Under our present law, Revisal, sec. 3113, “A written will with witnesses must have been prepared in the testator’s lifetime and signed by him or some other person in his presence and by his direction, and subscribed in his pres*77ence by two witnesses at least/’ etc. Construing this law, tbe courts haA^e beld “that it is not necessary always that the testator should sign the will in the presence of the witnesses; it is sufficient that the will be acknowledged by the testator in their presence, the will being physically present and identified.” In re Herring’s Will, 152 N. C., 258; Nickerson v. Brick, 66 Mass., 332. Nor is it required that the witnesses should subscribe to the will in the presence of each other. In re Herring’s Will, supra; Payne v. Payne, 54 Ark., 415; Eulbecks v. Granberry, 3 N. C., 232; Gardner on Wills, p. 217. With these authoritative interpretations in mind, it is always required that in order to a valid written will with witnesses the same should, as stated, be signed by the testator or some other person in his presence and by his direction, and subscribed in his presence by at least two witnesses.
In regard to the proper probate, the method by which these essential facts should be established, the statute, sec. 3127, makes provision as follows: “In case of a written will, with witnesses, on the oath of at least two of the subscribing witnesses, if living; but when any one or more of the subscribing Avit-nesses to such will are dead, or reside out of the State, or cannot after due diligence be found within the State, or are insane or otherwise incompetent to testify, then such proof may be taken of the handwriting, both of the testator and of the Avitness or witnesses so dead, absent, insane, or incompetent, and also of such other circumstances as Avill satisfy the clerk of the Superior Court of the genuineness and the due execution of such will. In all cases where the testator executed the will by making his mark, and where any one or more of the subscribing Avitnesses are dead or reside out of the State, or are insane or otherwise incompetent to testify, it shall not be necessary to prove the handwriting of the testator, but proof of the handwriting of the subscribing witness or witnesses so dead, absent, insane, or incompetent shall be sufficient. The xorobate 0f all wills heretofore taken in compliance Avith the requirements of this section are hereby declared to be valid.”
It will thus be noted that when any one of the subscribing witnesses survives or is competent to testify, proof may be taken *78of tbe bandwriting, both of tbe testator and tbe other witness or witnesses, and of sucb other circumstances as shall satisfy tbe clerk of the Superior Court of tbe genuineness and tbe due execution of sucb will, with tbe proviso that when tbe testator has signed by making bis mark, proof of bis bandwriting is not necessary. According to tbe express provisions of tbe law, therefore, whenever tbe facts indicated have been properly established before tbe clerk, be may adjudge tbe will to be duly .proven and record tbe same, and, when sucb testimony is offered on an issue of devisa/oit vel non, it affords 'evidence from which tbe will may be established by tbe jury, and it is not required, as contended by tbe caveators, that, in order to a valid probate, tbe surviving witness should testify that be saw tbe other witness subscribe bis name to tbe instrument. In the case of Thomas’ Will, 111 N. C., 412, one of tbe authorities relied upon by tbe caveators to sustain their position, tbe original will was lost, and, in tbe endeavor to prove tbe will, it was shown, apparently without exception, that when tbe instrument was offered for probate in common form one of tbe subscribing witnesses bad testified to having subscribed tbe same as witness, and another, who was not a subscribing witness, testified that tbe signature of tbe other subscribing witness was in bis own proper bandwriting. This, with proof of tbe death of J. W. Thomas, tbe other subscribing witness, was tbe entire evidence offered on tbe issue. There was no evidence offered as to tbe bandwriting of tbe testatrix, and Associate Justice Avery, delivering tbe opinion denying probate, said: “The propounders failed to produce any witness who bad ever seen tbe signature of Ada W. Thomas to tbe original will or tbe signature of .either of tbe witnesses, or that would testify to their genuineness. Indeed, tbe only evidence offered to show tbe loss of tbe original paper was that of D. 0. Mangum, who last saw it in possession of tbe sole legatee and devisee.” And in R. R. v. Mining Co., another case to which we were referred by counsel, tbe Court only held that a certificate of probate in another State, disposing of property in this State, would not suffice here when it did not affirmatively appear that tbe provisions of our statute bad been complied with as to tbe due execution *79of a will. Revisal, sec. 3133. The other authorities relied upon were chiefly cases under the old Revised Statutes, where proof in common form was permissible by one of the subscribing witnesses, and it was held that when proof of that character was resorted to, the witness who was examined, if his evidence was set out, should appear to have testified to the proper attestation of the other witness (In re Thomas, supra,; Blount v. Patton, 9 N. C., 237) ; but these decisions do not bear on the requirements of the present statute, nor should they be allowed to control the positive provisions of our present law as to the proper probate of a will. While the-ruling on this question favors the propounder’s position, it would probably not avail them on this record, because we find no direct evidence admitted by the court as to the handwriting of D. W. Watson and the genuineness of his signature to the will. Such testimony was had before the clerk, on the probate in common form, but this was ruled out by his Honor, and the authorities seem to hold that in the trial of an issue of devisa/oit vel non, on caveat duly entered, the proof as to the formal execution of the will shall be made de novo. In re Hedgepeth, 150 N. C., 245; In re Thomas, 111 N. C., supra, 416. True, the witness D. E. Hin-son testified that Mr. Watson brought this will (Exhibit A) to him, “all fixed up,” and “spoke of it as his will,” and this undoubtedly is a relevant circumstance; but the statute seems to require that, when the will purports to be signed by the testator himself, and only one of the subscribing witnesses is alive and competent, that some evidence should be introduced as to the handwriting of the testator or the genuineness of the signature. Without further reference to this feature of the case, we are of opinion that the propounders are entitled to a new trial of the cause by reason of the fact that from the form of the issues and the charge and rulings of the court, the execution of the second will (Exhibit B) has been allowed no effect whatever as to the validity of the first (Exhibit A). It is well recognized in this State that a will, properly executed, may so refer to another unattested will or other written paper or document as to incorporate the defective instrument and make the same a part of the perfect will, the conditions being that the paper *80referred to shall be in existence at the time the second will is executed, and tbe reference to it shall be in terms so clear and distinct that from a perusal of the second will or with the aid of parol or other proper testimony, full assurance is given that the identity of the extrinsic paper has been correctly ascertained. The principle is sometimes referred to as “The doctrine of incorporation by reference,” and is very well stated by Chief Justice Gray in Newton v. Seaman’s Friend Society, 130 Mass., 91, as follows: “If a will, executed and witnessed as required by statute, incorporates in itself by reference any document or paper not so executed and witnessed, whether the paper referred to be in the form of a will or codicil, or of a deed or indenture, or of a mere list or memorandum, the paper referred to, if it.was in existence at the time of the execution of the will, and is identified by clear and satisfactory proof as. the> paper referred to therein, takes effect as part of the will, and should he admitted to probate as such.” While there are some discrepancies in the application of the principle to the facts of the different cases, this statement is in accord with the great weight of authority here and in other jurisdictions in this country and in England, where the subject has been very much considered. Siler v. Dorsett, 108 N. C., 300; Bailey v. Bailey, 52 N. C., 44; Chambers v. McDaniel, 28 N. C., 226; Bullock v. Bullock, 17 N. C., 307; Thayer v. Wellington, 9 Allen, 85 Am. Dec., 761; Allen v. Maddox, 11 Moore P. C., 427; 14 English Rep. Reprint, 757; Smart v. Prujean, 6 Ves. Chan., 559; 1 Redfield on Wills, p. 262; Theobald on Wills, p. 50; 1 Jarman on Wills, 5 Am. Ed., p. 265. And the position, we think, should undoubtedly prevail in the present instance. In the opening clause of Exhibit B the declaration is, “I do make and declare this addition to my last will and testament,. and this addition is in no wise to interfere with my former will,” and, in the closing paragraph the language is: “I hereby appoint my beloved wife, Laura Hinson Watson, my lawful executrix of this addition to my former will, according to the true meaning of the same, but-in no wise is my former will to be affected by this addition, but the said will to stand as at present intended.” At the time of the execution of this later instrument, *81tbe testator said to one of tbe subscribing witnesses “that be bad made a will already and did not want tbis in any way to interfere witb bis former will,” and tbe witness D. E. Hinson testified that Mr. Watson brought tbe first paper, Exbibit A, to bim “all fixed,” bad tbe witness to erase tbe name of M. T. Hinson as executor, be having died, and' insert tbe name of D. D. Hinson, and asked witness to put it away and keep it for bim, as tbe witness bad a more secure place for tbe purpose, and tbis tbe witness did, and banded same to propounder’s attorney after Mr. Watson’s death. Tbe second instrument clearly refers to tbe one former will, and there is no evidence or suggestion that any other will bad ever been made or prepared by or for tbe testator. On tbis record, therefore, we are of opinion that tbe references in Exhibit B to tbe extrinsic paper are sufficiently clear and definite to permit that parol or other proper proof should be received as to tbe identity of Exhibit A, and that a perusal of tbe second will and tbe facts in evidence dehors afford testimony of a kind and character to require that tbe question as to such identity should be determined by tbe jury. If, on a second bearing, this Exhibit A should not be declared a valid will, as an original and separate proposition, then, on tbe first or some issue properly responsive, tbe question should be decided whether Exhibit A is tbe “former will” referred to in Exhibit B, and has its identity been established by clear and satisfactory proof. In tbe appeal of W. J. Bryan, 77 Conn., 240, reported witb an elaborate note in 68 L. R. A., p. 353, to which we were referred by counsel, while tbe principle of “incorporation by reference” is stated in somewhat more exacting terms than in some of the other decisions, tbe doctrine is recognized as tbe basis of tbe Court’s ruling. In that appeal tbe sum of $50,000 was given “in trust for the purposes set forth in a sealed letter which will be found witb tbis will,” and Chief Justice Torrance, delivering tbe opinion, thus states tbe ratio decidendi of the case: “There is not in tbe language quoted, nor anywhere else in tbe will, any clear, explicit, unambiguous reference to any specific document as one existing and known to the testator at tbe time bis will was executed. *82Any sealed letter or any number of them, setting forth the purposes of the trust, made by anybody at any time after the will was executed, and ‘found with the will,’ would each fully and accurately answer the reference; and if we assume that the reference calls for a letter from the testator, it -is answered by such a letter or letters made at any time after the will was drawn. The reference is ‘so vague as to be incapable of being applied to any instrument in particular’ as a document existing at the time of the execution of the will.” And like statement will serve to distinguish a recent case in our own reports of Freeman v. Shields, 158 N. C., 123.
All the authorities agree that, in order to a proper application of the principle, the paper referred to should be in existence at the time the valid instrument is executed; to hold otherwise would be to repeal the statutory requirements as to valid execution of written wills. An extrinsic paper could not be incorporated with the proper formalities unless it then existed. But the decision in any aspect should not control or affect the disposition of the present appeal when it appears, as heretofore stated, that the second and valid will makes clear and distinct reference to “my former will,” directing further that said will is to stand as first intended, and with'no evidence or suggestion that there had ever been more than one such former will or any other paper of that character. The case of Bryan v. Bigelow, 77 Conn., p. 604, reported with a full note in 107 Am. St. Reports, p. 64, simply holds that “the letter” referred to, having been properly rejected as a constituent part of the will, was not efficient as a declaration of trust, and has no bearing on the question presented on this record.
We are not inadvertent to the fact stated by the witness, that at the request of the testator, and just before he put away the first will, he erased the name of M. T. Hinson, who had been designated as executor, and inserted that of D. D. Hinson. Under our statute and decisions construing laws of similar import, if M. T. Hinson had been alive at the time, the effect of the erasure would or might have amounted to a partial revocation of the will, to wit, as to the designation of the first named executor, but the insertion of the name of X). D. Hinson, *83not baying been properly made or witnessed, would be inoperative. Tbe first named executor, being dead at tbe time, tbe erasure of tbe one name and insertion of tbe other is without any effect on tbe instrument; tbe result being that, if Exhibit A is properly established, tbe testator would have died testate' as to tbe property disposed of therein, but without naming an executor. Revisal, sec. 3115; In re Shelton’s Will, 143 N. C., 218; Bigelow v. Gillette, 123 Mass., 102.
For tbe error heretofore indicated, tbe propounders are entitled to a new trial of tbe cause, and it is so ordered.
New trial.