Cousin v. Cousin

Full title: COUSIN v. COUSIN

Court: United States Court of Appeals, Eighth Circuit

Date published: Nov 13, 1951

192 F.2d 377 (8th Cir. 1951)

Opinion

No. 14401.

November 13, 1951.

E. Marshall Thomas, Dubuque, Iowa (Francis J. O'Connor, Dubuque, Iowa, on the brief), for appellant.

H.F. Reynolds, Dubuque, Iowa (Louis F. Fautsch, Dubuque, Iowa, on the brief), for appellees.

Before SANBORN, THOMAS, and COLLET, Circuit Judges.

COLLET, Circuit Judge.

The jurisdiction of this action to set aside a will results from diversity of citizenship and the amount involved. The plaintiff is the son of the decedent. Defendants are decedent's second wife and her two daughters by a former marriage. The purported will gave the entire estate to defendants with the exception of $1 to plaintiff. The parties will be referred to as they appeared in the trial court. There was a jury trial. At the close of all the evidence the trial court directed a verdict for the defendants sustaining the will. The sole issue presented at the trial was whether the signature of the testator to the will was a forgery. The primary question for determination on this appeal is whether the plaintiff made a submissible case, and the principal question in determining that issue is the probative force to be given to opinion evidence by a handwriting expert and two lay witnesses that the signature was not that of the alleged testator, and whether that testimony and certain facts and circumstances were sufficient to present a jury question, in the face of the direct testimony of the subscribing witnesses that the signature was genuine. The issue involves the ascertainment and application of the law of Iowa.

By statute the Legislature of Iowa has provided that evidence respecting handwriting may be given by comparison made by experts or by the jury with writings of the same person which are proved to be genuine. I.C.A. § 622.25. Hammond v. Wolf, 78 Iowa 227, 42 N.W. 778. But the Iowa Supreme Court has, in giving effect to that statutory provision, placed its own evaluation on the comparative weight to be given such testimony when it is opposed by direct testimony that the signature was genuine.

The first case we find on the subject is that of Borland v. Walrath, 33 Iowa 130. That case involved the genuineness of the signature to a mortgage. The party whose signature was in question was living at the time of the trial and testified that she did not sign or acknowledge the mortgage. She supported that testimony by evidence of facts and circumstances consistent with her denial of the signature, such as the condition of her health at the time, her absence from the place where she was said to have signed it, and the absence of any circumstance, as the court put it, "tending to cast suspicion upon her credibility." The evidence of the plaintiffs, who were seeking to enforce the mortgage and who were contending that the signature was genuine, consisted of the testimony of the notary public who certified to the acknowledgment, and of "certain skilled witnesses" who testified to the genuineness of her alleged signature upon a comparison thereof of signatures admitted to have been made by her. The court noted that the notary did not pretend to state from memory that the acknowledgment was made before him, but based his testimony that the acknowledgment was genuine upon his habit of making certificates of acknowledgment only when the acknowledgments were actually made before him. The court then said, 33 Iowa loc. cit. 132: "The case assumes this attitude: The direct and positive evidence of the defendant, denying her signature, is to be weighed against the certificate of acknowledgment and the uncontradicted evidence of experts as to the similarity of the signature in question to defendant's genuine writing, together with their opinion based thereon."

It was conceded that the certificate of acknowledgment made a prima facie case of the genuineness of the signature and that the party seeking to deny the signature "ought to make a clear case against the certificate of the officer in order to overthrow the instrument." And it was further said that "Public policy demands that instruments in writing pertaining to the titles of real estate, which are authenticated in the manner pointed out by the law, should not be lightly set aside. But they cannot be sustained against the positive and explicit evidence of credible witnesses." The court goes on to say that, 33 Iowa loc. cit. 133:

"The evidence as to the genuineness of the signature, based upon the comparison of handwriting and of the opinion of experts, is entitled to proper consideration and weight. It must be confessed, however, that it is of the lowest order of evidence, or of the most unsatisfactory character. It cannot be claimed that it ought to overthrow positive and direct evidence of credible witnesses who testify from their personal knowledge. It is most used and is most useful in cases of conflict between witnesses as corroborating testimony.

"On the one hand we have the signature to the mortgage sustained as genuine by the certificate of acknowledgment and by the comparison of handwritings, upon which are based opinions of experts; on the other we have the positive evidence of the defendant, whose credibility is not doubted, corroborated in a degree by other testimony. In our opinion the preponderance is in favor of defendant. We are free to admit that we are not without doubts, and it is probable that questions of this character can never be determined with absolute convictions of certainty. We feel, however, that it is safer to give credit to the positive evidence of a credible witness than to disregard it upon presumptions that are not of the highest order."

The action was before the court without a jury. The trial court found that the signature was not genuine. The Supreme Court appears to have weighed the evidence and found with an expressed hesitancy that the preponderance was against the expert testimony and the notary's certificate. In the comparison of this case with later expressions of the Iowa court, particularly Butman v. Christy, 197 Iowa, 661, 198 N.W. 314, ante, we deem it of significance that in this case the court treated the expert testimony supported by the notary's certificate as sufficient to present a doubtful question on the weight of the evidence.

The next case called to our attention dealing with the question was Whitaker v. Parker, 42 Iowa 585. That case involved the genuineness of a signature to a note. The defendant denied that he signed the note. The signature upon the note was compared with other signatures of his, admitted to be genuine, and witnesses who were shown to be experts testified for the defendant, declaring that in their opinion the name signed to the note was not written by defendant. The plaintiff testified that the note was signed by defendant in his presence, and he was corroborated to some extent by facts and circumstances disclosed by the evidence. The defendant's denial of his signature was also corroborated by facts and circumstances testified to. That was a jury case, and the court instructed the jury that although the statute provided that evidence respecting handwriting may be given by comparison made by experts or by the jury with writings of the same person which are proved to be genuine, yet evidence of this character, while proper for the consideration of the jury, was of the "lowest order of evidence, or evidence of the most unsatisfactory character." And that "It cannot be claimed that it ought to overthrow positive and direct evidence of credible witnesses who testify from their personal knowledge, but it is most useful in cases of conflict between witnesses as corroborating testimony." Evidently that instruction was drawn in accordance with the language of the opinion in Borland v. Walrath, supra. The jury found the signature to be genuine. The Supreme Court approved the instruction and affirmed over the protest of plaintiff that the instruction practically destroyed any value of expert testimony. In its opinion the court recognized the fact that under the Iowa statute the opinion of handwriting experts was admissible and that the weight to be given that testimony "is to be determined just as the force of all other evidence." It went on to say: "The effect, then, which all evidence has upon the mind is determined by observation and experience, the only original instructors of wisdom. These teach that the evidence of experts is of the lowest order and of the most unsatisfactory character." Citing Borland v. Walrath, supra.

In Hammond v. Wolf, supra [ 78 Iowa 227, 42 N.W. 779], the question involved was whether the signature to a note was genuine, which the defendant whose name appeared on the note denied. The note was lost after suit was brought on it, and two lawyers, who as partners brought the suit and had examined the note and the signature, were offered as witnesses to testify that the signature appearing on the lost note sued on, when compared with admittedly genuine signatures of the defendant, was genuine. Their testimony was excluded upon the ground that "the signature in question [was] not in court." The Supreme Court held that under the statute of Iowa permitting evidence respecting handwriting to be given by comparison 42 N.W. loc. cit. 780, this evidence was admissible, but went on to observe that "The testimony of experts in regard to the genuineness of writings has not been considered as a rule to be of a satisfactory character, (Whitaker v. Parker, supra; Borland v. Walrath, supra;) but it is admissible for what it is worth."

The next case cited in the chronological development of the law of Iowa on this question is that of Butman v. Christy. This case has a long and interesting history in the Iowa courts. It first appears as Christy v. Buttman, 194 Iowa 262, 189 N.W. 671, wherein the question for disposition by the Supreme Court was whether the evidence was sufficient to warrant the submission of the question of undue influence in the making of the will to the jury. The jury had found that the will was obtained as a result of undue and improper influence on the testator, and the judgment of the court below was that the will be set aside. The Supreme Court reversed and remanded the case on the ground that the evidence did not make a submissible case on the issue of undue influence. When the case was retried, the charge of undue influence was dropped but by amendment of the pleadings it was charged that the signature to the will was a forgery. Again the jury found against the validity of the will and the case went back to the Iowa Supreme Court on the question, among others, of the propriety of the trial court's refusal to permit cross-examination of an expert witness on handwriting to show that he was paid for testifying and how much. While the case was reversed on that ground, the court reviewed the evidence offered in support of the charge of forgery and stated in the concluding paragraph of its opinion "that, unless upon a retrial additional evidence should be produced, it would be the duty of the court to direct a verdict for proponents." In view of that statement and the subsequent history of the case, it is important to note what the evidence was on this occasion in support of the charge of forgery. That evidence was summarized by the court in the following language, Butman v. Christy, 197 Iowa 661, 198 N.W. 314, 317, 318: "The only issue submitted to the jury was the question of the genuineness of the signature of Retta C. Angell to the will. Upon the issue the only evidence offered was the testimony of the subscribing witnesses and the executor, who testified that the will was signed by the testatrix in their presence and the testimony of an expert on handwriting, who testified that, by comparison with three proved signatures, it was not her signature, and two proved signatures and a photographic copy of a third offered for the purpose of comparison by the jury."

In commenting upon the value of the testimony of the expert on handwriting, and the comparison of signatures, the court said, 198 N.W. loc. cit. 318: "While the only evidence that tends to support the verdict is made competent by statute, yet its value is so slight and its character so unsatisfactory that juries are usually and properly instructed that it ought not to be permitted to outweigh the positive and direct testimony of credible witnesses who testify from personal knowledge. The jury in this case was so instructed, but it appears to us to have entirely disregarded the admonition. There is not a single extraneous fact or circumstance that in any degree tends to corroborate the expert witness in saying that the signature is not genuine." (Italics supplied.)

The case was remanded and retried. Again there was a verdict of a jury against the validity of the will, on the sole ground that the signature thereto was a forgery. On the third appeal to the Supreme Court it is reported as Buttman v. Christy, Iowa, 208 N.W. 218. The questions presented on the third appeal were whether the trial court should have directed a verdict for proponents, whether the verdict was a result of passion and prejudice, and whether the verdict was contrary to the evidence and instructions of the court. In disposing of the appeal the Supreme Court recalled that it had theretofore directed the lower court to dismiss the action upon a retrial unless additional testimony was introduced. It then stated: "One handwriting expert was added to the list of witnesses, and three persons interested with the contestants testified that they were familiar with the handwriting of testatrix, and that the signature in question was not hers. This testimony was not introduced upon the previous trials. We shall not undertake to review the evidence in detail. There is nothing in the record to indicate passion and prejudice on the part of the jury, or that the proponents did not have a fair trial. There is no reason to believe that a fourth trial would result differently from the prior trials, and the evidence was sufficient upon the last trial to require submission of the issue to the jury. (Italics supplied.) There must be an end to litigation. The court cannot assume to exercise the functions of a jury. (Italics supplied.) So far as the record shows, the case was fairly tried. No reversible error is shown in the record, and the verdict of the jury is affirmed."

The question was before the Iowa Supreme Court again in Keeney v. Arp De La Gardee, 212 Iowa 45, 235 N.W. 745, 748. A jury instruction (referred to as No. 8) was given by the trial court in substantially the same form as that heretofore referred to in Whitaker v. Parker, and included the statement that "while it is proper to consider such evidence, [expert testimony on handwriting] and it is proper to remark that it is the lowest order of evidence or evidence of the most unsatisfactory character, being in fact the result only of a comparison of the controverted with the genuine signature of the defendant and is less satisfactory in character than is positive testimony and such as ought not overthrow the positive and direct testimony of a credible witness who testifies from personal knowledge, but it is most useful in cases of conflict between witnesses, as corroborating testimony." The complaint was, as it was in the Whitaker case, that the instruction was unduly hostile to expert testimony and to the weight to be given thereto. The court disposed of that objection upon the ground that because the complaining defendant had given direct testimony that his signature was not genuine, in contradiction of the expert's testimony that the signature was genuine, the instruction was not prejudicial to him and that the defendant had invited the giving of the instruction by a request for one substantially similar thereto. The court then went on to observe that the language of the instruction as given conformed to the previous holdings of the court and was in literal conformity to the one given in Whitaker v. Parker, commenting that "Its correctness has been uniformly held ever since." Citing Hammond v. Wolf, 78 Iowa 227, 42 N.W. 778. But the court in this case added the following language which we deem of significance: "By our foregoing pronouncement we are not to be understood as holding that the rule set forth in instruction 8 is to be applied unqualifiedly to all forms of expert opinion evidence in forgery cases. We are not unmindful of the fact that, since our original pronouncement in the Whitaker Case [Whitaker v. Parker], 42 Iowa, [585], 586, much progress has been made in the means and methods of detecting forgeries. By microscopic inspection, and by magnified photographs and sometimes by chemical tests, the expert may be able to discover and to demonstrate the existence of facts which negative the genuineness of the signature. Such facts, when proved, become substantive evidence, rather than mere expert opinion. How far, if at all, the pronouncement in instruction 8 should be applied or qualified in such case we have no occasion now to decide. Nor do we foreclose such question. It is not involved in this record."

In Brien v. Davidson, 225 Iowa 595, 281 N.W. 150, 152, 282 N.W. 480, the action was by representatives of an estate to recover assets assigned to others by the decedent by written instruments, before his death, which the estate claimed were forgeries. The assignees of the property involved in the assignments contended that the signatures thereto were valid and genuine. The questioned signatures and admittedly genuine signatures were enlarged by photographic or other processes, and differences in shading, in line quality, in line tremor or the absence thereof, and many other details were pointed out by expert witnesses on handwriting. Three such experts testified that the signatures were genuine and three experts testified they were forgeries. The trial court found that the signatures on the assignments were forgeries. In that finding he was materially aided by the following circumstance. It was found that of the six questioned signatures, five of them could be perfectly superimposed one upon the other, and that the sixth one could be so superimposed except for a slight break in the signature, explainable by the paper slipping during the process of forging. It was also noted that some alterations in each of the instruments appeared to have been made on two typewriters which were kept in the office of one of the defendants charged with the forgeries, and that the dates on each of the questioned documents, except one, had been altered. The questioned assignments were all acknowledged. The court held that the notary's certificate was presumptively true, that the presumption was a strong one, and that while the certificate was not conclusive, clear and convincing evidence was necessary to contradict it. The proponents of the validity of the signatures, relying upon the above stated presumption resulting from the notary's certificate, attacked the sufficiency of the expert handwriting testimony to overthrow the presumption. In advancing that contention the plaintiffs asserted in the language of the Whitaker case "that evidence of expert witnesses based on comparison of signatures is of the lowest order and most unsatisfactory, is not conclusive, and is the merest guesswork." In disposing of that contention the court referred to its former decisions in Murphy v. Murphy, 146 Iowa 255, 125 N.W. 191; Keeney v. Arp De La Gardee, supra, and Fekete v. Fekete, 323 Ill. 468, 154 N.E. 209; In re Burtis' Will, 43 Misc. 437, 89 N.Y.S. 441, as illustrative of the recognition by the courts of the progress that has been made in the scientific detection of forgeries by experts. Referring to the Keeney case it said: "We recognized the advance in the field of expert testimony when, in Keeney v. [Arp] De La Gardee, 212 Iowa 45, 235 N.W. 745, Judge Evans, speaking for the court, had this to say:" — quoting what we have heretofore quoted above from the opinion in the Keeney case on the advancement of the science. It then quoted at some length from the opinion In re Burtis' Will, supra, the following portion of which we deem significant: "But this does not signify that the evidence of those witnesses [expert witnesses] must be disregarded because they disagree. Nor is it important specially which side has the greater number of expert or lay witnesses sworn in his behalf. It is the nature or character of the testimony given by the witness which is important. In the case of expert witnesses, their opinions are valuable only in so far as they point out satisfactory reasons for the ultimate conclusion of the witness. If the witness simply testifies that he believes the signature genuine, or not genuine, as the case may be, and gives no reasons for reaching his conclusion, his opinion is valueless, and the court will not consider it. If he gives reasons for his opinion, then it is the duty of the court to examine into and analyze those reasons, and determine the correctness or incorrectness of the opinion, and not simply consider the conclusion of the witness alone."

The court affirmed the trial court's judgment holding that the signatures were forgeries upon the expert testimony and the corroborating facts noted, in spite of the presumption arising from the notary's certificate.

In the case of In re Early's Estate, 234 Iowa 570, 13 N.W.2d 328, the will involved was made in 1915. By a combination of peculiar circumstances, the question of its validity did not arise until 1942, when application was first made that it be admitted to probate. The testator died in February, 1917. The attesting witnesses were both dead at the time of the trial. It was not denied that the signature of the testator was genuine. The controversy was whether the signatures of the attesting witnesses were genuine and hence whether the will had been properly attested. The genuineness of the signature of the attesting witnesses was established by the testimony of a long-time associate of each of the attesting witnesses that it was their opinion that the signature of each of the attesting witnesses was genuine. The Supreme Court affirmed the judgment of the trial court holding the will valid upon that evidence of the genuineness of those signatures.

The last case we find on the subject before us is that of Hoover v. Hoover, 238 Iowa 88, 26 N.W.2d 98, 101. The validity of the will involved in that case was challenged upon the ground, among others, that the signature to the codicil was a forgery. At the close of the evidence, the trial court directed a verdict sustaining the validity of the codicil, and plaintiff, on appeal, contended that the question of the validity of the codicil should have been submitted to the jury, because a handwriting expert witness testified that the signature to the codicil was a forgery. On that question the court said: "We have held that where the testimony of an expert witness on handwriting stands alone, and there are no extraneous facts or circumstances tending to support his opinion, the trial court should direct the verdict where there is direct and positive testimony of the subscribing witnesses who saw the testatrix sign the will in question." (Citing Butman v. Christy, 197 Iowa 661, 198 N.W. 314.) "In the instant case the will was a proven or probated instrument. The attestation clause and signatures of the subscribing witnesses refute the handwriting expert's opinion. There are no facts or circumstances that tend to corroborate the expert. (Italics supplied.) Other evidence produced by plaintiffs seems to oppose the expert testimony. * * * We will not review the testimony [of the handwriting expert] for there are other considerations that support the trial court's ruling."

The court then held, for reasons not material here, that contestant did not have a contestable interest in the codicil and made no further comment upon this particular contention. It was further contended, however, that since the handwriting expert also testified that two pages of the will were written at a different time than the page which contained the decedent's signature that the question of whether these two pages had been substituted should have been submitted to the jury. The expert's testimony on that subject was to the effect that all of the pages were typed on the same typewriter but that the inking on the last page containing the signature was lighter, that the left-hand margin was closer to the edge of the paper than on the two pages in dispute, and that certain perforations appeared on the pages in dispute which did not appear on the last page containing the signature. The only significance of the court's ruling on this contention to our problem lies in the fact that the court pointed out that this was not a proper subject for expert testimony because such things as those were as easily determined by the jury as by the so-called expert. And we would not deem that holding of significance were it not for the fact that it throws some light upon the changed attitude of the Iowa courts to expert testimony, in the light of the development of the art of scientific examination of signatures. In that respect the result reached in the Hoover case and the reason therefor give some indication that one of the reasons for the low opinion of handwriting experts' testimony expressed in the early cases was the fact that in the early days of the handwriting experts, their opinion was not fortified by or with modern methods of examination and comparisons but consisted almost entirely of the bare statement of the expert's opinion that the signature was genuine or not genuine, as a result of comparison of signatures, which was practically as easy for a nonexpert to determine.

The foregoing analysis of the Iowa cases examined results in the following conclusions:

First, that the opinion of a handwriting expert is admissible by statute and must be given consideration by the trier of the facts;

Second, that the testimony of a handwriting expert witness, standing alone, with no extraneous facts or circumstances tending to support his opinion, will not make a submissible case when opposed by direct and positive testimony of the subscribing witnesses who saw the instrument signed;

Third, that when the handwriting expert's testimony is supported by the testimony of other persons familiar with the handwriting in question, such evidence does present a submissible issue. Buttman v. Christy, Iowa, 208 N.W. 218.

Fourth, where handwriting expert testimony is supported by other facts and circumstances, a submissible case is made. Buttman v. Christy, Iowa, 208 N.W. 218; Brien v. Davidson, supra; Keeney v. Arp De La Gardee, supra.

In view of the foregoing conclusions as to the state of the law in Iowa on the subject in question, we must necessarily review briefly the evidence in this case to determine whether it meets either of the requirements above noted.

In testing the sufficiency of the evidence to require the submission of the issue to the jury, we must view the evidence in the light most favorable to the plaintiff, as there is no way to determine what the jury's reaction to that evidence would have been when a verdict is directed. Necessarily, therefore, we state the facts from that viewpoint.

John B. Cousin was married to Florence Cousin in 1917. To that marriage one child was born, the plaintiff, John T. Cousin. The family lived on a farm near Dubuque, Iowa, and the evidence shows a normal affectionate relationship between the father and son. In May, 1930, John B. and Florence Cousin were divorced, upon the petition of the latter. The mother was awarded plaintiff's custody and $4,500 alimony. Thereafter, plaintiff and his mother lived together in Dubuque until 1936 and then moved to the state of Connecticut. In January, 1931, Mr. Cousin married Mae Annette Cousin. She had two daughters by a former marriage. There were no children born to this marriage. The son, plaintiff John T. Cousin, was in the military service for several years during World War II, serving 42 months overseas. Upon his return from the army, he and his wife visited his father in Dubuque in 1945 and again, with their young son, in 1947. There never was any rift or unpleasantness between him and his father. John B. Cousin died March 13, 1949. The purported will of John B. Cousin was offered for probate by his widow, Mae Annette Cousin, on April 25, 1949. It was entirely in her handwriting, with the exception of three blanks left for the insertion of the name "John B. Cousin". One of those blanks was in the opening line of the will, one in the attestation clause, and the other was the blank for the signature to the will. Mae Annette Cousin testified that she wrote the will at the farm home on February 8, 1933, which is the date of the purported signature of the testator. She says that she was in the home during all of that day but did not see the will signed. One of the subscribing witnesses to the will was her brother-in-law and the other a close friend of her family. After Mr. Cousin's death, and prior to the offering of the will for probate, Mae Annette, on March 21, 1949, filed a sworn petition for administration in the District Court for Dubuque County, Iowa, in which she stated that: "John B. Cousin late of said county died therein on or about the 13th day of March, 1949, leaving no Will, so far as known, or believed by the petitioner. * * *" Upon the trial the foregoing facts were developed by the evidence. In addition thereto, a handwriting expert, shown to be competent, testified at length concerning the genuineness of the signature to the will. Many admittedly genuine signatures of John B. Cousin were produced. Enlarged photostatic copies of them were compared with the signature appearing on the will, and based upon a number of characteristics of the letters in the admittedly genuine signatures and the signature to the will, the handwriting expert testified the signature to the will was not genuine. That the name John B. Cousin appearing in the places left blank in the first line and the attesting clause by Mae Annette when she prepared the will was written by the same person who signed the will and that it was not written by John B. Cousin. Two lay witnesses shown to be familiar with the signature of John B. Cousin testified that the signature to the will was not his. There was also evidence that on two occasions prior to his death and after the date of the purported will, Mr. Cousin stated that he had no will. In addition to those facts, plaintiff insists that the signatures of the attesting witnesses appear brighter and fresher than other handwriting in the will, including the signature to it, and that that fact supports the conclusion of forgery. Opposed to that evidence was the testimony of the subscribing witnesses and evidence offered by defendant in explanation of extraneous facts relied upon by plaintiff.

The testimony of the subscribing witnesses and other evidence of defendants may have been sufficient to convince the jury that the signature was genuine, had that question been left to the jury. With that we cannot now be concerned. It is our province to determine whether plaintiff's evidence, viewed in the light most favorable to him, would have supported a verdict in his favor. And in determining that question we must bear in mind the rule of law in Iowa that it requires more than a scintilla of evidence to support a verdict. In re Kenny's Estate, 233 Iowa 600, 10 N.W.2d 73. But certainly there was more than a scintilla of evidence in support of the charge of forgery. And it appears reasonably clear to us that the evidence in this case measures up to the requirements of the Supreme Court of Iowa on the third appeal of the Buttman case, 208 N.W. 218, and also the requirements stated by that court in Brien v. Davidson, supra, and Keeney v. Arp De La Gardee, supra.

It has been heretofore observed that on the second appeal of the Butman case the Supreme Court directed the trial court to direct a verdict in favor of the validity of the will if, upon retrial, there was no evidence supporting the charge of forgery other than the unsupported testimony of the handwriting expert, and that when the case went back on the third appeal, the only additional evidence was the testimony of another handwriting expert and three persons interested with the contestants, who testified that the signature was not genuine.

We have the positive declaration of the Iowa Supreme Court in Brien v. Davidson, supra, 281 N.W. loc. cit. 153, that "Nor is it important specially which side has the greater number of expert or lay witnesses sworn in his behalf." We must conclude that the all-important factor is whether a handwriting expert's testimony is supported by other evidence. And the opinion of the Supreme Court on the third appeal of the Buttman case, in our judgment, definitely holds that it is supported when accompanied by the testimony of other witnesses on handwriting competent to testify on the subject.

In addition to the support we find in the opinion on the third appeal of the Buttman case for the conclusion that the evidence was sufficient to make a submissible case, the sufficiency of the evidence to require submission to the jury is adequate under the more recent declarations of the Supreme Court of Iowa. Again, as heretofore noted, that court held in Brien v. Davidson, supra, and Keeney v. Arp De La Gardee, supra, that when the testimony of the expert is supported by other facts and circumstances, a submissible case is made. And in the Brien v. Davidson case the court quoted with approval from First National Bank of Iowa City v. Hartsock, 202 Iowa 603, 210 N.W. 919, and Jones v. McGruder, 87 Va. 360, 12 S.E. 792, 798, the statement that: "A transaction may, of itself and by itself, furnish the most satisfactory proof of fraud, so conclusive as to outweigh the answer of the defendants, and even the evidence of witnesses. The circumstances attending and following a transaction are often of such a character as not to leave even a shadow of doubt as to the real object and motive of the parties engaged in it. * * * These often furnish more conclusive evidence than the most direct testimony."

In the case at bar, in addition to the testimony of the handwriting expert and the lay witnesses relative to the lack of genuineness of the signature, the evidence on behalf of the plaintiff showed that Mrs. Cousin, who admittedly wrote the will and was in the home at the time it was supposed to have been signed, under oath stated soon after the testator's death that he died intestate; that the relationship between testator and his son was shown to have been a normal affectionate relationship of father and son; that on two occasions testator, after the date of the purported will, stated that he had made no will (which, as will be hereafter noted, may not have been admissible), and other facts which we have related above. These are facts and circumstances supporting the testimony of the expert and other witnesses on handwriting that the signature was not genuine, and bring this case within the rule announced by the decisions of the Iowa Supreme Court. We must conclude, therefore, that the evidence required the submission of the issue of the genuineness of the will to the jury.

But the defendants contend that the facts and circumstances related above are not such extraneous facts and circumstances as the Supreme Court of Iowa had in mind when it declared that expert testimony supported by facts and circumstances would make a submissible issue. It is argued that only such scientific facts as a chemical test to show that the paper upon which the will was written was not as old as the purported signature, and the like, are the kind of facts and circumstances which will support handwriting experts' testimony. We do not agree. In Keeney v. Arp De La Gardee, supra, [ 212 Iowa 45, 235 N.W. 249] the court said: "By microscopic inspection, and by magnified photographs and sometimes by chemical tests, the expert may be able to discover and to demonstrate the existence of facts which negative the genuineness of the signature. Such facts, when proved, become substantive evidence, rather than mere expert opinion." The implication from that language would seem to be that if the expert's testimony developed such "substantive evidence", that within the expert's testimony alone would be found the necessary corroboration of and support for his expert opinion. There was much testimony from magnified photographs and microscopic inspection in the expert's testimony. But whether it alone was sufficient under the above quoted language of the Keeney case to give the necessary support to his opinion we need not say, for in Brien v. Davidson, supra, the court, quoting from other cases, said: "A transaction may, of itself and by itself, furnish the most satisfactory proof of fraud, so conclusive as to outweigh the answer of the defendants, and even the evidence of witnesses. The circumstances attending and following a transaction are often of such a character as not to leave even a shadow of doubt as to the real object and motive of the parties engaged in it. * * * These often furnish more conclusive evidence than the most direct testimony." (Italics supplied.) There can be no doubt from that language that the court was speaking of a field of facts and circumstances broader than mere scientific tests and the like.

Plaintiff contends that it was error for the trial court to refuse to permit John B. Cousin's first wife to testify that she told him in 1944 that his son had been writing him letters from overseas and asking why he had not answered them. Whereupon he told her that his son's letters had been intercepted by his wife, Mae Annette Cousin, and that he immediately wrote his son a letter. The proffered testimony was incorporated in an offer of proof. The purpose of the offer was to prove the relationship existing between the father and the son.

Plaintiff further complains of error on the part of the trial court in excluding two letters written by John B. Cousin to his former wife in 1934, indicating love and affection on the part of the father for his son. These exhibits were also offered for the purpose of showing the relationship between decedent and his son.

The court admitted testimony which in its opinion had a direct bearing upon the state of mind of the decedent toward his son. This was evidently upon the basis of Iowa decisions holding that in cases involving the question of mental capacity and undue influence, evidence of this character was admissible, not to prove the fact of incapacity or undue influence, but to demonstate the state of mind of the testator or devisor. Johnson v. Johnson, 134 Iowa 33, 111 N.W. 430: In re Rogers' Estate, 229 Iowa 781, 295 N.W. 103; Graham v. Courtright, 180 Iowa 394, 161 N.W. 774, loc. cit. 777; Zinkula v. Zinkula, 171 Iowa 287, 154 N.W. 158.

We give to the opinion of a trial judge on questions involving interpretation of the law of his own state great deference. Our examination of the Iowa authorities on these questions leads us to the conclusion that the result of the ruling of the trial court was correct. But we express some doubt concerning the question of whether the principles supporting the admission of such testimony in cases involving mental capacity and undue influence exist in a case like this involving only the question of the genuineness of a signature to a will. It would seem that by analogy, at least, cases such as Johnston v. Linder, 168 Iowa 441, 143 N.W. 410, and Reed v. Reed, 225 Iowa 773, 281 N.W. 444, 447, in which latter case the court said, in passing upon the admissibility of the statement of a testator to the scrivener of the will, explaining the reason for making the will as she did, that: "While her statement in this respect was hearsay and was receivable only as indicating the mental condition of Mrs. Reed, * * *" tend to indicate that evidence consisting of declarations of a testator, indicative of his state of mind, would be hearsay and not admissible in an action involving only the question of forgery.

The defendants contend that the evidence of the decedent's declarations that he had made no will was inadmissible in an action involving only the question of forgery for the reason that such declarations are only admissible in cases involving mental capacity and undue influence. Defendants cite no authority on the subject from Iowa and state that the question has not been determined by the Iowa Supreme Court, but rely upon Throckmorton v. Holt, 180 U.S. 552, 21 S.Ct. 474, 45 L.Ed. 663. Throckmorton v. Holt definitely holds that such declarations are not admissible. If such evidence is admissible under either the statutes of the United States or the rules of evidence applied in the courts of the United States, or under the rules of evidence applied in the courts of general jurisdiction of the state in which the United States court is held, under Rule 56 of the Rules of Civil Procedure for the District Courts of the United States, 28 U.S.C.A., it should be admitted.

The uncertainty of the rules of evidence applied in the courts of Iowa, coupled with our policy of deference to the judgment of the trial court concerning the law of his own state, prompts us to leave these contentions concerning the admissibility of this evidence for the present to the informed judgment of the learned trial judge. Of course, if there is no established rule of evidence in Iowa on the subject and these questions should again arise, it will be our duty to determine them upon general principles. But since the cause must be remanded for a new trial, the trial judge will have an opportunity to consider those questions prior thereto with greater care than was heretofore possible in the middle of a trial. And we deem it to be in the interest of appropriate deference to the judgment of the trial court concerning matters of local law to utilize any reasonably available and practicable means of securing the trial judge's considered judgment on such matters.

Convinced as we are from our laborious examination of all the available authorities in Iowa on the subject, that plaintiff's evidence required the submission of the issue of forgery to the jury, we must, in spite of our deference to the trial judge's learning and his conclusion, remand the cause for a new trial. It is so ordered.