial. Their testimony cannot be reconciled and many of their own statements are inconsistent, although it is quite natural that in the excitement which followed the explosion there would be confusion and that their versions as to just what happened might differ. Some of them were quite ignorant of the English language in which they were questioned; others did not have a definite idea as to the locations of the No. 5 dry cargo space and Nos. 7 and 8 tanks. McNulty, whom I did see and hear testify, did not impress me as giving a reliable account as to what occurred or exactly what he saw. On the other hand, the reading of the testimony of Russell, the third assistant engineer, left me with the distinct impression that he was honest and intelligent. He was within some 40 feet of where all agreed the explosion did occur. He definitely places the explosion as taking place in the No. 5 dry cargo space and states that the coaming on this space was crumpled up immediately after the first explosion and that smoke and flames were coming from the after starboard corner of that space; also that the deck on the starboard side of this space was perfectly solid after this explosion. This is corroborated by several others, although some of the petitioner's witnesses say that flames and smoke came from the vicinity of the Nos. 7 and 8 tanks. After going over all the testimony quite carefully and comparing details, there is a convincing weight of evidence, in my judgment, to the effect that the first explosion occurred in the No. 5 dry cargo space, and I so find.

The contention of the claimant is that the loss of its cargo was due to the petitioner's own neglect; that therefore petitioner is deprived of the benefit of the fire and limitation statutes even if they apply; also that these statutes do not apply because the contract of carriage involved is petitioner's personal contract; further, that the petitioner in any event waived the protection of the fire statute. In particular, the claimant contends that Russell created a short circuit at panel box "F" on the electric line running from the panel box through the No. 5 dry cargo space and this ignited gas or vapor which had leaked into this space from the oil in the surrounding tanks, producing the explosion.

Certain interrogatories were propounded by claimant to petitioner and a motion was argued before Judge Bondy respecting them, and on August 15, 1935, Judge Bondy entered the following order:

*164 "* * * Ordered that the burden of proof in the limitation of liability proceedings herein is on the claimant, Atlantic Refining Company, to prove that the loss was occasioned by negligence; if negligence is found, the burden of going forward then shifts to the petitioner, Kellogg Steamship Corporation, to show that the negligence was without its privity or knowledge; that the burden of proof under the Fire Statute set forth in the petition herein is upon the claimant, Atlantic Refining Company, to show that the fire which destroyed the cargo was occasioned through the design or neglect of the petitioner. * * *"

This, of course, constitutes the law of the case in this court and the fundamental question to be decided is whether the claimant has shown that the loss was occasioned through the design or neglect of the petitioner.

The petitioner denies that it was negligent in any respect and says that the electric wiring system was properly installed and was in good condition; that Russell's act did not and could not have produced an arc or spark in the No. 5 dry cargo space; also that there was no accumulation of gas or vapor from the oil in the No. 5 dry cargo space to ignite and explode.

To support its contention the claimant must show that the explosion was due to the ignition of explosive gas in the No. 5 dry cargo space by a spark or other electrical manifestation from the electric wiring. But these requirements are met if conditions and circumstances be shown from which the inference could reasonably be drawn that such factors existed and produced the explosion. Standard Oil Co. of New York v. R. L. Pitcher Co., 289 F. 678 (C.C.A.1). And to entitle it to recover claimant must prove that this occurred through the neglect of the petitioner.

The flash at the panel box immediately followed by the explosion in the No. 5 dry cargo space certainly suggests that they were related and that the spark produced at the panel box and there manifested by a flame traveled along the electric wire passing through the No. 5 dry cargo space and ignited something in there causing the explosion. Nobody saw or could see just how the explosion took place for the electric wire was out of sight and the No. 5 dry cargo space was a closed chamber, but this does not necessarily mean that the cause of explosion may not be determined with a fair degree of accuracy for circumstantial evidence or a reasonable inference drawn from the circumstances is good proof, if convincing.

It is definitely established that the No. 5 dry cargo space was empty and closed; that its one ventilator was permanently covered; that there was nothing in it to explode unless oil or vapor from the oil leaked into it from the tanks of oil which surrounded it. Hobbs crude oil, which the tanks contained, has a gasoline content of 36 per cent. and is classed as one of the lighter crudes and the vapor from this oil has a much lower viscosity than liquids and will pass through openings too small for liquids to flow through. The proportion by volume of this vapor necessary to form an explosive mixture when mingled with air is stated by the various chemists and engineers to run roughly from 1.5 per cent. to 6.5 per cent. The side bulkheads of the No. 5 dry cargo space contained some 3,000 rivets. If some of these rivets had become loose, a sufficient amount of gas would have leaked into the No. 5 dry cargo space within a comparatively short time to have produced an explosive mixture there, and the testimony is that there is a tendency for rivets to loosen through the working of a ship in a sea way.

In October, previous to the explosion, when a survey was made of the Doris Kellogg, 245 loose and 1,638 slightly leaking rivets were found in various parts of the ship. These were then repaired, but it is not unlikely that in the interval others had worked loose. The examination of the rivets in the No. 5 dry cargo space bulkheads in October was merely a visual one. The surrounding Nos. 7 and 8 tanks had not been subjected to a hydrostatic test since December, 1931, and this is the customary test. On the eight previous voyages prior to October, 1932, she had carried molasses. After that and just previous to the one on which the explosion occurred, she had made a voyage to San Francisco and brought back fish and perilla oils. The statement of the chemist representing the consignee of the fish oils that he made an examination of the bulkheads and dry cargo spaces when she was discharging at Bayway on her previous voyage and saw no signs of leakage, is not as reassuring as to the condition of the rivets as a careful inspection made by an expert before she left Texas on this return voyage *165 would have been. The gas or vapor from Hobbs crude oil will penetrate when fish oil will not.

"* * * Not only is fuel oil much more penetrating, but it is more difficult to contain than water." The Turret Crown (C.C.A.) 297 F. 766, at page 775.

One witness, McNulty, testified that he had spent five or ten minutes in the No. 5 dry cargo space two days previous to the explosion using a flashlight and going around slowly to see if there were any leaks "or anything like that," and that he had done this each day and found no signs of leakage and detected no smell of oil or gas. However, in view of his statements on other matters and the impression I received while he was testifying, I am unable to accept his testimony as reliable. Henderson, the chief officer, testified that he went down in the No. 5 dry cargo space the day before the explosion as McNulty was engaged in some other work and that there was no smell of gas in it. He also said that on the morning of the explosion he went into Nos. 4 and 5 dry cargo spaces; that his purpose was to look over these spaces with the idea of trying out some new paints on the bulkheads; and that he wanted to see if they needed brushing or wiping down before he started painting. In the light of the testimony that the bulkheads did not need painting, also that these spaces are dark and difficult of access an unlikely place to select for his purposehis story seems improbable. However, this vapor is heavier than air and would sink to the lower part of the compartment, and because these two say they smelled nothing is not proof that there was no gas in the No. 5 dry cargo space. In Lancashire Shipping Co. v. Morse Dry Dock & Repair Co. (D. C.) 43 F.(2d) 750, at page 757, affirmed (C. C.A.) 48 F.(2d) 1077, Judge Campbell said, "That Whitehead and Jepson were not overcome by the fumes does not indicate to me that an explosive mixture of gasoline vapor could not have been present, because the vapor was heavier than air and was accumulated in the tunnel well below the wooden flooring which, though not closely fitted, would furnish adequate protection from cross currents of air and was below the level of the feet of men walking on the flooring, and it may well be true that Whitehead and Jepson told the truth when they said they were not conscious of the smell of gasoline, as gasoline vapor has an anaesthetic action on the olfactory nerves."

I think that the only reasonable inference which can be drawn from all the circumstances, in the absence of any other explanation, is that there was an accumulation of explosive vapor in the No. 5 dry cargo space which had seeped in from the surrounding tanks of oil.

The circumstances, it seems to me, lead quite irresistably to the conclusion that this accumulation of vapor was ignited by a spark from the electric wires running from the panel box "F" through the No. 5 dry cargo space. The explosion immediately followed the flash at the panel box, and there was nothing else in this empty space to ignite it.

Russell was burned about the face and shoulders by a flame which suddenly shot out from the panel box while he was working at the box with a screw driver. He does not remember just what he did. He undoubtedly produced a short circuit either accidently by bridging with the screw driver two of the bus bars in the panel box, which are located about an inch apart, or through making a defective splice in the broken riding cable. Petitioner urges that the panel box was of the closed front type in which the bus bars are concealed by the cover which is fastened on with screws or swung on hinges; but even so, if that cover had been unscrewed or had been swung open, the bus bars would have been exposed and Russell testified that he saw the wires inside the panel box turned back and taped, which indicates that the cover, if there was one, was open while he was working.

There has been much testimony by electrical engineers and others as to whether or not a short circuit at the panel box could have produced a spark on the wire running from there through the No. 5 dry cargo space. Those called by the claimant say that in their opinions that it could have done so, while those called by petitioner express just as definite opinions to the contrary. The petitioner's experts contend that no spark nor arc could have been created on the system which would burn a hole through the iron conduit as the melting point of the fuse or the copper wire is below that of the iron conduit. However, fuses have fixed time "lags" of several seconds, and a short circuit of the bus bars would create a heavy overload on the feeder from the panel box to the main *166 switchboard, and barring interruptions the overload would go up to the capacity of the generator. A section of a conduit pipe was offered in evidence which had been burned through in a system and under conditions comparable to these in question, and the testimony is convincing that in spite of protective system, damage accompanied by sparks not infrequently results when a short circuit occurs if there is a loose connection or the insulation becomes worn. The weight of the testimony is that a short circuit in a conduit may cause a manifestation of electrical energy or heat which can burn through a conduit of the character in the No. 5 dry cargo space. I think that is what happened here and that it ignited gas in that space. This seems to me to be the only reasonable inference in the absence of any other means of ignition in this closed empty space. The iron pipes or conduits through which the wires in the No. 5 dry cargo space ran were at least eleven years old, having been installed not later than 1921, when the Doris Kellogg was converted into a tanker. The lead and armored cable is generally regarded as more satisfactory for vessels than the iron pipe conduit system as there is a tendency for the moisture to condense in the pipes and cause the rubber insulation on the wire to deteriorate. After the completion of her previous voyage from San Francisco, several sections of her wiring system were ripped out and replaced with lead and armored cable; but the conduits with their wires were left in the No. 5 dry cargo space.

"It was not necessary that the plaintiffs show that a static spark was actually produced; but it was sufficient if conditions and circumstances were shown from which the inference could be reasonably drawn that a static spark was produced." Standard Oil Co. v. R. L. Pitcher Co., supra, 289 F. 678, at page 683. See, also, Fitzgerald v. Brooklyn Inst. of Arts & Sciences, 175 App.Div. 554, 162 N.Y.S. 625; Leathem Smith-Putnam Nav. Co. v. Osby et al., 79 F.(2d) 280 (C.C.A.7); The New Berne, 80 F.(2d) 244 (C.C.A.4).

Assuming that some careless act of Russell at the panel box resulted in igniting the vapor in the No. 5 dry cargo space, his act is not the "neglect" of the petitioner necessary to establish liability under the fire statute (46 U.S.C.A. § 182). For the words "neglect of such owner" in the statute mean negligence of the owner personally, or in the case of a corporate owner, negligence of its managing agents, as distinguished from that of the ship's officers. Earle & Stoddart v. Wilson Line, 287 U.S. 420, 53 S. Ct. 200, 77 L. Ed. 403.

Nor is there sufficient proof to charge petitioner with such "neglect" for failure to have reasonable inspections made of the bulkheads surrounding the No. 5 dry cargo space. However, I think that the petitioner is chargeable and directly chargeable with "neglect" in maintaining the unnecessary fire hazard in the No. 5 dry cargo space. Running electric wires through this empty space, which had little or no ventilation and surrounded by tanks filled with Hobbs crude oil containing a high percentage of explosive gas, certainly is not free from danger; in fact, it was clearly potentially dangerous and subjected the cargo to undue risk. In the event of the quite reasonable contingencies the working loose of some of the rivets in one of the bulkheads allowing gas to come in and a short circuit, or other accident to the wiring system, neither unusual, an explosion was to be anticipated. Where it was known that the consequences of such a potential hazard may be so disastrous, it is reasonable to require that commensurate care will be used to avoid the existence of such risk.

Although the Doris Kellogg may have been safe for the carriage of other cargo, the question now presented is whether she was properly equipped for the transportation of Hobbs crude oil, which she undertook to carry. Church Cooperage Co. et al. v. Pinkney et al., 170 F. 266 (C. C.A.2); The Southwark, 191 U.S. 1, 24 S. Ct. 1, 48 L. Ed. 65. And it was incumbent upon the petitioner to use the degree of care commensurate with the dangerous nature of the cargo. The Willfaro (D.C.) 9 F.(2d) 940, affirmed Williams S. S. Co. Inc. v. Wilbur et al., 9 F.(2d) 622 (C.C.A. 9).

In International Mercantile Marine S. S. Co. v. W. &. A. Fletcher Co., 296 F. 855, 858 (C.C.A.2), the steamer St. Louis while undergoing repairs in the respondent's yard, was badly damaged by fire. Employees were removing varnish from a staircase, with a volatile liquid varnish remover. Cotton waste, soaked with benzine, was used to wipe off the remover and was then thrown down on deck. Other employees nearby were removing paint with an open flame torch. The fire started *167 in the cotton waste. The court held the respondent liable, saying, 296 F. 855, at page 858:

"* * * What caused the removersoaked waste to blaze is the vital inquiry. The only cause suggested by the evidence is the blowtorch, and the maintenance of that probable cause in proximity to so much inflammable material was itself negligence. Liability is measured by the known dangers to be guarded against, and if care according to the circumstances is wanting, the natural inference is that injury accrues from the known danger it is caused by the lack of care. The blowtorch near remover and waste was negligence, the danger of fire was well known, and we find adequate cause proximately existing in that negligence for the ensuing loss."

In Christopher et al. v. Grueby et al., 40 F.(2d) 8 (C.C.A.1), the fishing schooner Commonwealth was lost as a result of a fire and several members of her crew were killed and injured. She was equipped with a lighting plant run by a four-cylinder gasoline engine which frequently backfired. It was necessary to pour the gasoline into the tank through a small orifice, and gasoline was often spilled on the floor in the operation. The court held the petitioner liable and denied limitation, saying [40 F.(2d) 8, at page 13]:

"Whether the fire was started by a spark from a wire, or by a spark from the uncovered generator brushes, or from the back-fire of the engine, or from some other source, it can reasonably be found that the negligent situation, which the owners had created and/or suffered to exist in the engine room, and their failure to furnish adequate means `ready for immediate use' for `promptly and effectually extinguishing burning gasoline,' one or both, were the cause of the loss of the ship, of the lives of 12 of the crew, and of the injury of three others.

"If the engineer was careless in pouring gasoline when the engine or engines were running, nevertheless the owners could reasonably have foreseen that he might do so, and their negligence in suffering this fire hazard to exist in the engine room could be found to be the proximate cause of the disaster, and we think it was."

In International Marine Co. v. Fels (D.C.) 164 F. 337, the owner of the steamer Haverford sued to recover damages resulting from an explosion in the No. 3 hold of the vessel, in which 1,000 boxes of respondent's soap had been stowed. On the ship's arrival in Liverpool the hatches were removed and shortly afterward a violent explosion occurred. The court found that the soap contained naphtha and despite the wrappings, gave off a vapor which, when mixed with air in the right proportions, was highly explosive if brought into contact with a flame or spark. It held that the explosion was caused by the vapor and said (164 F. 337, at page 340):

"It appears that the naphtha was the only commodity in the hold that was of an explosive nature, and in view of the explosion that took place, which was of such a character as might readily have been produced by naphtha, it does not seem that the fact of there being no definite proof of the manner of ignition, should prevent the only rational explanation of the explosion being adopted, without regard to the details."

Several naval architects and other experts were called by the claimant. All stated that there was a potential danger in running electric feeder wires through such spaces on oil tankers because of the possibility of the accumulation of gas and their ignition by an arc or spark from the wiring caused by short circuits, etc. None of the disinterested witnesses called by the petitioner were willing to give their unqualified approval of such practice. According to the testimony, there are but few tankers where the electric light feeders are carried below deck and these, like the Doris Kellogg, are converted dry cargo carriers. A number of shelter deck tankers have feeder wires in their shelter deck; but the shelter deck usually extends nearly the length of the vessel and has full ventilation from hatches and ventilators. The shelter deck is not comparable to the No. 5 dry cargo space.

That the petitioner realized the danger to the cargo from electric wires is indicated from the instructions given by Mr. Haslam, its marine superintendent, to cut off any circuits when not in use to reduce the risk of explosion. The wiring on the Ruth Kellogg, a sister ship, was removed and run along the fore and aft bridges. In 1931 the then master of the Doris Kellogg said that he "didn't like the idea" of the electric wires under deck and that on one voyage when he carried crude oil he strung temporary wiring above deck *168 from the bridge deck to the running lights. He also testified that in 1931 he talked to Mr. Haslem about the wiring being under deck and about changes which would be made in the future. This conversation was not denied.

Section 37 of the American Bureau of Shipping Rules (1917 issue) provides in article I: "In vessels carrying petroleum or any substance giving off an explosive gas no permanent conductors must be installed in any compartment liable to contain such fumes."

In "Recommended Practice for Electrical Installations on Shipboard, Marine Rules," published by American Institute of Electrical Engineers, 1920 Edition, it is provided on p. 26 that, "Feeders of every description should be run with a view to avoiding heated spaces such as engine and fire rooms and galley, spaces containing gases such as fire rooms, pump rooms, summer tanks, spaces where damage may occur such as cargo spaces and exposed sides of houses."

The 1930 Edition of American Institute of Electrical Engineers provides, page 38, as follows: "Feeders of every description should be run with a view to avoiding spaces where heat and gases may be encountered such as galleys, fire rooms, pump rooms and oil or `summer tanks'."

It is plain that the wiring on the Doris Kellogg was not in accordance with these rules. While such standards of safety, which are urged by the petitioner to be mere recommendations, are not controlling, they do reflect the experience and judgment of experts and should be taken into consideration, together with all the facts and other accepted standards if there be any in deciding whether petitioner was negligent. The Vestris (D.C.) 60 F.(2d) 273. The American Bureau of Shipping (1926) recognizes that these rules are sound, for its article 82 reads, "The foregoing rules are considered minimum requirements and equivalent materials or arrangements will be accepted if specifically approved. In the absence of any requirement for electrical and interior communication systems on shipboard the current Marine Rules of the American Institute of Electrical Engineers will be acceptable."

The petitioner urges that 1926 rules of the American Bureau of Shipping and the 1930 Marine Rules of American Institute of Electrical Engineers were not in force either when the Doris Kellogg was built in 1920, nor when, about a year later, she was converted into a tanker, so should not be considered. But in determining whether the petitioner was guilty of neglect in 1932 in maintaining its wiring system, it is proper to take into consideration with the other facts that it was then generally recognized as a result of experience, which these rules reflect, that such a system was unsafe. Atlantic Transport Co. et al. v. Rosenberg Bros. & Co. et al., 34 F.(2d) 843 (C.C.A.9).

The petitioner submits as evidence that the Doris Kellogg was in fit condition; that the surveyor for the American Bureau of Shipping recommended in October, 1932, that she retain her classification of "A-I," which she did. But it does not appear that her electrical system was surveyed at that time, however. The American Bureau of Shipping Rules provide for a special survey every four years, which includes machinery, electric system, and equipment. The Doris Kellogg "No. 3 survey" was begun in October, but it was not completed when she sailed for San Francisco and the survey of her entire electrical installation remained to be carried out later. Certificates of surveyors and inspectors are to be valued in the light of what the actual facts disclosed. The Vestris (D.C.) 60 F.(2d) 273. See, also, Bank Line v. Porter, 25 F.(2d) 845 (C.C.A.4); The Abbazia (D.C.) 127 F. 495; The Negus (D.C.) 298 F. 749.

The petitioner urges the following theory as to the cause of the explosion: That the sulphur in the oil combined with the iron in No. 7 tank to form iron sulphide, a pyrophoric compound; that with the increase of the ullage on the day before the explosion, the iron sulphide on the sides of the tank was exposed to oxygen, and that this created sufficient heat to ignite the vapor over the oil in the tank. But as I find that the explosion occurred in the No. 5 dry cargo space and not in the No. 7 tank, there is no reason for discussing the merits of this theory.

Mr. Haslam, vice president and marine superintendent of the petitioner, was fully aware of these conditions on the Doris Kellogg and that is sufficient to charge the petitioner with the neglect under the fire statute (R.S. § 4282, 46 U.S.C. A. § 182). United States v. Charbonnier, 45 F.(2d) 174 (C.C.A.4); Arkell & Douglas, *169 Inc., v. United States, 13 F.(2d) 555 (C.C.A.2).

And as the loss occurred with the petitioner's privity and knowledge, it is not entitled to limitation of liability under sections 4283-4287 of the Revised Statutes (46 U.S.C.A. §§ 183-187).

I think the claimant is entitled to recover its loss.

Accordingly, the petition is denied with reference to a commissioner as to amount of its damage. If the foregoing be deemed insufficient, compliance with Admiralty Rule 46½, 28 U.S.C.A. following section 723, findings and conclusions may be settled on notice.