SherwiN, J.
The evidence is voluminous, and an attempt to set forth in detail sufficient thereof to show its trend in support of the adverse claims of the parties hereto would extend this opinion beyond any length profitable to the profession or to the parties. All of the parties to the action are children of the deceased — his wife having died before he did — and the testimony as to their father’s physical and mental condition during the period of his life material to the question before us is in conflict. We have given the entire record the care which the importance of the litigation demands, and reach the conclusion that the verdict has such support in the evidence that we should not say, as a matter of law, that it is wrong.
The will in question was executed on the 18th day of June, 1897 — a little over four months before Mr. Kirsher’s ■death. He was then past eighty years of age, and physically weak. For many years prior thereto, and prwgr *340to 1894, he had been totally blind. On the 11th day of ' July, 1894, he suffered a stroke of apoplexy, and was treated therefor by Dr. Henry Matter, who testified that when he reached him, soon after the attack, Mr. Kirsher knew him, and was able to, and did, talk intelligently with him and others. Dr. Matter also testified that he did not consider the stroke a severe one, and that Mr. Kirsher partially recovered therefrom, though never enough so to. walk without help. There is but little conflict in the. testimony as to Mr. Kirsher’s physical condition after the. stroke, but the contestants claim that it was immediately followed by senile dementia which continued until his-death. It may be said in this connection, however, that the contestants’ witnesses are not agreed as to the' duration of the effect of the stroke upon the deceased, though there is testimony supporting the contestants’ theory. On the other hand, the decided weight of the testimony, as-we view it, shows that no general or settled mental disability was caused by the stroke, or followed it. On this branch of the case the court gave this instruction:
“If, considering as directed in instruction No. 7 — the-last preceding instruction — you have found that the said Peter Kirsher at any time prior to the date of the execution of said will was of unsound mind, then his mental unsoundness is presumed to continue, unless a recovery or restoration is shown, and the burden is upon the defendants to show such recovery, and that at the very time of’ the execution of the instrument in question the said Peter-Kirsher was of sound mind, as elsewhere defined in these-instructions. If you find by a preponderance of the evidence that the said Peter Kirsher for a longer or -shorter-time before the execution of the will was of unsound mind,, yet, if you further find by a preponderance of the evidence that at the very time of the execution of the will the said Peter Kirsher was of sound mind, then your verdict will be for the defendants. But if you find, as hereinbefore-*341instructed, that at the time of the execution of the will the said Peter Ki'rsher was of unsound mind, then youi verdict will be for the plaintiffs.”
Instruction 7, referred to in the foregoing paragraph, was a general one, directing the jury what might be considered in determining the. mental condition of the deceased at the time the will was executed, and instructing as to the weight of the testimony.
Primarily, every person is presumed to be sane until the contrary is proved, and the burden of proof of insanity rests in the first instance upon the party alleging it. It x. testambn-Ftyfpre-pac' sumption, 1s eduahy as true that, when settled and general unsoundness of mind is proved, a presumption arises in favor of its continued existence. Corbit v. Smith, 7 Iowa, 65; Blake v. Rourke, 74 Iowa, 523; Bever v. Spangler, 93 Iowa, 601. But this court-has never held (and, so far as we have examined the cases, no other) that proof of insanity at a stated period, without reference to the particular circumstances connected therewith, is sufficient to authorize the inference of insanity at a remote subsequent period. Temporary mental aberration is not uncommon, and the causes thereof are numerous, among which, science and observation have taught us, are all forms of violent disease, including appolexy. In Trish v. Newell, 62 Ill. 196 (14 Am. Rep. 79), it is said, “It is no more a presumption of law that one rendered unconscious and incapable of mental action by a stroke of paralysis will continue so for four months thereafter, than that he would if the same effect was produced by a blow on the head.” The instruction given failed to recognize the distinction we have pointed out, and permitted the jury to infer insanity at the time the will was excuted from “the testator’s mental condition immediately following the stroke, and without requiring it to find that there was then a settled condition of mental unsoundness. In other words, the jury was told that if it *342found mental unsoundness two years before, whether habitual or temporary, it could presume that Mr. Kirsher was insane when he executed his will. That such is not the rule is held by our own cases. supra, and by the weight of authority. JBuswell on Insanity, 213. Lawson on Presumptive Evidence, 179.
The court gave the following instruction: “A number of physicians have been called as medical experts; that is, they have given their opinions, baser] upon hypo-2, expert instruction, thetical questions put to them. You will carefully consider this testimony, and give it the weight you may' think it justily entitled to. The weight and value of such testimony depends upon whether the statements of facts, of which such experts have not personal knowledge, but which they accept as true for the purposes of answering the question propounded them, are in material and important particulars correct, fair, and impartial, then such testimony may [be of great value, but if you find such statements of facts are in material and important particulars incorrect, unfair, partial, and untrue, , then you should attach little or no weight to'such testimony.” It hardly needs the citation of authority to show that this instruction is erroneous. The practical experience of lawyers and courts has so often demonstrated the fact that a very slight change in the hypothesis will change completely the answer of an expert witness, that it is unnecessary to say that the value of answers to such questions must be based solely upon the truth of the facts upon which they are based, and that, if the facts are not found to be as stated, the answers are of no value, and cannot be considered at all. Hall v. Rankin, 87 Iowa, 264; Schouler on Wills (3d Ed.) section 207. If the instruction had not said that “little or no weight should be given to the testimony if the hypothesis were found untrue,” it might be said that the jury would understand that it was not to consider answers to such quesiion, under *343the rule stated in Bever v. Spangler, supra, where no instruction on the subject was given. But the language here, by implication, at least, authorized the consideration of the testimony of the experts in any event.
The instructions given as to the weight of the testimony and the credibility of the witnesses were without error, and, in a general way, covered the questions. If 3. instruction. the defendants desired more specific ones any phase 0f the case, they should have asked them.
The issue of undue influence was withdrawn by an instruction, and it was not necessary to state further that the evidence on that subject was also withdrawn. The 4. withdrawal of evidence, iury understood that as well without further clii'eotion as it would had a volume been written on the subject.
We doubt whether Dr. Moenck was in any proper sense so connected with the deceased as to become his physician, within the meaning of section 4608 of the Code; but, if he was his testimony was competent. Winters v. Winters, 102 Iowa, 53. The weight to be given his testimony, under the circumstances, was a matter for the jury to determine.
It is claimed that much of the testimony given by the plaintiffs themselves was incompetent, under Code, section 4604.' Without calling specific attention to each witness against whom the objection is made, or reciting the testimony objected to, we may say that we find no serious error in any of the rulings. It is, perhaps, strange that witnesses could testify that the deceased did not recognize them or seem to know them, when he was totally blind and hence could not see them, and when, as they say, they had no conversation with him; but the court cannot say, as a matter of law, t at the witnesses had conversations with the deceased at the time in question, when they all testify that they did not. It was for the jury to say *344whether, under the facts stated, the deceased did or did not, or could or could not, recognize them. Denning v. Butcher, 91 Iowa, 425.
Objection was made to answers of Mary Pontious as to what she did with her personal earnings during her s. personal transaction. father’s life. The answers were competent, unqer the rule in McElhenney v. Hendricks, 82 Iowa, 659.
There was no error in admitting the testimony of Edwards. He had seen and observed the .deceased on different occasions during the time it is claimed he was insane, 6. non-expert opmion. and) after stating the facts connected therewith, was permitted to say that he thought “he was not of sound mind.” In re Fenton's Will, 97 Iowa, 198.
Dr. Burkhart, who at one time had. been the family physician of the deceased, testified that he attended him as such up until February, 1894; that from 1892 until that 7 Evidence: value of. tíjne was physically and mentally weak, and became worse all of the time He detailed facts on which he partially based his conclusion,, and was then permittéd to testify that when he last treated him he thought he was “unsound in his mind.” This testimony was competent, and its weight for the jury. The doctor was evidently mistaken about treating Mr. Kirsher for a ■stroke of paralysis in 1892, or at any other time, but this only affected the value of his testimony.
The hypothetical questions put to the expert witnesses, were substantially in accord with the well-settled rule. They need only be based upon what the evidence tends to 8. hypothetical questions. Prove> aQd they nee(i not cover all of that. A caref uj[ examination of the questions and evidence shows that the various elements of the questions are supported by the evidence.
The court taxed all of the costs, except attorney’s fees, to the estate, and the appellants complain because their *345attorney’s fees were not so taxed, while the contestants 9 taxation of costs. complain because the other costs were so taxed. We shall not disturb the order, however. As we have heretofore said, all of the parties to this litigation are children of the deceased. The property was substantially all given to the defendants by the will, which was duly executed, and admitted to probate, after due notice, without objection or question. It was thereby determined that it was the valid last will an-d testament -of the deceased, and until the order was set aside it was ■conclusive upon all parties. When this action was commenced to determine the true character of the will,, it may be presumed, at least, that the proponents were acting in good faith in defending, and with no other object in view than to settle that question according to the very right of the matter. If this is true, there is no reason apparent to us why the estate should not bear the reasonable court expense incident to such a trial. There is ■certainly no impropriety in attempting to uphold what is in good faith believed to be a valid will, and to tax the costs to the defendants for so doing would be to punish them unjustly. On the other hand, if each party pays its •own attorneys, the matter is equalized, and neither will suffer in the final distribution of the estate. On this subject generally, see In re Carman's Will, (Iowa) 48 N. W. Rep. 985; Meeker v. Meeker, 74 Iowa, 352; Howard v. Smith, 78 Iowa, 73; Perkins v. Perkins, 116 Iowa, 253.
The appellants’ motion to transfer the case to the ■equity side of the docket for trial was overruled, and error is assigned thereon. Code, section 3283, provides io. setting bateeof aewm. that, after a will has been produced, the clerk or court shall fix a day for proving it, and that, “ when the probate of a will is contested, either party to the contest shall be entitled to a jury trial thereon.” Section 3296 of the Code provides that the pro-late of a foreign or domestic will “shall be conclusive as *346to the due execution thereof, until set aside by an original Or appellate proceeding.” The contention of the appellants is that whenever a will has been probated, whether with or without a contest, the order admitting it to probate can only be set aside- in an equitable action. The exact reverse of this contention was held in Leighton v. Orr, 44 Iowa, 679, and it has been the practice in this state-for morO than a quarter of a century; and, if there were no judicial construction of the statute, we would hestitate-long before departing from so venerable a rule. Nor do-the cases cited by the appellants seem to announce a contrary doctrine. The appellants’ motion to dismiss the cross-appeal is overruled.
The appellees filed an amendment to the abstract, containing about one hundred and fifty pages of testimony, set forth largely in the form of questions and answers. This-form was not necessary for a proper understanding of the matter testified to, and greatly extended the record and' amendment, the principal part of which is a repetition of the appellants’ abstract, varied a little in manner of expression. The motion to strike this amendment is overruled, because we have found therein some material testimony not found in the original abstract; but there is so little of this that we order three-fourths of the cost of the-amendment taxed to the appellees.
The judgment will stand aeeirmed on .the plaintiffs” appeal, and on the defendants’ it is reversed.
Bishop, O. J., taking no part.