(after stating the facts as above). The District Judge examined into the cause of collision, and, from the testimony offered, incomplete as it is, concluded that the collision was due to faulty navigation. It was determined that the loss was not covered by 'the war risk policy, but was cohered by the marine policy, and the libel was accordingly dismissed. The provisions of the war risk policy assume all the consequences of hostilities or warlike operations, while the f. c. and s. clause of the marine policy excludes all consequences of hostilities or warlike operations. The appellee’s war risk policy covers “acts of kings, princes, and people authorized by and in prosecution of hostilities between belligerent nations.”
In order to impose liability under the war risk clause policy, “all forms of hostilities or warlike operations of whatever kind” must consist of some form or kind of hostility or warlike operations which have proximately caused the loss. Remote consequences of hostilities cannot become a recoverable loss, even if they may be said to be proximately caused by soniething itself ascribable as a consequence of hostilities. British Steamship v. The King, [1921] 1 A. C. 99, 107, 131.
The phrase “all consequences” has been held by the British *979courts as meaning “proximate consequences only.” Ionides v. Universal Marine Ins. Co., 14 C. B. (N. S.) 259 (1863); Anderson v. Martin, [1908] A. C. 334, 339. The loss was by a collision at sea, and this prima facie is considered a marine peril. To establish a loss which falls upon the war risk underwriter, the burden must be borne by him who attempts it, and in this case the appellant. Therefore the question is presented: What is mean by the phrase “hostilities and warlike operations,” as used in the war risk policy? There is little of aid in the authorities established by the courts of this country. The question has been considered by the British courts in a considerable number of cases. There the authorities may be summarized as holding that “hostilities” is intended to describe an actual operation, offensive or defensive, in the conduct of war, and “warlike operations” as operations in time of war. The peril must be due directly to some hostile action, if it be considered a warlike risk. If the peril be a maritime risk, and is but aggravated or increased by the operation relied upon as a warlike operation, then the risk is not a war risk. British S. S. v. The King, [1921] A. C. 99, 133. There it was said:
“If the operation relied upon as a warlike operation is one which creates no new risk, but only aggravates or increases an existing maritime risk, by removing something which, but for the war, would have been a safeguard against the risk, then the risk is not a war risk. But, if the peril be directly due to hostile action, it is a war risk.”
When one vessel of one of the convoys was attacked by a submarine, the course of the convoy was altered four points to the right; but before the convoys met, and this collision ensued, it was again altered four points to the left. This latter change must be deemed to have neutralized the former. The trial judge, who exhaustively examined the navigation of the vessels, concluded that the torpedoing of the vessel in the west-bound convoy was 5 hours before the meeting of the east-bound fleet, and did not affect the question of the collision. or its cause. It did slow up the convoy making progress towards its point of destination. It was about 30 nautical miles from the scene of the torpedoing of the Merida to the point of collision, and about 5 hours were consumed in making that distance. It is explained that there was zigzagging of the vessels in making at least part of this distance. We conclude that this attack, occurring, as it did, some 5 hours previous to the time of the collision, has no bearing upon the cause of the collision.
The appellant’s contention is that this engagement was a warlike operation, because (first) the vessels were sailing without lights; (second) that they were proceeding in convoy; and (third) that the Napoli carried a cargo intended for warlike use. Where vessels proceeded at sea during this war period, it was the custom to sail without lights. These vessels were all operated without lights. The voyages upon which each of the vessels were engaged would, if in time of peace, be treated as an ordinary maritime adventure. It would not become a warlike operation, within the intention of the terms of the policy, because of the fact that, as a precaution against possible attack or capture, the masters of the vessels did not show lights during the night, *980and even though the consequences of such action meant the concealment of the vessels and their liability to collide. It may be imprudent navigation to take this risk, or it may be blameworthy from other points of view; but, if it is done in obedience to lawful commands, it cannot be considered a warlike operation. The object, of course, is to avoid an ememy’s attack; but no enemy was present at the time of the collision. The purposes of the adventures of the ships were peaceable. Neither vessel was doing a warlike act, and those who issued the order to the navigators of the vessels did not consider their orders to be warlike, even though performed in a war period. In a word, nothing of actual hostilities was present at the time of the collision.
Therefore we conclude that sailing without lights was neither a warlike operation nor was it the consequence of one. The reason for sailing without lights was due to the submarine peril, but the collision was due to faulty navigation, which cannot be said to be the proximate consequence of the failure to have lights. It appears that the lights were turned on just before the collision as the convoys met. The authorities of the House of Fords of England are fully in accord in pronouncing that sailing without lights does not constitute a warlike operation. The Petersham Case, Britain S. S. Co. v. The King, [1921] 1 A. C. 99. In the Ionides Case, 14 C. B. (N. S.) 259, a cargo was placed on board the ship Fhrwood, insured by a policy warranted free from all consequences of hostilities. On her way from New Orleans to New York in 1861, her master, supposing that he had passed Cape Harteras, instead of keeping his course N. N. E., changed it to W., and went ashore about 10 miles south of the Cape. 'Until the outbreak of the Civil War, a light had always been maintained at Cape Harteras; but it had been extinguished by the Confederate authorities for hostile purposes. The court of first instance in England held the proximate cause of the loss was the peril of the sea and that the underwriter was therefore liable. This decision was by the Court: of Common Pleas. It later met with approval in the Petersham Case, supra, and was approved by Mr. Justice Brown, when a District Judge, in Richelieu & Ohio Navigation Co. v. Boston Marine Ins. Co. (C. C.) 26 Fed. 605, and later in The Titania (D. C.) 19 Fed. 104, by Judge Addison Brown. It involved the application of the doctrine of proximate cause. There, there was hostile action in the matter of the extinguishing of Pe lights by the Confederate troops, and there was a maritime casualty due to stranding, both of which had bearing upon the liability under the marine policy which warranted “free from all consequences of hostilities,” and the question whether the proximate cause of the loss was a marine peril or hostilities was presented. It always becomes necessary to distinguish between causes and conditions as well as proximate and remote causation. The distinction must always be maintained between warlike operations and acts done in the course of a war.
The maxim, “causa próxima non remota spectator,” as a guide in this determination, compels the conclusion that the absence of light does not justify the claim of the appellant that that of itself consti*981tutes a warlike operation. Nor was it a warlike operation because the vessels were sailing in convoy. The vessels were not required to sail in convoy. It was optional with each master to become a member of the convoy. It proved to be the custom of the war period. If a vessel joined a convoy, it was required to conform to the instructions as to courses and management given by competent naval authorities. It perhaps meant an increase of the maritime risk of collision. It was sailing in company with other vessels for the salce of protection in case of danger of attack; but each vessel was still on a peaceful adventure. It was the duty of the naval officers and the escorting vessels to diligently perform their duties of convoying and protecting the ships, and to defend the ships in case of attack, and not to 'burden them or expose them to hazard.
But at the time of the collision none of these things which it was the duty of the escorting vessels or naval officers to perform took place. There-was no requirement at that time for their services. It was no command of the naval officer which brought about the collision. It has been held by a divided court in England that sailing in convoy is not a warlike operation. British India Steam Navigation Co. v. Green, [1919] 1 K. B. 632; [1919] 2 K; B. 670; [1921] 1 A. C. 99. In the House of Lords, the majority of the Lords declined to hold that the vessels convoyed become identified with the ships of war protecting them. The east and west bound convoys were each navigating for itself. The course of the convoys might cross in any event, and it was obviously impossible for the allied naval authorities to have laid out their courses so that they would not meet. Each sailed in complete ignorance of the other’s sailing, from all that the record shows, and we find nothing concerning the convoys’ courses which might be said to constitute a warlike operation.
Nor can it be said that fhe character of the cargo made the adventure a warlike operation. The nature of the operation, and not the character of the cargo or the persons in charge, has been held to be the material thing to be considered in England. Owners of S. S. Marchtrove v. The King, 36 Times L. R. 108. There a vessel chartered to the United States Navy, carrying munitions for the American Army in France, was held to be only a cargo ship, and in Harrison, Ltd., v. Shipping Comptroller, [1921] 1 K. B. 122, the vessel stranded on her way from Salonika to Tranto, having on board hospital stores for the British government and carrying a few British troops, and it was held that she was not engaged in a warlike operation, since her aim was to get as peacefully and expeditiously as she could to her harbor of destination. The war risk would be the same, no matter what the character of the cargo, as would the marine risk. “The dominant feature of the ship and the dominant object of the voyage must be looked at.” Harrison, Ltd., v. Shipping Comptroller, supra.
The District Court held that both the navigators of the Napoli and the Lamington “failed in their ship management to.exercisp the ordinary care and skill of their calling,” and concluded that the proximate cause of the collision must therefore fall upon the marine underwriters. An examination of such testimony as has been offered *982leads us to agree with this conclusion. Mr. Justice Hill, who considered the case in the British courts in a suit by the owners of the cargo on the Napoli against the owner of the Lamington (the opinion is part of the record), found no fault on the part of the Lamington, but said that “the immediate cause of the collision was the porting of the Napoli and nothing else.” Both judges have concluded that it was the faulty navigation which brought about the collision, and not a warlike operation.
We think the loss is one that must be sustained by the marine underwriter, and not by the war risk underwriter, and that the decree below must be affirmed.
ROGERS, Circuit Judge. I concur in the conclusion announced in the foregoing opinion.