The court below, upon the petition of Woltman, duly made an order admitting a certain document to probate as the last will of the decedent. From this order Walter Manchester, a brother and heir of the decedent, appeals.

The document referred to was wholly in the handwriting of thé decedent. The only objection presented upon this appeal is that it was not signed by the decedent.

The document began as follows:

The name of the decedent does not appear in or on the paper anywhere, except in the opening clause as above shown. This document was folded by the decedent and placed in an envelope, which was then sealed and indorsed by the decedent, in her own handwriting, with the words, “My Will, Ida Matilda Manchester. ’ ’ In that condition it was, by her direction, placed in her safe deposit box, where it was found after her death. These are all the facts bearing upon the question of its execution as a will.

“An olographic will is one that is entirely written, dated, and signed by the hand of the testator himself. It is subject to no other form, and may be made in or out of this state, and need not be witnessed.” (Civ. Code, see. 1277.)

The will in question fits this description in all respects except the signing thereof. It was not “signed” according to the meaning of that word in ordinary usage. To sign, as applied to a document, is defined as follows: “To affix a signature thereto; to ratify by hand or seal; to subscribe in one’s own handwriting.” (Webster’s Dictionary.) Unquestionably, as used in the above-quoted section, it means the signature of the testator in his own handwriting written somewhere in or upon the document, with the intention by so writing it to authenticate the document. The name written at another place than the end of the document, and not for the purpose of authenticating it and indicating its completion, but merely to identify the person who is making the will, cannot be deemed to be a name “signed” to the document, unless that word is given a meaning entirely different from that which it is generally understood to have. (7 Words & Phrases, p. 6508.) The Civil Code itself provides that words “are construed according to the context and the approved usage of the language.” (Civ. Code, sec. 13.) If this be done, a document in which the name of the person making it appears only in the beginning thereof, and by way of recital to designate that person as the maker, in the manner above shown, cannot be said to have been signed by the maker. (Estate of Walker, 110 Cal. 393 , [ 52 Am. St. Rep. 104 , 30 L. R. A. 460, 42 Pac. 815 ].) We have here the additional indication of lack of that completion essential to a will in .the closing words, “whereunto I hereby set my hand this,” etc. These are apt words to precede a signature in attestation of a will or deed, and they tend to show that the decedent intended to sign immediately below, but failed to carry out that intention.

The defect is not cured by the words “My Will” indorsed on the envelope and signed by the decedent. The manifest purpose of that indorsement was to state that the paper within the sealed envelope was the will of Ida Matilda Manchester. It shows that the decedent believed that the inclosed document was her will, and indicates that she believed that it was lawfully executed and valid. But her belief that it was a valid will, properly executed, does not make it so. The power to dispose of one’s property by will and the mode by which it may be exercised are matters under legislative control, and the mode prescribed by the statute must be followed, or there is no will. “For the purpose of determining whether a will had been properly executed, the intention of the testator in executing it is entitled to no consideration. For that purpose the court can consider only the intention of the legislature as expressed in the language of the statute, and whether the will as presented shows a compliance with the statute.” (Estate of Seaman, 146 Cal. 460 , [ 106 Am. St. Rep. 53 , 2 Ann. Cas. 726, 80 Pac. 701 ]; Estate of Walker, 110 Cal. 391 , [ 52 Am. St. Rep. 104 , 30 L. R. A. 460, 42 Pac. 815 ].) If the decedent had used a printed form of will, filled the blank spaces as she desired, and had duly signed it, without attesting witnesses, the document would show her testamentary intentions as satisfactorily as the document she here prepared, but it would not have been a valid will. (Estate of Billings, 64 Cal. 427 , [ 1 Pac. 701 ].) The act of signing is essential. An intention proven by other means will not serve the purpose.

There is no force in the argument that the indorsement on the envelope can be treated as a part of the document inclosed, as if it were an additional page attached thereto to give room for the signature. The circumstances are utterly inconsistent with the idea that the indorsement was made for any such purpose. They show that it constituted no part of the paper designated as “My Will” within the sealed envelope. (Warwick v. Warwick, 86 Va. 596 , [6 L. R. A. 775, 10 S. E. 843 ].)

It is claimed that the decisions in Estate of Stratton, 112 Cal. 513 , [ 44 Pac. 1028 ], and Estate of Gamp, 134 Cal. 233 , [ 66 Pac. 227 ], are opposed to the foregoing conclusions. In Estate of Stratton the olographic will was duly signed at the end, but below the signature were the words, “My husband Thomas Stratton.” Discussing the effect of these words as part of the document, the court held that they were intended to name the husband as a legatee; that they were merely displaced by the testator, and should be read as though written at the beginning or end of the last clause, in which the legatee was designated only by the pronoun “he.” In this discussion the court said: “The statute does not require that an olographic will shall be ‘subscribed by the testator at the end thereof.’ It is sufficient that it be ‘signed’ by him and this signing may be at the beginning or in any part of the docu merit, ’ ’ citing Estate of Walker, 110 Cal. 391 , [ 52 Am. St. Rep. 104 , 30 L. R. A. 460, 42 Pac. 815 ].

In Estate of Gamp, proof that the final clause of an olographic will, including the signature of the testator at the end thereof, had been torn off and destroyed after the testator’s death was duly made by two credible witnesses, as required by section 1339 of the Code of Civil Procedure, and this was held to be sufficient proof of the authenticity of the will. The initial clause also contained the name of the testator. The court said that this also established the fact that the document was intended by him to be his last will, and that “the writing by him of his name in that clause was itself a sufficient signature. ’ ’

The declarations in these cases are entirely consistent with the rule stated in Estate of Walker, that “To sign an instrument or document is to make any mark upon it in token of knowledge, approval, acceptance, or obligation. The signature is the sign thus made.” In neither of them was there any discussion, or need of discussion, as to the kind of proof permissible to establish the fact that the signature appearing in an olographic will, though not at the usual place at the end of the document, was nevertheless intended as a signature in execution thereof. They do not in the least control our decision of this question. The true rule, as we conceive it to be, is that, wherever placed, the fact that it was intended as an executing signature must satisfactorily appear on the face of the document itself. If it is at the end of the document, the universal custom of mankind forces the conclusion that it was appended as an execution, if nothing to the contrary appears. If placed elsewhere, it is for the court to say, from an inspection of the whole document, its language as well as its form, and the relative position of its parts, whether or not there is a positive and satisfactory inference from the document itself that the signature was so placed with the intent that it should there serve as a token of execution. If such inference thus appears, the execution may be considered as proven by such signature.

This rule has the support of the better reasoned decisions of other jurisdictions. Thus, in Waller v. Waller, 1 Graft. (Va.) 474, [42 Am. Dee. 564], where the subject is elaborately discussed, the court says that the signing of the will is required as proof “that the act is a complete, concluded act,” and that “the signing must be such as upon the face and from the frame of the instrument appears to have been in-, tended to give it authenticity. It must appear that the name so written was regarded as a signature, that the instrument was regarded as complete without' further signature and the paper itself must show this.” In Warwick v. Warwick, 86 Ya. 596, [6 L. R. A. 775, 10 S. E. 843 ], the will began as follows: “I, Abraham Warwick, declare this to be my last will and testament,” but the testator’s name did not occur elsewhere in the will, and there was no declaration in the document that the signature at the beginning was written for the purpose of authentication or execution. The court held that the signature was not “affixed in such a manner as to make it manifest that the name was intended as a signature.” To the same effect are Catlett v. Catlett, 55 Mo. 339 ; Crutcher v. Crutcher, 30 Tenn. (11 Humph.) 384; Armani’s Succession, 43 La. Ann. 415 , [ 26 Am. St. Rep. 183 , 9 South. 50 ]; In re Tyrrell’s Estate, 17 Ariz. 418 , [ 153 Pac. 767 ]; French v. French, 14 W. Va. 479 . And in the recent decision in Estate of Dombrowski, 163 Cal. 295 , [ 125 Pac. 235 ], we said: “No doubt a subscription or signing of any kind will not constitute a valid execution of a will unless made with the intention, on the part of the testator, of finally and completely authenticating the will.” This was said of an attested will, but that it was not supposed to state a rule confined to attested wills is shown by the fact that Waller v. Waller, 1 Gratt. (Va.) 474, [ 42 Am. Dec. 564 ], is cited as an application of it.

The few decisions to the effect that where the name of the testator appears only in the initial clause of an olographic will, the will may be considered as “signed” within the meaning of statutes similar to ours, are all based on the supposed authority of the English ease of Lemayne v. Stanley, 3 Lev. 1, [83 Eng. Reprint, 545], decided in 1681. The will there considered was an attested will. The subsequent eases following it take it as absolute authority, and do not properly distinguish or limit the scope of its language. The case has been criticised in England and limited strictly to cases where witnesses saw the will written or where it was subsequently acknowledged before them. (Morrison v. Turnour, 18 Ves. Jr. 175, [34 Eng. Reprint, 284]; Coles v. Trecothick, 9 Ves. Jr. 234, [32 Eng. Reprint, 598].) To overcome its effect in England the statutes of 1 Viet., 15 Viet., and 16 Viet, were enacted, changing the law so as to require that “all wills be signed at the end thereof.” In view of this consequence of that decision, it is obvious that its authority should not be considered as controlling. The rule we have stated is, in our judgment, supported by sounder reasons and is less likely to produce evil results.

Our conclusion is that the document in question was not signed by the testator, as required by section 1277, and that it is not entitled to probate.

The order is reversed.

Henshaw, 'J., Lorigan, J., Melvin, J., gloss, J., Lawlor, J., and Angellotti, C. J., concurred.