The finding shows that the plaintiff's house in Ellington in this State was heated by a furnace equipped with an oil-burner system, located in *Page 731 the cellar. In the night of November 19th, 1924, damage was done to the house and contents by smoke and soot caused by a defect in or by defective action of the furnace and oil-heating device. The house and contents were insured in a total sum of $13,000 under the standard form of fire insurance policies, "against all direct loss or damage by fire" occasioned to the house and its contents. When the fire was discovered by the plaintiff he at once turned off the switch and thus shut off the oil supply of the heater, and by this means put out the fire. The trial court has expressly found that the fire which caused this damage was all the time inside and never outside the oil-heating furnace. By motion to correct the finding, the plaintiff seeks to establish that the fire had escaped from the furnace and had ignited a board on the ceiling of the cellar above the furnace; that this board was aflame when he discovered the difficulty, and that he extinguished this flame by using the fire extinguisher. This was the substance of his own testimony at the trial as certified to this court. On the other hand, evidence given by two insurance adjusters who made an examination of the premises, is also certified, and they say there was no charring or other evidence of the presence of any flame outside the furnace itself. This of course tended directly to contradict the testimony of the plaintiff upon this point. There was introduced, however, a piece of circumstantial evidence — the board which the plaintiff claimed was aflame when he entered the cellar. The plaintiff claimed that the condition of this board showed the direct action of flame, while the witnesses for the defense insisted that it did not, but that its condition was the result of smoke and soot only. It is clear that the question whether there was or was not fire outside the furnace was a disputed one. Upon the whole evidence the trial court reached *Page 732 the conclusion that this claim of the plaintiff was not established. Were all the evidence before us, a correction of this finding could be made by this court, but this board, which may well have had an important bearing upon the conclusion of the trial court, is not before us. We therefore cannot decide that the conclusion was an unreasonable one under the circumstances. The trial court also refused to find some subordinate facts as testified to by the plaintiff, to the effect that the "oil burner and furnace were a mass of flames when the fire was discovered," and that "the fire extinguisher was played on a board nailed to the ceiling." It was the province of the trial court to determine whether the plaintiff's testimony was correct or was mistaken or untrue. The finding that there was no fire outside the furnace, however, is necessarily against the plaintiff's evidence in these particulars. For reasons already stated, we cannot say that the conclusion is an unreasonable one upon all the evidence, and paragraph four of the finding of the trial court must therefore stand, save that we strike out the words "and also used a fire extinguisher," and change the word "agencies" to "agency." In this situation, therefore, the only remaining question is whether, upon the facts disclosed by the findings, the fire was one for the results of which the defendants are liable under the terms of their policy, or, in other words, whether the plaintiff's damage, caused by the smoke and soot from a fire which was confined to the furnace, was a "direct loss or damage by fire" within the proper meaning of that term as used in the policy.
It is not entirely clear whether the plaintiff claims that this fire — even though it did not escape from the furnace — was a "hostile fire." Two contentions only are made in the brief; that the finding by correction should show the fire did escape and that so escaping *Page 733 it created liability as a hostile fire even though it did not in fact burn any property but did damage by soot and smoke only.
Since the finding must stand, it conclusively appears that the flame did not escape from the furnace and the damage which was caused was the result of the defective operation of the fire in the interior of the furnace. The reasons of appeal may, however, be construed as somewhat broader than the brief, and the arguments of counsel also assumed a wider range, and since it is also a question of considerable public interest and importance, we shall consider whether a fire thus confined to the furnace can be said to be a "hostile fire" and within the terms of the policy.
The term "direct loss or damage by fire," contained in the policy, raises two questions — what is a "fire" as the term is thus used, and what is a "direct loss or damage" as there meant?
The word "fire," in insurance, does not have the technical meaning which is developed from a scientific analysis of its nature and properties, but more nearly that which conforms to the popular understanding of the word. "It is rather an effect, than an elementary principle; it is the effect of combustion, and is equivalent to ignition or burning; yet heat is not `fire.' Timber may be contracted from the heat of the sun, and a loss be thereby sustained, yet it is not a loss by `fire,'" but "there may be a loss from heat, and `fire,' nevertheless, be the proximate cause of such loss." 4 Joyce on Insurance, p. 4719. So a loss by the breaking of electrical machinery as the result of the direct operation of fire upon it, has been held to be a loss from "fire," though nothing was in fact consumed.Lynn Gas Electric Co. v. Meriden Fire Ins. Co.,158 Mass. 570, 33 N.E. 690. In Richards on Insurance Law (3d Ed.) at page 284, it is said: "Loss by fire *Page 734 means the result of the ignition of the property insured or of some substance near it. Thus, the action of fire in charring, scorching, cracking, smoking or heating may be included though no flame be seen." And citing, in support of the latter statement, Scripture v.Lowell Mut. Fire Ins. Co., 64 Mass. (10 Cush.) 356, 57 Amer. Dec. 111; Singleton v. Phenix Ins. Co.,132 N.Y. 298, 30 N.E. 839.
In a case in the United States Circuit Court for the District of Connecticut, it appeared that a steamer was injured by a collision and sank to her promenade deck, but continued to float until a fire broke out which resulted in destroying her buoyancy so that she sank. She was raised and repaired at a cost of $84,000. It was found that if the fire had not broken out the boat would not have sunk but could have been towed to a dock and repaired for $15,000. It was held that the difference between these figures — $69,000 — was the loss proximately caused by the fire, though much of it, of course, was not the result of actual burning.Norwich N.Y. Transportation Co. v. Western Mass.Ins. Co., 34 Conn. 561.
It was early held that the word "fire" as used in policies of fire insurance meant "hostile fire" as distinguished from "friendly fire." The leading case is that of Austin v. Drew, 4 Campb. 360, 128 English Reprint, 1104. In that case nothing was consumed by fire, but, by improper management of the heating apparatus, heat, smoke and soot were emitted without any escape of the fire from the place where it was designed to be and the loss by heat, smoke and soot was held not covered by a policy against loss by fire.
In both the Lynn Gas Electric Co. and Norwich New York Transportation Co. cases, supra, the fire which caused the loss had escaped from the place where it was intended to be, and so was a "hostile fire." *Page 735
In a case arising in Georgia, a fire was built in a stove on the ground floor of the building, and the stovepipe extended through the ceiling and through the second story of the building. The pipe became disengaged at the ceiling of the second floor. Smoke and soot escaped into the second-story room and damaged goods situated there, but it was not shown that fire occurred outside the stove. The court held that the fire was a "friendly fire," and that the damage was not a loss by "fire." The court said: "It does not appear . . . that there was any fire in or about the building, except in the stove where it was intended to be built. This fire did not spread from where it was built and intended to remain. It was, therefore, all the time during the alleged injury and damage to the goods, what is termed in the books a `friendly' and not a `hostile' fire." The court added: "It is true that there is sound authority for the proposition that an insured can recover loss occasioned by smoke, soot, etc., thrown out by a fire; but we think in these cases it will be found that such matter causing injury was the product of a hostile fire." Cannon v. Phenix Ins.Co., 110 Ga. 563, 567, 35 S.E. 775, 776. We find few and scant criticisms of the rule. One occurs in Wisconsin, where it was argued that Austin v. Drew was not, when properly analyzed, an authority for the distinction. A very exhaustive dissenting opinion in that case, however, maintains the contrary and seems well supported by the leading cases and textbooks.O'Connor v. Queen Ins. Co., 140 Wis. 388, 395,122 N.W. 1038. In that case the majority opinion suggests the test to be, in effect, whether "the fire was extraordinary and unusual, unsuitable for the purpose intended, and in a measure, uncontrollable . . . inherently dangerous because of unsuitable material used." *Page 736 This was clearly based upon the facts of that particular case.
While the use of the term "hostile fire" does suggest, superficially, any fire which by getting beyond control or operating in a manner different from that expected or designed, causes damage, as in the Wisconsin case, yet when fully considered, it becomes obvious that the wide range of circumstances upon which liability would thus depend would tend to render each case mostly a law unto itself and introduce elements of great uncertainty in determining the rights of the parties. Whether changing instrumentalities of heating and lighting, use of electrical current, etc., may sometime bring about a modification of the rule, we do not need to anticipate. Though the present distinction may seem arbitrary, yet it is of long standing, makes for certainty in the ascertainment of rights, and has been acted upon in the writing of so vast a number of insurance contracts throughout this country, that its soundness may not, at this time, be questioned.
The rule has had well nigh universal approval in those courts in this country where the question has directly arisen and by text-writers generally. Way v.Abington Mutual Fire Ins. Co. (1896) 166 Mass. 67,43 N.E. 1032; McGraw v. Home Ins. Co. (1914)93 Kan. 482, 144 P. 821; American Towing Co. v. GermanFire Ins. Co., 74 Md. 25, 21 A. 553; Gibbons v.German Ins. Sav. Institution (1889) 30 Ill. App. 263;Cannon v. Phenix Ins. Co. (1900) 110 Ga. 563,35 S.E. 775; Fitzgerald v. German-American Ins. Co. (1899) 30 Misc. 72, 62 N.Y.S. 824; Hansen v. LeMars Mutual Ins. Asso. (1922) 193 Iowa 1,186 N.W. 468; Richards on Insurance (3d Ed.) § 231; Beaumont on Insurance, p. 37; 1 Wood on Fire Insurance (2d Ed.) p. 236.
The conclusions of the trial court, that the fire in *Page 737 question was not a "hostile fire" and not one which was within the contemplation of the policy, were correct. The judgment for the defendant was not erroneous.
There is no error.
In this opinion the other judges concurred.