Opinion by
Mr. Justice Moschzisker,Counsel for the appellant states the question involved to be, “whether a testamentary paper is signed at the end thereof so as to constitute a valid will under the Act of April 8, 1833, P. L. 249, when it contains a disposing clause after the signature, which additional clause is not signed, and there is no evidence to'show when the additional clause was written?” The answer depends upon the rule of presumption to be applied.
The court below, in its opinion disposing of the case, said: “It appears from the testimony of two competent witnesses that the signature to the paper offered for probate is Mrs. Taylor’s, and that it is in her handwriting; but there is no oral evidence as to whether the testamentary disposition following her signature was written after or before she signed the paper. ...” The court ruled, “It must be presumed that the clause was written after she had signed her name. ... So much of the writing as precedes the signature of Mrs. Taylor should be admitted to probate as her will.”
The appellant contends that, in the absence of evidence, the presumption is that the words following the signature were written prior to the execution of the will, and that the court erred in deciding otherwise; in support of this contention he relies upon Hays v. Harden, 6 Pa. 409, which is the first case in our reports upon this subject. The syllabus correctly states the purport of that decision: “A testamentary paper, to constitute a will under the act of 1833, must be signed at the end thereof; hence a paper which contains, after the signature of testator, a clause stating his reason for making the devise, which clause was not signed at the end thereof, is not a valid will.” A careful consideration of that case and of our *349subsequent decisions fails to convince us of the correctness of the appellant’s position. Although the report in the Hays case neglects to state whether or not there was any evidence to show when the subsequent clause was written, Mr. Chief Justice Gibson says in his opinion: "It is evident that the testator considered the whole paper one will.”
When the subject was again before this court in Heise v. Heise, 31 Pa. 246, we decided that the addition of an unexecuted codicil would not have the effect of revoking a perfect will, and in referring to the Hays case we said: “If it be true, that a different interpretation has been given to the statute, and that interpretation has become a rule of property, we should not feel at liberty to assert any other. . . . We do not, however, understand that such is the fact. Hays v. Harden, 6 Pa. 409, is the first case in which the question was mooted. A reference to the facts of that case will reveal that the decision had no reference to the effect of an imperfectly executed codicil or writing, made after the completion and publication of a perfect will; such a will is protected from destruction by the statute except in certain defined modes. The alleged will in that case, commencing with a recital, contained a clause appointing executors and directing the payment of his debts; then anothei devising, in the following words, ‘All my leasehold estate to Abraham Hays. Witness my hand and seal this 26th day of September, 1844. John Hays, [sbal]’ This was immediately followed by another testamentary disposition, concluding, ‘Signed, Sealed, and Delivered by the above named John Hays to be his last will and testament, in the presence of us, who at his request and in his presence have subscribed as wit-nesess thereto.’ Then followed the attestation of two witnesses, who proved their signatures and that of the testator. All this took place at one time. There never was a perfect instrument made and published by John Hays as such. It was not a case of subsequent repeal, but of original defective execution.” This shows that as far *350back as 1858 we understood it to have been a fixed fact in the Hays case that the writing of the entire paper “took place at one time,”' and there is nothing in the report of that case or any of the subsequent cases to indicate that our decision rested upon any rule of presumption.
In Wineland’s Appeal, 118 Pa. 37, there was evidence tó show that the added clause was written before the will was signed, and the will was declared invalid; while in Saunders v. Samarreg Co., 205 Pa. 632, there was evidence to show that the addition was made after the signing of the will, and the will was held to be valid. In Beaumont’s Estate, 216 Pa. 350, the matter after’ the signature was not testamentary in character, and following Wikoff’s Appeal, 15 Pa. 281, we decided that it did not invalidate the will. The last case on the subject is Teed’s Estate, 225 Pa. 633, where the will was written upon the fourth page of a sheet of letter paper and signed at the foot of that page. At the top of the third page there was an unsigned and unattested clause appointing an executor. The court below treated the addition as an unexecuted codicil, and sustained the probate of the will. In affirming, we said: “In the absence of evidence the presumption would be.'that such an addition to the will as that here shown, was made after the signature of the testatrix, and that of the subscribing witnesses had been put on the paper. The general rule is thus stated in 1 Underhill on Wills (1900), sec. 268, where it is said: 'Unattested alterations in a will, though proved to have been made by the testator, in the absence of any evidence showing when they were made, will be presumed to have been made after execution, and consequently, unless properly attested, will not operate as a partial revocation of the will.’ And again in Page on Wills (1901), sec. 432, it is said: 'An alteration in a will is, as a general rule, presumed, in the absence of evidence, to have been made by the testator after the execution of the will. This rule is different from that often said to obtain in the case of- other written in*351struments for this reason: In the case of other instruments it is a civil wrong, if not a crime, to alter a written instrument. In the case of wills the testator may alter the will as much as he pleases, without wronging anyone.’ ”
The appellant argues that the determination of the point under discussion was not necessary to the Teed case, and all that we have quoted should be taken as dicta. Be this as it may, we are convinced of the soundness of the view there expressed and see no reason to depart from it. When a testamentary writing is found beneath the signature of a testator, the probability is that such writing was not there at the time the will was executed; for reason and general practice assure us that if it had been part of the will, the testator would have signed beneath it and not above it; so, in the absence of evidence, it is proper to presume that the writing was added after the execution of the will.
The assignments of error are all overruled and the decree of the orphans’ court is affirmed.
Mr. Justice Mestbjezat dissents.