— Lewis Sober executed before two witnesses, in the form prescribed by statute, a will which was in the handwriting of another. Subsequently he attached to said will the following codicil:—

Underneath were the words: “Witness, George Penlington, 123 O’Farrell Street.” All except these latter words is in the handwriting of the testator. The court below admitted the will to probate, but refused to admit the codicil. Two points are made by the respondent in support of the action of the court with reference to the codicil.

1. It is argued that the fact that the codicil was attested by one witness shows that it was not the intention. of the testator to make an olographic codicil; that it must be presumed that he intended to make an attested codicil, and left the execution incomplete. We see no force in the argument. It will be observed that there is no attestation clause, such as is usual in attested wills and codicils, no statement that the testator signed the will in the presence of the witness, and declared, at the time of signing, that the instrument was his will. If there had been such a clause, signed by only one witness, or unsigned by anybody, there might be more *479 force in the position, although we do not desire to be understood as expressing or intimating any opinion upon such a case. Here are simply the words, “Witness, George Penlington, 123 O’Farrell Street.” These words do not seem to us to indicate an intention to make an attested will or codicil. It may be that the testator caused them to be placed there for the purpose of supplying proof of his own handwriting, or in other words, proof that the codicil was olographic. Many testators execute their wills long before death, and they may be persons who are not well known in the community. It may therefore be difficult to prove the handwriting of a document purporting to be a will, found among their papers after death. In view of this, it would be a natural and proper precaution to furnish means of making the required proof. And it seems to us that courts should presume that the intention of the testator was that of a reasonable and prudent man under the circumstances, and should not adopt a somewhat strained construction to defeat what there is every reason to believe was the desire of the testator. The witness clause is not, under the circumstances, to be considered as a portion of the will, which is no more affected by it than any other document which does not require attestation, such as a deed or contract would be. The case of Power v. Davis, 3 McAr. 162, is not in point. There the testator appended a formal attestation clause in his own handwriting, but left such clause unsigned by anybody. There is no room here for the argument made in that case.

2. It is contended that an attested will cannot be revoked by an olographic codicil. The position of the respondent is based upon the fact that the olographic document refers to the other. The argument is, that the olographic codicil cannot be understood without referring to the attested will; that the latter is in contemplation of law "a part ” of the former, and that therefore it *480 does not come within the definition of an olographic will, which is a will “entirely written, dated, and signed by the hand of the testator himself.” This argument, if allowed to prevail, would have far-reaching consequences. It would exclude all reference to any document which was not entirely written, dated, and signed by the testator himself. For example, an olographic will which devised all the right, title, and interest of the testator under a will of his ancestor, or under a marriage settlement or other contract drawn by another hand, would be invalid. Pushed to its logical conclusion, the argument would exclude all reference to deeds or other documents for descriptions of property, etc.

A similar objection was at one time made to references in attested wills to documents which were not attested. Wills have always been required to be executed with formalities of some kind or other; and it was argued that documents which were not executed with these formalities could not be referred to. But considerations of practical convenience prevailed over technical reasoning. And it became well settled that an attested will could refer to documents which were not attested. (See Newton v. Seaman’s Friend Society, 130 Mass. 91 ; 39 Am. Rep. 433 ; Baker’s Appeal, 107 Pa. St. 381; Brown v. v. Clark, 77 N. Y. 377 ; Fickle v. Snep, 97 Ind. 291 ; 49 Am. Rep. 449 ; Fessler v. Simpson, 58 Ind. 87 ; Gerrish v. Gerrish, 8 Or. 351 ; 34 Am. Rep. 585 ; and see Estate of Shillaber, 74 Cal. 144 .) Now, if an attested will can refer to a document which is not attested, we see no good reason why an olographic will may not refer to a document which is not in the handwriting of the testator. The only difference between an olographic and an attested will is in the form of the execution. The statute has prescribed two forms in which written wills may be executed. In each case the instrument must be signed by the testator. But the formality of witnesses is dispensed with if the instrument is all in the handwriting *481 of the testator himself. One form is the precise equivalent of the other. Whatever would be good as an attested will or codicil is good as an olographic one, if written, dated, and signed by the hand of the testator. And whatever may be done in or by the one may be done in or by the other. Therefore, if the formalities of attestation are not required in a document referred to' by an attested will or codicil, the corresponding formalities are not required in a document referred to by an olographic will or codicil.

The argument made for the respondent rests upon the proposition that a document referred to by a will or contract is in contemplation of law " a part” of the document making the reference. And if this be granted in the sense in which it is asserted, a very plausible argument can be built upon it. But as has been said in other connections, it is not true as a matter of physical fact that the two documents are one and the same. The law for some purposes—mainly of construction—regards one as a part of the other. But this fiction ought not to be extended to absurd or unjust consequences. (See City of Napa v. Easterby, 76 Cal. 222 ; Bull v. Coe, 77 Cal. 54 .) And we think it should not be extended so as to cover the present case. The will which the statute requires to be attested by witnesses in the one case, and written by the testator in the other, is the document signed by him. The documents referred to are for the purpose of explaining the meaning of the will and aiding in its construction.

The conclusion Which we have reached is sustained by authority. The decision in Estate of Shillaber, 74 Cal. 144 , does not quite cover the case. For although the language of the opinion sanctions the reference, the decision was that the document referred to could be rejected as unnecessary, and the will left to stand without it. But Estate of Skerrett, 67 Cal. 587 , goes further than is required here. In that case the deceased in his *482 lifetime signed a deed purporting to convey certain property to his sister. This deed was never delivered, and therefore did not convey the property. It could not take effect as a will, because it was not witnessed as such, and was not in the handwriting of the deceased. But a copy of the deed was found after the death in an envelope, with a letter addressed to the sister, and written, dated, and signed by the deceased himself. This letter was insufficient, standing by itself, as a will. But it referred to the deed, and stated in substance that said deed was intended to be a provision for the sister in case of the death of the writer. The court held that although neither document, taken by itself, was sufficient as a will, yet both, taken together, should be admitted to pro-, bate as such. The point is not reasoned out at length,' but the opinion is clear, and the decision proceeds upon the principle above discussed. The courts of Louisiana, which are governed by a similar statute, seem to act upon the same theory. (See Miller v. Miller, 82 La. Ann. 441 .) And we think that the rule is a sound and salutary one. If testators are to be encouraged by a statute like ours to draw their own wills, the courts should not adopt, upon purely technical reasoning, a construction which would result in invalidating such wills in half the cases. ■

We therefore advise that the judgment and order' appealed from be reversed, with directions to admit the codicil to probate in connection with the will.

Belcher, C. C., and Foote, C., concurred.

For the reasons given in the foregoing opinion, the judgment and order appealed from are reversed, with directions to admit the codicil to probate in connection with the will.