The plaintiff in error urges two propositions, either one of which being found in its favor, would result in a reversal of the judgments: 1. That the proximate cause of the fire, was the explosion, the lightning being only the remote cause, and the loss is, therefore, not within the terms of the lightning clause of the policy. 2. That whether the lightning clause taken alone, would, under the facts, create a liability or not, yet when that provision is considered in connection with the entire policy, it is plain that the loss which occurred was not, within the contemplation of the parties at the time of the. making of the contract, one which was intended to be covered.

1. Respecting the first proposition, it may be said that undoubtedly the rule is, that the proximate and not the remote cause of. the loss is *584to be regarded in determining liability. As said by Lord Bacon: ‘‘It were infinite for the law to judge the causes of causes, and their impulsions one of another; therefore, it contenteth itself with the immediate cause, and judgeth of acts by that without looking to any further degree.” And it is contended here, with much effect, that the true rule is, that where a new cause has intervened between the fact accomplished and the alleged cause, such 'new cause must be considered the real cause; that in this case the lightning-striking the powder house was inadequate to produce the disruption of the insured property without the intervention of some other and nearer cause; that the force and energy which produced the mischief came, not from the lightning, but from the explosion, and therefore the explosion was a new cause which intervened and hence must be regarded as the proximate cause. While on the other hand, it is insisted that the law seeks the first efficient cause, which will be regarded as the causa prójima, however many other agencies may have intervened, and that, in this case, the lightning was the efficient cause and the other merely incidental, and therefore, the mere agent or instrument through which the cause operated.

Attention has been called to a formidable array of decisions, pro and con, giving a review of the question of proximate and remote cause as the same has arisen and been decided in a great variety of cases, and these decisions bring- before the mind, as a subject of study, the general doctrine of proximate and remote causes. But we would regard it as unprofitable labor to seek through the cases for a satisfactory expression of the rule, since no general rule will be found suited to all *585conditions, and each case, as it arises, must, after all, be decided upon the special facts belonging to it, and often upon the very nicest discriminations. And it seems not worth while to pursue the point in considering the present case, because, as it appears to us, there is no necessity for such inquiry, inasmuch as the case may be satisfactorily disposed of upon the second proposition.

2. It is contended, in support of the judgment below, that, inasmuch as the lightning clause is not a part of the original policy, but is attached thereto as a modification, it must, therefore, control where it is inconsistent with other portions of the policy, and that it is inconsistent with that part of section two which relates to loss by explosion.

■ It is a rule of construction, founded in reason and resting upon abundant authority, that the meaning of the contract is to be gathered from a consideration of all its parts, and that no provision is to be wholly disregarded because inconsistent with other provisions unless no other reasonable construction is possible; and that a special provision will be held to override a general provision only where the two cannot stand together. If reasonable effect can be given to both, then both are to be retained. Are the two provisions referred to irreconcilably inconsistent? The lightning clause insures against loss or damage caused “by lightning to the interest of the assured in the property described;” but it is “subject in all other respects to the terms and conditions of the policy.” That is, while affording protection to the property insured from the lightning, the ocher terms of the policy are to have full effect. Recurring now to the other provision involved, *586we find that the insurance “does not apply to or cover any loss caused by explosion, unless fire ensues, and then the loss or damage by fire only.” Here there was no fire..

We think that these two claüses are not inconsistent, but that each can be given effect without destroying the other. Construed together, they make the company liable for any damage to the building and contents in case the same were injured by lightning, but that in no event, would the company be liable if the loss were occasioned by an explosion. The provision is against loss by lightning to the property insured, subject to the terms of the policy; i. e., provided the loss is not occasioned by an explosion. This, it seems to us, gives a reasonable construction to each clause, and does no violence to any part of the contract. We think, also, without stopping to refine upon the doctrine of proximate and remote causes, that, within the meaning of these provisions, the loss in this ease was by explosion, and not by lightning.

And this it is reasonable to assume, must have been the understanding of the parties in the making of this contract, for-, while it is unlikely that either had actually in mind the extent of the peril from the proximity of the powder house across the way, yet no more apt language could have been used to exclude liability for this very peril had the parties contracted with full knowledge of its existence and dangerous character. Construed with reference to the subject matter, the language used is equivalent to a declaration on the part of the company that it will not be held for any loss, whether it comes within the general peril of lightning or not, and without undertaking to consider whether it does or not, if such loss *587occurs by explosion, unless fire ensues. If fire follow an explosion, then liability attaches; if not, there is none. Nor can it reasonably be urged that the insured did not understand the meaning of the language of this provision, for it is obvious.

He could not as a reasonable man, in the face of such an exception, have expected the company to be liable for any loss, save from consequent fire, if such loss should accrue from explosion. Although the explosion of gunpowder by means of lightning happens but rarely, yet it is a possible peril and sometimes occurs, which fact may account for the company declining to take such risk, while its infrequency may account for the willingness of the insured himself to bear it. But whether the latter actually had the extent of this risk in mind or not, when he entered into the contract, he must be held in law to have assented to an exception which, upon its face, takes risks by explosion out of the perils insured against.

That destruction by explosion, of a house seventy-one feet away from one struck by lightning, should be deemed a natural' result of the lightning, is at least a doubtful proposition. But be that as it may, when there follows in a policy, after a lightning clause, a provision which distinctly excludes liability for loss by explosion, it appears plain that within the contemplation of the parties at the time of the making of the contract, a loss by explosion could not have been understood to be embraced within the protection of the policy.

The' conclusions stated are sustained by abundant authority. True it is that cases are to be found which declare principles of construction which, if applied here, would make the company *588liable for this loss, if its liability were measured wholly by the lightning clause. But in no case which has come within our observation, and we have examined a great many, has a liability been found to attach where there was a provision excluding liability for loss by explosion, and the loss was caused by fire, or as here, by lightning, taking effect in a distant building, and the damage being wrought to the insured property by an explosion, produced by the fire or the lightning without either of the latter agenciés coming in contact with the insured property: Everett v. The London Assurance, 115 E. C. L., 126; Caballero v. Ins. Co., 15 La. An., 217; St. John v. Ins. Co., 11 N. Y., 516; Briggs v. Ins. Co., 53 N. Y., 446; Montgomery v. Ins. Co., 16 B. Mon., 427; Heuer v. Ins. Co., 144 Ill., 393.

Judgments of the circuit court and of the court of common pleas reversed, and judgment for plaintiff in error. t