delivered the opinion of the court.

This is the sequel to Hugo v. Clark, 125 Va. 126, 99 S. E. 521. The question is whether a certain paper dated December 27, 1916, is the true last will and testament of Cyrus Warden, deceased. After the case was here before, it was twice tried in the circuit court. The first trial resulted in a verdict for the will. The court set this verdict aside, and at the next trial there was a verdict against the will, upon which the judgment now under review was entered.

The paper in question purported to bequeath and devise all of Warden’s property to Eva Catherine Clark, the proponent and contestee below and the plaintiff in error here. It is admitted that the paper was duly executed and published as a will, and that it became and was the true last will and testament of Cyrus Warden from its date until a certain day in March, 1917, on which latter day the contestants claim that it was revoked by a subsequent inconsistent will. If this contention is sound, the contestants, who are Warden’s heirs and distributees, will take the estate, because *102he destroyed the second will a few days before his death, and died intestate. We will designate the two papers as the December will and the March will, respectively.

The March will was prepared by the direction of Warden, was signed by him and acknowledged as his will before two attesting witnesses. It is conceded to have been upon its face, and in the form of its execution and acknowledgment, strictly in conformity with the requirements of the law with respect to a valid will. But the contention is made, on behalf of Mrs. Clark, that the paper was not in fact executed by Warden with the requisite testamentary intent ; that he did not intend it unconditionally as a will, either at the time he signed and acknowledged it, or at any time prior to its voluntary destruction by him; and that, therefore, it never did become, in legal contemplation, his will.

[1, 2] We shall first consider the case as made at the first trial. It seems to be the contention of counsel for contestees that under the present Code the action of the court in setting aside the first verdict cannot be inquired into here, but we think otherwise. It is true that section 6363 of the Code of 1919 (which deals with the rule of decision in this court where a new trial has been granted or refused below purely upon the evidence) omits the provision contained in that section as it formerly stood (Code 1904, sec. 3484), that “when there have been two trials in the lower court, * * the rule of decision shall be for the appellate court to look first to the evidence and proceedings on the first trial, and if it discovers that the court erred in setting aside the verdict on that trial, it shall set aside and annul all proceedings subsequent to said verdict and enter judgment thereon.” But this omission was due to the fact that under the provisions of section 6251 of the Code of 1919, if a verdict is set aside because contrary to the evidence or without evidence to support it, the trial court will usually enter -final *103judgment without a new trial, and it was deemed unneces^ sary to retain the old rule, because in future two new trials upon the evidence would not ordinarily occur. (See Revisors5 Notes to sec. 6363.) In this case, however, the first trial was had before the Code of 1919 went into effect, so that the trial court, under the practice then prevailing, could not enter a final judgment for either party. The result .is that the contestee, who won under the first verdict and lost under the second, is here with two exceptions, the first to the action of the court in granting the contestants a new trial, and the second to its action in refusing her motion to set aside the second verdict and enter final judgment in her favor. It is manifest that the revisors had no purpose of cutting off any rights which parties might have become entitled to under the former practice. A contrary holding would be at variance with any reasonably clear warrant ifi the statute itself, and not within the expressed purpose of the change, stated by the revisors as follows: “The object is to end the action at once and put the losing party to his writ of error, thus avoiding the temptation to perjury and in many cases the unnecessary expense of a second trial.” (Revisor’s notes, sec. 6251.)

The contestants having shown that the March will was regular in form as to its execution and acknowledgment, "the contestee, to rebut the presumption of an animus testakdi which would thus arise, relies chiefly upon the testimony of Mr. Ivor A. Page, the lawyer who drew the paper. Mr. Page was the only person (save the attesting witnesses) to whom Warden ever spoke about this will, and the only person with whom he ever discussed its legal effect as a testamentary document: He was called as a witness by the contestants,' and on direct examination, after having stated that the circumstances surrounding the preparation and execution of the paper were more peculiar than those relating to any will or paper he had ever drawn, further testified *104■in part as follows: “He (Warden) said: ‘Mr. Page, I have known you forty-odd years, and I have come to you to draw a paper which may be my will and it may not/ ” And further: “Warden said: ‘A certain thing may happen that this will not be my will. I have drawn it, and if it does happen it will be my will/ After, he told me what to do I drew the paper. I took a copy of it and I said: ‘Mr. Warden, if anything happens I have a copy of this paper/ and he said: ‘Dont take that; I don’t want you to because” this may not be my will,’ and I deliberately tore it up and threw it in the wastebasket.” On cross-examination, Mr. Page further testified in part as follows:

“Q. Mr. Warden told you Judge Ackiss had drawn a will for him?

“A. Absolutely, he told me he had Judge Ackiss prepare a will.

“Q. And the paper he asked you to prepare he told you might or might not be his will ?

“A. He did that.

“Q. He acknowledged it with that statement fresh in your mind?

“A. Yes sir.

“Q. He acknowledged it, in the fact of those other facts, that it might or might not be his will?

“A. Yes.

********

“Q. It was acknowledged with you knowing that it might or might not be his will ?

“Q. Absolutely. He told me it might not. He told me the reason.

■ * * * * * * * *

“Q. Did he tell you the will would be effective under certain conditions?

“A. Yes sir.

*105“Q. And would not be effective except under certain conditions?

“A. Yes sir.

“Q. So there was no intention to revoke that will except in the event of the happening of this event?

“A. That is what he told me. He said: ‘This may or may not be my will. If this happens it will be, and if it don’t it won’t.’ ”

The March will was prepared just one month before Warden died. Three or four days before he died he was in the office of Judge Ackiss and asked the latter this question: “Are you sure that will that you prepared will stand?” and was assured by Judge Ackiss, in reply, that the paper then inquired about, which was the December will, would stand It satisfactorily appears that what Warden had in mind at the time of this inquiry was a possible contest of the December will by his heirs. The December will prepared by Judge Ackiss was delivered to him by Warden for safekeeping, was in his possession at the time of the above conversation, and remained with him until the testator’s death.

It may be added, merely to explain what otherwise might appear as an omission or oversight in the recital of the testimony, that while Mr. Page said he “told me the reason,” Warden did not in fact ever explain his reason for wanting to prepare a paper which he might thereafter adopt as his will. It does appear in the evidence that Warden was much attached to Mrs. Clark, a daughter of one of his former friends, lived with her for some time, and died at her home; and it further appears that he was not on good terms with his next of kin.

The question involved in this case becomes at last a very narrow one, but it is not free from difficulty. Did Cyrus Warden ever have a second will? Or, to put it differently, was the so-called March will a testamentary document in such a legal sense as that it can bé treated as a will sub*106sequent to, and therefore destructive of, the December will?

[3, 4] The legal rules and principles which must control our decision are clear. A will is revoked by a subsequent inconsistent will; and after such revocation the destruction of the latter will does not have the effect of reviving the former, even though the testator so intends. Code 1919, sec. 5234; Rudisill v. Rodes, 29 Gratt. (70 Va.) 147. The decisive and narrow question, therefore, is whether the March paper ever in truth became Mr. Warden’s will. The mere fact of the formal signature and acknowledgment, though strictly in accordance with the requirements of the statute, is not conclusive upon this question. There must have been a testamentary intent. In McBride v. McBride, 26 Gratt. (67 Va.) 476, 481, this court, speaking through Judge Staples, said: “It must satisfactorily appear that he intended the very paper to be his will. Unless it does so appear, the paper must be rejected, however correct it may be in its form, however comprehensive in its details, however conformable to the otherwise, declared intentions of the party, and although it may have been signed by him with all due solemnity.”

In Early v. Arnold, 119 Va. 500, at page 502, 89 S. E. 900, at p. 901, Judge Harrison, delivering the opinion of the court, said: “It is true that no particular language or form is necessary to constitute a valid will, and that the surrounding circumstances are admissible to explain, in a proper cáse, the intention of the testator; but one may execute a paper with every formality known to the law and by it devise all his property, but unless he intends that very paper to take effect as a will, it is no will.”

In 40 Cyc., p. 1077, it is said: “Animus testandi is essential to the validity of a will; therefore in order to render a document operative as a will it must have been intended to *107operate as a will when it was executed, or later adopted as such.”

In Boofter v. Rogers, 9 Gill (Md.) 44, 53, 52 Am. Dec. 680, 683, this language is used in the opinion: “In order to make a paper the last will and testament of a deceased person, at the time it is written, it must appear that such person possessed the animus testandi at that time. A paper, although not a last will and testament at the time it is written, may be made such afterwards by adoption.”

In 2 Schouler on Wills, 5th Ed., sec. 278, the author says: “Wills, to be valid, require of course the genuine animus testandi; to the extent, at least, of intending a disposition whose legal effect the court may safely pronounce testamentary. The mind should act freely and understandingly to this intent; and therefore it may be shown in evidence to vitiate an alleged will, not only that it was the offspring of an unsound mind, of essential error, or of coercion, but that it was written in jest, or without any idea of making an operative will. Such jests, however, are unsafe ones; and parol evidence tending to prove that a paper expressed and executed with all solemn formalities as a will was not so intended, or was only to operate under certain reservations not expressed on its face, is very little encouraged by authority.”

In Fleming v. Morrison, 107 Mass. 120, 72 N. E. 499, 105 Am. St. Rep. 386, it was held that although the instrument purports to be a will and to be signed by the testator and attested and subscribed by all necessary witnesses, it may be proved to be not -of a testamentary character by evidence that the testator told the witness before whom he first acknowledged it before they parted that the instrument was a fake will made for a purpose.

In a note to In re Kennedy, 159 Mich. 548, 124 N. W. 516, 28 L. R. A. (N. S.) 417, 134 Am. St. Rep. 746, found in 18 Ann. Cas. at page 892, it is said: “The weight of authority *108supports the rule that evidence is admissible to show that an instrument purporting to be a will, and properly executed as such, was not intended to operate as a will, but was executed with some other purpose or object in view. Lister v. Smith, 10 Jur. N. S. (Eng.) 107, 3 Sw. & Tr. 282, 33 L. J. P. 29, 9 L. T. N. S. 578, 12 W. R. 319; In Goods of Nosworthy, 11 Jur. N. S. (Eng.) 570, 4 Sw. & Tr. 44, 34 L. J. P. 145; Nichols v. Nichols, 2 Phill. Ecc. (Eng.) 180; Davis v. Rogers, 1 Houst. (Del.) 44; Fleming v. Morrison, 187 Mass. 120, 72 N. E. 499, 105 Am. St. Rep. 386. See also Trevelyan v. Trevelyan, 1 Phill. Ecc. (Eng.) 149; Boehm v. Kress, 179 Pa. St. 386, 36 Atl. 226.

“The momentous consequences of permitting parol evidence thus to outweigh the sanction of a solemn act are obvious. It has a tendency to place all wills at the mercy of a parol story that the testator did not mean what he said. On the other hand, if the fact is plainly and conclusively made out that the paper which appears to be the record of a testamentary act was in reality the offspring of a jest, or the result of a contrivance to effect some collateral object, and never seriously intended as a disposition of property, it is not reasonable that the court should turn it into an effective instrument. And such, no doubt, is the law. There must be the animus testandi.” Sir J. P. Wilde in Lister v. Smith, cited supra.

“Generally the animus testandi is the natural and primary inference from the act of signing and formal publication; but this inference may be weakened or destroyed by any attending circumstances of sufficient force, by evidence of the weakness or incapacity of the testator to do, or of the want of intention actually to do, what he seems to do by that act. This is not admitting parol evidence to vary the-will, but to ascertain whether it is really the will of the decedent.” Davis v. Rogers, 1 Houst. (Del.) 44.

*109[5] We have found no case involving facts exactly similar to those of the instant case, but we think it follows as a matter of reason and principle from the foregoing authorities that while the cases will necessarily be few and exceptional, yet where the facts justify it an instrument which on its face is testamentary may be shown by parol not to have been such. This, of course, does not mean that it will be permissible to introduce evidence to show that the operation of a paper executed with testamentary intent, and having once become a testamentary paper absolute in its terms, can be shown by extrinsic evidence to depend upon conditions not therein expressed. Perhaps the best and most succinct statement of the rule'is to be found in 30 Am. & Eng. Enc., 2nd Ed., page 580, 581, as follows: “Evidence is admissible * * to show that an instrument, on the face of it testamentary, was not written animo testandi; but not to show that the operation of a will, absolute in its terms, depends upon conditions.”

[6] The evidence clearly tended to show that the so-called March will at all times lacked the essential animus testandi. This being true, and the evidence having been substantially the same at both trials, and there having been no errors of law upon the first, we are of opinion that the first verdict should be upheld.

This is not a case in which we are permitting parol evidence to vary or contradict the terms of a written instrument, It is somewhat analogous to the case of Whitaker & Fowle v. Lane, 128 Va. 317, 104 S. E. 252, 10 A. L. R. 1157, in which we held that it was not a violation of the rule against parol evidence to permit such evidence for the purpose of showing that a sealed instrument, absolute on its face, was delivered to the grantee or obligee on an oral condition precedent to its effectiveness.

Counsel for the contestants quote and rely upon the following from the case of Cody v. Conley, 27 Gratt. (68 Va:) *110313, 322: “Ordinarily where the instrument is executed with all the requisite formalities it will be presumed to have been done animo'testandi, notwithstanding that it may be expressed to have been made to avoid the contingency of dying intestate, in case the testator should not return from a contemplated journey. In such a case, in order to render the instrument contingent in its operation, it should clearly appear by its language that it was not intended to remain as an operative will, except in the event of the failure to return.”

But in Cody v. Conly there was no question oí a testamentary intent. The question there was as to a conditional operation of a testamentary document. The difference is narrow but substantial. In the Cody Case there was admittedly a testamentary act, while in the case at bar the testamentary character of the paper is distinctly challenged.

The case of Sewell v. Slingluff, 57 Md. 537, appears to be at variance with the conclusion which we have reached as to the law of the ease; but there is this in the opinion which supports our view, and which appears to us' to be in conflict with much that is said in the opinion in that case: “There are three essential requisites for every good and valid will; and these requisites áre, perfect testamentary capacity, the intention to dispose of property in the event of death and the form required by the statute. One of these requisites for every perfect will — the intention to dispose of property in the event of death — is what the law terms the animus testandi, and is thus defined in 2 Shepherd Touch., 204; ‘The second thing required to the making of a good testament is that he that doth make it, have at the time of making it animus testandi, i. e., a mind to dispose, a firm and advised determination to make a testament; otherwise, the testament will be void.’ And he then goes on to say that if a man jestingly, and not seriously, writes or says that such a one shall have his goods, this is no will.”

*111The case of Brown v. Avery, 63 Fla. 355, 376, 58 So. 34, Ann. Cas. 1914 A, 90, also holds against the admissibility of parol evidence to prove that one instrument purporting to be a will, and properly executed as such, was not so intended. The opinion in that case contains a quite full citation of the authorities on both sides of the question.

It is insisted that the first verdict ought to be set aside for error in the instructions, and for error in allowing certain evidence over the objection of the contestants. We cannot accede to this view. The instructions were substantially in accord with the views herein expressed. It is true that two instructions offered by the contestants, and substantially right in themselves, were refused, but the principles announced therein were amply covered by others given in their behalf by the court. It is also true that two other instructions asked for by the contestants were refused as asked, and given in modified form, but the modifications were in accord with the views which we have expressed in this opinion.

[7] The evidence which is said to have been improperly admitted at the first trial was the testimony of Judge Ackiss hereinbefore set out with reference to Mr. Warden’s inquiry of him about the December will, and its efficacy as against a contest. We think this testimony was properly admitted. The question is whether the March paper ever in fact became a will. The evidence tended to show that it was not ever so intended by Mr. Warden, and that was the vital question before the jury. The inquiry which he made of Judge Ackiss, along with the fact that the December will still remained in the possession of Judge Ackiss, and that the March will was subsequently destroyed by Mr. Warden, were proper circumstances for the consideration of the jury as tending to show the mental attitude of the testator towards the March paper.

*112[8] The case is a most unusual one. Its like will seldom arise; and this decision as a precedent is not likely to be nearly so far-reaching as counsel for the' contestants seem to fear. There can be no doubt that this decision accomplishes substantial justice in the sense that it permits a testator of admittedly sound mind to do with his property as he pleases. Courts cannot always have the satisfaction of feeling that decisions arrived at by them have reached the right of the individual case, because they must have due regard for the legal proverb that hard cases make bad law, but they must never lose sight of the fact that justice in the individual case should always be a primary consideration, and should be brought about if that result can be attained without sacrifice of sound and just general principles.

Upon the whole case, we are of opinion that the court erred in setting aside the verdict at the first trial, and we shall accordingly enter up a judgment thereon in favor of the contestee.

Reversed.