During the trial the court erroneously kept from the jury much of the evidence that the caveators had a right to use in one way or another, and a new trial is required. The court’s most serious and pervasive error, accomplished by a series of rulings, was in keeping from the jury evidence or information in any form which indicated that the propounder and some of his witnesses, who testified that the testator was of sound mind when the will was executed, had expressed a contrary opinion or taken a contrary position four years earlier. The contrary opinion or position was expressed or taken in connection with a special proceeding filed in court 1 December 1976 for the purpose of having the testator declared mentally incompetent to manage his affairs. The petition in that proceeding, signed by the propounder and caveators alike, was drafted by Attorney Thomas E. Strickland, who also prepared the will, testified as a witness for the propounder, and wrote some group letters to the caveators and propounder about the proceeding in which the testator’s inability to understand certain important business problems that he then had was mentioned. Before the trial began, pursuant to an oral motion in limine made by the propounder, the court entered an order prohibiting the caveators from alluding to the proceeding in their opening statement or during the cross-examination of the propounder. The order was later expanded to prevent the caveators from questioning propounder as to his opinion of the deceased’s mental capacity in 1976 and from cross-examining Attorney Strickland about either the special proceeding or the letters that he wrote in connection with it. The court’s only basis for all these orders was that the events involved took place four years before the will was made and were therefore too remote to bear upon the testator’s mental capacity when he executed the will.
*535Though the ability of a person to make a will is determined as of the date of the will’s execution, evidence concerning his mental condition at other times is admissible if it tends to show the testator’s mental capacity at the time the document was executed. “Evidence of mental condition before and after the critical time is admissible, provided it is not too remote to justify an inference that the same condition existed at the latter time.” 1 Brandis, N.C. Evidence § 127 (1982). Whether evidence of a testator’s mental condition at another time is too remote depends upon the circumstances involved. As Justice Brogden, writing for our Supreme Court, said: “No precise or mathematical definition can be fashioned. . . . The best that appellate courts can do in dealing with the subtle processes of the mind is to interpret facts in such cases by the rule of reason and common sense.” In re Will of Hargrove, 206 N.C. 307, 311-12, 173 S.E. 577, 579-80 (1934).
The excluded evidence tended to show that in 1976 when he was 86 years old the testator’s mental faculties had deteriorated to the point of deficiency because of the aging process and chronic ill health; and because of the circumstances involved it clearly justifies an inference, we think, that the testator’s mental condition had not changed for the better four years later. Thus, the evidence was not too remote to bear upon the issues being tried and all of the exclusionary orders entered on that basis were erroneous. But even if the excluded evidence had been too remote to be received for substantive purposes, it was not thereby automatically rendered unusable for all purposes, as the trial court apparently assumed. No citation of authority is needed for the elemental proposition that under our system of jurisprudence, depending upon the circumstances of the case, evidence that is inadmissible as substantive evidence is often admissible or usable for other purposes. One proper purpose otherwise inadmissible evidence is frequently used for is to undermine the credibility of opposing witnesses; and remote or not, caveators had a right to cross-examine the propounder and his witnesses about the contrary positions they had taken and the contrary statements they had made concerning the testator’s incompetency at an earlier time. 1 Brandis, N.C. Evidence §§ 42, 46 (1982).
Also, apparently, on the grounds of remoteness, the court erroneously refused to permit the caveators to introduce into evidence duly authenticated records of the testator’s hospitaliza*536tion on some fourteen different occasions between September, 1971 and January, 1977. Those records tend to show that throughout that period the testator was suffering from “generalized” or “cerebral” arteriosclerosis. This condition, as is commonly known, is associated with the aging process, is not reversible, tends to progress once it starts, and can impair one’s mental and physical faculties. The records also tend to show that at different times during the several year period covered by them that the testator suffered from several other serious and debilitating illnesses, some of considerable duration, including hepatitis, diverticulosis of the colon, osteoarthritis, carcinoma of the pancreas, pye-lonephritis, prostatitis and urinary bladder retention. That testator’s circulatory system had been impaired and probably getting worse for several years has some tendency to show that by the time the will was written he was incapable of executing it. And that his health generally had been breaking down for the same period complements and strengthens the evidence of impaired circulation, and also tends to show that he was capable of being imposed upon and unduly influenced by others.
Though the court permitted the caveators to introduce part of the record of testator’s last hospitalization, covering the period August 15, 1980 to August 22, 1980, the nurses’ notes were erroneously excluded therefrom. The basis for this exclusion is not recorded and we know of no sound basis that could have been stated. The evidence certainly is not too remote — the period covered began but nine days after the will was executed — and had the tendency to show that the confused, disoriented condition recorded therein also existed nine days or so earlier when the will was signed. The court also erroneously excluded certain other evidence offered by the caveators, which tended to show that the testator had treated his feeble spouse and her property irrationally in 1976. Under the circumstances, such evidence was relevant to the testator’s mental capacity when the will was executed. In re Will of Hinton, 180 N.C. 206, 104 S.E. 341 (1920).
 Finally, the order directing a verdict against the caveators on the undue influence issue was erroneously entered. In addition to the testator’s advanced age and feeble health, the caveators’ evidence tended to establish that: When the will was written the testator was living in the propounder’s home; he was estranged from his other children, the caveators, and they had little access *537to him; he was taken to the lawyer’s office by the propounder and executed the will in his presence; the will in effect disinherited five of his six children for no known reason. This evidence was sufficient in our opinion to establish a prima facie case on this issue. See In re Will of Mueller, 170 N.C. 28, 86 S.E. 719 (1915).
This matter is remanded to the Superior Court for a new trial consistent with this opinion.
Chief Judge VAUGHN and Judge WHICHARD concur.