delivered' the opinion of the court:
Leopold Dollander owned at his death considerable farm land and some town property in Henry county. He died testate April 22, 1914, leaving his widow, Rosalie Dollander, and his seven children, the appellees, and the five appellants, who are the children of his deceased daughter, Mary Duyvetter. The only question raised in this case is whether the grandchildren take per stirpes or per capita under the second section of the will. The master and the trial court decided in favor of appellees, holding that the grandchildren took per stirpes and not per capita. •
The second section of the will reads as follows: “I give, devise and bequeath unto my beloved wife, Rosalie Dollander, the use of all my real and personal property during her life or so long as she remains my widow. Upon her death or in event of her re-marriage all my said property shall be vested in my children, Leonie Dhaemers, the children of Mary Duyvetter, deceased, Nellie Cathlyn, Angelina Afinóse, Charles, Martin, Jacob, Frank and Mandus Dollander, share and share alike.” The testator signed by mark in the presence of three witnesses. The only other provisions of the will were the first paragraph, which provided for the payment of just debts and funeral expenses, and the third paragraph, which named the executor and revoked all former wills. The widow relinquished her in*276terests in and to the property before the beginning of this proceeding.
It is argued by counsel for appellees that the will indicates that it was drawn by someone not well versed in the drafting of a will. However, we-do not think the form or wording of the will, outside of section 2, gives any assistance in' the construction of said section. If the will be construed so that appellants take per capita the share of each grandchild would be one-twelfth; if it be construed as held by the trial court, that they take as a class, each . grandchild’s share would be one-fortieth. The real and personal property amounted to between $40,000 and $50,000.
On the hearing before the master in chancery the executor, who drafted the will, testified as a witness, and was asked if there was any question, when he was drawing the will, as to who were the children of the deceased daughter, Mary Duyvetter, and what was said about 'that, and the witness was permitted to answer, subject to objection, that the testator wished those children to have their deceased mother’s share. There is no argument by counsel for appellees that this testimony was competent, and we think it was clearly incompetent. (Hawhe v. Chicago and Western Indiana Railroad Co. 165 Ill. 561; Alford v. Bennett, 279 id. 375.) The paramount rule to be followed in construing a will is to ascertain the intention of the testator and follow the intention thus ascertained unless contrary to some settled rules of law. (Howe v. Hodge, 152 Ill. 252; Wardner v. Baptist Memorial Board, 232 id. 606.) This intention is to be gathered from the words and expression used in the will itself. Yet a court in construing a will is not bound to shut its eyes to the state of facts • under which the document was made. “On the contrary, an investigation of such facts often materially aids in elucidating the scheme of disposition which occupied the mind of the testator. To this end it is obviously essential that the judicial expositor should place himself as fully as pos*277sible in the situation of the person whose language he has to interpret, and, guided by the light thus thrown on the testamentary scheme, he may find himself justified in departing from a strict construction of the testator’s language without allowing conjectural interpretation to usurp the place of judicial exposition.” 1 Jarman on Wills, (Sweet’s 6th ed.) 503; Abrahams v. Sanders, 274 Ill. 452, and authorities there cited; 30 Am. & Eng. Ency. of Law, (2d ed.) 666.
Counsel concede that the testator left seven children and the children of a deceased daughter. The wording of the second section would seem .to indicate that he had his children particularly in mind by this section. The will gives the names of all of his children but does not give the particular names of any of the grandchildren. It simply groups them all together as “the children of Mary Duyvetter,” and the arrangement of the names of his children, interwoven with the reference to the children of his deceased daughter as a class and without naming them, would indicate that he was thinking of the grandchildren as a class, representing the deceased daughter, rather than as individuals who were to receive separate and exclusive shares of his property.
The authorities as to the construction of wills which provide that property shall be divided equally among the named children of the testator and the unnamed children of a deceased child are in hopeless conflict. (Perdue v. Starkey Heirs, Ann. Cas. 1916C, (Va.) 409, and cases cited in note.) When the bequest is to an individual or several named individuals, and to others as a class, the latter generally take per stirpes; but this rule yields when the testator used language indicating an intention that the members of the class should share equally with the named individuals. (30 Am. & Eng. Ency. of Law,—2d ed.—727. To the same effect is 40 Cyc. 1491.) “The nature of descent *278and distribution under the statute is important, because, unless the intention of testator is clearly manifest in his will, the courts construe a will by which property is to be divided among a specified class as’ contemplating a division in analogy to the Statute of Descent and Distribution.” (Page on Wills,—1901 ed.—643, 644, and cases cited in note, among others Kelley v. Vigas, 112 Ill. 242.) The author then goes on to say: “Thus, when the persons designated stand in equal degree or degrees of relationship to testator and the devise or bequest is to inure to the benefit of all of them, the court will order a division per capita, * * * while if the devisees or legatees stand in unequal degrees of relationship to testator the law favors a construction which results in a distribution per stirpes among the beneficiaries.” (See, also, to a similar effect, the reasoning in 1 Schouler on Wills,—5th ed.—sec. 540.) When the words “equally,” “equal among,” “share and share alike,” or other similar words, are used to indicate an equal division among a class, the persons among whom the division is to be made are usually held to take per capita unless a contrary intention is discoverable from the will. This would seem to be the fair conclusion drawn from the decisions in this court. (See Kelley v. Vigas, supra; Pitney v. Brown, 44 Ill. 363; McCartney v. Osburn, 118 id. 403; Auger v. Tatham, 191 id. 296; Welch v. Wheelock, 242 id. 380; Straw v. Barnes, 250 id. 481.) The presumption, however, in favor of a per capita distribution yields readily in favor of a faint indication of the testator that the distribution shall be per stirpes. If from the will, as a. whole, a different intention appears such intention will control, notwithstanding such words. • A leading authority on this question is doubtless Jarman on Wills, who states : “Where a gift is to the children of several persons, whether it be to the children of A and B or to the children of A and the children of B, they take per capita and not per stirpes; * * * but this mode of construction will yield to a *279very faint glimpse of a different intention in the context.” (2 Jarman on Wills,—Bigelow’s 6th Am. ed.—205, 206.)
A reference to decisions in various States will show that the courts have frequently departed from the general rule on the very slightest indication in the will that a per stirpes distribution was intended as to some of the legatees rather than a distribution per capita. In the case of Eyer v. Beck, 70 Mich. 179, there was a provision in the will dividing the property equally “among my heirs, to-wit, John Beck, the children of Christian Beck, Jr., deceased, Elizabeth Eicher, Gottlieb Beck, Peter Beck and Magdalena Eyer.” In that case it was said (p. 180) : “The arguments on both sides were based on decisions in England and America, some of which were in point on each theory, a part holding that such legacy went by representation and a part that it made a distribution per capita among children and the three grandchildren. It appears incidentally that at the time of the distribution in the probate court one of the daughters of testator had died leaving six children, who were testator’s grandchildren, each of whom, under the order of the circuit court, gets only one-sixth of the amount allowed to each of the three grandchildren referred to in the will. * * * The cardinal principle of interpretation of wills is to carry out the intention of the testator if it is lawful and if it can be-discovered. If the language used has been so fully established by construction as to make a rule of property courts cannot very well depart from the settled interpretation, although testimony, if introduced, might show the testator did not really so understand it. But this is a rule of law and not of interpretation and is enforced only on that account. It would be an untrue presumption, in point of fact, in most cases, because we all know that the abstruse rules set up by precedents are often not much known to even the legal fraternity who have not had their attention called to them, and the unprofessional world is not familiar with law books.” The opinion then goes on *280to discuss other branches of the case and the decisions thereon, and states (p. 181) : “They are very generally based either on an unreasoned acceptance of rulings which they deem analogous or upon considerations outside of the language itself; and it is worthy of remark that the leading cases which sustain a distribution per capita intimate that a very small indication of an intent to the contrary would change the rule. This will does not name the three grandchildren who are claimed to stand on the footing of children. It gives no reason why they are nearer to the testator in affection than his other grandchildren, and he must have known that his named children might all, or any of them, die during his lifetime, leaving descendants. In case there was any real preference for one set over the rest it would seem that the testator would have expressed it, so as to show that he deliberately preferred them. * * * But where he makes a difference between those presumptively equal in his regard it seems to us that it ought to be made clear by his language. * * * The laws of inheritance are generally understood by intelligent people. It is understood that when there are children and grandchildren to take, the grandchildren by one child take in the aggregate only that child’s portion.”
The only difference 'between the will construed by the Michigan court and the will here construed is that the Michigan will used the word “heirs” and this one used the word “children,” and it would seem clear from what has been already stated that the testator was trying to put the children in a class by themselves and the grandchildren in another class. There is nothing in the will itself to indicate that the testator intended to favor his deceased daughter’s children more than he did the children of any of his other children.
In Lyon v. Acker, 33 Conn. 222, the will provided: “I give, devise and bequeath to my three daughters and the children of my son Samuel A. Lyon, viz., to Mary F. Acker, *281wife of Abram Acker, Susan E. Voorhis, wife of William Voorhis, Josephine Lyon, and the children of my son Samuel A. Lyon, my house and homestead where I now reside, to them and their assigns forever, share and share alike.” It was argued in that case that to give effect to the words “share and share alike” could only be done by holding that the grandchildren were to take per capita. The court said (p. 224) : “The names and number of these children do not appear in the will. They are referred to only as a class,—a fact which seems to indicate that she intended to make all the grandchildren only equal to one of the daughters; at least it seems to us that this view of the case gives effect to the words in question as well as the other. If so, they do not require us to adopt the construction contended for by the petitioners. The effect of such a construction would be to give to the children of Samuel four-sevenths of the property devised and to the three daughters the remaining three-sevenths. That the testatrix intended to discriminate against her daughters and in favor of. her son’s children will certainly not be presumed, and when we look at the language of the will or the circumstances of the case we fail to discover anything to indicate the existence of any such intention. Had she made her son the devisee it would require clear and unmistakable language to give him four-sevenths of the property. Has she done any more, or did she intend to do any more, than to give the father’s share to the children? We think not. Again, the claim of the respondents is in accordance with the general laws of distribution. We think it a sound rule that when a devise, or legacy is given to heirs or their representatives, courts will apply the general principles governing the descent of estates unless a contrary intention appears.” The same may be said on all the questions here raised.
In Page on Wills (1901 ed. sec. 554) it is said: “By the terms of a will one person may be balanced against a class. Thus, a provision that an income should be ‘equally *282divided’ between the widow of the testator and the heirs of the testator’s mother was held to mean that the widow should receive half the income.” See to a similar effect, Lachland’s Heirs v. Downing’s Exrs. 50 Ky. 32, where the will provided: “All the residue of my estate, whether real, personal or mixed, not herein otherwise disposed of, I desire may be ecpially divided, after my death, between my brother, John Downing, my two sisters, Elizabeth Cameron and Nancy Gibson, and the children of sister Nelly Each-land, to them and their children forever, it being my desire that the portions allotted to my brother John and my two sisters and the children of my deceased sister, Nelly Each-land, shall be made as nearly equal as possible, both in kind and in amount.” See, also, Balcom v. Haynes, 96 Mass. 204, where the will provided: “Sixthly, I give and devise to my brothers, John W. Haynes, Amos Elaynes and Charles Haynes, and my sisters, Susan Boyd, wife of Stephen Boyd, Ruth Boyd, wife of Warren Boyd, and the heirs of Lydia Walkup, and their heirs, respectively, all the rest and residue of my real and personal estate, * * * to be divided in equal shares between them.” It was held that the heirs of Lydia Walkup, who was the testator’s sister, should be treated as a class and receive only the amount that would have gone to the sister had she been living. In that case the context of 2 Jarman on Wills, heretofore quoted, was referred to, and it was Held, that the construction of the will yields to a very faint glimpse of a different intention on the part of the testator. The same court in' Perkins v. Stearns, 163 Mass. 247, construed a will of somewhat similar character, holding that certain legatees took per stirpes, as a class, and not per capita. A similar reasoning is to be found in Ruggles v. Randall, 70 Conn. 44, Raymond v. Hillhouse, 29 Am. Rep. (45 Conn.) 688, Wood v. Robertson, 113 Ind. 323, Henry v. Thomas, 118 id. 23, In re Risk’s Appeal, 52 Pa. St. 269, and White v. Holland, 92 Ga. 216.
*283In MacLean v. Williams, 59 L. R. A. (Ga.) 125, the court said (p. 127) : “If all the heirs-at-law stand in the same relation to the decedent they take equally per capita. If some stand in different degrees from others they take per stirpesf but they take equally nevertheless. The estate, in either event, is divided into shares, and equal shares, although in the one case each share goes to an individual and in the other casé the equal shares go to a class of individuals. The Statute of Distributions sets forth the settled policy of the law as to where the estate of a decedent shall go. While a testator is allowed to ignore, either in part or altogether, the rules laid down in that statute, it will not be presumed that it was the intention of the testator to disregard the law as it is contained in the statute in any part, unless the terms of the will are such as to make this intention manifest.” The opinion cites as upholding this doctrine, Page on Wills, sec. 556.
An interesting discussion of the old rules of law as laid down in Jarman on Wills with reference to taking per stirpes or per capita under similar provisions is found in Roome v. Counter, 10 Am. Dec. (N. J.) 390, where the writer of the opinion states that most of the earlier authorities were based upon Blackler v. Webb, 2 P. Wms. 383, in which the lord chancellor intimated that he was in doubt as to the proper decision; and also upon the case of Phillips v. Garth, 3 Brown’s Ch. 64, where Buller, sitting as justice for the lord chancellor, recognized the case of Blackler v. Webb as good law and decreed accordingly, but on appeal to the lord chancellor, and finding, upon the argument, that he leaned much the other way, the cause was compromised by the parties, with the advice, no doubt, of the able counsel who advocated the cause. The writer of the opinion in Roome v. Counter reached the conclusion that neither of those decisions should necessarily be followed or have weight except in cases that were exactly like the cases there decided, and in that case, in construing a will *284which provided “that all the remainder of my movable estate shall be equally divided, that is to say, Henry Counter, and the heirs of my son Peter Counter, Anna Roome, Susannah Berry, Elizabeth Dodd and Sarah Counter,” it was held that the heirs of Peter Counter, the deceased son, should take per stirpes.
A reading of the authorities in the note already cited in Perdue v. Starkey Heirs, supra, in'the notes to Jarman on Wills, Page on Wills and Schouler on Wills, in 30 Am. & Eng. Ency. of Law, and in 40 Cyc. will show that the decisions on this question are in irreconcilable conflict. The decisions of this court in Kelley v. Vigas, McCartney v. Osburn, Welch v. Wheelock and Straw v. Barnes, supra, all indicate that the courts generally will depart from the rule laid down by Jarman on Wills if there is the faintest intention to show that the class,—in this case the grandchildren,—might be considered as taking per stirpes rather than per capita. Where there is an ambiguity existing'in a will, unless there is a manifest intention to the contrary, “the presumption that the testator intended that his property should go in accordance with the laws of. descent and distribution will be applied as an aid in construing the will; hence such a construction should be given the will as favors the heirs-at-law or next of kin in preference to disinheritance, or to strangers or persons not so closely related to the testator.” (40 Cyc. 1412. See to the same effect, 30 Am. & Eng. Ency. of Law,—2d ed.—668, and cases cited.) The reasoning in Straw v. Barnes, supra, would tend to support the same conclusion.
In Ferrer v. Pyne, 81 N. Y. 281, the court said (p. 283) : “In Powell on Devises (vol. 2, p. 331,) it is said that where a gift is made to a person described as standing" in a certain relation to the testator and to the children of another person standing in the same relation, as to my brother A and the children of my brother B, A only takes a share equal to one of the children of B; and this position is *285abundantly sustained by the authority of English cases [citing them] and to some extent by the courts of this country. Yet if the case stood upon the words of the residuary clause, alone, we should find great difficulty in confirming, by the sanction of this court, a construction opposed to the apparent meaning of the language used by the testator and "at variance with the natural disposition of mankind. We find the testator calling to mind his children, their names, their relations to others by marriage. * * * The living children are named by him, while the children of the daughters who are dead are spoken of, not by name, but as ‘the son of Isabel’ or ‘the children of Irene,’ evidently giving to them the place as recipients of his bounty which Isabel or Irene, if living, would have filled. He designates the children of Irene as a class and not as individuals, remembers them not in their own persons but as representatives of their parent and substitutes them in her place. We are unable to discover any intent to bestow upon them any greater or more numerous marks of his affection than their parent would, if living, have received. The rule referred to has in modern times been applied with reluctance by some courts because it had become a rule of property and by others out of deference to its supposed authority, but in many, if not in all, cases with open protest, while by others it has been wholly rejected.” The reasoning of this opinion on this point is quoted with approval in the note to Perdue v. Starkey Heirs, supra, on page 414.
While, perhaps, the conclusion we have reached may not be supported by the majority of the American and English authorities, judging them only by number, we think it is fully supported by the weight of authority where the question has been exhaustively considered and by the sounder reasoning,—that is, that in a will worded as is this one, the grandchildren are to be considered as a class and should take, under the will, per stirpes and not per capita.
The decree of the circuit court will therefore be affirmed.
Decree affirmed.