This action was brought upon a policy of marine insurance. Judgment for respondent was affirmed by the Circuit Court of Appeals, 4 Cir., 89 F.2d 545, and certiorari was granted. 302 U.S. 664, 58 S.Ct. 10, 82 L.Ed. —-.

Petitioner was the owner of a cargo of bananas aboard the Norwegian steamship Smaragd. While proceeding up Chesapeake Bay to Baltimore on July 21, 1935, the vessel stranded and before she could be floated the entire cargo of bananas became overripe and rotted, causing a total loss. Petitioner held a floating policy of insurance which had been issued by respondent on June 23, 1933. The general coverage clause of the policy embraced perils of the sea.

To the declaration setting forth these facts and claiming that the loss was within the coverage of the policy, respondent filed four pleas, the first two pleading the general issue and the two others being special pleas. Petitioner joined issue on the first and second pleas and demurred to the third and fourth. The District Court overruled the demurrer to the third plea and, as that decision was considered by the parties and the court to be conclusive of the issue, final judgment was entered for respondent.

The third plea, thus sustained, set forth a rider which for an additional premium, had been added to the policy on April 4, 1934, and had been canceled on January 25, 1935, before the loss occurred. The rider amended the policy so as to provide that the coverage should be 'free of particular average unless the vessel be stranded, sunk, burned, on fire or in collision, in any, all or several of which events the insurers are liable for such loss by decay, injury or damage to the fruit as is occasioned thereby or occurs during or in consequence of delay resulting therefrom.' The insurers also assumed liability for such loss in consequence of delay resulting from breakage of shaft, loss of blades from propeller, and derangement or breakage of machinery or rudder and/or stern post, 'whether or not the vessel be stranded, sunk, burned or in collision,' provided that the loss amounted to 10 per cent. after deducting 5 per cent. for ordinary loss.

On the cancellation of the rider it was agreed that the policy should have the same coverage as prior thereto, and the premium rate was reduced from 60 cents to the original rate of 25 cents on the $100 of risk. 89 F.2d 545, at page 546.

In affirming the judgment, the Circuit Court of Appeals observed that the judgment had been entered upon the pleadings and that petitioner had conceded that its right to recover depended upon the construction of the policy. 89 F.2d at pages 545, 546. The appellate court then examined the policy and after a review of authorities in this country and in England held that the loss was not within the general coverage clause relating to perils of the sea. The court in concluding its opinion referred to the rider as showing that the parties had interpreted that clause in the same way. Id., page 549. We granted certiorari because of the importance of the principal question thus determined by the Court of Appeals, a question which had not been decided by this Court and as to which the decisions of other courts were said to be in conflict.

We are met by two preliminary questions. The first of these is with respect to the sufficiency of the declaration. It is suggested that the declaration does not allege that the bananas were shipped in sound condition and that they would have been merchantable at the end of a normal voyage, and that there is no allegation as to the duration of the delay. It does not appear that these questions were raised in the District Court and they were not dealt with by the Court of Appeals, which evidently assumed the sufficiency of the declaration to present the main question as to the interpretation of the general coverage clause. Both courts below have decided the case upon the assumption that the fruit was in sound condition when shipped and would have been merchantable at the end of the voyage had it not been for the stranding and the consequent delay. In view of this course of proceedings we make the same assumption. If any question as to the condition of the cargo or length of the delay and its effect had been presented in the trial court, it might have been met by amendment of the declaration and the issue could have been tried; and if the main question, upon the assumption stated, has been wrongly decided and the case is remanded to the District Court, there will still be opportunity to try any other issues of fact or law which may properly be presented.

The other preliminary question is with respect to the effect of the rider above mentioned. We do not regard that indorsement as either controlling or persuasive. Manifestly it did not affect the application of the general coverage clause. That clause had been in effect for a long period before the rider and by express agreement that clause remained in effect after the rider was canceled. The rider covered losses not embraced in the marine perils against which the policy insured and it also covered losses which were already covered by the policy, and there was room for difference of opinion as to exactly how far the rider overlapped. The actual views of the parties when the rider was obtained and the reasons for its cancellation are not shown with any definiteness. For all that appears the insured may have been advised and may have assumed that the coverage clause of the policy gave protection in such a case as is here involved and may not have desired to continue to pay the additional premium for the other losses described. We are not called upon to speculate as to the state of mind of the insured's officers and one asserted ambiguity is not to be cured by another. If the general coverage clause permits recovery, we see nothing to defeat it in what was done in connection with the rider.

Petitioner thus states the main question broadly: Does the general clause of the marine cargo insurance policy, insuring for loss caused by perils of the sea, cover a loss where a marine peril, viz., stranding, has so delayed the voyage that the cargo has become a total loss?

Respondent contends (1) that deterioration of perishable cargo caused through inherent vice while the vessel is delayed by a sea peril, is not, without more, covered under a marine policy which does not expressly insure against such deterioration; and (2) that in order to recover for such deterioration it must be shown that the adventure was not merely delayed but frustrated by reason of the vessel's forced departure from the course of her voyage, as, for example, where the vessel has put into a port of distress and remained there for a period which constituted a virtual abandonment or frustration of the voyage.

In considering these contentions, we start with the fact that the vessel, while proceeding up Chesapeake Bay, stranded. That is alleged and conceded. Stranding is a peril of the sea. Liverpool & Great Western Steam Co. v. Phenix Insurance Co., 129 U.S. 397, 438, 9 S.Ct. 469, 32 L.Ed. 788; Richelieu & Ontario Navigation Company v. Boston Marine Insurance Co., 136 U.S. 408, 421, 10 S.Ct. 934, 34 L.Ed. 398; Arnould on Marine Insurance, 11th ed., § 816; Winter on Marine Insurance, 2d Ed., p. 156. Loss through stranding was within the coverage of the policy. This, as the court below observed, was expressly recognized in the warranties against particular average in which loss by stranding was excepted. 89 F.2d 545, at pages 546, 547. And it was the stranding which caused the delay. The case is not one of the mere lengthening of a voyage due to the ordinary vicissitudes of wind and wave against which the underwriter does not insure, Jordan v. Warren Insurance Co., Fed.Cas.No.7,524, 1 Story 342, 352, and we are not called upon to determine in what circumstances other than those now presented delay may be considered to be due to a peril of the sea within the meaning of the policy.

The cargo was perishable fruit. Respondent insists that its decay was caused by inherent vice which began to operate as soon as the fruit was picked. But, although perishable, the cargo was insured for the voyage against sea perils and the sole question is whether a sea peril caused the loss. As we have said, we must assume for the present purpose, in view of the way in which the case was presented and determined below, that the fruit was sound when shipped and would have been merchantable on arrival after a normal voyage, and that had it not been for the delay due to the stranding of the vessel, the loss would not have occurred despite the perishable nature of the cargo.

We are not impressed by the argument that to permit recovery it must appear that the adventure was not merely delayed but that it was frustrated through a forced departure of the vessel from the course of her voyage and the putting into a port of distress. So far as the cargo in question was concerned, the adventure was frustrated by the stranding, and the cargo became a total loss before the vessel could be floated. That loss was just as complete as if the vessel had been compelled to put into a port and the voyage had then been abandoned.

The sole question is whether in these circumstances the stranding should be regarded as the proximate cause of the loss. Respondent contends that decay or inherent vice was the proximate cause. It is true that the doctrine of proximate cause is applied strictly in cases of marine insurance. But in that class of cases, as well as in others, the proximate cause is the efficient cause and not a merely incidental cause which may be nearer in time to the result. AEtna Insurance Company v. Boon, 95 U.S. 117, 130, 24 L.Ed. 395; Arnould on Insurance, 11th ed., § 783.

The subject was discussed in an illuminating way by Lord Shaw in his judgment in Leyland Shipping Company v. Norwich Union Fire Insurance Society, (1918) A.C. 350, 368—371. He said:

'To treat proxima causa as the cause which is nearest in time is out of the question. Causes are spoken of as if they were as distinct from one another as beads in a row or links in a chain, but—if this metaphysical topic has to be referred to—it is not wholly so. The chain of causation is a handy expression, but the figure is inadequate. Causation is not a chain, but a net. At each point influences, forces, events, precedent and simultaneous, meet; and the radiation from each point extends infinitely. At the point where these various influences meet it is for the judgment as upon a matter of fact to declare which of the causes thus joined at the point of effect was the proximate and which was the remote cause.

'What does 'proximate' here mean? To treat proximate cause as if it was the cause which is proximate in time is, as I have said, out of the question. The cause which is truly proximate is that which is proximate in efficiency. That efficiency may have been preserved although other causes may meantime have sprung up which have yet not destroyed it, or truly impaired it, and it may culminate in a result of which it still remains the real efficient cause to which the event can be ascribed.'

There, in a policy insuring a ship, there was a warranty of exception in case of hostilities or warlike operations, and the question was whether the loss of the vessel fell within the exception. The vessel was torpedoed and sustained severe injuries, but succeeded in making the outer harbor of the port of Havre. Notwithstanding all efforts by pumping and otherwise, she bumped, broke her back, and sank. It was contended that she perished by a peril of the sea because sea water entered the gash in her side which the torpedo made. The entry of the sea water was indeed a peril of the sea and was proximate in time to the sinking. But as 'proximate cause is an expression referring to the efficiency as an operating factor upon the result' it was held that 'the real efficient cause' of the sinking of the vessel was that she was torpedoed and hence that the loss was within the exception.

In the Leyland Case the House of Lords approved the decision of the Court of Appeal in Reischer v. Borwick, (1894) 2 Q.B. 548, where the policy covered collision with any object but excluded perils of the sea. The ship struck a snag which made a hole in her. She was anchored and the leak was temporarily plugged. Then, while she was being towed towards the nearest dock for repair, the water burst through the hole and she had to be run aground and abandoned. The contention was that the proximate cause of the damage was the excepted marine peril of the inrush of the sea water. But the Court of Appeal held that the proximate cause was the collision with the snag.

The same principle applies although within the network of causation there may be found the operation of natural forces to which a disaster, within the coverage of the policy, has given play. The decision of Justice Story in Magoun v. New England Marine Insurance Company, Fed.Cas.No.8,961, 1 Story 157, is an illustration. In that case the question was whether the loss was due to a restraint and detainment of government within the words of a policy insuring the vessel and freight. It appeared that there had been an arrest and detainment by the authorities of New Granada; that the vessel had been restored, but that, when restored, it was found, from her long exposure to the weather in a hot climate, in an open roadstead, that she had been so damaged that she could not perform the voyage without great repairs which would cost more than the vessel was worth; that hides belonging to the cargo had become rotten and were thrown overboard; and that no other vessel could be found to carry the residue of the cargo to destination. The vessel was accordingly abandoned to the underwriters, and Justice Story held that the abandonment was good and the underwriters were liable for a total loss. He said: 'The argument is, that the injury to the vessel, by the long delay and exposure to the climate, was the immediate cause of the loss, and the seizure and detainment the remote cause only; and that, therefore, the rule applies, 'Causa proxima, non remota, spectatur,' and the underwriters are not liable for injury by mere wear and tear, or by delays in the voyage, or by worms, or by exposure to the climate. But it appears to me, that this is not a correct exposition of the rule. All the consequences naturally flowing from the peril insured against, or incident thereto, are properly attributable to the peril itself.'

In support, Justice Story referred to the decision of the Court in Peters v. Warren Insurance Co., 14 Pet. 99, 10 L.Ed. 371, in which he delivered the opinion. In that case he put the following illustration, 14 Pet. 99, at page 110, 10 L.Ed. 371: 'Suppose a perishable cargo is greatly damaged by the perils of the sea, and it should, in consequence thereof, long afterwards, and before arrival at the port of destination, become gradually so putrescent as to be required to be thrown overboard for the safety of the crew: the immediate cause of the loss would be the act of the master and crew; but there is no doubt, that the underwriters would be liable for a total loss, upon the ground that the operative cause was the perils of the sea.'

And in the same case, Justice Story took occasion to observe that if there be any commercial contract, which, more than any other, 'requires the application of sound common sense and practical reasoning in the exposition of it,' it is 'certainly a policy of insurance.' Id., 14 Pet. 99, at page 109, 10 L.Ed. 371.

If we apply this principle of the 'real efficient cause' to the instant case, it can hardly be doubted that upon the facts assumed the loss would be within the coverage of the policy. Indeed this is not strongly contested, but it is insisted that the case is controlled by certain precedents to which we should give heed in dealing with an ancient form of words. These precedents are found in certain English cases to which the Circuit Court of Appeals referred. The court recognized that a number of American cases had taken a different view, but thought that, in the absence of a contrary decision by this Court or any federal court, the cited English cases should be followed in the view that in the field of marine insurance 'it is highly desirable that our decisions be kept in harmony with those of England.' 89 F.2d 545, at page 549.

Reference is made to the case of Taylor v. Dunbar, L.R. 4 C.P. 206. There meat shipped at Hamburgh for London was delayed on the voyage by tempestuous weather and solely by reason of such delay became putrid and was necessarily thrown overboard. The court held that it was not a loss by perils of the sea. Judge Keating said that the facts showed 'beyond a doubt that the proximate cause of the loss of the meat was the delay in the prosecution of the voyage,' and 'that delay was occasioned by tempestuous weather.' But he held 'that a loss by the unexpected duration of the voyage, though that be caused by perils of the sea,' did not entitle the assured to recover. Judge Montague Smith, concurring, said that if it were held 'that a loss by delay, caused by bad weather or the prudence of the captain in anchoring to avoid it, was a loss by perils of the sea,' the court would 'be opening a door to claims for losses which never were intended to be covered by insurance, not only in the case of perishable goods but in the case of goods of all other descriptions.'

The case of Pink v. Fleming, L.R. 25 Q.B.D. 396, upon which chief reliance is placed, was a case of collision. It was necessary for the damaged ship to put into a port for repairs and for that purpose to discharge a portion of the goods insured, consisting of fruit. The goods were reshipped but on arrival it was found that being of a perishable nature they had been damaged by the handling necessary for their discharge and reshipment and by the delay. The Court of Appeal held that the collision was not the proximate cause of the loss and that there could be no recovery on the policy. Lord Esher said: 'The collision may be said to have been a cause, and an effective cause, of the ship's putting into a port and of repairs being necessary. For the purpose of such repairs it was necessary to remove the fruit, and such removal necessarily caused damage to it. The agent, however, which proximately caused the damage to the fruit was the handling, though no doubt the cause of the handling was the repairs, and the cause of the repairs was the collision. According to the English law of marine insurance only the last cause can be regarded. There is nothing in the policy to say that the underwriters will be liable for loss occasioned by that. To connect the loss with any peril mentioned in the policy the plaintiffs must go back two steps, and that, according to English law, they are not entitled to do.'

The court thought that the case was governed by Taylor v. Dunbar, supra. Lord Esher added: 'With regard to the American authorities, the American law on the subject seems to differ materially from our law, and therefore it is not necessary to consider them.'1

It seems that neither of these cases went to the House of Lords, and we find it impossible to reconcile Lord Esher's ruling, 'that according to the English law of marine insurance only the last cause can be regarded,' with the elaborate exposition of the doctrine of proximate cause which has been given by the House of Lords in Leyland Shipping Company v. Norwich Union Fire Insurance Society, supra, from which we have quoted. And it is recognized in England that 'Passages in Lord Esher's Judgment in Pink v. Fleming * * * to the effect that only the cause last in time can be looked to, cannot now be supported.' Arnould on Marine Insurance, 11th ed., § 783, note (r). So far as the English rule, however, relates to losses, on ship or goods, 'proximately caused by delay,' 'although the delay be caused by a peril insured against,' it has been embodied in statute which apparently applies to all goods, perishable or otherwise, unless the policy otherwise provides. That statute of course is controlling in relation to English practice in that class of cases. Marine Insurance Act, 1906, sec. 55(2)(b). L.R. Statutes, 6 Edward VII, p. 227. There is no statute applicable here which so restricts the doctrine of proximate cause, as we understand it and as it is set forth in the Leyland Case, and the weight of American authority is contrary to the doctrine of Pink v. Fleming.

In williams v. Smith, 1804, 2 Caines, N.Y., 1, 2 Am.Dec. 209, the action was on a policy of insurance covering a cargo consisting chiefly of naval stores, including tar in barrels, on a voyage from New York to Algiers. The vessel experienced severe weather which resulted in such serious damage that it was compelled to put into Cadiz as a port of distress. There more than half of her landing was taken out and the vessel moved to the usual place for repairing. While there an epidemic fever broke out which prevented all business and made it impossible to obtain permits for taking the cargo from the place where it had been landed. Meanwhile the vessel was driven to sea by a storm and sustained further injuries and on returning to Cadiz it was found that the cargo, both on shore and on board, from the heat of the climate and violence of the gale, was deteriorated more than one half of its original value. As the whole would not have produced enough to fit the ship for the completion of her voyage, the vessel and the cargo were abandoned to the underwriters. The court charged the jury, among other things, that 'any damages which arose in consequence of the fever at Cadiz, were within the perils of the policy.' The jury brought in a verdict for a total loss. The court denied a new trial, Judge Kent stating in his opinion 'that the damage resulting from the pestilence at Cadiz' was covered by the policy. The court found it unnecessary to decide 'whether a pestilence is a peril direct within the policy,' but held that it formed 'a sound excuse for delay at Cadiz' and 'if the consequence of that delay was a deterioration of the subject insured, the insurer must be answerable for the loss.' In Tudor v. New England Marine Insurance Company, 12 Cush., Mass., 554, the suit was upon a policy of insurance on a cargo of ice shipped from Boston to Calcutta. It was agreed that the ice was properly packed and surrounded with nonconductors of heat. Under the stipulations of the policy, the insurer could be liable, if at all, only for a total loss. There was also a clause that there should be no liability 'for ice melting in consequence of putting into port.' By perils of the sea, the vessel sprung a leak which increased and it became necessary to put away for a port. On taking out her cargo, in order to ascertain her condition and the practicability of repairs, it was found that the ice had settled; a portion of it having been melted by the sea water which had come in contact with it by reason of the leak. The ice was taken out and sold for a very small sum in comparison with its estimated value at the port of delivery. The voyage was abandoned. It was denied that the loss was occasioned by a peril of the sea. The insured had judgment for a total loss. The court considered it to be well settled that if an article insured as free from average be 'placed in such a condition, that in consequence of inevitable deterioration or decay, it cannot be carried to the port of destination, but will necessarily, before the completion of the voyage, be wholly destroyed, and it is accordingly sold, at an intermediate port,' this would constitute a total loss within the meaning of the policy. The court also held that the exception of the risk of ice 'melting in consequence of putting into port' did not include 'a loss occasioned by the melting of the ice from other causes, or a combination of other causes.' The real cause was the injury to the vessel which made it necessary to take out the ice in a port in the tropics. See, also, Musgrave v. Mannheim Insurance Co., 32 Nova Scotia Rep. 405.

Cory v. Boylston Insurance Company, 107 Mass. 140, 9 Am.Rep. 14, is cited by respondent as an answer to the Tudor Case, but the latter was not overruled or even mentioned. In the Cory Case a vessel with wine on board met with severe gales which prolonged her voyage and caused her to ship much sea water, and upon her arrival at the port of destination the cases of wine were found to be more or less wet either by the sea water or by the steam and dampness generated in the hold by the presence of the sea water and the changes of climate through which the vessel had passed. The case was controlled by special provisions of the policy. The policy provided that the insurers should not be liable for loss by leakage unless occasioned by stranding or collision, or 'for damage or injury to goods by dampness, rust, change of flavor, or by being spotted, discolored, musty or mouldy, unless the same be caused by actual contact of sea water with the articles damaged, occasioned by sea perils.' The court thought that the latter clause was inserted with a knowledge of the decisions in Baker v. Manufacturers' Insurance Co., 12 Gray 603, and Montoya v. London Assurance Co., 6 Exch. 451. To bring a case within that clause the court held that it was not enough 'that perils of the sea should be the efficient, and within the rule laid down in the previous decisions, the proximate cause, by which the sea water was shipped,' but that the sea water 'must come into actual contact with the articles, for the damage to which the underwriters are sought to be charged.'

Respondent also cites Perry v. Cobb, 88 Me. 435, 34 A. 278, 280, 49 L.R.A. 389, where the insurance was on a cargo of lime shipped by a brig from Rockland to New York. The voyage was prolonged on account of rough weather but the vessel arrived tight and only slightly damaged. No sea water had reached the cargo unless in a few instances where a hatch had been taken off or once when the cabin was flooded. That damage was far below the partial loss which had been excepted from the coverage. The remaining damage was 'from the shrinking of the staves of the barrels, and slacking up of the cooperage,' causing a loss of contents and making the barrels insecure for hoisting; a condition which was claimed to have resulted from the rolling and pitching of the vessel caused by the storms of an unusually protracted voyage. The court said that 'All authorities agree that a protracted voyage is not a sea peril, within a marine policy, because it is not an unusual event, but one of the natural incidents to sea transit'; that the evidence was conflicting 'as to the proximate cause for the condition of the cargo upon its arrival'; that the associates, who in this case had insured each other and to whom it was agreed that the question of liability should be submitted, were men of large experience in burning and shipping lime; and that their decision against the plaintiff 'must have great weight upon the fact as to whether the condition of the cargo, upon its arrival in New York, was other than what might have been expected from ordinary sea weather at that time of year,' without any 'unusual sea peril.' The conclusion was that the principal damage to the cargo came 'from its own inherent qualities, excited by the longcontinued transit.' Id., 88 Me. 435, at pages 449, 450, 34 A. 278, 282, 49 L.R.A. 389.

In the case of Bond v. The Superb, 3 Fed.Cas. p. 845, No. 1,624, 1 Wall.Jr. 355, also cited by respondent and the court below, the decision of Mr. Justice Grier at circuit turned upon the question of liability for general average and in this view had distinguishing features.

Fourteen years ago a case closely resembling the one at bar came before the New York courts. Brandyce v. United States Lloyds, The Corsicana, 207 App.Div. 665, 203 N.Y.S. 10, 11; Id., 239 N.Y. 573, 147 N.E. 201. The action was upon policies covering a cargo of potatoes insured against perils of the sea. The vessel, in consequence of a collision at sea with some unknown object, was compelled to put into Charleston for repairs where it was found necessary to discharge the cargo. After repairs the vessel resumed her voyage but, on account of the delay, the potatoes because of sprouting and rot had to be sold. They were not injured directly in the collision or touched by sea water. The precise question presented was 'whether loss by natural deterioration, during a delay in the voyage caused by a sea peril, is a loss by sea perils, within the meaning and intent of the policy of insurance.' The Appellate Division reviewed the authorities in England and in this country. The court held that mere delay on the voyage, as a result of which cargo is spoiled or damaged, was not a ground for recovery. But the court found that the collision, the sea peril insured against, was the real cause of the loss. The court said: 'The evidence indicates that, if the Corsicana had not been damaged by reason of sea perils, the potatoes would have arrived sound. The proximate cause of the loss, therefore, was the sea peril, because it was the efficient dominant cause, which, although incidentally involving delay, placed the cargo in such a condition that, because of inevitable deterioration or decay, it could not be reshipped and carried to its destination.' The Court of Appeals of New York stated the question in the same way and affirmed the judgment.

We lay on one side cases of protracted voyages caused by storms and the special questions to which their varied circumstances give rise. Such a case is not before us. The instant case is one of stranding, a sea peril insured against, and we think that the well-settled doctrine of proximate cause meaning the real efficient cause of the loss, requires the conclusion that, upon the assumptions of fact we stated at the outset, the loss of the cargo was within the general coverage clause of the policy.

The judgment is reversed and the cause is remanded for further proceedings in conformity with this opinion.

It is so ordered.

Mr. Justice McREYNOLDS and Mr. Justice SUTHERLAND are of opinion that the case was correctly decided by the court below on grounds adequately stated.