This is a difficult question even in a scientific point of view; or rather, science acknowledges no such thing as fire. But by fire, in the common use of the word, is probably meant flame. Flame, however, is only hydrogen-gas heated to redness, or whiteness. We do not, however, call a cannon-ball, heated to redness, or even to whiteness, fire; and yet it cannot be doubted, that if red-hot iron, or any substance sufficiently heated, coming *in contact with property insured, injured it by heat, in a certain way, this would be a loss under a policy against fire. But this injury must reach a certain extent to come under the policy. Perhaps a rule which has been applied in the trial of persons charged with arson, might be found applicable here. If a person charged with this crime were proved to have put kindlings upon a floor, and set fire to them, with a purpose of burning a house, and the fire was extinguished or burnt out without affecting the floor, the crime would be only attempted and not committed. But what operation of the fire upon the floor would suffice to constitute the crime? It has been said, that the floor must be charred, or, in other words, the surface of the floor must be changed by heat from the condition of wood into that of charcoal. Such a rule would meet cases which sometimes arise in respect to property insured. It would be equivalent to this: that insurers against fire are not liable unless there be ignition or combustion. It is certain that very great injury may be caused by fire, without either ignition or combustion. In one case, a sugar-house, with its contents, was insured against fire, and in each story sugar in a certain state of preparation was deposited, for the purpose of being refined, and for this purpose a certain degree of heat was necessary. To obtain this there was a chimney running up through the whole building, with a register in it on each story, whereby more or less heat could be introduced at pleasure into the rooms. At the top of the chimney was a register, which was closed at night, that the heat might be retained in the building. This register was, by the negligence of a servant, left shut one morning when the fires were lighted; and consequently the smoke and heat were forced into the rooms where the sugars were drying, and they were very much injured thereby. Held, that the insurers were not liable. (k) But if there is an extraordinary fire, the insurers are clearly liable for the direct effects of it, as where furniture or pictures are injured by the heat, although they do not actually ignite. (l) Where there was insurance on a theatre, "not to cover any loss or damage by fire which may originate in the theatre proper," and a fire outside the building heated the wall so much as to cause the interior to burn, the insurers were held. (ll)

* It was formerly supposed that lightning was fire; and then all injuries by lightning might be regarded as injuries by fire. Now it is known, as a matter of science, that lightning is not fire, and that the light or flash of lightning arises from the shock upon the air. But the same shock produces great heat, wherever it falls; and therefore a house struck by lightning is frequently set on fire. If, however, a house be destroyed by lightning, but without ignition, insurers against fire are not liable ; (m) 1 nor if a house falls and becomes mere rubbish, and then takes fire and burns up. (mm) 2

A similar question has arisen in cases of explosion. Here, it seems to be settled by authority, that if the explosion be caused by gunpowder, it is a loss by fire; (n) 3 and the same rule would

(k) Austin v. Drew, 4 Camp. 360, Holt, N. P. 126, 6 Taunt. 426, 2 Marsh. 130.

(l) Case v. Hartford Ins. Co. 13 Ill. 676. See also Scripture v. Lowell Ins. Co. 10 Cush. 356.

(ll) Sohier v. Norwich Ins. Co. 11 Allen, 336.

(m) Babcock v. Montgomery Co. Ins.

Co. 6 Barb. 637, 4 Comst. 326; Kenniston v. Mer. Co. Ins. Co. 14 N. H. 341.

(mm) Nave v. Home Ins. Co. 37 Mo. 430.

(n) Scripture v. Lowell Ins Co. 10 Cush. 356; Waters v. Merchants Ins. Co. 11 Pet. 213, 225; Grim v. Phoenix Ins. Co. 13 Johns. 451.

1 Where a tornado, accompanied by electrical disturbance, destroyed insured property, it was held, in an action on the policy, which covered loss by lightning, that the jury most decide whether lightning was an active agent in the destruction. Spensley v. Lancashire Ins. Co. 54 Wis. 433.

2 As where the floors and roof have fallen in, and the walls only are standing. Huck v. Globe Ins. Co. 127 Mass. 306. See Breuner v. Liverpool Ins. Co. 51 Cal. 101. But where a building is standing on its posts, though out of plumb and abandoned, it has not fallen. Firemen's Ins. Co. v. Congregation, etc. 80 Ill. 558.

3 Where an explosion occurred from vapor, given off from material in process of manufacture coming into contact with a lamp, the insurers were held liable for the loss undoubtedly be applied if the explosion were caused by the burning of saltpetre or any other combustible substance. But a violent explosion may injure things at a considerable distance, by the mere shock; and this would not be injury by fire. Thus, where the damage was caused by the explosion of a powdermagazine a mile distant, the insurers were not held. (nn) And where a warehouse was insured against fire with an exception of explosion, and an explosion took place in a neighboring building which set it on fire, from which the insured building caught and was destroyed, the insurers were held not liable, the rule of causa proximo, non remota not applying. (no) The explosion of a steamboiler is not a loss by fire. (o) The distinction taken is this: that gunpowder explodes by combustion, and steam by expansion without combustion.