delivered the opinion of the Court.
Certain paper-writings, purporting to be a will of Sarah Ann Griffith, deceased, dated the 20th of December, 1875, and a codicil attached thereto, dated the 7th of March, 1876, were offered for probate, in the Orphans’ Court of Baltimore City.
By these papers, the testatrix gave to her • daughters Emma Coleman and Sarah Ann Ruddach, one thousand dollars each, and after the payment of certain legacies of about ten thousand dollars to other persons, she gave the rest of her property, amounting to at least one hundred and sixty thousand dollars, to her son David Griffith, and her daughters Mary E. Farnandis and Alverda Griffith.
On the petition of the appellees, grandchildren of the testatrix, issues involving fraud and undue influence, were *478sent to the Baltimore City Court for trial; and this appeal comes to us upon exceptions to the rulings of the Court below.
As to the question presented by the first exception whether the right of cross-examination extends to the whole case, or is limited to the matters in regard to which the witness has been examined in chief, there is a difference between the practice in this country and that which obtains in England. There, if a witness is called to prove any facts connected with the case, he becomes a witness for all purposes, and the other side may cross-examine him in regard to all matters relevant to the issues before the jury.
In this country the Supreme Court has decided, that this right is limited to facts and circumstances connected with the matter stated by the witness in his direct examination; and if the other side proposes to examine “him respecting other matters, they must do so by making him their own witness. Phila. & Trenton R. R. Co. vs. Simpson, 14 Pet., 448; Harrison vs. Rowan, 3 Washington Ct. Court, 580; Ellmaker vs. Brinckley, 16 Leigh & Rawle, 77. And this seems to us to be the better practice. There is no good reason why a witness called by one side to prove certain facts, should be considered a witness of that side in regard to other matters foreign to and in no manner connected luith the facts proven, and which the other side may desire to offer in evidence.
In this case however, the question proposed to the witness is strictly within the rule laid down by the Supreme Court. The memorandum of instructions, and the rough draft of the will, and the will itself, had been offered in evidence by the plaintiffs. The witness Venable, had referred to all of these papers in his direct examination; he had testified that the memorandum of-instructions had been delivered to him by Romulus Griffith, that finding some difficulty in understanding it he sent back for fuller *479instructions, and that these instructions were also communicated by Griffith ; under these circumstances it was competent on cross-examination to ask the witness, whether he made a fair copy of the rough draft embracing the additional instructions and submitted it to the testatrix, and whether it was approved by her as being in conformity with the instructions she had given.
These facts were germane to and connected with the circumstances under which the will was prepared, and in regard to which the witness had testified in his examination in chief. We should not, however, reverse the judgment on this ground, because it appears in the subsequent progress of the case that the defendants had the benefit of this evidence, and they suffered, therefore, no injury by the ruling of the Court.
In the second exception, the defendants offered to prove by the same witness, that the instructions for the codicil were given to him by Romulus Griffith as coming from the testatrix, what those instructions were, and that he prepared the codicil in accordance with said instructions, and as prepared by him, it was read to and approved by the testatrix. But the offer does not propose to confirm this alleged statement of Romulus, by showing that the witness afterwards repeated it to the testatrix, and that she admitted it to he true, hut only to show that she approved and executed a codicil that had been prepared according to her alleged instructions as received by Romulus. This might be true, and yet the fact that the testatrix had directed Romulus to communicate the ^instructions would, after all, rest upon the witness’ statement of what Romulus told him. It was, in fact, an attempt to prove by the witness that Romulus told him what the testatrix’s instructions were, or, in other words, to prove by the witness that Romulus told him what the testatrix had said. There was no error, therefore, in sustaining the objection to this question.
*480The questions arising under the third exception are of considerable importance in the trial of testamentary cases, and not altogether free of difficulty. How far, and for what purposes, the declarations of a testator, made after the execution of his will, may be offered in evidence under issues of fraud and undue influence, the decisions are conflicting. We do not propose to examine the many cases in which the subject has been considered, for this has already been done in Waterman vs. Whitney, 1 Kernan, 168, in Boylan vs. Meeker, 4 Butcher, 274, and other cases, and it is unnecessary for us to do more than state the conclusions we have reached.
Where such declarations are made so remote as not to constitute a part of the res gestee, they cannot he offered as independent evidence to prove the charge of fraud, or to show the external acts of undue influence, or attempts to influence the testator to make a will in a particular direction.
If offered for this purpose, they are inadmissible on two grounds. 1st. As mere hearsay evidence, which by reason of the death of the party whose statement is offered, can never he explained or contradicted by him. 2nd. It is inconsistent with the Statute of Frauds to permit a will executed with all the formalities required by the statute, to be impeached, or its validity in any manner impaired, by the parol declarations of the testator made after the execution of the will.
But the question, whether a will is the free and voluntary act of the testator, or the offspring of fraud, whereby his judgment was misled, or of influences operating upon him, in consequence of which his will was made subordinate to that of another, depends upon whether he had intelligence enough to detect the fraud, and strength of will enough to resist the influences brought to hear upon him.
The character and degree of the fraud practised, and the influence exerted, involve therefore necessarily, to some *481■extent, the physical and mental condition of the testator •at the time of the execution of the will. The influence that would he unlawful if exerted upon one advanced in years and in declining health, of a weak and vacillating will, might he altogether unavailing with one in robust health and of firm and resolute purpose. Any evidence, therefore, which tends to prove the precise mental condition of the testator, and to place him before the jury just as he was when the will was made, is admissible; and for this purpose the declarations of a testator may in some cases be the most satisfactory proof. It is a common practice to admit such testimony under issues involving testamentary capacity, and upon the same grounds it ought to he received under issues of fraud and undue influence, provided they are made sufficiently near in time, as to justify •a reasonable inference, that the mental condition which they are intended to denote, existed at the time of the execution of the will. Such evidence, it is true, may have an effect beyond that for which it can be legitimately offered; and although not competent to prove the facts upon which the charges of fraud and undue influence are founded, they may nevertheless tend to bias or prejudice the mind of the jury.
This objection, however, applies also to other species of evidence, which is competent for one purpose, hut not ■competent for another, and if it he admissible under the general rules of evidence, we cannot exclude it on that ground.
We are of opinion therefore, that the declarations of the testatrix to Mary Whitman, in November, 1876, “that she had made a will,” “that she was dissatisfied with it,” “that she had been persuaded to make it,” “that she was sorry she hadn’t let the law make a will for her, as it had for her husband, so that the children would have shared alike,” “that she had done great injustice to her other children and to her grandchildren,” naming them; *482“ and was troubled about it,” and “ sometimes tempted to-destroy it,” and other like declarations were admissible for the purpose of proving the mental condition of the testatrix at the time of the execution of the will and codicil, but for no other purpose.
We are of opinion also, that her declarations in regard to her testamentary intentions made in May, 1875, some-months before the execution of the will and codicil, and before any improper influences are supposed to have operated upon her, are admissible. Evidence of this character may be offered either to rebut the charges of fraud and undue influence, by showing that the will is consistent, with the long cherished wishes of a testator, or that it is contrary to well settled convictions of what he thought-was a just and proper disposition of his property among others standing in the same natural relation with those benefited by the will. The weight to be given to such testimony is a question for the jury. Denmead’s Appeal, 29 Conn., 309; Converse vs. Allen, 4 Allen, 512; Turner vs. Clemson, 15 N. J. Equity, 243.
We come now to the questions of law involved in the instructions granted and those refused by the Court. And here it is necessary to understand precisely the issues the jury had to try. That the will and codicil bore the genuine signatures of the testatrix; and that she was of sound mind and capable of making a testamentary disposition of her property, are admitted. The issues of undue influence were abandoned by the plaintiffs, and the only questions before the jury were whether the will and codicil were procured by fraud. The burden of proof was upon the plaintiffs. If they offered no evidence, or if it was insufficient to satisfy the jury that the papers offered for probate were procured by fraud practised upon the deceased,, then the defendants were entitled to a verdict.
The Court, however, instructed the jury, that if they found certain facts set forth in the plaintiffs’ first prayer, *483the burden of proof was upon the defendants to satisfy the jury, that the will and codicil were in all respects the free, voluntary and intelligent acts of the testatrix, and if they failed to do so, the plaintiffs were entitled to a verdict. In other words, assuming these facts to he true, the burden of proof which originally rested upon the plaintiffs, was shifted to the defendants. This was equivalent to saying, that the facts enumerated in the prayer amounted in law to a presumption of fraud, because on no other ground could a jury find a verdict for the plaintiffs.
How what are the facts thus relied on to support this presumption ?
First, that Eomulus Griffith was the son-in-law, agent and attorney-in-fact of the testatrix, and entrusted with the management of her affairs; that he employed his own counsel to draw the will and codicil, was consulted by the testatrix as to the making of said instruments, and otherwise took an active part in the preparation and execution of the same, by which his wife and himself were largely benefited. And it is insisted, that the doctrine recognized by Courts of equity, in dealing with matters of gift and contract, between parties standing in a fiduciary relation to each other, should be applied if not in its broadest sense, yet in some qualified manner to gifts under wills. It is true, that such Courts always watch with suspicion, transactions between persons standing in this relation, and although such transactions are not treated as altogether void, yet the burden of proof is upon the party holding the relation to show, that the influence necessarily arising from the confidence thereby reposed, has not been abused. And hence it is, that in gifts inter vivos between such persons, it is incumbent on the donee to show, that it was the free and voluntary act of the donor. But there is an obvious difference between a gift, whereby the donor strips himself of the enjoyment of his property whilst living, and a gift by will, which takes effect only from the death *484of the testator. And in Parfitt vs. Lawless, Law Rep., Probate and Divorce, vol. 2, 468, where the testatrix gave all her property to a Catholic priest, her confessor, it was expressly decided, that the doctrine of confidential relations as recognized by Courts of equity, in dealing with gifts or contracts inter vivos, had no application to gifts under a will.
Lord Penzance says: “In the cases of gifts or other transactions inter vivos, it is considered by Courts of equity, that the natural influence which such relations as those in question involve, exerted by those who possess it, to obtain a benefit for themselves, is an undue influence. The law regarding wills is very different from this. The natural influence of the parent or guardian over the child, or the husband over the wife, or the attorney over the client, may lawfully be exerted to obtain a will or legacy, so long as the testator thoroughly understands what he is doing, and is a free agent.”
But in addition to the relation which Romulus Griffith bore to the testatrix we have the further fact, that he was consulted by her in regard to making the will, that he employed his own counsel to draw it, and that his wife and himself were largely benefited by it. The fact that a party is largely benefited by a will prepared by himself, or in the preparation of which he takes an active part, is. nothing more than a suspicious circumstance, of more or less weight according to the facts of each particular case. If the testamentary capacity be doubtful, or if the party benefited be a stranger not allied by ties of kindred to the testator, these and other like facts would tend of course to increase the suspicion.
They are, however, but facts and circumstances to be considered by the jury in determining the question of fraud. But fraud in this case being a question of fact to be found by the jury, and not one of law to be inferred by the Court, it was error to instruct the jury that the facts *485thus relied ou by the plaintiffs created a presumption of fraud, or in any manner affected the burden of proof, which from the beginning to the end of this case rested upon the plaintiffs.
Then again, whatever suspicions these facts might excite or whatever explanation they might require, the undisputed facts show not only that the testatrix was fully competent to make a will, hut that the papers offered for probate were read and explained to her, and that she expressed herself fully satisfied with their contents. Whatever suspicion, therefore, the facts thus relied on might have excited, it was fully explained and removed by «the uncontradicted evidence offered by the defendants. And it is no answer to say, that an instruction to that effect, might have been asked by the defendants. It was misleading the jury to say, that certain facts tended to create a presumption of fraud, and therefore to shift the burden of proof .upon the defendants, if other facts in regard to which there was no conflict, showed that no such presumption could arise.
But then again it is said, there is a material variance between the will and the written memorandum of instructions, and this also is relied on as evidence of fraud. The testatrix was entitled to commissions amounting to about seven thousand dollars from her husband’s estate. By her will, she gave to Romulus Griffith and wife a sum equal in amount “to the commissions which she should receive during her life, and be entitled to at the time of her death.” By the memorandum of instructions, she gave to Griffith and wife “her portion of the commissions on her husband’s estate, if the said estate should not have been settled and commissions received before her death.” And it is insisted, that even if the testatrix had received these commissions in her life-time, and had paid them to Griffith and wife, yet they would he entitled under the will to an additional equal sum at her death. Now if it he conceded *486that there is a variance between the will and the memorandum in this respect, there is not a particle of evidence to show that it was made with a fraudulent purpose. The will was drawn by the witness Venable, and he says, if there is a variance, it was altogether unintentional on his part, and arises from an unhappy use of language. The appellants expressly disclaim any intention to impute improper motives to him, and if this variance escaped his attention, a lawyer and draftsman, it would he unjust and unreasonablé to impute knowledge of it to Romulus Griffith, because the will was read in his hearing. Taking, therefore, all these facts together, and adding thereto every fair and reasonable inference, we are of opinion that they are not sufficient to create a presumption of fraud, or to shift the burden of proof from the plaintiffs to the defendants.
In addition to this, the instruction was clearly erroneous, in submitting to the jury under the circumstances surrounding this case, the question whether the testatrix fully understood the provisions and effect of the will and codicil. Knowledge of its contents is, of course, essential to the validity of every will, hut where the testamentary capacity is unquestioned, such knowledge, as a general rule, will he inferred from the execution of the will itself. If there he suspicious circumstances surrounding the preparation and execution of a will tending to rehut this presumption, it may he proper to require additional proof of some kind that the testator did in fact understand its contents. But in this case it was not only admitted that the testatrix was of sound mind and capable of making a will, hut the evidence shows that both the will and codicil were read and explained to, and approved by her. Under such circumstances as these, the law imputes knowledge and the door of inquiry is closed. There may he exceptions to this rule, it is true, arising in cases of fraud practised upon a testator in the preparation and execution of his will, hut it is not *487■alleged here that one instrument was prepared and read to the testatrix, and another and a different one substituted in its place; nor is it alleged that any fraud was practised in reading the will or codicil. On the contrary, these papers were read to her, line by line, just as they were written ; and being in the full possession of all her faculties, there is nothing to except this case from the ■operation of the general rule as we have stated it.
Then again, there is another objection to this prayer, and one which applies with equal force to the third prayer. Whether the will and codicil were procured by fraud, were questions of fact to be found by the jury, and not questions of law to be inferred or presumed by the Court. The affirmative of the issue was upon the plaintiffs, and these instructions assume that if evidence is offered sufficiently strong to make out a prima facie case of fraud, the burden of proof was shifted to the defendants, and unless they offered evidence rebutting the prima facie case thus made, the jury were obliged to find that the will and codicil were procured by fraud. The Court thus undertook to say us matter of law, that the plaintiffs had proved the affirmative, or, in other words, had proved that the will and codicil were jDrocured by fraud, instead of leaving this question to be found by the jury. Whether the defendants offered any evidence or not, the affirmative of the issue was none the less ultimately upon the plaintiffs, and it was incumbent on them to satisfy the jury that it had been established. To entitle the plaintiffs to a verdict, the jury must find as matter of fact that the will and codicil were procured by fraud.
The second prayer is based upon the theory, that the testatrix was induced to discriminate in her will and codicil against certain of her children and grandchildren, in consequence of false representations made to her by those benefited by the will and codicil, such persons knowing that' these representations were false, and knowing that *488they would, and intending that they should, influence the testatrix in making this discrimination. In other words, that a downright fraud was practised upon her, in consequence of which, certain members of her family were excluded from a fair participation in her property. Objection is made to this prayer in the first place, on the ground that there was no evidence to warrant the Court in submitting to the jury, the finding of the facts on which it is based. It is not the office of a Judge ‘to weigh or balance conflicting evidence, however strongly the evidence on the one side may preponderate. That is the duty of the jury. If no evidence is offered, or if it is not such, as one in reason and fairness could find from it the facts sought to he established, the’ Court ought not to submit the finding of such facts to the jury. Without meaning to intimate any opinion in regard to the weight of the evidence, or the conclusion a jury ought to draw from it, it was sufficient, I think, to warrant the Court in submitting the facts set forth in this prayer to the jury. Upon this question the Judges who heard this case are equally divided in opinion, and the prayer would therefore he affirmed by a divided Court,.hut for one or two minor objections in regard to which we all agree.
Where a will is assailed on the ground of false representations instilled into the mind of the testator, it must appear that such representations were made for the purpose of influencing the testator in making his will. Now, in this prayer, the jury are required to find that the representations therein set forth, were made for the purpose of influencing the testatrix in making her will, hut they are not required to find that they were made for this purpose in regard to the codicil.
Then again it instructs the jury if they find certain' facts, “they may find that the will and codicil were obtained by fraud,” and their verdict “must” he for the plaintiffs. In order to avoid any misconstruction, it should *489have been qualified by adding after the words “were obtained by fraud” the words “and if they shall so find,” their verdict must be for the plaintiffs. As the case will be remanded for a new trial the prayer in these respects can he corrected.
It was also urged, that the several prayers offered by the plaintiffs were erroneous, because they assume if fraud was practised upon the testatrix in making her will and codicil, the entire will and codicil were void, although it might appear from the evidence, that the fraud affected only certa/in parts of the will and codicil. And it was insisted, that if the fraud did not affect the entire will and codicil, the jury by their verdict should have segregated the parts thus affected by the fraud. This is a question of importance, and so far as our information extends, is now for the first time raised in this Court. Whether under proper issues framed for the purpose of presenting the question, a jury may find part of a will void on account of fraud and another part good, it is unnecessary for us to decide. No such issue was sent by the Orphans’ Court to the Court below for trial, and the question is not therefore presented by this record. The issues before the jury were, whether the will and codicil were procured by fraud ? And their verdict must of course be responsive to these issues. Where a will is assailed on the ground of fraud and undue influence, if it is proposed that the jury shall find whether such fraud and undue influence affects the whole will or certain parts only, and what parts, it seems to us, there ought to he a separate issue framed presenting directly that question to the jury.
The plaintiffs’ fourth, fifth and sixth prayers announce well-settled principles, and there was no error in granting them.
The issues of undue influence having been abandoned and no longer therefore before the jury, the Court properly refused to instruct the jury in regard to the law applicable to such issues.
*490(Decided 7th February, 1879.)There was error in granting the plaintiffs’ first and third prayers, and also in granting the second as it was presented, and the judgment will he reversed and a new trial awarded.
Judgment reversed, and new trial awarded.