after stating the case as above, delivered the opinion of ihe court.
It is first contended by learned counsel for the plaintiff that there is nothing before us for review; that the so-called “Agreed Statement of Facts” is only a concession of certain independent and separate facts, which were offered in evidence as a basis for a general finding; and that, inasmuch as there was no objection made to the introduction of such facts in evidence or exceptions saved to the ruling of the court thereon, the doctrine announced in Barnard v. Randle (C. C. A.) 110 Fed. 906, and cases therein cited, is applicable. Undoubtedly it is true, as settled by a long line of authority, that where evidence is heard in an action at law, and a general finding made thereon, an exception to such finding alone presents nothing for review. But such is not the case now before us. The judgment entry and bill of exceptions both clearly disclose that the cause was submitted to the court upon an agreed statement of facts, signed by counsel for the respective parties, filed and made part of the record, and that no other evidence whatever was heard at the trial. It is of no significance that counsel at the trial formally offered in evidence the facts so agreed upon or any of them. Such practice, if adopted, did not change the essential character of the submission. It was a submission of facts agreed upon in writing for the judgment of the court as a conclusion of law thereon, and as such is the equivalent of a special verdict, presenting questions of law alone for the consideration of the court. Its conclusion thereon is subject to review by this court. Supervisors v. Kennicott, 103 U. S. 554, 26 L. Ed. 486; Bond v. Dustin, 112 U. S. 604, 5 Sup. Ct. 296, 28 L. Ed. 835; Lehnen v. Dickson, 148 U. S. 71, 13 Sup. Ct. 481, 37 L. Ed. 373; Cudahy Packing Co. v. Sioux Nat. Bank, 16 C. C. A. 409, 69 Fed. 782. Guided by the foregoing authorities, our sole duty is to determine whether the trial court, reached the correct conclusion of law from the facts so agreed upon.
It is next contended that we are foreclosed from any consideration of the force and effect of the suicide clause in question, because the policies in suit, being Iowa contracts, do not contain in their bodies the agreement exonerating the insurer from liability in case of suicide. Attention is called to the act of the general assembly of Iowa approved April 17, 1890, entitled “An act to prevent discrimination in life insurance.” Laws 1890, p. 49. Section 1 of this act is as follows:
“No life insurance company doing business in Iowa shall make or permit any distinction or discrimination in favor of individuals between insurants of the same class and equal expectations of life in the amount or payment of premiums or rates charged for policies of life or endowment insurance, or in 'the dividends or other benefits payable thereon, or in any other of the terms and conditions of the contract it makes; nor shall any such company or any agent thereof make any contract of insurance or agreement as to such contract, other than is plainly expressed in the policy issued thereon; nor shall any such company or agent pay or allow, or offer to pay or allow, as inducement to insurance any rebate of premium payable on the policy, or *272any special favor or advantage in the dividends or other benefit to accrue thereon, or any valuable consideration or inducement whatever not specified in the policy contract of insurance.”
It is contended by plaintiff’s counsel that the clause relied upon by defendant to defeat recovery in this action, namely, “I also warrant and do agree that I will not die by my own act, whether sane or insane, during the said period of two years,” being found only in the application made by Kelly for insurance, is not so “plainly expressed in the policy” as to be a valid and enforceable agreement, within the purview of that act.
The contention, as we understand it, is that the “policy,” within the purview of the act, is that particular paper signed by the insurer which contains its promise, and nothing else, and particularly that it does not include any of the agreements found in the proposition for insurance usually denominated the “application,” even though the same be attached to the other paper, and by express stipulation therein made part of the contract. This contention, in our opinion, is narrow and technical, and ignores the rule of construction of contracts, requiring a consideration of all its provisions, wherever found, to determine the intention of the parties. The stipulations of a paper, referred to in a contract „as the consideration upon which it is made and by express terms made part of it, are as binding upon the contracting parties as if the same were bodily incorporated therein.
The act of Iowa, supra, in our opinion, creates no exception to the foregoing general rule governing the interpretation of contracts. That act was obviously intended for three purposes: (i) To prevent discriminations in favor of particular insurants; (2) to secure that certainty with respect to the rights and duties of the parties which is always best attained by written agreements; (3) to provide a ready and available method by which the insured or assured may at all times have before them the covenants and agreements which they are required to observe or perform. Society v. Puryear’s Adm’r (Ky.) 59 S. W. 15. That the foregoing is the true interpretation of the act in question is, in our opinion, also conclusively shown by subsequent legislation in Iowa.
By section 1819 of the Code of Iowa, enacted by the 26th general assembly at its extra session, in 1897, it is enacted as follows:
“All life insurance companies or associations organized or doing business in tliis state under tbe provisions of the preceding chapters shall upon the 'issue of any policy, attach to' such policy or indorse thereon a true copy of any application or representation of the assured, which by the terms of such policy are made a part thereof, or of the contract of insurance, or referred to therein, or which may in any manner affect the validity of such policy, or, upon reinstatement of a lapsed policy shall attach to the renewal receipt a true copy of all representations made by the assured upon which the renewal or reinstatement is made. The omission so to do shall not render the policy invalid, but if any company or association neglect to comply with the requirements of this section it shall forever be precluded from pleading, alleging or proving such application or representations, or any part thereof, or the falsity thereon, or any part thereof, in any action upon such policy, and the plaintiff in any such action shall not be required, in order to' recover against such company or association, either to plead or prove such application or representation, but may do so at his option.”
*273The last-mentioned act was passed while the act of 1890 was on the statute book of Iowa, and both are found in the revision of 1897. They must therefore be construed together, and given full force and effect if possible. If the act of 1890 is to be construed as plaintiff s counsel contend, the act of 1897 is meaningless.
Its provision for attaching a true copy of the application was obviously intended to furnish the certainty of, and familiarity with, the terms of the whole contract which was generally contemplated by the act of 1890, and the special prohibition against reliance upon any of the terms and stipulations found in an application, unless a true copy of the same be attached to the policy, by necessary implication permit such reliance if a copy is so attached. In our opinion, it is perfectly clear that the legislature of Iowa by the latter act recognized an application for insurance when a copy of the same is attached to the policy at the time it is delivered, as a constituent part of the contract or policy of insurance itself. A legislative construction has therefore been given to the act of 1890 in full harmony with what seems to us to have been its obvious meaning. It follows that plaintiff’s contention to the contrary cannot be sustained.
It is next contended that defendant cannot avail itself of the insured’s agreement against suicide, as found in the application for the second policy of $5,000, because of an alleged variance between the policy as issued and that applied for. The application was for insurance payable to Robert P. Mulock. The policy, as issued, made Kelly, his executors, administrators, or assigns, the beneficiary. No explanation is found in the record of this alleged variance, but in argument it was stated that as Mulock, by whom the policy was to be taken as collateral security for some obligation of Kelly, had apparently no insurable interest in his life, the change was made in order that the policy might be made an available collateral by assignment. However this may be, the fact appears that immediately after the receipt of the policy by Kelly he assigned the same to Mulock. In this way the purpose contemplated by the application was accomplished. Mulock got the policy exactly as applied for.
Not only so, but both Kelly and Mulock are dearly estopped from' contending that the policy was not issued in conformity to the appli-' cation. Kelly accepted the policy, which contained a statement that it was issued in consideration of the application, a copy of which was attached to it; in other words, that the policy was issued in consideration of the very application which is now sought to be repudiated. Whatever variance there was between the application and policy must be presumed to have been made with Kelly’s full consent and approval. Neither he nor any one claiming under him can now be heard to repudiate the application so acted upon by the company and recognized by him. Insurance Co. v. Myers (decided at the present term of this court; C. C. A.) 112 Fed. 846.
The next contention requiring consideration by us is that the plaintiff acquired vested rights at least in the $2,500 policy, in which she was originally named as sole beneficiary, immediately upon its issue, which could not have been affected by any subsequent conduct of the insured; that even though Kelly agreed as a part of the consideration *274of the contract of insurance not to kill himself, sane or insane, within the period of two years, and even if he violated that agreement, such .violation cannot be invoked against the claim of the plaintiff.
It cannot be disputed that plaintiff, who was Kelly’s .wife and beneficiary in the policy in question, had a certain vested interest in the policy immediately upon its issue; such an interest, in fact, that neither Kelly nor the insurer, nor both, could, by appointment or agreement take from her without her consent. Her rights were created by the contract, and she, as one of the parties thereto, must, on familiar principles, consent to any deprivation, modification, or change of such rights before the same can be accomplished. Bank v. Hume, 128 U. S. 195, 206, 9 Sup. Ct. 41, 32 L. Ed. 370, and cases cited. But this well-recognized principle falls far short of sustaining plaintiff’s contention in this case. The question still remains, with what rights was she vested? This obviously depends upon the terms and conditions of the contract creating them. The husband assumed to act as her agent in the negotiation of a contract intended to be beneficial to her. He gave a consideration ’ therefor, consisting partly of certain ex-ecutory promises. He secured the promise from the insurance company to pay money to the wife, in case of his death, by promising that such death should not, for two years at least, be the result of his own act, sane or insane. All this was so done as to disclose a clear intention on the part of both that no risk against such death should be assumed by the company.
The wife, by asserting a claim on the policy, ratifies and affirms the contract as made by her agent, and that, too, subject to all its terms and conditions. She cannot avail herself of the promise to pay her the amount of the policy, and simultaneously repudiate the promise made by her husband, which was given to the insurer as a consideration for its undertaking. Neither can she enlarge the obligation of the insurer beyond the scope of that undertaken by it. Baker v. Insurance Co., 43 N. Y. 283; Pitt v. Insurance Co., 100 Mass. 500.
These conclusions would seem to be the necessary' result of well-recognized principles of agency and contract. But our attention is called to the case of Seiler v. Association, 105 Iowa, 87, 74 N. W. 941, 43 L. R. A. 537, to which special consideration was given by the learned trial judge. In that case the supreme court of Iowa held that where a policy of life insurance contains no stipulation against suicide, and is taken out in good faith, it is not avoided, as against a third party named as beneficiary in the policy, by the fact that the insured, while sane, purposely took his own life. In so holding it attempted to distinguish that case from Ritter v. Insurance Co., 169 U. S. 139, 18 Sup. Ct. 300, 42 L. Ed. 693, Id., 17 C. C. A. 537, 70 Fed. 954, 42 L. R. A. 583, which decided that the personal representative of an insured, who, when sane, deliberately killed himself with intent to secure to his estate the amount of insurance he had effected on his own life, could not recover on the policy though it contained no provision at all respecting suicide. The supreme court of Iowa held that even though the intentional suicide by an insured might constitute a defense to a suit brought on a policy payable to his estate, containing no provision against such suicide, it would be no defense in *275such case if the policy was by its terms payable to a third party as beneficiary.
It is urged that in so far as the $2,500 policy involved in the present case is concerned, wherein the wile is named as the beneficiary, the express agreement not to commit suicide is of no greater obligatory force than the implied one would have been if there had been no express one, and that, therefore, the Iowa case is directly in point. This may or may not be so, but, in any event, neither the case itself,. nor the argument deduced therefrom, is fully persuasive to our minds. A view contrary to the Iowa doctrine-is taken in Hopkins v. Assurance Co. (C. C.) 94 Fed. 729, and affirmed by the court of appeals of the Third circuit, 40 C. C. A. 1 (99 Fed. 199). See, also, the case of Dean v. Insurance Co., 4 Allen, 96, 99.
Moreover, it seems to us that if there be an implied agreement on the part of every insured not to intentionally kill himself for the purpose of enforcing the liability under a policr, — and such, in our opinion, is the rule laid down in Ritter v. Insurance Co., supra,—such agreement inheres in and forms a part of the contract, and is as much a condition to liability as if it were written out into an express agreement ; and, that being so, for the reasons airea dy pointed out a third party, claiming under such a policy of insurance made for her benefit, ratifies and adopts the implied as well as the express conditions and limitations of the contract. Rut we are not forced to any refinement of logic to support the conclusion readied in this case. There is here an express agreement, showing that the contracting parties had the subject fully in mind, and came to a definite understanding to the effect that the insurer did not undertake to assume the risk of suicide for at least two years, and this agreement must be enforced.
The ease of Fitch v. Insurance Co., 59 N. Y. 557, 17 Am. Rep. 372, relied on by the supreme court of Iowa in the Seiler Case, makes it clear that the court of appeals of New York considered that force and effect should be given to an express condition of a policy against suicide, even as against a third party, who might be the beneficiary. It there says:
“The policy contained no simulation that it should he void in case of the death oC the insured by suicide. It was not taken out for the benefit of Fitch Fthe insured], but of his wife and children. Although they were bound by his representations, and any fraud he may have committed in taking out the policy, the policy having been obtained through his agency, yet they were not bound by any acts or declarations done or made by him after the issue of the policy, unless such acts were in violation of some condition of the policy.”
The case of Kerr v. Association, 39 Minn. 174, 39 N. W. 312, 12 Am. St. Rep. 631, is also relied upon by the supreme court of Iowa as sustaining its conclusion in the Seiler Case. A reference to that case shows that the provision of the policy made the basis of a defense was as follows:
“If the assured shall die in, or in consequence of, the violation of any criminal law of any country, state, or territory in which the assured may be, this certificate shall be null and void.”
The defendant in that case offered to prove on the trial that Kerr, the insured, had • committed the crime of forgery in Minnesota, and *276fled to Canada to escape arrest, where he was ultimately discovered and apprehended by detectives, and, to avoid being brought back to Minnesota for trial, shot and killed himself. The supreme court held that that evidence was not admissible, saying:
“His death in Canada cannot be treated as the proximate result of liis crime in Minnesota. * * * And the fact of his suicide is not in itself to be construed as occurring in or growing out of a violation of law, within the meaning of the policy.”
The court then remarked as follows:
“In the law of insurance, suicide is not, as a rule, recognized as a ground of exemption from liability or for forfeiture of a policy issued for the beneiit of a third person, unless it is expressly so provided in the policy.”
The supreme court of Iowa also seems to think that' a different result might follow if the contract of insurance contained an express provision exonerating the company in case of suicide. It says, after reviewing the cases to which reference has just been made, as follows: .
“We wish now to add a few words on principle by way of emphasis of a thought already expressed. It is not the wrongdoer who makes claim here, nor any representative whose rights are to be measured by those of the wrongdoer, but persons who acquired an interest at the time the policy was taken out, and who are not in any way responsible for the loss under it. The defendant might well have guarded against this contingency in its contract Not having done so, we think it is now in no position to complain.”
How different is the case now under consideration by us!
As already pointed out, the plaintiff’s right, even in the policy of $2,500, was to be measured by those of the wrongdoer, and the insurer in the present case has in the contract specifically guarded against the very contingency which the supreme court of Iowa said might have been guarded against in the contract. We unhesitatingly reach the conclusion that, on principle as well as authority, no such vested rights were created by the $2,500 policy in favor of the plaintiff in this case as to relieve her against the consequence of self-destruction by the insured.
The other policy for $5,000, in which plaintiff acquired no rights until after the death of the insured, raises no such question as has just been discussed with reference to the $2,500 policy. She, by accepting an assignment from the beneficiary, directly or indirectly, after the death of the insured, is confessedly made subject to all the infirmities inherent in the contract as originally made.
The learned trial judge in his opinion (C. C.; 109 Fed. 56) places great confidence in plaintiff’s right to recover upon the principle that the covenant of the defendant to pay the amount of the policy in question is an independent covenant, and not at all dependent upon the covenant of the insured not to kill himself. We are unable to concur in this view. The covenant to pay is obviously dependent upon whether the death insured against occurs, and that death, as already seen, is one exclusive of suicide within two years. Not only is this so, but the promise not to kill himself within two years was by agreement made part of the consideration moving the insurer to make this promise. .
*277As said by Mr. Justice Miller in announcing the opinion of the supreme court in Construction Co. v. Seymour, 91 U. S. 646, 650, 23 L. Ed. 341:
“Whore a specified thing is to be done by one party as the consideration of the tiling to be done by the other, it is undeniably the general rule that the covenants are mutual, and are dependent, if they are to be performed at the same time; and, if by the terms or nature of the contract one is first to be performed as the condition of the obligation of the other, that which is first to be performed must be done or tendered before that party can sustain a suit against the other.”
In the case of Loud v. Water Co., 153 U. S. 564, 14 Sup. Ct. 928, 38 L. Ed. 822, Mr. Justice Jackson, speaking for the supreme court, says;
“Tiie question whether covenants are dependent or independent must be determined in each case upon the proper construction to be placed on the language employed by the parties to express their agreement. * * * The question in each case is. which intent is disclosed by the language employed in the contract?”
Applying the test last stated, we have no difficulty in reaching the conclusion that the parties to the contract of insurance in question obviously and plainly intended that the covenants should be mutual and dependent, and not independent of each other.
It is next contended that the contracts of insurance sued on in this case are by their own terms embodied exclusively in the policy, ■ and do not comprehend the applications made by the insured therefor. An argument to this effect is drawn from the fact that the policies state on their face that the payment was to depend upon the following condition: “⅜ ⅜ * The annual premium * * ⅜ shall he paid in advance on the delivery of this policy, and thereafter, ⅜ ⅜ ⅜ on a certain day, every year during the continuation of this contract;” “and subject to the provisions, requirements, and benefits stated on the back of this policy, which are hereby referred to and made a part hereof.” The maxim, “Expressio unius exclusio alterius,” is invoked, and it is claimed that because the provisions of the application are not found on the back of the policy their stipulations and contents are not a part of the contract of insurance. We cannot agree to any such view. We have already considered this general question in disposing of the argument based on the Iowa statutes, but it may not be improper to observe further that the company said in each of the policies that it was issued in consideration of the application made for it, and that such application was made a part of the contract. It thus clearly appears that by the same token by which the provisions on the back of the policy were made part of the contract the application therefor was also made a part of it. Nothing can be clearer than this.
Kelly, by accepting a policy which by its terms incorporated the application as a part of the contract, necessarily admitted that such was a fact. -He had already said the same thing in the application signed by him, namely, that it was offered to the company as a consideration of the contract, which he was to get. The entire application having been so made a part of the policy by agreement of the parties must be so treated by us, and the agreements, found in the application *278as well as those found in the policy proper, must be considered in determining the true import and meaning of the contract. Ritter v. Insurance Co., 169 U. S. 139, 18 Sup. Ct. 300, 42 L. Ed. 693; Insurance Co. v. McConkey, 127 U. S. 661, 8 Sup. Ct. 1360, 32 L. Ed. 308; First Nat. Bank v. Hartford Fire Ins. Co., 95 U. S. 673, 24 L. Ed. 563; Insurance Co. v. Webb, 45 C. C. A. 648, 106 Fed. 808; American Credit Indemnity Co. v. Carrollton Furniture Mfg. Co., 36 C. C. A. 671, 95 Fed. 111.
Accordingly, treating the application and all its terms and provisions as a part of each contract entered into between the insurance company and Kelly, what does it mean ? In answering this question there does not seen to be any necessity for resort to technical distinctions between representations and warranties or affirmative and promissory warranties. The cardinal rule to be observed in construing all contracts is to determine, from a consideration of the four corners of the instrument or instruments creating it, what was the intention of the parties to the same. Insurance Co. v. Gridley, 100 U. S. 614, 615, 25 L. Ed. 746; Long v. Timms, 107 Mo. 512, 519, 17 S. W. 898. Subjecting the contracts in question to this test, it is very apparent, as we have already indicated in disposing of other branches of the case, that the death of Kelly by suicide at any time within two years after the date ’ of the policy, whether sane or insane, was not a risk assumed by the insurer at all. The parties to the contract, when made, the insurer speaking for itself, and Kelly speaking for himself and as agent for the beneficiary, so agreed. This clearly enough appears from the following: The company agreed, in effect, that in consideration of the representations and agreements found in the applications for insurance, made by Kelly, it would upon the death of Kelly pay a certain sufn of money to the beneficiaries named in the policies. Kelly agreed that such death should not occur by suicide within two years at least. The fair and reasonable import of these mutual agreements, in our opinion, is that the death insured against ivas such a death as might occur at any time in the future, excepting that, however, by suicide, sane or insane, if the same should occur within two years.
Kelly, in his proposition for insurance called “Application,” said:
“I hereby warrant and agree not to reside or travel in any part of the torrid zone, and not to engage in any specially hazardous occupation or employment, during the next two years following the date of issue of the policy for which application is hereby made, and also not to engage in any military or naval service in time of war, during the continuance of the policy, without first obtaining permission from this company. I also warrant and agree that I will not die by my own act, whether sane or insane, during the said period of two' years.”
The company in its acceptance of the proposition, under the heading “Incontestability,” found on the back of the policy, said :
‘‘It is hereby further promised and agreed that after two years from date hereof the only conditions which shall be binding upon the holder of this policy are that he shall pay the premiums at the times and place and in the manner stipulated in the said policy, and that the requirements of the company as to age and military or naval service in time of war shall be observed, and that in all other respects, if this policy matures after the expiration of the said two years, the payment of the sum insured by this policy shall not be disputed.”
*279This last-mentioned covenant or agreement on the part of the company obviously has reference to the agreements of Kelly found in the proposition above quoted. It cannot escape observation that the parties treated the agreements referred to as conditions of continuing liability on the part of the company. They are referred to as “conditions,” and it is agreed that the only ones which shall be binding after two years are those relating to engaging in military or naval service in time of war. In all other respects it was agreed that “if this policy matures after the expiration of the said two years the payment of the sum insured by this policy shall not be disputed.” This carries the reasonable implication that, if the policy should mature by the death of the insured within two years, the policies might be disputed for breach of any of the “conditions” upon which liability depended.
The clear import and meaning of all this is that the policy was issued and the company’s obligation of payment assumed on certain specified conditions, among them that Kelly should not reside or travel in any part of the torrid zone, or engage in any of the specially hazardous occupations recited in the application, or die by his own act, whether sane or insane, within the period of two years after the date of the policy. The wisdom of conditions of this kind is not for us to con - sider, but it is perfectly obvious that they were intended to prevent devising schemes to defraud an insurance company by securing insurance in contemplation of immediate exposure to probable death or immediate purpose to take one’s own life. However this may be, the parties, by language admitting of no other rational meaning, agreed upon it, and that puts an end to our inquiry. The insured, acting for himself in one application, and as agent for the proposed beneficiary in the other, warranted and agreed, as a consideration for the execution of the policies, that he would not die by his own act, whether sane or insane, during the period of two years after the date of the policies. This, by reference to the incontestibility clause, already quoted, was by agreement made a condition to continuing liability.
Cut we do not wish to be understood as holding that it required any express agreement that the clause in question should be treated as a condition to liability. In our opinion, it, having been offered to and accepted by the company as a consideration of the assumption of the risk and for the continuance of liability for two years at least, is, in and of itself, a promissory warranty, requiring the insured to strictly conform thereto, in order to hold the company to a liability on the. policies in favor of the beneficiaries. Failure to observe the stipulation of warranty, resulting, as in this case, in the death of the insured, undoubtedly absolves the insurance company from liability.
For the reasons already given, we cannot agree with counsel for plaintiff that it is necessary to find in the policy itself an express stipulation that in case of failure to observe the warranty the policy should be void. Such is the conclusive result of such failure determined and fixed by the law itself. While we have endeavored to answer the argument of counsel based upon the technical doctrines of condition and warranty, we prefer to place our determination of this *280case upon the broad proposition that the contracts of insurance, when fairly and- reasonably construed, show that death of the insured by suicide, sane or insane, was a risk not undertaken by the insurer at all. There is no merit in the contention that a return of the premiums paid by Kelly was a prerequisite to a defense by the insurance company. The company earned,the premiums paid by Kelly for the risk which it agreed to assume, and which it did assume and carry until Kelly’s death.. This risk embraced death from practically all other causes but suicide. Cases where fraud may have been so practiced in the negotiations as to render the contract voidable at the instance of the company, or cases where no risk at all ever attached, are totally inapplicable to the facts disclosed in this case, and afford no warrant for plaintiff’s contention.
It is next contended that the agreement of Kelly not to die by his own act while insane was impossible of performance, and known to-be so by both parties to the contract, and therefore void. The argument is that self-destruction by an insane person is not his act, but rather the result of an irresistible impulse, over which he had no control, and therefore not within his power to prevent, and that an agreement to prevent it falls within that class of agreements which are void because of impossibility of performance. This argument is obviously founded on the doctrine taught by the case of Insurance Co. v. Terry, 15 Wall. 580, 21 L. Ed. 236. It is there held that where the condition is that, “if the insured sh^Il die by his own hand” (without the qualifying words sane or insane), the policy shall be void, the defense must show affirmatively that the death was the result of an intentional act of a responsible being, and not the act of one driven by an insane impulse, over which he had no control.
The history of insurance litigation shows that the decision in the last-mentioned case brought about other and different stipulations in policies of life insurance, exonerating insurers from liability in case of suicide by the insured, whether he was sane or insane at the time of committing the act. Stipulations of the latter kind have been frequently before the courts, and been pronounced valid and enforceable-on the distinct ground that they create an excepted risk; in other words, that a clause of that kind found in a policy of insurance evinces-a clear intention on the part of both parties to the contract that self-inflicted death by the insured, whether sane or insane at the time, was not one of the risks assumed by the insurer.
In the case of Bigelow v. Insurance Co., 93 U. S. 284, 23 L. Ed. 918, Mr. Justice Davis, in delivering the opinion of the court, says:
“There has been a great diversity of judicial opinion as to whether self-destruction by a man, in a fit of insanity, is within the condition of a life policy, where the words of exemption are that the insured ‘shall commit suicide,’ or ‘shall die by his own hand.’ But since the decision in Insurance Co. v. Terry, 15 Wall. 580, 21 L. Ed. 236, the question is no longer an open one in this court. In that case the words avoiding -the policy were, ‘shall die by his own hand,’ and -we held that they referred to an act of criminal self-destruction, and did not apply to an insane person who took his own life. ' But the insurers in this case have gone further, and sought to avoid altogether this class of risks. If they have succeeded in doing so, it is our duty to give effect to the contract, as neither the policy of the *281law nor sound morals forbid thorn to make It. If they are at liberty to stipulate against hazardous occupations, unhealthy climates, or death by the hands of the law, or in consequence oí injuries received when intoxicated, surely It is competent for them to stipulate against intentional self-destruction, whether it be the voluntary act of an accountable moral agent or not. It is not perceived why they cannot limit their liability, if the assured is in proper language told of the extent of the limitation, and it is not against public policy. The words of this stipulation, ‘shall die by suicide (sane or insane),’ must receive a reasonable construction. If they be taken in a strictly literal sense, their meaning might admit of discussion, but it is obvious that they were not so used. ‘Shall die by his own hand, sane or insane,’ is doubtless a more accurate mode of expression, but it does not more clearly declare the intention of the parties. * * * Nothing can be clearer than that the words ‘sane or insane’ were introduced for the purpose of excepting from the operation of the policy any intended self-destruction, whether the insured was of sound mind or in a state of insanity.”
In the case of Insurance Co. v. McConkey, 127 U. S. 661, 8 Sup. Ct. 1360, 32 L. Ed. 308, the policy sued on contained a clause that “no claim shall be made under this policy when the death or injury may have been caused by * * ⅜ suicide (felonious or otherwise, sane or insane).” Mr. Justice Harlan, in delivering the opinion of the court in that case, said if the insured “commit suicide then the law was for the company, because the policy by its terms did not extend to or cover self-destruction, whether the insured was at the time sane or insane.” See, to the same effect, the case of Scarth v. Society, 75 Iowa, 346, 39 N. W. 658. There are many other cases, both state and federal, to which attention might be called, which announce the same doctrine, but we do not deem it necessary to pursue this inquiry further.
Applying the rule governing the • interpretation of contracts herein-before referred to, and seeking to give a reasonable interpretation to the clause now under consideration, consonant with the manifest intention of the parties as disclosed by all the provisions of the policies, we can only reach one conclusion: that the insured, Kelly, not only agreed that he would not die by his own act, whether sane or insane, within the period of two years, but, in effect, agreed, as already stated, that the risk actually assumed by the company excluded death by suicide within two years. The facts of the case therefore do not warrant the application of the rule rendering contracts void which are impossible of performance, and so known by both parties to it.
The ability and persistence with which learned counsel for plaintiff, both in oral argument and brief, pressed the points already considered upon the attention of the court, caused us to enter upon a discussion of questions more at length than their intrinsic difficulty, in the light of controlling authority, probably required.
The result reached, after a full consideration of all questions presented, is that as a conclusion of law, dediicible from the agreed statement of facts on which the case was submitted, the plaintiff cannot recover. The judgment of the trial court must be reversed and remanded, with directions to render a judgment in favor of the defendant.