The opinion of the court was delivered by
Mason, J.:Probate of the will of Peter Kirby was refused by the probate court, but granted by the district court on appeal, and this proceeding is brought to review that order. The will was written by his oldest child, Mary Kirby Sellards, who was the executrix and a principal beneficiary, on five sheets of paper, which were separate at the time of its execution, but fastened together with a pin when offered for probate, the second and fourth pages being interchanged. The signature of the testator and the witnesses appeared only on the last sheet, and the only direct testimony as to the identity of the others was that of Mrs. Sellards. The opponent of the will contends that the proof in this respect failed by reason of the incompetence of the witness, and also that a presumption of undue influence arose from the relation of the draftsman to the testator.
*293The statute (Gen. Stat. 1868, eh. 117, § 11; Gen. Stat. 1909, § 9786) provides that “if a devise or bequest be given to a person who is a witness to the will, and the will can not otherwise be proved than by, the testimony of such witness, the devise or bequest shall be void, and the witness shall be competent to give testimony of the execution of the will in like manner as if such devise or bequest had not been made.” Clearly, however, the words “witness to the will” refer to an attesting witness. This accords with the purpose and history of the legislation. That the statute distinguishes between a witness to the will and one who testifies in its support when it is offered for probate is made apparent by the original language of the succeeding section, providing that “the court shall cause the witnesses to such will, and such other witnesses or any person interested in having the same admitted to probate as may desire, to come before such court.” (Gen. Stat. 1868, ch. 117, § 12; Gen. Stat. 1901, § 7948.) Mrs. Sellards was not a witness to the will in such a sense that her competency was affected by her interest. Moreover, her testimony was not necessary to identify the unsigned sheets. As the first four pages happened to end with incomplete sentences, the connection of the subject matter made as effective a union of the several parts as could have been accomplished by their physical attachment. “It is a rudimental principle that a will may be made on distinct papers. ... It is -sufficient that they are connected by their internal sense, by coherence or adaptation of parts.” (Wikoff’s Appeal, 15 Pa. St. 281, 290. See, also, 30 A. & E. Encycl. of L. 580.) The sequence of the pages was indicated by figures placed at the top of each, as well as by the connection of the language, and the fact that they were fastened together in other than the proper order is not important.
A further contention is that the fact that the will was written by the testator’s daughter, who was one of *294the principal beneficiaries, created a presumption of undue influence, which was not rebutted by any evidence. There is some apparent conflict in the authorities as to the effect to be given to the circumstance that a will is drafted by a legatee. In section 137 of Underhill on the Law of Wills it is said:
“Many cases seem to hold that the circumstance that a party draws a will, . . . under which he takes a legacy, . . . raises a presumption of undue influence exerted by him which must be rebutted by clear and satisfactory evidence. . . . But the safer and more correct statement of the rule is that such a condition of affairs creates no presumption, but merely raises a suspicion which ought to appeal to the vigilance of the court. . It is a fact to be considered with other facts. It is undoubtedly a suspicious fact, but its weight depends, not solely upon its character, but upon the facts and circumstances with which it is connected. In some cases it would have no weight at all. Thus, if it appear that the testator had testamentary capacity, that he dictated his will and knew its contents at the date of its execution, and that it was executed in the statutory manner, the mere fact that the will was written by the sole beneficiary would not be enough, unless coupled with other extremely suspicious facts, to overthrow it, or, taken alone, to cast the slightest suspicion upon it.”
What justly creates suspicion is not that the draftsman is a legatee, or even a sole legatee, but that he receives an unreasonable or unnatural benefit. This idea is thus expressed in section 245 of the third edition of Schouler on Wills:
, “The circumstance that a party who derives under ■the will a disproportionate benefit or a benefit to which he had no natural claim is the party who drew it lends disfavor to the instrument, and may turn the scale against its admission to probate. . . . Our later cases appear to rule that wherever the testator’s draftsman or manager of the execution may be thought worthy of some generous token, undue influence and fraud are not to be presumed from the fact that the will gives him a legacy or executorship accordingly. *295The extent of his benefit as compared with that of natural objects of one’s bounty is a matter of some consequence.”
So, also, the suspicion that attaches to a will written by a beneficiary is heightened where he sustains a confidential relation toward the testator. That condition is sometimes said to create an actual presumption against the will. Thus Mr. Underhill says:
“Where it appears from the evidence that the testator and the beneficiary stood in confidential relations toward one another, and it also appears that the legatee . . . was the draughtsman of the will, . . . circumstances are shown from which a presumption of undue influence or fraud arises, which the proponent of the will has the burden of proof to overthrow, and to show that the will was the free and voluntary act of the testator.” (1 Underhill, Law of Wills, § 145.)
On the other hand a portion of a note in the same work reads:
“The fact that the draughtsman of a will, who is a legatee at the same time, stands in a confidential relation to the testator, does not, in the absence of affirmative proof of fraud or coercion, raise a presumption against the voluntary character of the will.” (1 Underhill, Law of Wills, note 3 to § 137.)
And the rule is thus stated in volume 29 of the American and English Encyclopedia of Law, at page 124: '
“Undue influence is not generally presumed in the case of a legacy or devise made by a client in favor of his attorney from the mere relation itself, or from the circumstance that the will was drawn by the attorney; at least where the legacy or devise is not disproportionately large.”
Perhaps an unnecesary difficulty is created by an effort to say at just what point the union of a number of suspicious circumstances, no one of which is enough in itself to defeat probate, shall be deemed to give rise to an actual presumption that a will was the result of *296undue influence. The real question in each case is. whether all the circumstances so far as shown are such as to lead the court to believe that in fact the will does, not actually express the voluntary purpose of the testator. The substance of Kirby’s will was that his property was to go to his wife for her life, and then, except for a specific legacy to each of two granddaughters, to. his three children. There is nothing in this disposition to suggest any overreaching on the part of Mrs. Sellards. Nor was she shown to have occupied a position, of confidence and trust toward her father beyond that arising from their, relationship. If it had appeared that she had been intrusted with the general management of his business, or that he was in the habit of deferring to her judgment in his affairs, a different question would be presented. We think that her writing the will created no presumption of undue influence. No question of incapacity or restraint appears to have-been raised at the hearing. The controversy' turned chiefly upon whether there had been a revocation. But in seeking to counteract evidence in support of the will its opponent introduced affidavits of the attesting witnesses, which had been read in the probate court, stating among other things that when Kirby signed the will he was of sound mind and memory and not under any restraint.
A recent statute makes this provision:
“In all actions to contest a will, if it shall appear that such will was written or prepared by the sole or principal beneficiary in such will, who, at the time of writing or preparing the same, was the confidential agent or legal adviser of the testator, or who occupied at the time any other position of confidence or trust to such testator, such will shall not be held to be valid unless it shall be affirmatively shown that the testator had read or knew the contents of such will, and had independent advice with reference thereto.” (Laws 1907, ch. 430. § 1; Gen. Stat. 1909, § 9787.)
This rule by its terms relates to actions to contest a *297will, but even if extended by interpretation to cover proceedings for its admission to probate the present, case would not be affected. The requirement that the testator must be shown to have had independent advice-clearly implies that in order to be deemed to occupy a position of confidence or trust to the testator the draftsman must be someone to whom he naturally looks, for counsel. That situation does not follow from the mere relationship of daughter and father.
The fact being established that the will had been duly executed, a more difficult question is whether the-, evidence showed that it had been revoked. The testator, in the presence of the subscribing witnesses, signed it with pen and ink below the attestation clause,, and not elsewhere. He kept it in his own possession for a few weeks and then delivered it to the executrix,, who retained it until after his death. When she offered it for probate his ink signature had been partly obliterated by means of knife scratches, and his name had been written in pencil j ust above the attestation clause, either by himself or by someone who closely imitated his handwriting — a matter concerning which the opinions of witnesses differed somewhat. The law permits a will to be revoked “by the testator tearing, canceling, obliterating or destroying the same with the intention of revoking it, by the testator himself, or by some-person in his presence or by his direction.” (Gen. Stat. 1868, ch. 117, § 87; Gen. Stat-. 1909, § 9814.) In order to accomplish revocation under this provision it is necessary that the testator, with an intention to revoke, cause some physical defacement of the will, adapted to give expression to that purpose. If the other conditions were proved there would be no difficulty in saying that here a sufficient actual defacement was shown, for while the signature was not rendered wholly illegible the partial erasure of it was of such a character as to suggest a purpose to discredit it. Under the English act requiring “burning, tearing or other*298wise destroying,” a mere crossing out of the signature, leaving it still legible, is held not to be enough (1 Williams, Executors, 10th ed., pp. 101, 102; 1 Jarman, Wills, 6th ed., p.*116), but the rule is otherwise where the statute includes the word “canceling.” (Estate of Olmsted, 122 Cal. 224; Matter of Hopkins, 172 N. Y. 360. See, also, Woodfill et al. v. Patton et al., 76 Ind. 575.) Except as already stated, the evidence in this case affords no light as to when, with what purpose or by whom the changes were made. The general rule is that where a mutilated will is found among the testator’s effects the presumption arises .that the mutilation was his own act, done with a revoking purpose. (1 Underhill, Law of Wills, §§ 231, 232; 1 Redfield, Law of Wills, 3d ed., p. *307; 30 A. & E. Encycl. of L. 635; 28 Am. St. Rep. 351, note.) Here the will had been out of the possession of the testator for about a year at the time of his death, but as the proponent must be deemed to assert that it was not changed while in her custody the situation is doubtless the same as though it had never left the testator’s hands. The partial erasure of the signature is not so unequivocally a cancellation as the running of a pen through it, but would doubtless be sufficient to warrant the application of the rule referred to if it were not for the penciled addition. We must regard this as a genuine signature, for the trial court evidently so treated it, and the evidence supports that view. Doubtless, if the testator erased his first signature intending an unconditional revocation thereby, a subsequent unattested signing could not revive the will. (30 A. & E. Encycl. of L. 657.) But the condition of the instrument affords no presumption that such was the case. In In the Goods of William King, deceased, 2 Rob. Ecc. (Eng.) 403, a will was admitted to probate under these circumstances :
“After the death of the testator the will was found with his signature, as at first written, opposite to the center of the attestation clause, erased; but subse*299quently to that erasure the testator’s signature was written a short distance beneath the place where the original signature stood.” (Page 403.)
The only evidence regarding the change was that it was made after the will was executed and witnessed, and before it was found after the testator’s death. The ■opinion reads: .
“The original signature of the deceased has been erased, but by whom and with what motive it is not •easy to determine. It is manifest, however, from the facts and circumstances deposed to, that the erasure was not made by the testator animo revocandi, as required by the wills act. On that ground, therefore, I decide this case: in the probate the original signature must be restored, and the second signature omitted.” (Page 404.)
Citing that case, in section 392 of the third edition of 'Schouler on Wills the author says:
“If a will should show the testator’s signature struck through with a pen and another signature written and left, the natural presumption would be that the original erasure was not made with the intention to revoke at all, but was connected in some way with the final execution by the signature substituted.”'
In In re Wood’s Will, 11 N. Y. Supp. 157, a will was admitted to probate which had been found in the testatrix’s safe, with the signature erased first by drawing lines over it, and then by nearly erasing such lines and the original signature, but with her name rewritten ’by her over the erasure. It was not shown when or why the change was made. In the opinion it was said:
“The proponents having proved the due execution of the will, it is entitled to probate, unless the contestants qjrove its revocation by some one of the modes pointed •out by the statute. ... If the will had been found in her safe, carefully preserved among the valuable papers of the testatrix, with her signature erased, it would have been a fair and reasonable presumption that she erased the signature animo revocandi; and it would then be lacking in one of the statutory requirements of a valid will — the signature of the decedent at *300the end thereof. But when found with the signature carefully restored, no such presumption arises. In the absence of all proof, how can I find that it was made with the intent to revoke, when the instrument was preserved by her with her signature carefully restored? An intention to revoke a will not fully consummated is no revocation. ... It may be that Mrs. Wood drew the lines through her name with the intention of revoking the will, but immediately and before the act was completed, changed her mind, erased the marks, and restored her signature. To sustain the theory of the learned counsel for the contestants, I must find that the erasure was made by the testatrix herself, understanding^, freely, and voluntarily, with no other purpose than to destroy- her will, and that it was done at some time previous to the act of rewriting her name, and this finding is asked for in the absence of proof and with the burden resting upon the contestants to establish the fact of revocation.” (Page 158.)
In the present case the one fact concerning the testator’s intention that is definitely established is that at one time he wished to dispose of his property according to the terms of the will and complied with every requirement of the law to give effect to that desire. What his subsequent purposes were is a matter of speculation. That a few weeks later he delivered the will to the testatrix has some tendency to show that he then regarded it as a valid instrument, although he must already have made the change if he made it at all. In that event, as reasonable a supposition as any is that when he executed the will he failed to notice the space left for his signature above the attestation clause, and therefore inadvertently signed below; that afterward he discovered the inadvertence, and in order to correct it attempted to erase the original signature and affixed the other in the more natural place, intending, not revocation, but ratification. Of course a signing below the attestation clause was sufficient (17 L. R. A., n. s., 354, note), but the testator, on the matter being drawn to his attention, may easily have supposed otherwise. Upon this hypothesis it is not necessary to re*301sort to the principle of “dependent relative revocation” to save the will, for- the mutilation was without effect unless accompanied by a purpose to revoke. Another way of accounting for the condition of the will seems not unreasonable — the testator may have started to erase his signature, changed his mind before completing the erasure, and then rewritten his name, not as a new execution, but as evidence that he did not intend revocation.
The judgment is affirmed.