he knew nothing of the matter except from a conversation with Mrs. Heap. The court ruled that at this stage, and until further reason for its relevancy appeared, he should exclude the question.

During the trial the appellants offered in evidence three assessment lists made out and sworn to by Mr. Cushman, as the agent of Mrs. Heap, of her taxable property for the years 1883, 1885, and 1886. This testimony was excluded.

The appellants further offered to lay in the inventory and distribution of the Lord estate, as tending to show what the amount of property was that came to Mrs. Heap in 1866, either directly or indirectly. On objection, it was stated that it was proposed to follow it up with proof that she had the same property in 1879, and it was claimed as tending to show the amount of her property to have been thirty thousand dollars or upwards. The court declined to receive the evidence, but stated that if counsel had evidence as to her property at later dates, he would consider and rule upon the question when such evidence should be offered.

We cannot well understand why all the evidence so preBented should not have been received. In was incumbent upon the appellants to show the existence of a confidential relation between Cushman and the testatrix. To this extent, undoubtedly, the burden was upon them; and they had assumed and were endeavoring to support it. The court, indeed, as we shall hereafter see, went further, and held that proof of the existence of such relation did not shift or raise the appellants' burden, upon the issue of undue influence. Clearly, the facts and circumstances surrounding the relations between Cushman and the testatrix had a direct, positive, and important bearing upon this question, and the amount, situation, and character of the property with which Cushman was intrusted, during any and all portions of his stewardship, as well as the degree of knowledge which the testatrix possessed in regard to the same, and the conduct of Cushman in imparting information upon the subject to her or in withholding it from her, were relevant and important. Evidence upon these points would have disclosed the degree of confidence reposed. It was one of the elements by which it would be made to appear how far the provisions of the will in his favor were reasonable, and in fact, to use its own language, "only a just reward for his faithful care." Without this element, the remaining factors might be insufficient for the solution of the

problem. We think the evidence was admissible, and its exclusion erroneous.

During the trial, a witness for the appellants having undertaken to compare the mind and memory of the testatrix, as regarded the amount of property she was worth and the disposition she wished to make of it, with that of an average child of seven or eight years, upon objection and motion made to strike out that portion of the answer, the court granted the motion, saying that the witness might state her opinion of the strength of Mrs. Heap's memory, or of the extent of her knowledge of her property, or the like, but not comparatively; which the witness did. Although we should have hardly been disposed to grant a new trial on this ground alone, since it does not appear very probable that any injury resulted to the appellants in consequence of this ruling, we nevertheless think that the witness should have been allowed to make the comparison. It may be correct, as the appellees claim, that there is no such thing as an average child, and that the mind and memory of a child is incapable of perfect measurement. This is true of almost any standard. It would be true of the standard which the law prescribes for the determination of reasonable care and prudence. It is nevertheless true that such a comparison carries with it, to an ordinary apprehension, a greater approximation to certainty than any merely general and abstract statement. The words "childish," and "second childhood," are used in common parlance, and so used because they fittingly and adequately express a clearly defined idea. That eminent jurist Judge Seymour, in a charge to a jury, reported in the appendix to 39 Connecticut, 591, says that the main difficulty in questions of this sort arises from the want of a definite measure of mental capacity; and then adds: "I can give you no precise rule, but I think it clear that if the prisoner's perception of consequences and effects was only such as is common to children of tender years, he ought to be acquitted."

It is found that "before the appellants closed their case they requested the counsel for the appellees to produce and deliver to them for their examination all books kept by Mr. Cushman, and containing accounts of his stewardship with Mrs. Heap, as her agent. The counsel for the appellees declined to do so. The counsel for the appellants thereupon asked the court to order such production. The court declined

to do so. Mr. Cushman was at the time, and throughout the trial, in court."

We find no error in this. With the existing statutory provisions in regard to discovery (Acts of 1889, o. 22), and with the power of parties to compel the attendance of witnesses and adverse parties for examination, and to call them when, as in this case, present, the exercise of the extraordinary power invoked is best intrusted to the sound discretion of the trial court, to be granted or withheld as circumstances seem to warrant; and there we prefer to leave it.

During the argument, one of the appellants' counsel claimed the right to read, as a part of his argument to the jury, paragraph 38, section 1, page 509, and paragraph 88, section 37, page 526, of Redfield on Wills; but upon objection being made, the court declined to permit him to do so. This ruling is so clearly in accordance with the decision of this court in Baldwin's Appeal, 44 Conn. 87, that it seems at first thought strange that the question should again be raised. We are, however, aware that the later decision in State v. Hoyt, 46 Conn. 330, has been regarded by some members of the profession as somewhat modifying the doctrine of the earlier case; especially since, although the former was a civil and the lat ter a criminal case, yet as the subject-matter of the proposed reading in the latter belonged exclusively to the realm of fact, and not of law, it is perhaps difficult to see how, on principle and in reason, a distinction can be drawn between the two classes of cases; and this has doubtless led to some question. And as the rule upon this subject is one calling for very fre quent application in the trial of causes, it seems most important that it should be definitely and positively stated, and clearly and fully understood. We will therefore say that the ruling in State v. Hoyt, 46 Conn. 330, must be regarded as confined exclusively to cases where the plea of insanity is interposed in behalf of persons indicted or informed against; the allowance in such cases forming an exception, based upon a practice which the majority of the court in that case felt had so hardened into a rule that they were not at liberty to abrogate it; that in all other cases the decision in Baldwin's Appeal, 44 Conn. 37, fully applies; and that facts relevant to the cause cannot, except within the limits of certain defined and recognized exceptions, be proved by reading from published books, or given to the jury except under the sanction of an oath and the test of cross-examination; while the reading to the court

or jury of such books of science, art, or purely technical knowledge, necessary for an accurate apprehension of any relevant matter, as have been shown in evidence to be recognized by experts as standard authority, and by such proof in effect incorporated in and made part of the testimony of such witnesses, may be allowed or refused by the trial court in the exercise of its judicial discretion. To this extent the allowance of such reading would seem to fall within the reason of the decision in Tompkins v. West, 56 Conn. 478.

The appellants filed twenty written requests to charge. These requests were only partially complied with. Some of those refused were clearly incorrect, while others appear to possess merit. For the sake of greater brevity, we will, however, confine our further consideration to two points in the charge as actually made. On the question of testamentary capacity, the court, in addition to declaring the rule correctly as this court has established it, that the law merely requires that the testator should be possessed of sufficient intelligence and memory to fairly and rationally know and comprehend the effect of what he is doing, and the nature and condition of his property, to understand who are or should be the natural objects of his bounty, and his relations to them, the manner in which he wishes to distribute it among or withhold it from them, and the scope and bearing of the provisions of the will he is making, further said: "Mere physical weakness or disease, old age, eccentricities, blunted perceptions, weakening judgment, failing memory or mind, are not necessarily inconsistent with testamentary capacity. One's memory may be failing, and yet his mind not be unsound. One's mental powers may be weakening, and still sufficient testamentary capacity remain to make a will."

This is also undoubtedly true; but such facts are admissible in evidence upon the question of capacity, and it was mainly by the proof of their existence that the appellants sought to establish the want of such capacity. The court, therefore, in charging as it did, and in entirely failing to make reference in any portion of the charge to the significance of such facts as evidence, would seem rather, in effect, to have withdrawn them from under the eyes of the jury, and from their consideration of them as such evidence, and thereby may, and we fear must, have misled the jury, to the injury of the appellants.

Upon the question of undue influence, the court charged the

jury as follows: "If a paper is executed with the requisite formalities of a will, and the person signing it is shown to have sufficient capacity to make a will, the presumption is, that it was executed freely and without undue or improper influence, until the contrary appears, and the burden of proof is therefore upon the party alleging undue influence. That burden, however, becomes shifted whenever the person who draughts the will or participates in procuring its provisions from the testator also occupies a relation of special confidence toward the testator, and at the same time is made specially benefited by the terms of the will. The reason of this rule is, that his confidential relations may have enabled him to exercise a controlling influence over the testator in his own behalf in procuring the provisions of the will from the testator. The burden is therefore placed upon him to show, by a fair preponderance of proof, the fairness of his own conduct. This rule applies only when the beneficiary or the confidential person draughts the will, or takes part in procuring from the testator its terms and provisions for some one else to reduce to writing. It does not apply where the confidential relation only exists between the testator and the beneficiary. There must be some participation in the procurement of the will."

We think the jury would have been fully warranted, from the evidence disclosed by the finding of the nature, character, and extent of the confidential relation between the beneficiary and the testatrix, in the inference and presumption that such beneficiary did take part in procuring from the testatrix the terms and provisions of the will, without direct and positive proof of the fact, and that their right to draw such inference should have been stated to them by the court. The mere existence of a confidential relation would not, indeed, in all cases, and necessarily, raise such presumption, especially when it appeared that the opportunity of familiar and secret communication and intercourse between the testator and the beneficiary, at a time proximate to the execution of the will, was wanting; but whenever a legacy is given to an attorney, confidential adviser, guardian, or other person sustaining a relation of special confidence to a testator, or whenever the individual who prepares the instrument or conducts its execution, not being a relative who would in the absence of a will be an heir, derives a benefit from its provisions, in either instance the surrounding circumstances may be such that a presumption, similar in character in each case, would naturally