Decedent Louie Kalouse made the following bequests in his will executed on August 25, 1970, during a hospital stay:
Article II
I do hereby give and bequeath my organ to Louise Nespor of Oxford Junction, Iowa, to be hers absolutely.
Article III
I do hereby give and bequeath my old antique doll to Jessie Guthrie and Grace Houstman to be theirs absolutely.
Article IV
I do hereby give and bequeath all of my old albums, pictures, and photographs to my first cousins to be divided among them as equally as possible.
Article V
I do further hereby give, bequeath and devise all the rest, residue, and remainder of my property, real and personal, of every kind and character and wherever situated unto my first cousins on both my father’s and mother’s side of my family, and to Frank Nespor, in equal shares, share and share, alike, with the share going to Frank Nespor to be equal with that of my other first cousins.
Kalouse died on November 26, 1976, at age 79, survived by 24 first cousins plus Frank Nespor. Thirteen other first cousins had died before the will was executed, and five first cousins died after the will was executed but before Kalouse died. In this will construction action the trial court held that Article V creates a class gift and that the antilapse statute therefore does not apply. On appeal several heirs of predeceased first cousins contest the trial court’s holding.
I. Class gift. In construing a will we apply the following rules stated in Elkader Production Credit Association v. Eulberg, 251 N.W.2d 234, 237 (Iowa 1977):
(1) [T]estator’s intent is the polestar and if expressed shall control; (2) it must be gleaned from a consideration of all language contained in the will, the scheme of distribution, and facts and circumstances surrounding the making of the will; and (3) technical rules of construction should be resorted to only if the will is clearly ambiguous, conflicting, or testator’s intent is for any reason uncertain. See e. g., In re Estate of Spencer, 232 N.W.2d 491, 495 (Iowa 1975).
Article V left the residue of Kalouse’s property to “first cousins on both my father’s and mother’s side of my family, and to Frank Nespor, in equal shares, share and share, alike . . ..” Kalouse’s first cousins were his closest living relatives. Frank Nespor was a half-brother of a first cousin but was not a blood relative of the decedent.
If Kalouse had excluded Frank Nes-por and left his property to “my first cousins on both my father’s and mother’s side of my family,” a class gift would have clearly resulted. See Smith v. Harris, 227 Iowa 127, 131, 287 N.W. 255, 257 (1939) (“surviving children”); White v. Wachovia Bank & Trust Co., 251 F.Supp. 155, 159 (M.D.N.C.1966) (“surviving brother and sisters or their legal representatives”: “Ordinarily a gift to persons who are not named or numbered in the language of a gift but are designated therein only in general terms, as by relationship to the testator or another, is a gift to a class.”); Lacy v. Murdock, 147 Neb. 242, 246, 22 N.W.2d 713, 716 (1946) (“children”); In re Estate of Ransom, 89 N.J.Super. 224, 230, 214 A.2d 521, 524 (1965) (“grandchildren”); Green v. Green, 9 Ohio Misc. 15, 18, 221 N.E.2d 388, 391 (1966) (“lineal descendants of my son, Richard C. Green, per stirpes”); Sanderson v. First National Bank, 446 S.W.2d 720, 724, 726 *101(Tex.Civ.App.1969) (“sisters”); Annot., 61 A.L.R.2d 212, 237-40 (1958). A class gift “is a gift to two or more persons who are not named and who have one or more characteristics in common by which they are indicated or who answer to a general description.” In re Estate of Coryell, 174 Neb. 603, 608, 118 N.W.2d 1002, 1005 (1963).
The trial court’s comments on the class gift issue are pertinent:
However, in the case involved the only naming was that of Frank Nespor. There is no evidence whether the testator was simply thinking of the particular persons who were alive on the date of the execution of the will, or if he was thinking of the future group who would survive him. It may well be that testator had not even gone through the mental processes to make that determination. However, it is plain from the wording of his will and the mentioning of his first cousins that these were the people he wished to inherit his property, along with Frank Nespor. This general theme of disposition would indicate that it would be thwarted by providing for specific bequests to individuals, rather than to the class first cousins. Except for the mention of Frank Nespor there is no working [s/c ] in the will that indicates any specific individual or even a number of any individuals. Although Frank Nespor is named individually, the use of his name was to place him in the same classification with the other first cousins.
It is noteworthy that decedent mentions “first cousins” three times in Articles IV and V. This is especially true when he does not mention heirs at law, second cousins, children or spouses of first cousins, relatives, deceased first cousins, or first cousins “alive at this time.”
It well may have been more equitable of the decedent to have named all of his heirs at law, or to have included the children of deceased first cousins. In this regard the Iowa Supreme Court, in In re Estate of Fairley, 159 N.W.2d 286, 288, stated: “Although our purpose is to arrive at the true intent of the testatrix, experience has demonstrated the advisability of adhering to established rules for the construction of wills rather than freeing judges to operate on broad principles of equity and justice.”
The gift to “first cousins on both my father’s and mother’s side of my family” was a gift to several people, the beneficiaries were unnamed, they were all equally related to testator, their number was subject to change by decrease, and all except Nespor, to be considered later, answered to the general description of first cousins. The will itself refers to first cousins several times, and nothing in the will suggests that the issue of first cousins were to be included. See Note, Class Gifts in Iowa, 21 Drake L.Rev. 167, 169 (1971) (“The class of ‘children’ only includes issue of the first generation, and does not include grandchildren, unless a contrary intent is expressed in the will. Similarly, ‘nieces and nephews’ does not include grandnieces and grandnephews.”). This court stated in Parish v. Welton, 194 Iowa 1274, 1277-78, 190 N.W. 947, 949 (1922):
In the instant case, the testator, at the time of drawing his will, knew of the conditions with which he was dealing; that at said time three of his children were deceased; and that they had left children surviving them who were the grandchildren of the testator. Knowing this situation, he drew the will providing for his property to be divided equally among all of his children. He designated his “children” as the class which should be the beneficiaries of his will. He made no provision that the children of any of the deceased children should participate. He made no provision that any of his grandchildren should participate. The property was to be divided equally among all of his “children.”
We cannot extend the word “children” to include grandchildren, any more than in the Nicholson’s Will, In re case [115 Iowa 493, 88 N.W. 1064 (1902)] could we extend the word “nephews” to include grandnephews. It is undoubtedly true that the word “children” may, in some instances, be construed to include grand*102children, where it is evident from the context of the will that such was the plain intent of the testator. Bowker v. Bowker, 148 Mass. 198, 19 N.E. 213. But such a situation does not confront us here, for there is nothing in the context of the will to indicate that the intent of the testator was to include grandchildren in the term “children.” The general rule is that the word “children,” when used in wills, is to be understood and construed in its primary sense, and always so where there are persons in existence answering such meaning of the word. Under such circumstances, the word “children” does not include grandchildren or any others than the immediate descendants of the ancestor of the first degree. Yates v. Shern, 84 Minn. 161, 166, 86 N.W. 1004; Palmer v. Horn, 84 N.Y. 516; Mowatt v. Carow, 7 Paige (N.Y.) 328; Carpenter v. Perkins, 83 Conn. 11, 74 A. 1062; Thomas v. Levering, 73 Md. 451, 21 A. 367; In re Long’s Estate, 228 Pa. 594, 77 A. 924.
(Emphasis added.)
As in Parish, the testator here was aw'are of the conditions with which he was dealing —13 of his first cousins were dead when testator executed his will in 1970, including Vince Pegoriek (died 1917 or 1918), Albert Pegoriek (died 1928), Tillie Riches (died 1932), Tillie Koranda (died 1939), and Charles Riches (died 1943). We find the conclusion difficult to draw that Kalouse “intended” to include these people, some of whom had been dead for 30 to 50 years. Even though a number of the first cousins had descendants, Kalouse nevertheless designated “first cousins” as the class of beneficiaries. Thus the result here must be the same as in Parish — only first cousins take and heirs of first cousins do not take.
The inclusion of Frank Nespor did not change the class gift to a gift to individuals. In the context of gifts to a group and a named individual the gift to the group is still a gift to a class; the addition of the named individual merely raises a question as to the status of that individual. 3 Restatement of Property § 284 & Comment a at 1492 (1940); Annot., 61 A.L.R.2d at 292, 293. Compare Spencer v. Adams, 211 Mass. 291, 97 N.E. 743 (1912) (distributed as if single class), with In re Pierce’s Estate, 177 Wis. 104, 188 N.W. 78 (1922) (held to be gift to class and gift to individual). Hence the gift to the group here is a class gift, and the only question is whether Frank Nespor is included in the class. But his gift is not questioned on appeal, and so no issue arises as to him. We note however that the will states, “with the share going to Frank Nespor to be equal with that of my other first cousins,” indicating a single class gift. (Emphasis added.)
The trial court was right that testator’s first cousins on his father’s and mother’s sides and Nespor constitute a class.
II. Extrinsic evidence. The heirs of first cousins argue a contrary result is required because of extrinsic evidence in the form of the scrivener’s deposition testimony. That testimony includes the following:
A. Anyway, after that I wanted to find out what other person did you want your property to go to, and, well, he said he didn’t really — really care what happened. He had no close relatives. And he thought that his first cousins ought to get the property and that was the best he could do. He said — I asked him if — do you want to name these people, and he said, no, he didn’t want to name them, because he was afraid he might forget somebody and leave them out, because he wanted all of them to share pretty much equally. And then, kind of as an afterthought, he thought of Frank Nespor, and he thought that Frank should get the same share as any of these cousins should get. He wanted him to be treated the same, was my impression.
Frank was not related, as I understood, so he wanted to name him specifically as being — getting shares one of the cousins did.
Q. Were you acquainted with Frank Nespor? A. Yes, I — I knew who he was through the office. I think we did his income tax, too.
Q. Now, you say that Frank Nespor was more or less of an afterthought. *103Would you elaborate on that? A. Well, we were talking about relatives, and he couldn’t remember all of his first cousins or didn’t want to try to remember all their names, but he wanted them to share, and then — and then he said that Frank should have a share, too. I can remember him saying that, that Frank Nespor was a friend of his and had done a lot of things for him, and he wanted to have him share like his relatives would. And I think he named Frank as an executor, too, as I recall.
Q. Was any consideration given, Mr. Story, to leaving anything to the first cousins who had predeceased Mr. Kalouse at the time he made the Will? A. Well, I have no specific recollection as to that. I know that I asked him specifically about naming people, and he told me he didn’t want to name anybody, because he didn’t want to leave anybody out. He wanted to be sure that all the first cousins were included. But I can’t recall him or either of us talking about what would happen if one of them died before he did. (Emphasis added.)
Before considering the import of this evidence, we set forth some testimony of the scrivener relating to the time the will was executed, which is pertinent to an understanding of the case:
Q. And where was — where did you find Mr. Kalouse? A. He was in a room in the hospital. It seemed to me like it was a private room. I don’t recall anybody else being in there with him. But at the time that I came in he was — he was not in the bed, but he was sitting on the edge of the bed. He was able to stand up and walk around. I recall he had some kind of a bandage or corset around his middle, because apparently he had broken ribs or gored by a bull or something.
Q. Now, did you learn from him or from any other source why he was in the hospital? A. He told me he had been gored by a bull. Maybe — I’m not sure I learned it anywhere else either, but I know he did say that.
Q. Can you be a little more specific as to his condition of health at that particular time when you saw him at St. Luke’s Hospital? A. Well, when I saw him, I mean, mentally, I mean, he appeared to be alert and knew what was going on. He had, I think, some pain probably from the ribs or whatever it was that was injured about his middle. But mentally he seemed to be same as I had seen him before.
Q. Did you visit with him about— about this Will that he wanted you to prepare? A. Yes. When I got there, I told him that they had called us to come down there to see him, because he wanted to make a Will, and he said that was true, that he felt he should have a Will, and so then we did visit somewhat as to what he wanted to do by way of a Will, and after I ascertained what he wanted, then I did sit down and write this out in my handwriting, and then I read that to him, also, and asked if that’s what he wanted, and then we called in a nurse to be a witness, and he did sign it before myself and this other nurse.
Q. Now, why was it that you wanted to do it this way, which might be a little unusual, rather than to go back and prepare it and submit it to him later? A. Well; my understanding was — when I went down there was that it was something urgent, needed to be done right away. Now, when I got there, I’m not sure if that was necessarily true. I mean, as far as looking at him, he appeared to be all right. But I didn’t check with his doctor to find out what his condition was.
Q. Was this his wish, to have it done immediately? A. Yeah, he wanted it done right away.
Q. Did he give you any explanation as to why he wanted it done immediately? A. Well, apparently I took it his doctor had told him maybe that he ought to have one made. And if you knew Louie, why, he’s not the kind of fella that wanted to make a Will before, when he’s in the office, probably something he would *104put off a long time, but it was my impression somebody told him he ought to have one right away, so he called us to have it made.
At the outset, we are not persuaded that the deposition, taken at face value, shows Kalouse did not intend a class gift. Ka-louse stated he did not want to leave out any first cousins, but that does not answer the question of whether he intended a class gift. The scrivener testified, “But I can’t recall him or either of us talking about what would happen if one of them died before he did.” When asked to name individuals, Kalouse refused; this is consistent with a class gift.
Assuming however that the scrivener’s testimony shows gifts to individuals were intended, the heirs’ problem is the principle against varying wills by extrinsic evidence. The general rule is that “extrinsic evidence is not admissible to vary, contradict or add to the terms of the will, or to show a different intention on the part of the testator from that disclosed by the language of the will.” See Wagg v. Mickelwait, 165 N.W.2d 829, 831 (Iowa 1969); In re Estate of Winslow, 259 Iowa 1316, 1323, 147 N.W.2d 814, 818 (1967) (scrivener’s testimony excluded); In re Estate of Hogan, 259 Iowa 887, 890, 146 N.W.2d 257, 258 (1966); In re Estate of Stonebrook, 258 Iowa 1062, 1073, 141 N.W.2d 531, 537 (1966); Bankers Trust Co. v. Allen, 251 Iowa 938, 944, 945, 135 N.W.2d 607, 611 (1965) (“Evidence of testator’s intention as an independent fact, divorced from the words of the will, is clearly inadmissible. Courts will not, from oral testimony, make a will testator perhaps intended to, but in fact did not, make.” Also, “[CJourts cannot draw [testator’s] will to carry out a possible intent not expressed in the will. A contrary holding would nullify the requirement that wills be in writing.” (Emphasis added.)); In re Estate of Miller, 243 Iowa 920, 929, 54 N.W.2d 433, 438 (1952) (scrivener’s testimony excluded). Accord, Tamm, Inc. v. Pildis, 249 N.W.2d 823, 831 (Iowa 1976) (contract); Sanderson v. First National Bank, 446 S.W.2d at 723 (class gift). See also In re Estate of Lepley, 235 Iowa 664, 672, 17 N.W.2d 526, 529-30 (1945) (“‘However clearly an intention not expressed in the will may be proved by extrinsic' evidence, the rule of law requiring wills to be in writing stands as an insuperable barrier against carrying the intention thus proved into execution.’ ” (Emphasis added.)).
Extrinsic evidence used in this manner violates the parol evidence rule. Tamm, Inc., 249 N.W.2d at 831, 834; Egan v. Egan, 212 N.W.2d 461, 464 (Iowa 1973); Pappas v. Hauser, 197 N.W.2d 607, 611 (Iowa 1972). This holds true in the class gift context. As stated in 96 C.J.S. Wills § 693, at 22-23 (1957):
The only universal rule for determining whether testamentary gifts to several persons are gifts to them as a class or as individuals is to ascertain the intention of the testator as expressed in the instrument to be construed. .
Primarily, the determination of the question depends on the language of the will, but it is not absolutely controlled and limited thereby; the substance and intent, rather than the words, are to control; as aids thereto, the general scope of the will, the general purpose of the testator, the particular language used, the relationship of the parties, and the surrounding circumstances may all be considered, subject always to the limitation, prevailing generally as to the availability of extrinsic facts and circumstances as aids to construction, that such matters cannot be resorted to in order to defeat the plain and unambiguous language of the will. (Emphasis added.)
The editor states regarding decisions on “first” and “second cousins” in Annot., 94 A.L.R. 26, 111 (1935):
Where there were persons in existence who fitted the description of “first cousins” and “second cousins,” such description was held to be unambiguous, so that it could not be shown that the testator intended by the use of those terms to designate persons other than those included in the ordinary acceptation of the terms. (Citations omitted.)
*105Use of the extrinsic evidence involved here would violate the parol evidence rule. Where a will is involved the parole evidence rule applies with at least as much vigor as in contract cases, although authority exists for applying the rule more strictly in wills cases since wills must be in writing. As stated in Annot, 94 A.L.R. at 30:
It is stated in some cases that the rules relating to the admissibility of parol evidence in connection with wills are the same as those relating to other written instruments. Hanner v. Moulton, (1885; C.C.) 23 F. 5 (affirmed in (1891) 138 U.S. 486, 11 S.Ct. 408), 34 L.Ed. 1032; Tucker v. Seaman’s Aid Soc. (1843) 7 Met. (Mass.) 188; Crosson v. Dwyer (1895) 9 Tex.Civ.App. 482, 30 S.W. 929.
However, it is stated in Smith v. Holden (1897) 58 Kan. 535, 50 P. 447, that more strictness is observed in the reception of parol evidence of expressions of a testator’s intention than in cases of like evidence explanatory of contracts inter vivos. And in Robinson v. Ramsey (1925) 161 Ga. 1, 129 S.E. 837, the rule excluding parol evidence to contradict, add to, or vary the terms of a written instrument is said to be especially applicable in the construction of so solemn an instrument as a will.
There would seem to be some reason for making the parol-evidence rule more stringent with regard to wills than with regard to other instruments, at least those not required by law to be in writing, for, in the case of wills, the person offering such evidence is confronted not only with the common-law rule against the admission of parol evidence to vary or add to a written instrument, but also with the Statute of Frauds and the Statute of Wills, both requiring testamentary instruments to be in writing and the latter setting up certain formalities to be observed in the execution of a will. (Emphasis added.)
See also 80 Am.Jur.2d Wills § 1279, at 388 (1975) (“The rules for the admission and exclusion of parol evidence in regard to wills are essentially the same as those which prevail in regard to contracts in general.”).
In this case the extrinsic evidence was admitted by stipulation. An evidentia-ry rule exists that evidence received without objection becomes part of the evidence in the case and is usable as proof to the extent of its rational persuasiveness. Tamm, Inc., 249 N.W.2d at 833 (citing McCormick’s Handbook of the Law of Evidence § 54, at 125-26 (2d ed. E. Cleary 1972)). This rule, however, does not apply to evidence which is used in violation of the parol evidence rule, since that rule is a substantive law rule. Jarvis v. Cunliffe, 140 Conn. 297, 299, 99 A.2d 126, 127 (1953) (“[A] written contract may not be varied by parol and this is a rule of substantive law. . even though evidence of an oral agreement is before a trier, it may not be given the effect of varying a written contract which was intended to incorporate all of the terms of the agreement.”); 9 J. Wigmore, Evidence § 2400, at 3 (3d ed. 1940 & Supp.1979); Restatement (Second) of Contracts § 239, Comment a at 546 (Tent. Drafts Nos. 1-7, 1973). This court has specifically dealt with the application of the principle. The court has stated in Randolph v. Fireman’s Fund Insurance Co., 255 Iowa 943, 949, 124 N.W.2d 528, 531 (1963): “The rule against varying, modifying or contradicting written instruments by parol evidence is one of substantive law rather than of evidence, and such evidence will be disregarded even though not objected to when offered.”
The court has followed this rule in recent times. In 1976 we stated in Tamm, Inc., 249 N.W.2d at 834-35:
The McCuens’ testimony concerned conversations they had with Mr. Hunt about termination of the driveway arrangement. For the same reasons asserted in the previous division we believe this testimony violates the parol evidence rule. The testimony was not offered to assist in the interpretation of words of termination because there were none. It was clearly an attempt to add to contractual language words not previously there *106or referred to. It is true there were no objections to this testimony, but there is case law support for the proposition a court should sua sponte apply the parol evidence rule. In Randolph v. Fireman’s Fund Ins. Co., 255 Iowa 943, 949, 124 N.W.2d 528, 531, 8 A.L.R.3d 907, 913-914, this court said:
“Since we think the policy is clear in its intent that no farm employees are covered unless listed and a premium charged, we have no occasion to give any effect to the parol evidence introduced. It was conflicting; but whether objected to or not, under the circumstances here it has no weight. The rule against varying, modifying or contradicting written instruments by parol evidence is one of substantive law rather than of evidence, and such evidence will be disregarded even though not objected to when offered. . . . [citing authorities].” (Emphasis supplied.) See also Williams v. Williams, 251 Iowa 260, 264, 100 N.W.2d 185, 188.
Therefore, their testimony is entitled to no consideration in arriving at our decision.
See also Annot., 81 A.L.R.3d 249, 264-65 (1977).
The heirs apparently believe the rules stated in Tamm, Inc. and Randolph do not apply here since the evidence was admitted by stipulation. On the contrary, since the parol evidence rule is a substantive law rule, the following statement is pertinent from 73 Am.Jur.2d Stipulations § 5, at 539-40 (1974):
While, ordinarily, courts are bound by stipulations of litigants, that rule cannot be invoked to bind or circumscribe a court in its determination of questions of law. It has generally been stated that the resolution of questions of law rests upon the court uninfluenced by stipulations of the parties, and accordingly, virtually all jurisdictions recognize that stipulations as to the law are invalid and ineffective. The same rule applies to legal conclusions arising from stipulated facts. It has thus been held that it is not competent for the parties or their attorneys to determine by stipulation questions as to the existence or proper construction or application of a statute; as to the validity of a legislative enactment; or as to the validity or effect of a written instrument.
See also Aubuchon v. Bender, 44 Mo. 560, 570 (1869) (deed; “No agreed statement of facts can fix a conclusion of law. The relationship and death are facts to be admitted, but who were his heirs is a question of law which the court is bound to declare.”).
From the authorities we conclude that the extrinsic evidence of Kalouse's declarations to the scrivener may not be used to construe the will.
III. Antilapse statute. Then who took at the death of Kalouse? The parties argue the question of the application of the antilapse statute to the Article V bequest. This statute provides in section 633.273, The Code 1977: “If a devisee die before the testator, his heirs shall inherit the property devised to him, unless from the terms of the will, the intent is clear and explicit to the contrary.”
The application of the antilapse statute depends on whether the bequest was to a class or to individuals. “Generally if the gift is to a class the surviving members take the whole gift, if to named individuals there is no such right of survivorship to the other named beneficiaries but the heirs of the deceased beneficiary take his share under the antilapse statute . . . .” Gunn v. Wagner, 242 Iowa 1001, 1010, 48 N.W.2d 292, 297 (1951); In re Estate of Huston, 224 Iowa 420, 425, 275 N.W. 149, 151 (1937) (“It is the well-known rule of law in this and most other jurisdictions that unless a contrary intent is indicated by a will, a devise to a class includes only such members as are living at the time of the testator’s death, when the will takes effect.”); 1A D. McCarty, Iowa Probate § 942, at 423 (2d ed. 1964). Since Kalouse left the residue of his property to a class, the antilapse statute does not apply as to predeceasing cousins. Thus, first cousins who died before the will was executed are not included, nor are their *107heirs. Parish v. Welton ; Note, 21 Drake L.Rev. at 169-70. Likewise, first cousins who died after the will was executed and before testator died, and their heirs, are not included. Redinbaugh v. Redinbaugh, 199 Iowa 1053, 1054, 203 N.W. 246, 246-47 (1925); Note, 21 Drake L.Rev. at 170-71. We decline to change these rules.
The trial court correctly construed the Kalouse will. Each of the 24 first cousins alive on the date Kalouse died and Frank Nespor receive one twenty-fifth of the residue of the estate.
AFFIRMED.
All Justices concur except HARRIS and McGIVERIN, JJ., who dissent, and REES, J., who takes no part.