Bleckley, Chief Justice.
There is nothing to distinguish this, case from Duffie v. Corridon, 40 Ga. 122, except that in the execution and attestation of this will there was but one transaction, the witnesses all subscribing the unsigned will in the presence of the testator, and he, at the same time and place and immediately after they affixed their signatures, signing the document in their presence. In Duffie v. Corridon there were two interviews, at the first of which two of the witnesses (together -with another who was not afterwards present) subscribed, and at the second, the testator and the third witness. But is this difference in the facts of the two cases material ? The *381doctrine distinctly held by the court in ruling Duffie v. Corridon is, that until the testator signs, there is nothing to attest, the signature of the testator being the principal if not the only matter to which the attestation contemplated by law applies. It is obvious that if this be the true reason why the witnesses cannot subscribe their names until after the testator has signed his, it is of no consequence when the form of attesting an unsigned will is gone through with, whether on the same occasion of the testator’s added signature or on a previous occasion. In either case the attesting act would be performed when there was no signature in existence to be attested, and therefore no subject-matter to which the act could apply. To witness a future event is equally impossible, whether it occur the next moment or the next week.
We rule the present case on the authority of the prior one above cited, being satisfied, after careful examination, that to abide by the principle of that decision we must regard the order of time in which the respective signatures occur, rather than- the interval of time by which they are separated. The manifest teaching of Duffie v. Corridon is that the testator must sign first. That teaching is not followed but directly violated when the witnesses sign first. Judgment affirmed.
Note. — Where signing by one or more of the witnesses precedes that of the testator, and is on a different occasion, the will is not validly attested: Reed v. Watson, 27 Ind. 443 ; Hindmarsh v. Charlton, 8 H. L. Cas. 160. But when all (witnesses and testator) sign on the same occasion, authorities differ. The will is held not good in Olding’s case, 2 Curt. 865 ; Byrd’s case, 3 Id. 117 ; Cooper v. Bockett, 3 Id. 648 ; Shaw v. Neville, 1 Jur. (N. S.) 408 ; Ragland v. Huntington, 1 Ired. L. 563 ; Jackson v. Jackson, 39 N. Y. 153 ; Sisters of Charity v. Kelly, 67 N. Y. 409 ; McMulkin’s case, 6 Dem. (N. Y.) 347 ; and see Pearson v. Pearson, L. R. 2 *382Prob. & D. 451 ; Fischer v. Popham, 3 Id. 246. In the following cases the attestation had an additional element of invalidity, namely as not being done in testator’s presence: Cox’s will, 1 Jones, 321 ; Boldry v. Parris, 2 Cush. 433 ; Chase v. Kittredge, 11 Allen, 49 (cases reviewed). The will is held good in Obrien v. Galagher, 25 Conn. 229, and in Miller v. McNeill, 35 Pa. St. 217. But in connection with this latter case, it must be observed that the Penn, law does not require subscribing witnesses. Statute quoted and cases cited in Frew v. Clarke, 80 Pa. St. 178, 179 ; 2 Brightly’s Purdon’s Dig. (ed. 1885), p. 1710, note (h). In Swift v. Wyley, 1 B Mon. 114, the decision rests partly upon acknowledgment of signatures by the witnesses It is miscited in Chisholm v. Ben, 7 B. Mon. 410, but followed in Sechrest v. Edwards, 4 Met. (Ky.) 163 The case of Vaughn v. Burford, 3 Bradf. (N. Y.) 78, is overruled by the N. Y. cases cited supra. Rosser v. Franklin, 6 Grat. 1, rests partly on the fact that the signature of the testator was by his request affixed before attestation. In Virginia too, witness may acknowledge signature : Sturdivant v. Birchett, 10 Grat. 67. And see Pollock v. Glassell, 2 Grat. 439. In general see 1 Jarman on Wills, *110 ; Schouler on Wills, §328 ; Beach on Wills, §43.