The opinion of the Court was delivered by

Wills in writing must be proved by two witnesses. This instrument, purporting to be the last will and testament of Jacob Lewis, was proved by three subscribing witnesses; one of whom, and his character, was unimpeachcd, and his testimony unco'ntradicted, proved that the will was dictated by the testator, put in writing in form by his direction, read to hint paragraph by paragraph, and approved of by him, as being conformable entirely to his directions. This was the requisite proof. The execution was *495legally proved ; the instrument duly authenticated. The plaintiff made out in the first instance.his proof on the issue of devisavit vel non; the onus of disproving it; proving that a will, the contents of which were unknown to the testator, was imposed on him, and he by practice induced to sign a will containing dispositions of which he was ignorant, lay on the defendant ; for there was no evidence tending to she w that this was not the real testamentary disposition of the testator, unless the want of a sound mind when he signed the will, was made out in proof. The Court confounded the legality of the execution, with the question of sanity, the imputed fraud, with the authentication of the writing. The authentication of the instru-" ment by the.requisite number of persons, was a mere abstract question of law to be decided by the Court; the sanity, and all questions of fraud by the jury. 1 Sm. L. 40. But here the question of requisite proof was entirely submitted to them; and before they were to inquire into these facts, they were informed by the Court, they must be satisfied they had the requisite proof of the execution of the will; and taking the whole charge together, it is manifest that the Court was of opinion, that it was equally incumbent on a plaintiff to prove knowledge of the contents by the testator, by the same medium of affirmative testimony, and the same number of witnesses as the execution of the will. In this the Court erred. If further proof, were required than was given in this case, few wills could be proved. Of all instruments executed by men, a will is that of which the party is most desirous of keeping the contents secret. In dispositions which are to take effect in a man’s life-time, secrecy is seldom desired ; but few men make known their dispositions by will. Where a man does not draw his own will, the friend or scrivener whom he employs is the only depository, generally, of its contents. An inquiry by a subscribing witness of the contents of a writing, or of the grantor, if he had knowledge of what he was going to sign, would be considered as most impertinent curiosity; in the case of a will most offensive. It would be something like the separate examination of a feme covert, as to her knowledge of the content^ of a deed she was about to acknowledge. Formerly as the law stood, the interest of a witness was affected by his attestation from his presumed knowledge of the contents; but this forced presumption, contrary in *496most instances to the truth, no longer prevails; because a witness is only called to authenticate the instrument, and is not presumed to be, as in practice he is not, privy to the contents. Welford v. Bezely, 1 Ves. 6. Powell on Mort. 186. Under the statute of frauds, if a testator acknowledges the signature to be his hand-writing, it is sufficient, though the subscribing witnesses did not see the testator sign. 2 P. Wms. 253. 2 Ves.4t54<. Where a will is not signed by the testator, and is not of his own hand-writing, there must be proof by two witnesses that he knew the contents; but where a man signs any instrument, deed, or will, to which there are subscribing witnesses, proof by them is the primary proof of the execution, and if that can be had no other can be resorted to. There cannot be a doubt, but if this were an instrument by deed or other written engagement, that the mere signature of the party, though illiterate, or even blind, proved by the subscribing witnesses, would be deemed as sufficient execution, and the only thing to be proved would be, that the person was not imposed on. In this respect wills do not differ from other instruments, except in requiring two witnesses to authenticate them. In the present case, the fact was completely established by one witness, who took the instructions from the mouth of the testator, and read the will over after they were reduced to form, paragraph by paragraph to the testator, who said, “it is just as I told you yesterday.” This evidence would be sufficient, though but by one witness; the execution of the will being proved by two, one witness is sufficient to rebut the imputation of fraud. The possibility of practice on the testator by imposing on him a will, of the contents of which he was ignorant, might exist in the case of an illiterate or blind man, but it would be removed by the testimony of one witness as well as of one hundred, if he was believed, that the will was truly read to him. An argument of a most curious kind at one time produced a verdict against a will. One of the subscribing witnesses swore that an alteration was made without the testator’s knowledge. It was insisted on, that if the jury believed him, it was not the testator’s will, and if they did not believe him, he was not a credible witness, and therefore there were not three credible witnesses to the will; so that whether the witness was believed or disbelieved, stiff his evidence destroyed the will. But *497this mode of reasoning was considered most fallacious, and a mere trap to catch the jury. 1 Br. Civ. is1 Adm. Law, 285. If a will of land is lost or secreted, parol proof of its contents by one witness would be sufficient, the proof of the execution being made by the requisite number. So on the other hand, parol evidence may be admitted, and that by one witness, that one will was substituted in the room of another, in order to set it aside on the ground of fraud. The execution of a will might be made out by proof in the most formal manner; yet one witness might establish a fraud, that would overthrow all this; so one witness to rebut a fraud, might prevail against the proof by three or any other number, because proof of the existence or non-existence of the imputed fraud, in obtaining the will, depends not on the number qf the witnesses, but on. their credit. The Court here should have decided the question of law as to the legal execution of the will, and have instructed the j ury, that it had the requisite proof; was legally authenticated by the requisite number of witnesses; and if there was any evidence to impeach it on the ground of practice, or want of capacity in the testator at the time of the execution, have submitted these facts to the determination of the jury; whereas, they were instructed first to satisfy themselves of the requisite proof of the execution by the prescribed number of witnesses, including a knowledge of the contents by the testator, before they entered on the inquiry, as to the imputed fraud, or incapacity. In this the Court fell into error, and judgment is reversed, and a venire facias de novo awarded.

Judgment reversed, and-a venire facias de novo awarded.