DISSENTING OPINION OF

I dissent from the majority opinion.

As a question of fact I contend that there was no collision.

The plaintiff, who testified as a witness, said in part “that the accident occurred while passing the mound, it started there; that the overturning happened when the car had already passed the mound, it ended there; that he can not state with precision on which side the car fell; that his car might have fallen to the right, but it could also have happened that the right-hand side wheels climbed' over, thus overturning the car towards the left-hand side, one side being higher than the other.”

This part of the testimony of the plaintiff seems to have been overlooked by the majority of the court, for this is the *311only way to explain tlie erroneous conclusion of the court in holding that there had been a collision of the car of the plaintiff against a mound of earth on one side of Santurce’s new road. The facts in the case of Fabregas v. Porto Rican & American Ins. Co., 31 P.R.R. 632, are so different from those of this case that it can not apply here. That case refers to an automobile which fell over a. precipice and although the authorities do not agree, it was held that the impact of the automobile against the earth was a collision. I do not see the slightest parallel as to the facts so as to justify in any way its application to the case at bar.

The legal question which relates to the construction to be given to the clause covering the risk of collision is of the greatest importance. The clause reads:

“To indemnify the assured in the said warranties against loss by reason of damages or destruction of any of the automobiles enumerated and described in the said policy, including its operating equipment, while attached thereto, if caused solely by accidental collision with another object either moving or stationary, excluding, however: (a) Damages caused by the falling or turning* over of any of such automobile^, unless such fall or overturn is the direct result of, and immediately follows such accidental collision; (b) Damages or destruction caused directly or indirectly by fire.”

I have made all possible inquires without finding any authority directly construing, in connection with automobile insurance, a clause exactly or similarly worded as the clause which* is the subject of this suit, and it seems that in the absence of jurisprudence applicable thereto the tendency has been to apply as far as possible the principles established in connection with policies insuring risk by collision between vessels.

The ease of New York & Boston Dispatch Express Co. v. Trade & Merchants Insurance Co., 42 Am. Rep. 441, deals with a policy against fire covering certain goods carried in a boat. The goods were damaged on account of a collision with another vessel. A fire ensued and the boat sank, carry*312ing down the goods which, however were not destroyed by fire. Although the terms of the policy are inverted as compared with those of the case at bar, because in the former the risk covered by the policy was a fire and not a collision risk, the principle applicable is the same. The opinion of the court says:

“The defendant contends that the plaintiffs’ goods were injured by the sinking of the Ship, and as the water which flowed through the hole made by the collision caused her to sink, therefore the collision and not the fire caused the loss. Undoubtedly the injury occasioned by the collision would have caused the ves'sel to sink, and thereby have injured the plaintiffs’ property; and if that had been the only cause operating, the plaintiffs can not recover, for the insurance is not against collision, but only against fire- But if mean's and appliances were at hand by which that result could have been avoided, and the intervention of a new agency, namely, that of fire, prevented their use, then the fire was the proximate and immediate cause of the loss. It added a new element of destruction which rendered it impossible to control or prevent the consequences which would naturally follow from the collision.”

Other cases are cited in the above case and among them is that of Howard Ins. Co. v. Transportation Co., 12 Wall. 194, 42 Am. Rep. 443, and in the latter the reasoning of the court is as follows:

“The ves'sel of the defendant in error was injured by a collision, in consequence of which she filled rapidly with water, and a fire broke out. The jury found that the damage done by the sinking of the vessel was the natural result of the fire only. It was also found that the water would not have caused the ves'sel to sink below her promenade deck, had not some other cause of sinking supervened. Mr. Justice Strong, in a very able opinion, considers the question of proximate cause. And in answer to the claim of the plaintiff! in eiror, that the Sinking of the vessel was the result of two concurring causes, one the fire, and the other the water let in by the breach made by the collision, he says: ‘As the influx of the water was the direct and necessary consequence of the collision, it is argued that the collision was the predominating, and therefore the proximate cause of the loss. The argument overlooks the fact, distinctly found, that the damage resulting from the sinking of the vessel was the natural *313and necessary result of the fire only. If it be said that this was but an inference from facts previously found, it was not for that reason necessarily a mere legal conclusion. But we need not rely upon this. Apart from that finding, the other finding's, unquestionably of facts, show that neither the collision nor the presence of the water in the steamer’s hold was the predominating efficient cause of her going to the bottom. That result required the agency of the fire. It is found that the water would not have caused the ve'ssel to sink below her promenade deck, had not some other cause of sinking supervened. It would have expended its force at that point. The effects 'of the fire were necessary to give it additional efficiency. The fire was therefore the efficient predominating cause, as well as nearest in time to the catastrophe, which not only directly contributed to all the damage done, after the 'steamer had sunk to her promenade deck, but enlarged the destructive power of the water and rendered certain the submergence of the vessel.’ See also St. John v. American Ins. Co., 1 Kern. 516; Peters v. Warren Ins. Co., 14 Pet. 99.”

This last case had to be distinguished in the case of The G. R. Booth, 171 U. S. 450, cited by the majority, and it seems that necessarily it had to he done owing to the strength of the reasoning of Judge Strong.

Moreover, in view of the wording of the policy in the present case, the authorities cited by the majority who construe clauses that are different can not have the influence or scope sought to be given to it.

The policy in this case could not be clearer in its terms. It was thereby stipulated that only the injury caused by collision, as distinct from that caused directly or indirectly by fire, was covered by the policy. I do not conceive that the words “directly or indirectly caused by fire” refer, according to the majority, “to the cause or accident that set in motion all the other causes or results and was the proximate or efficient cause of the happening of the event.” This is stated on the strength.of the authorities which are not cited. It would be curious to hear of a case where a fire set in motion an automobile so as to make it collide. Perhaps what was meant, and it would have been clearer, is that the fire *314■was caused before the collision. The question would then be of itself so plain as to make any discussion thereon unnecessary. But this is what the policy did not mean. The meaning of the language used by the insurers is as follows: “We insure your property against collision, except in case of an ensuing fire; therefore, we will pay you damages resulting from a collision, but not damages caused by fire.” This is the literal meaning of the clause of the contract. The purpose of the clause thus established is to prevent fraud and surely this was the mental process of the parties when executing the contract. The effect of a collision generally is not the destruction of an automobile, but a fire destroys it totally. What the court has done is to modify the contract in such a manner as to impose on the company an obligation which does not belong to it — to make it liable for a fire risk without the payment on the part of the insured of any premium on that account. The clause of the contract has been disregarded and it is the imagination which has been in action by construing into .the clause a very different thing from that which the parties had in mind when contracting. See section 1250 of the Civil Code.

For the foregoing reasons I have come to the conclusion that the judgment dismissing the complaint and from which this appeal has been taken should have been affirmed.

delivered tbe opinion of tbe court.

On tbe 4tb of May, 1924, while tbe plaintiff was driving an automobile at night be struck a pile of eartb on tbe new *304highway in Santurce (carretera). The automobile was upset and as a consequence fire ensued and by reason of these two antecedent events, — the overturning and the fire, — the automobile was totally destroyed. Under the weight of authorities, when an automobile strikes a pile of earth on the highway the consequent impact is recognized to be a collision and if the automobile is insured against collision a recovery may generally be had. Fábregas v. Porto Rican & American Ins. Co., 31 P.R.R. 633, and cases; Pred v. Employers Indemnity Co., 35 A.L.R. 1003, and cases; Power Motor Co. v. United States Fire Ins. Co., 35 A.L.R. 1028, 223 Pac. 112; Interstate Casualty Co. v. Stewart, 208 Ala. 377, 44 So. 345, 26 A.L.R. 427, cases and note; Yorkshire Ins. Co. v. Bunch Morrow Motor Co., 212 Ala. 588, 103 So. 670; Wood v. Southern Casualty Co. (Tex.), 270 S. W. 1055.

The court below found that there had been a collision, as defined in the Fábregas case, supra.

The plaintiff did possess a policy of insurance to cover loss by collision, for which his predecessor in title had paid the sum of $63.75. Under the policy, in case of destruction of the automobile, the liability of the company was to be $100 less than the total value of the automobile destroyed. The insurance policy, however, contained the following clause:

“To indemnify the assured in the said warranties against loss by reason of damages or destruction of any of the automobiles enumerated and described in the said policy, including its operating equipment, while attached thereto, if caused solely by accidental collision with other object either moving or stationary, excluding, however: (a) Damages caused by the falling or turning over of any of 'such automobiles, unless such fall or overturn is the direct result, of, and immediately follows such accidental collision; (b) Damages or destruction caused directly or indirectly by fire.”'

The defendant company alleged that as the whole damage was practically caused by the fire it was in no way respon*305sible for the loss, and the District Court of San Jnan agreed ■with the defendant company.

We think the appellant is quite right when he maintains that the proximate canse of the loss in this case was not the fire but the collision. As the automobile was insured against collision and all the damage produced was the result or consequence of the collision, the’ said collision was the efficient cause of the damage in this case. In a certain limited sense the fire was the cause of the loss, but it was not the cause- of the loss within the intendment of the parties to this contract. The policy bolder insured against collision. He insured against collision and its natural consequences. Fire is a natural and probable consequence of a violent shock to a vehicle operated by an inflammable material like gasoline. When the company insured against collision it must have intended the natural and probable results of such an event. Within the meaning of. the policy the loss or damage was not caused directly or indirectly by fire, but was caused by the collision. The words “directly or indirectly caused by fire” under substantially all the authorities refer to the cause or accident that set in motion all the other causes or results and was the proximate or efficient cause of the happening of the event.

It is otherwise when one of the later causes is the matter plainly within the intention of the parties to an insurance policy, as shown in the opinion of the Supreme Court of the United States in Insurance Co. v. Transportation Co., 12 Wall. 194. The steamer Norwich was insured against fire. A collision caused Water to enter the furnace, spreading the fire, and the boat was burned and sunk. From the effects of the collision alone she would not have sunk; her sinking was immediately due to the fire. The insured recovered as for a loss by fire. The court said: “But it is well settled that when an efficient cause nearest the loss is a peril expressly insured against, the insurer is not to be relieved *306from responsibility by bis showing that tbe property was brought within that peril by a cause not mentioned in the contract. ’ ’ The effect of that decision is that even if a cause of loss is not the proximate cause of said loss, nevertheless the plaintiff may recover if he shows that the subsequent cause or event was the contingency expressly insured against. As we shall see, that case is the exception that points the rule.

Waters v. Merchants’ Louisville Ins. Co., 11 Pet. 213, is one of the earliest and leading cases on the proximate cause of loss in insurance policies. There was some question in that case whether an explosion caused by fire was a loss by fire or by explosion merely. The court was of the opinion that as the explosion was caused by fire the latter was the proximate cause of the loss, and also said: “It is a well-established principle of the common law, that in all cases of loss, we are to attribute it to the proximate cause and not to any remote cause; causa próxima, non remota spec-tatur.”

In the case of The G. R. Booth, 171 U. S. 450, Mr. Justice Gray was the spokesman of the court. To quote from the syllabus, the facts of the case were as follows:

“A provision in a bill of lading, that the carrier ‘shall not be liable for loss or damage eau'sed by the perils of the sea,’ or by ‘accidents of navigation,’ does not exempt the carrier from liability for damage to part of the cargo by sea water under these circumstances : While the ship was being unloaded at the dock in the port of her destination, a ca'se of detonators in her hold exploded, without fault of any one engaged in carrying or discharging the cargo, and the explosion made a large hole in the side of the ship, through which the water rapidly entered the hold, and damaged other goods. ’ ’

In, the course of the opinion the court said:

■“In the case at bar, the explosion of the ease of detonators, besides doing other damage, burst open the side of the ship below the water line, and the sea water rapidly flowed in through the opening made by the explosion, and injured the plaintiff’s sugar. *307Tbe explosion, in consequence of which, and through the hole made by which, the water immediately entered the ship, must be considered as the predominant, the efficient, the proximate, the responsible cause of the damage to the sugar, according to each of the tests laid down in the judgment of this court, above referred to. The damage to the sugar was an effect which proceeded inevitably, and 'of absolute necessity, from the explosion, and must therefore be ascribed to that cause. The explosion Concurred, as the efficient agent, with the water, at the instant when the water entered the ship. The inflow of the water, seeking a level by the mere force of gravitation, was not a new and independent cause but was a neces'sary and instantaneous result and effect of the bursting open of the ship’s side by the explosion. There being two concurrent eau'ses of the damage —the explosion of the detonators, and the inflow of the water— without any appreciable interval of time, or any possibility of distinguishing the amount of damage done by each, the explosion, as the cause which set the water in motion, and gave it its efficiency for harm at the time of the disaster, must be regarded as the predominant cause. It was the primary and efficient cause, the one that necessarily set the force of the water in operation; it was the superior or controlling agency, of which the water was the incident or instrument. The inflow of the sea water was not an intermediate cau'se, disconnected from the primary cause, and self-operating; it was not a new and independent cause of damage; but, on the contrary, it was an incident, a necessary incident and consequence, of the explosion; and it was one of a continuous chain of events brought into being by the explosion — events 'so linked together as to form one continuous whole.”

And further on the court said;

“Upon principle and authority, therefore, our conclusion is that the explosion, and not the sea water, was the proximate cause of the damage to the sugar, and» that this damage was not occasioned by the perils of the sea, within the exceptions in the bill of lading. ’ ’

The case of Waters v. Insurance Co., supra, was referred to, hut more particularly the court had referred to Insurance Co. v. Transportation Co., supra, quoting some of the language that we have already cited, and also quoting as follows:

“And certainly that cause which set the other in motion, and *308gave to it its efficiency for harm at the time of the disaster, must rank as predominant.”

Then Mr. Justice Gray went on to say:

“The rule was held to be inapplicable to that case, because the damage resulting from the fire, and that caused by the collision, apart from the fire, were clearly distinguished; and because the policy, exempting the insurers from liability for losses by fire by certain specified cause’s, covered losses by fire from all other causes, including collisions. But for those distinctions, the decision could hardly be reconciled with the earlier opinions already referred to, or with that delivered by the same able and careful judge in the later case of Insurance Co. v. Boon, 95 U. S. 117.”

We shall take no more time to review it, but the whole case merits perusal for its careful analysis of “proximate cause” in insurance cases and losses under bills of lading.

From the case of Wheeler v. Phenix Insurance Co., (N. Y.), reported in 38 L.R.A. (N. S.) page 474, on page 479 the exception contained in standard policies against fire insurance is reproduced as follows :

“This company shall not be liable for loss directly or indirectly by invasion, insurrection, riot, civil war or commotion, or military or usurped power, or by order of any civil authority; or by theft; or by neglect of the insured to use all reasonable means to save and preserve the property at and after a fire or when the property is endangered by fire in neighboring premises; or (unle'ss fire ensues, and, in that event, for the damage by fire only) by explosion of any kind.”

It was held in that case that when an explosion was caused by a fire in a building the insurer was not relieved from liability for the loss under a policy insuring against loss by fire, except loss caused directly or indirectly by explosion of any kind unless fire ensued, and in that event by fire only. As shown in that case and in the notes, when an explosion takes place in the insured property during the progress of a fire therein, such explosion is treated as a mere incident of *309■flae preceding fire, the fire being treated as tbe efficient cause; and tbe whole loss is tbe risk insured.

Cases are assembled also in 10 British Ruling Cases, 211; 22 R.C.L. 217; 133 A.S.R. 1091, 1093; Ann. Cas. 1913 a, 1300; 5 Ann. Cas. 782, and see also Mitchell v. Potomac Ins. Co., 183 U. S. 42.

One of tbe clearest cases' is to be found in Western Ins. Co. v. Skass (Col.), 171 Pac. 358. Counsel for tbe insurance company bad attempted to distinguish a previous case of tbe Supreme Court of that state. Tbe court said:

“Tbe rule stated in tbe opinion in tbe Hyman Case is in accord with most, if not all, of tbe decisions of various courts that have determined the liability of an insurer under a fire insurance policy containing this usual-explosion clause, where, as in the case at bar, a hostile fire preceded and caused the explosion and the explosion was an incident to such fire.”

And tbe court cited numerous cases, quoting particularly from Transatlantic Fire Insurance Co. v. Dorsey, 56 Md. 70, 40 Am. Rep. 403, and especially this language: “In such case, tbe fire is tbe direct and efficient cause of tbe loss, and tbe explosion but tbe incident, and if tbe insurers intend to exclude such liability, they must do so by plain and unambiguous terms.”

In Bird v. St. Paul Fire & Marine Ins. Co., 224 N. Y., on page 50 Mr. Justice Cardozo says: “There is no doubt that when fire spreads to an insured building and there causes an explosion, tbe insurer is liable for all tbe damage.” In that case fire caused an explosion and tbe explosion was communicated to a boat 1,000 feet away. Tbe court held that tbe cause of tbe damage was too remote and was not within tbe intention of tbe contracting parties. Quoting from Lord Shaw, said the court: “Tbe same cause producing tbe same effect may be proximate or remote as tbe contract of the parties seems to place it in light or shadow. That cause is to be held predominant which they would think of as pre*310dominant. A common-sense appraisement of every day forms of speech and modes of thought must tell us when to stop. ’ ’

It was with these precedents and antecedents that collision • insurance came into being. The clause exempting from fire was probably inserted bearing in mind the provisions of the standard fire insurance policy. Unquestionably when a man takes out an insurance policy against collision both the parties intend that he should be indemnified for losses by collision and the almost necessary incidents thereof, like a fire. To say that the effects of fire caused by a collision could be excluded is to state a contradiction in terms. A collision is a violent shock between two objects, and a violent rubbing together of two objects is liable to produce a flame. This was the method of our remote ancestors to obtain 'a fire. If the object was to exclude all fire damage, under the decisions the contract should have been couched in different language.

The judgment appealed from should be reversed and another rendered for the plaintiff in the sum of $1,100.