Whiteield, J.,
delivered the opinion of the court.
We do not think the court erred in refusing the peremptory instruction asked by the contestants, the appellants here; but it was clearly erroneous to have given the peremptory instruction asked by appellees, the proponents of the will, upon the testimony in this record. This was peculiarly a case for the jury, under proper instructions from the court, as to the law. A verdict for either party, on this testimony, should not have been disturbed by the court, under proper instructions as to the law. And, as the case must go back for a new trial, we refrain from the expression of any opinion on the facts of the case. Tacts are for the jury, the triers of issues of fact. But there are certain errors of law assigned which we will proceed to notice, that the case may, on the new trial, conform to the views herein expressed.
The contestants insist, in the first place, that the alleged will of Cornelius Sheehan was never, in fact, signed by him at all; that there is no signature of his, of any kind, to the will; that that which is called such is an illegible scrawl, no better that a [690]*690straight line would be. They further insist, along this line of contention, that section 1520, of the Code of 1892, qualifies section 4488 of the same code, so as to require a testator to sign his name in his “proper handwriting,” and that, hence, no one could sign his name for him, at his request, and that, if he be able to write, his “mark” will not do. This contention is not maintainable. Section 1520 of the code is found in the Chapter on “Definitions” and it means, manifestly, nothing more by the phrase, “shall always be in the proper handwriting of -such person,” than that the “signature” shall be written, as contradistinguished from engraved or lithographed or printed, in case the party be able to write; and, if he be not able to write, that it is meant that his “signature” in the cases provided by the statute, shall be made by “his proper mark,” and not “printed, engraved, or lithographed.” This statute does not require the signature to a will to be written out. A mark may still be adopted as a signature. The language in section 4488, “signed by the testator, . . . - or by some other person in his presence . . . and by his express direction,” demonstrates that the “signature” meant in section 4488 is not, necessarily, 'one written at all by the testator himself, and discriminates this case from Ferguson v. Board, 71 Miss., 524, 14 South., 81. It is this section which defines what is a proper signing of a will. The original of section 1520 is found in subdivision 19, sec. 4971, of the Devised Statutes of Wisconsin of 1878, and Justice Cassidy of the supreme court of Wisconsin, in his learned treatise on the Law of Wills, discussing this statute (secs. 63 and 118), says: “But a testator may properly sign by making his mark, independently of this statute, and, if he so sign, it may be unnecessary to prove that he could not write. If he could only sign by virtue of this statute, it might be necessary to prove, in the language of the statute, that he was unable to write. It follows that the testator may sign by his mark, as well as by writing his name — either his signature, initials, or a [691]*691fictitious name.” Any signature or mark signed by tbe testator, or by another in bis presence and at bis express direction, to tbe will, as and for bis completed signature, and acknowledged and adopted by him as such at tbe time, in tbe presence of tbe subscribing witnesses, is a sufficient signing within tbe meaning of Code 1892, § 4488. It was said in Re Plate's Estate, 148 Pa. St., 55, 23 Atl. 1038: “exactly wbat constitutes a signing has never been reduced to a judicial formula. Tbe principle upon which these cases proceeded was that whatever tbe testator .... was shown to have intended as bis signature was a valid signing, no matter bow imperfect, or unfinished, or fantastical, or illegible, or even false tbe separate characters or symbols be used might be, when critically judged.” And we think this clearly correct. It is not necessary that be should write out bis complete signature. It is enough that wbat be writes is intended for and adopted as bis complete signature. But it must be carefully noted that, in all that we have thus far said, we have been speaking only of tbe mere physical act of signing. One without testamentary capacity may write bis signature to an alleged will. Tbe testator must have testamentary capacity, and tbe act of signing must be in exercise of that testamentary capacity. Again, one may have testamentary capacity, and yet be unduly influenced to sign. Tbe testator’s signature must be made in tbe exercise of testamentary capacity, and in tbe absence of undue influence. Both these grounds tbe want of testamentary capacity, and tbe existence of undue influence — are earnestly insisted on by tbe contestants. And we wish to be understood as intimating no opinion as to tbe effect of tbe evidence on either issue, but merely and only as defining wbat signing is sufficient, in law, under section 4488. It is only necessary to add, in this connection, that tbe mere steadying, by tbe draughtsman of tbe will, of tbe band of a testator competent to make a will, and free from undue influence, is immaterial, provided the testator consciously performs, [692]*692himself, the act of writing his signature. Watson v. Pipes, 32 Miss., 466.
It is next earnestly insisted by appellants, with great wealth of learning and marked force of reasoning, that the declaration of testamentary intention testified to by Taft, Moran, and O’Elyn should not have been admitted. It is conceded that such declarations by a testator as to the disposition he intends to make of his property, before the making of the will, and his declarations, after he has made the will, of what he had intended to do, or has done, in the disposition by will of his property, are admissible, on the issue of testamentary capacity, to show the state of his mind, whether sound or unsound. Such declarations would seem to be declarations symptomatic of the condition of the mind, as exclamations forced by pain or disease from a suffering body are symptomatic of such pain and illustrative of its nature and cause. • Irrational exclamations may thus well be symptomatic in one case of a diseased mind, and exclamations of physical pain and anguish in the other case of a diseased body, and competent in both cases, for the same reason,' as primary evidence. 1 G-reenl. Ev., page 149, sec. 102, and note c.
But it is conceded that they should be admitted on that issue • — testamentary capacity — alone, and on the ground indicated alone, and that the instruction to the jury trying the issue devisavit vel non should restrict the jury to their consideration in this aspect alone, and that they are never competent to show undue influence in the procurement of the will, whether made before or after the will. To support this contention counsel cites many authorities, relying chiefly on 2 Whart. Ev., sec. 992. But this section only 'holds, properly understood, that the genuineness of the will conceded, and no issue of undue influence or testamentary capacity being involved, the “intent is to be drawn from the will, and not the will from the intent,” and that in such cases parol evidence is not competent to vax-y, [693]*693enlarge, or contradict the will, not in such case challenged as being the -testator’s will. This is clearly shown by the fact that Love v. Buchanan, 40 Miss., 758, and Gilliam v. Chancellor, 43 Miss., 437, are cited by Mr. Wharton to support the text. In these cases and in Magee v. McNeil, 41 Miss., 25, there was no question raised as to testamentary capacity or undue influence, but, it being admitted that the instrument was the will of the testator, the effort was to explain, alter, or contradict the terms of the will by parol; and it was very properly held that this could not be done. The effort in Love v. Buchanan was to show that the word “heirs” did not embrace one who was an heir; in Magee v. McNeil, to show that a will, conditional as written, was intended to be unconditional; in Gilliam v. Chancellor, the effort was to show that a sum of $5,000, secured by a marriage contract, was intended to be satisfied by a legacy of $5,000, and in Gilliam v. Brown, 43 Miss., 641, the effort was similarly to show that a legacy of $15,000 was intended to satisfy an indebtedness of the testator to his brother, the legatee. In the last two cases of satisfaction or ademption by the legacy, the court did not deem it necessary to a decision to decide the point as to the competency of parol testimony. And, to place beyond dispute what was meant by Mr. Wharton in the section cited by counsel, he expressly declares the law on the precise point under review to he (Section 1012, same Volume) as follows: “Hence, whenever a will is attached on the ground that it does not exhibit the testator’s real intent, he being in disturbed mind or under undue influence at the time it was executed, it is admissible to put in evidence his declarations in support of the will.” What is said by Mr. Greenleaf (2 Greenl. Ev., sec. 690), on this subject, to the effect that such declarations, before or after the will is made, are not even competent to show insanity, is shown by Seldon, J., in Waterman v. Whitney, 11 N. Y., 157, to be clearly unsound, as is pointed out by the accomplished editor [694]*694of the American Decisions in his learned note to Jackson v. Kniffen, 3 Am. Dec., at page 397. Learned counsel also relies strongly on the statement of Mr. Bigelow, in his note in 1 J arm. Wills (6th ed.), at page 71, to the following effect: “By the weight of authority, and in principle, the statements or declarations of the testator, whether made before or after or at the time of executing the supposed will, are not admissible in evidence to prove undue influence or fraud. . To receive evidence of that kind would be to receive heresay,” citing many authorities. But an examination of these authorities will show that they do not bear out Mr. Bigelow in his sweeping statement. One of the cases on which he mainly relies (Shailer v. Bumstead, 9Mass., 112) holds the direct contrary, as may be seen at page 120, 99 Mass. This case is clearly discriminated, and its true holding on this precise point is accurately stated in the note, above referred to, at page 397, of volume 3 of the American Decisions, where the annotator says: “In case of insanity or undue influence, the evidence takes a wider range, and here the subsequent as well as the prior declarations of the testator are admissible. This is fully borne out by Waterman v. Whitney, 11 N. Y., 157; Boylan v. Meeker, 28 N. J. Law, 274; Shailer v. Bumstead, 99 Mass., 112; McTaggart v. Thompson, 14 Pa. St. 149; Dennis v. Weeks, 51 Ga. 24. There is not in this instance the same objection to the evidence, as in the other cases, as being heresay. For this purpose it is direct and primary.” Now, Mr. Bigelow, distinctly concedes, in his note (page 71), that the declarations are not to be excluded because of the objection that the statute requires wills to be in writing, but solely because that they are heresay: “For the very question under consideration is whether there be a will, . . . the declarations would be admissible, if they were not heresay.” While the annotator of the third volume of American Decisions distinctly declares that, on principle and authority, they are not heresay on this issue, but “direct and primary evidence.” We [695]*695think this latter declaration the true rule, at least to the extent that the declarations are competent, is the sounder. It is curious that Mr. Bigelow, a most accurate law writer, should hare fallen into this misconception, when, in the very same note, in discussing the burden of proof on the issue of undue influence, he says at page 68: “The objection of undue influence may seem, at first, of the nature of a defense or confession and avoidance in common-law pleading. It virtually admits the testator’s competency, though not his entire soundness of body or mind, and the due execution of the instrument, but seeks to avoid the effect thereof. But, in reality, it denies that the supposed will is the will of the supposed testator. That is, it traverses; and, as it is for the one who offers the instrument for probate (an instrument under which the rights of widow, heirs, and next of kin, under the law, are disturbed) to prove it to be the true will of the alleged testator, it should follow, in principle, that the burden of proof, as well in regard to undue influence as in regard to capacity, should rest upon him”— a statement, the correctness of which we fully indorse and which seems logically to follow from another well settled principle, declared by Mr. Bigelow at page 66 of said note, towit: “It is true that, while the question of capacity must be kept distinct from; — that is, not confused with — the question of undue influence, the state of mind and body of the supposed testator at the time of executing the will may be, and commonly is, a material subject for consideration, in determining whether undue influence has been exercised upon him.” Now, if it is part of the prima facie case of the proponents of a will, in showing that it is the alleged testator’s will, to show prima facie that it is so for one reason, because he was not unduly influenced to make it; and if, as correctly stated by Mr. Bigelow,- the real inquiry on an issue of undue influence is whether the instrument be, in fact, the alleged testator’s will, then, clearly, the exclusion of the declarations must rest on the assumption that the [696]*696instrument is the testator’s will, the very point in issue. The very same reasoning which makes it improper to exclude such declarations as infringing the statute requiring a will to be in writing, because so to exclude them would assume that the instrument was the will of the alleged testator, equally makes it improper to exclude them on the ground of heresay, on an issue of undue influence, because such exclusion, in that view also, assumes that the instrument is the will of the alleged testator. The statement of the law on this point, in the note in the third volume of American Decisions, is therefore manifestly the sounder doctrine. And that it is, the following authorities, in addition to those cited in that volume, unanswerably demonstrate: 7 Am. & Eng. Enc. Law, p. 73; Collagan v. Burns, 47 Me., 449 (the best reasoned case we have seen); Roberts v. Trawick, 17 Ala., 55; Chappell v. Trent, 90 Va., 849, 19 S. E., 314; Cassody, Wills, secs. 316, 317; Boudinot v. Bradford, 2 Dall., 266; Durant v. Ashmore, 2 Rich. Law, 184; Steele v. Price, 5 B. Mon., 58, 66; Weeks v. McBeth, 14 Ala., 474; Lawyer v. Smith, 8 Mich., 411; Harring v. Allen, 25 Mich., 505; Patterson v. Hickey, 32 Ga., 156; Tynan v. Paschal, 27 Tex., 286; Renaud v. Pageot, 102 Mich., 568, 61 N. W., 3; McHugh v. Fitzgerald, 103 Mich., 21, 61 N. W., 354; Beach, Wills, sec. 115.
We have thus far said nothing of Tucker v. Whitehead, 59 Miss., 594, because learned counsel attempt to distinguish that case, in principle, from the authorities holding such declarations competent on an issue of undue influence, on the ground that that case was one as to the competency of declarations made by the testator, where the act of revocation — tearing off his name — was not shown to have been done by the testator. It is insisted that, in such case, the statutory act of revocation being equivocal, such declarations, not made at the time of tearing off the name, were admitted to make certain what otherwise was uncertain — whether the name was torn off by the testator or a [697]*697spoilator. It is true that only was the point decided, but the language of the court shows that the whole subject had been explored and the authorities cited go the whole length of holding the competency of the declarations on an issue of undue influence; the one emphasized by Judge Chalmers, Collagan v. Burns, 57 Me., 449, especially. It is noteworthy that five of the cases cited by Judge Chalmers are cited by Judge Cassody (Wills, sec. 316), to sustain -the proposition that it is “well settled that such declarations are admissible” upon the issue of undue influence. That case, therefore (Tucker v. Whitehead), is strongly persuasive of the view of this court at that time, and our researches confirm the soundness of the view indicated therein. So far as all that may be said touching the unreliability of such declarations is concerned, it may be replied, in the language of Judge Chalmers in that case, that “the evils which may spring from the introduction of parol proof in such a case are less than those which must be wrought by its exclusion,” and in the language of the supreme court of Maine in Collagan v. Burns, 57 Me., at page 458, that “the exclusion of evidence, relevant and material, from the fear that it may not receive its just degree of credence, is the resort of barbarism. Civilization hears, weighs, examines," compares, and then decides.” Indeed, it is worthy of special observation that it is held, in this last case, that “whether it be the making of a will or the destroying of one, the evidence in each case is alike, and for tlie same reason admissible.” So far as the objection that such declarations infringe the statute requiring wills to be made in writing is concerned, it is obvious that the statute also requires wills to be revoked in a particular mode, pointed out therein. In one case the inquiry is whether the instrument is still his will; and in the other, whether it ever was his will. But we mean to decide only that, on the issue of undue influence the declarations were competent. Many of the authorities holding such declarations incompetent on the issue of undue influ[698]*698ence are cases in which the only question was whether, the will being admitted to be the testator’s will, the proof was offered simply to contradict or vary its terms, and the decisions are malapropos as to the point here involved. It is sometimes said, accurately enoug’h, that such declarations are not competent as admissions against interest, not being, like declarations of one in possession of real estate that his possession is |>ermissive, characterizing his possession because the testator is not a party to the suit on an issue devisavit vel non — he has no interest. And it is sometimes loosely said, with far less accuracy (if, indeed, it be an accurate statement at all), that such declarations are no more competent than like declarations of a grantor in a deed, or the party to any written contract. But, as pointed out by •the editors of the American Leading Cases (Lawson v. Morrison, 2 Am. Lead. Cas., 488), and in Collagan v. Burns, 57 Me., 454 a will is ambulatory and no one has any vested rights thereunder, till death to be affected by such declarations. And it is sometimes said that these declarations are not evidence of the truth of the facts stated in them. But, as shown in Jackson v. Kniffen, 3 Am. Dec., 391, note, they are primary evidence of unsoundness of mind, and of undue influence as implicated in, and involved in such unsoundness of mind; the two things being, in the nature of things, logically, to some extent, indissolubly blended. It may be that the true solution of the apparent confusion is this: That what such declarations are evidence pf is, not in themselves alone that the testator did have the testamentary intentions he declared he had, for he may have wished to conceal his intentions, but that he did say he had the testamentary intentions testified to; and the jury are then to draw such inference as the whole evidence warrants, that they were or were not his real testamentary intentions, at the time of making the declarations, as compared with those set forth in the will, and looking to the change or absence of change in his condition, family, property, state of feelings, affections, etc., between the time [699]*699of making them and the will. The witnesses might speak truly in testifying that he did make the declarations and yet he may not, in fact, have had such testamentary, intentions. Whether his testamentary intentions were, at the time of making the declarations as declared, the jury are to determine, as stated, from the whole evidence. And if, from the whole evidence the jury believe they were really as declared at that time, an inference might legitimately be drawn that when the subsequent will conformed to them, they had continued down to the making of the will, and when the subsequent will did not conform to them, the testator had purposely misstated his intentions for some reason, or that he had changed his intentions, or that the will was not his will, but the product of undue influence. But the whole probative value of them is for the jury and not for the court. In the case at bar all of the testamentary declarations do not conform to the alleg’ed will, and this, of course, is a circumstance which the jury are to weigh in their consideration. Our conclusion is that the declarations of an alleged testator whether made before, after, or at the time of making the alleged will, as to his testamentary intentions, are both on the issue of testamentary capacity and of undue influence, competent evidence. The court did not err, therefore, in admitting them.
The next point which we shall notice is upon whom is the burden of proof as to testamentary capacity and undue influence. It will be seen, by reference to the elaborate note to Prentiss v. Bates (Mich.), 17 L. R. A., 494, (50 N. W. 637, 53 N. W., 153), that there is a great conflict of authority on the former question even; and, by reference to Baldwin v. Parker, 96 Am. Dec., 697, that like conflict of authority exists as to the latter. So far as the former question is concerned it is settled in this state, in Tucker v. Whitehead, 59 Miss., 594, that the burden is on the proponents throughout, and it is expressly so enacted by sec. 1825, Code 1892. As to the burden where the issue is undue influence, it is held, in Baldwin v. Parker, [700]*700supra, and in other authorities, that the burden is on the contestants. But to our minds, the reasoning is very unsatisfactory. On the other hand, it is squarely decided, in Prentiss v. Bates, supra (decided in 1892 by the supreme court of Michigan); in Chappell v. Trent, 90 Va., 849; 19 S. E., 314 (decided in December, 1893); in Runyan v. Price, 86 Am. Dec., 459; and apparently in Barry v. Butlin, 1 Curt.Ecc., 638, that the burden is on the proponents of the will. It is. correctly said by Baron Parke, in this last case, that the “strict meaning” of the term, “onus probandiis this, that if no evidence is’ offered by the party on whom the burden is cast, the issue must be found against him. Now, when the proponent of a will offers the will and the record of its probate, a presumption is thereby raised that the alleged testator had testamentary capacity; and this presumption satisfies the burden of proof in that respect; and the contestant must fail unless he overcomes this by proof on his part. But there is no shifting of the burden of proof, properly understood. And, just so, we think that the production of such will and such record of probate, under our statute, raises the presumption that the alleged will was the free and voluntary act of the alleged testator. It is not only necessary that the testator shall have testamentary capacity, but that that capacity shall be exercised freely and voluntarily. If either- its existence, or the freedom of its exercise is wanting, the instrument is not the alleged testator’s will. Both are essential parts of the proponent’s case. The issue is single-will or no will. And when we consider that the exercise of undue influence implies some degree of mental capacity to be overcome, and how indissolubly the two things are implicated the one in the other, it seems to us clear that the burden as to both is on the proponent throughout, and there is no shifting of this burden. But the burden in both respects is. satisfied by the introduction of the will and the record of the probated will, and the contestants must then offer proof to overcome the prima facie case thus made in both [701]*701respects. In the course of the opinion in Baldwin v. Parker, 96 Am. Dec., 704, quoting from Boyse v. Roseborough, 6 H. L. Cas., 49, it is said, “that where once it has been proved that a will has been executed with due solemnities by a person of competent understanding, and apparently a free agent, the burden of proving that it was executed under undue influence is on the party who alleges it,” and that '“undue influence is not to be presumed” ; and this is cited as an authority for the view we are combating. But the language, “apparently a free agent,” shows clearly that it is meant that the legal presumption where capacity exists, is that the testator was a free agent. And this presumption, of course, meets the burden, and puts the contestant to proof. “Undue influence is not to be presumed,” certainly. To put it in exactly equivalent phrase, it might be said, that the alleged testator was not a free agent is not to be presumed. On the contrary, capacity proved, the legal presumption is that he was a free agent — that the alleged will was his free and voluntary act. And this presumption satisfies the burden of proof as to undue influence, and the contestant is thus put to his proof. That is the whole meaning of that case, and of others wherein this question is obscurely and inaccurately treated. Mr. Bigelow has stated the true rule in the quotation we have above made from his note -to 1 Jarm. Wills. It is true he refers to three cases, apparently holding the contrary (see p. 68)— Potter's Appeal, 53 Mich., 106, 18 N. W., 575; In Re Martin, 98 N. Y., 193; and Armstrong v. Armstrong, 63 Wis., 162, 23 N. W., 407; but they fall clearly within the explanation there made by Mr. Bigelow, and are not in real conflict with the views we have expressed. See, specially, also, what Mr. Bigelow very accurately says as to the loose talk about the shifting of the burden of proof at page 65 of this very valuable note. The view we take, that the burden of proof as to undue influence — that is better phrased, that the testator executed the will freely and voluntarily — is upon the proponents throughout the trial [702]*702of an issue devisavit vel non is, we think, meant to be declared by our statute and is in harmony with our decisions along this general line, as, for instance, as to the burden of proof of sanity being on the state in criminal trials. Ford v. State, 73 Miss., 734, s. c., 19 South., 665. We, therefore, hold, on this proposition, that the burden of an issue devisavit vel non, is, both as to testamentary .capacity and undue influence — or, better, that the testator was a free agent — on the proponents throughout. The court, therefore, erred in holding that the burden was on the contestants.
In the the course of the trial the proponents offered the record of the probate of the will, and the will, and rested. The contestants then introduced their testimony as to testamentary capacity and undue influence and rested. The proponents then introduced testimony as to these two. matters in rebuttal. Contestants objected to this testimony as not in rebuttal, insisting that all the testimony proponents had on these points should have been offered in chief. The court correctly overruled this objection. That the course pursued in the conduct of this trial was correct is expressly decided in Runyan v. Price, 86 Am. Dec., 459, and In re Lehman's Will, 40 Minn., 371, 42 N. W., 286.
With respect to the rule as to allowing nonexperts to give their opinion as to the soundness of mind of the subject of inquiry, the true view is set forth in Wood v. State, 58 Miss., at page 742. They should testify as to the facts, acts, declarations, etc., of the subject of inquiry, and then give their opinions based upon these facts. Unless the attending physician (in this case, Dr. Lowry) is qualified as an expert, the rule applies to him, and should be observed. And this method of examination on the new trial should be rigidly adhered to as to all non-expert witnesses.
Most of the other very numerous assignments of error are to the action of the court in allowing proponent’s counsel to ask leading questions in chief. Very many of them are well taken. [703]*703It would too greatly protract this opinion to notice them seriatim. Care should be taken to avoid this error on the new trial. We remark, finally, that inasmuch as the learned court below was not called on to instruct the jury as to what constituted testamentary capacity or undue influence, we have nothing before us on either of these points — no action of his to review. Nor the errors indicated the judgment is reversed, and the cause remanded.
Reversed and remanded.