It seems to me to be a stretch of the power of the court to direct an exercise, partial or otherwise, of the power of sale contained in the will. This suit was brought merely for the purpose of obtaining a judicial construction of that instrument. The ■power of sale is discretionary and is one in trust. The execution of such a power ought not to be coerced, except by a judgment rendered in a suit brought for that purpose, and upon proof showing the necessity therefor. If the executrix improperly refuses to co-operate with her co-executors in effecting a sale, she . may be compelled to do so, or be removed by appropriate proceedings. I am also of opinion that the executrix is testamentary guardian of her infant children, and as such is entitled to the custody of their persons, and to the custody and management of their personal estate, and to receive the rents and profits of their real estate. Testamentary guardianship did not exist at common law, but rests upon statutes. (12 Car. 2, ch. 24; 2 R. S., 150, §§ 1, 2, 3; Schoul Dom. Rel., 393.) The power of testamentary disposition given, is to dispose of the custody and tuition of infants. The so-called wife of the testator was appointed guardian of .the persons only of her children. But such guardianship *149obviously involves the custody and tuition of them. I think, therefore, that her appointment was in substantial conformity with the statute. But, if otherwise, the third section of the Devised Statutes cited confers upon any person, to whom the custody only of a minor has been disposed of, the powers of a testamentary guardian. Whether security should, be required of the guardian, is a question which is not before us.
The provisions of the will in favor of the testator’s so-called wife have not been assailed. By some of those provisions the testator gave to her his dwelling-house, and the lot on which it stands, during her widowhood, and directed that if the same should be sold by his executors, with her consent in writing, the proceeds thereof should be held and invested, and the income applied to her use. The will further provided, that if the testator disposed of the said house and lot before his death, then in lieu thereof he gave to her the income of a principal sum, to be held and invested by his executors, equal to that at which he so disposed of said property. Intermediate the making of the will and the death of the testator, he mortgaged that property for the accommodation of his eldest son, and that mortgage is still outstanding. The gifts for the benefit of his so-called wife were expressly made in lieu of her dower, and she having accepted the same, the question arises whether she is bound to keep down the taxes and the interest which shall accrue upon the mortgage. I think she is. Having elected to accept the provisions of the will, she must take them according to their legal effect. The court cannot alter those provisions, or either of them, in order to obviate any change of circumstances which has occurred since the will was made. The mortgaging of the house and lot, although it was a disposition of that property in one sense, yet it was not such an one as calls for the exercise of the alternative provided by the will. The widow must therefore keep down the taxes on that property, and the interest on the mortgage thereon, and the will makes no provision for reimbursing her in any form. With respect to the terms of the judgment in other particulars, there has been no contest, save on the subject of the gifts for the benefit of children. The testator, by his will, in the clearest language, designates the defendant, Catharine Shields, as his wife; he *150creates trusts for the benefit of bis surviving children; for “each” of his sons, and “ each” of his daughters, separately, with gifts over to their children, and he appoints his so-called wife guardian of the persons of his infant children during their respective minorities. It appears that after the death of the tes-, tator, a woman, whose maiden name was Jane Valentine, brought an action in this court to recover her dower in the testator’s lands, on the ground that she was his Avidow; that all the parties to this action were parties defendant in that action, and that upon a trial of special issues, it was found and adjudged that the testator and said Jane were married in the year 1833; that said Jane was his widow, and that she was entitled to recover her dower in the lands of the testator. Upon the trial of this action it was proved that, at the date of the testator’s will, there were living a daughter' and a son, who were the offspring of the testator and said Jane, and who are still living; that the daughter was born in the year-1840, and that the son was born in the year 1848. It was also proved that on November 20th, 1848, the testator was formally married, by a priest of the Roman Catholic Church, to the said Catharine; that he had by her four sons and four daughters, all of whom, except one daughter, were ‘ bom before he made his will, and all of whom are living; that from the time of the marriage between the testator and said Catharine,, until his death, they lived together openly, and notoriously, as husband and wife, and together with their children, as they were successively born, formed one household; that said Jane was only fourteen years old in 1833, and that all connection, between her and the testator, ceased in 1848, after which she never saw him. The evidence tended very strongly to show that the testator never recognized said Jane as his wife, or their offspring as his lawful children, and that the kind of recognition which he did bestow upon them was such as indicated that his connection with said Jane was unlawful, and that he did not regard said children as legitimate.
Upon these and some minor facts the justice, at Special Term, found that the children referred to in the will were the issue of the testator’s marriage with said Catharine. The justice did not find that either marriage was lawful or unlawful, or that the issue of either marriage were legitimate or illegitimate. He was *151requested to find that the children of the testator and said Jane were the lawful issue of their said marriage, which request was refused, and the defendant excepted. Upon that question of fact it was not the duty of the justice to comply with the request, unless the judgment in favor of Jane, before referred to, operated as an estoppel against the children of Catharine. Treating the judgment as evidence merely, to be considered with the other evidence in the case, the request was properly refused. For it is not the right of a party to have a fact found, unless it has been proved by uncontroverted testimony. The judgment recovered by Jane was one in personam, and not one in rem. The question of the legitimacy of children was not involved in the case. That point, therefore, is still open to litigation. Upon the trial of this case; no direct evidence of either the non-existence or of the invalidity of the alleged marriage with Jane was given; but evidence was given, aliunde the judgment in the suit for dower, tending to show that the testator and said Jane were in fact married in the year 1833, and that the defendants Catharine J. Bonner and William B. Shields are the offspring of that marriage, and the only offspring thereof who were living when the testator made his will. Assuming, therefore, the existence of such marriage, and its continuance until the death of the testator, nevertheless the question presented, relative to the gifts to children, is one respecting the intention of the testator only. That intention is to be collected from all the provisions of the will, and the circumstances surrounding the testator at the time the will was made, and when ascertained must control the interpretation thereof.
That rale is too familiar to require a citation of authorities to sustain it. Prima facie the word “children” means legitimate children. But when it appears from the will itself, by express designation, or by necessary implication, that the intended objects of the testator’s bounty are illegitimate children, they will take under the description of “ children.” “ Necessary implication,” said Lord EldoN, in Wilkinson v. Adams (1 V. & B., 422-466), “ means not natural necessity, but so strong a probability of intention, that an intention contrary to that which is imputed to the testator cannot be supposed.” Such manifestation of the testator’s intention is as effectual as if he had-written the word “ illegiti*152mate ” before the word “ children.” In Wilkinson v. Adams, just cited (1 Y. & B., 422), the testator at the time he made his will had a wife living, but no children by her. Another woman, Ann Lewis, was living with him, by whom he had three natural children, all of whom were born before his will was made. All ^ these children, at the time when the testator made his will, had acquired the character and reputation of being his natural children by Aim Lewis. By his will he made a devise in these words, “to the children which I may have by the aforesaid Ann Lewis, and living at my decease,” etc. It was contended that the testator contemplated the death of his wife before his own death, a marriage with Ann Lewis, and his having children by her. But the judges, in an opinion sent to the lord chancellor, upheld the devise, upon the ground that from the material parts ol- the will it appeared that the testator meant the devise to operate in favor of his illegitimate children, born and to bo born by Ann Lewis, and they declared that the case of Cartwright v. Vawdry (5 Ves., 530), relied on by Mr. Perry in this case, could not be considered as a decision. Lord EldoN, after commenting on the case of Cartwright v. Vawdry and other cases, and explaining the former, decided that the illegitimate children were, upon the will itself, the whole taken together, entitled to take as devisees. In Beachcroft v. Beachroft (1 Madd., 430), Sir Thohas Plumee said: “I know of no rule of law which prevents illegitimate children claiming under a class or description as well as any other strangers. Such children are not prohibited from taking, as by the civil law, and I see no reason to prevent them taking under a general description.”
The rule thus stated has been applied in numerous cases, both prior and subsequent. (See Pow. Dev. pt. 2, chap. 17; Jar. on Wills, chap. 31.) Mr. Jarman states (vol. 2, p. 155), as a conclusion from all the cases, that illegitimate children “ are not objects of a gift to children, or issue of any other degree, unless a distinct intention to that effect be manifest upon the face of the will, and if by possibility legitimate children could have taken as a class under such gift, illegitimate children cannot.” The English courts have usually adhered, with a far greater degree of strictness and pertinacity than would in general be deemed allowable *153by the courts of tliis country, to any rule for the construction of wills, when once established. In some instances they have done so when an application of the rule required them to violate the manifest intention of the testator. A notable instance of this occurred in the case of Edmunds v. Fessey (7 Jur. [N. S.], 282). In that case there was a bequest “ to each of the sons and daughters of A. E. living' at my death. ” At the death of the testator there were living three sons and one daughter of A. B. One of the sons and the daughter were illegitimate. Sir JoiiN Eojiilly, notwithstanding he was of opinion that it was evident the testator intended to include all the children, held that the illegitimate daughter took the legacy, but that the illegitimate son was excluded. See the remarks of Mr. J. Rcdfield on this subject. (2 Redf. on Wills, 340 et seq.; 1 id,, 656 et seq.) In Williams on Executors, etc. (7th ed., vol. 2, p, 1099), it is stated that the rule of Mr. Jarman is not an invariable one, and that statement is confirmed by recent cases in England.
In Holt v. Sindrey (L. R., 7 Eq. Ca., 170), the testator made a bequest to his daughter Mary, the wife of John Lattimer, for life, and after her death to her children. Mary married one Elen-loy, but separated from him on the day of the marriage, and they never mot again. Within a year after they separated, and more than thirty years before the death of Flcniey, Mary married John D. Lattimer, and had issue by him seven children. Sir Joi-iN Stuart, Y. C., held that the illegitimate children of Mary took under the will, because they were clearly described therein as the objects of the gift. He said, “ the fact that, from ah unknown circumstance, the children begotten of Mary Lattimer were, although the testator did not know it, illegitimate, seems to me to have nothing to do with the question whether they are or are not sufficiently described in the will.” In Lepine v. Bean (L. R., 10 Eq., 160), the testator, William Bean, gave his real and personal estate to trustees in trust, to pay the income to his “ dear wife, Margaret Bean,” for life, provided she should so long continue his widow and unmarried, and after her death or second marriage, upon trust to divide the estate between all his children in equal shares. The testator was not married to Margaret Bean, but he had a wife named Elizabeth Bean living. She was seventy *154years old, aiul the testator had always lived apart from her, and he had had no children by her. Margaret Bean had always lived with the testator, and was recognized as his wife, and he had four children by her, two of whom had died before the date of the will — one was then alive, and the fourth was born afterwards. These children were all baptized in the name of the testator, and were always known by his name. There the Master of the Bolls, Sir JohN Boiiilly, held that after the death of Margaret Bean, if she continued unmarried, the estate would devolve on the surviving child of the testator and said Margaret. He said, “ lie calls her (Margaret) his wife all through the will, and he also calls the children ‘ my children.’ I think it is pretty clear that by these words he meant his children by the woman he called his wife.” In Crook v. Hill (L. R., 3 Ch. Div., 773; S. C. L. R., 6 Ch., 311), there was a gift to the testator’s daughter, Mary, with power of appointment to her children. She had had two children by one Crook. Mary was not the wife of Crook, although she had gone through the ceremony of marriage with him, nor could she become his wife, because Crook was her deceased sister’s husband. It was held that, as by the will the testator showed that he intended to recognize the children as legitimate, they were entitled to take under the gift to children, notwithstanding they were illegitimate. That decision was affirmed in the House of Lords (L. R., 6 H. L., Ca., 265). The doctrine thus laid down is fully recognized and approved in Dorin v. Darin (L. R., 17 Eq., 463; S. C. L. R., 7 H. L. Ca., 568; Laker v. Hordern, L. R., 1 Ch. Div., 644; Barlow v. Orde, L. R., 3 Priv. Coun. App., 164-189.)
Our own reports do not abound with cases analogous to that which is before the court. The rule as stated by Ch. Walworth, derived from the English cases then (1841) published is, that “ when there are legitimate children in existence at the time of making the will, so as to satisfy the words of the devise or bequest in their primary sense, an illegitimate- child cannot take under a general devise or bequest to children, as a class, unless there is something appearing upon the face of the will to show that the testator intended to include others besides legitimate children. (Collins v. Hoxie, 9 Paige, 87, 88.) A different rule was not applied in Gardner v. Heyer (2 Paige, 11), for in that case the testator *155was never married, and a devise to his illegitimate children was held to be good, and the chancellor cited with approval the case of Beachcroft v. Beachcroft (supra). In Cromer v. Pinckney (3 Barb. Ch., 466), the chancellor reiterated the rule in Oollins v. Iloxie, and enlarged it by giving effect to extrinsic circumstances existing when the will was made in determining the construction of the will.
The rule, thus established, simply carries out the intention of the testator, as manifested by the language which ho used, in conjunction with the circumstances which surrounded him at that time, instead of defeating that intention by attributing to words a technical meaning different from the sense in which the testator used them.
It is very true that in the cases cited there were no legitimate children to whom the description of the will could apply, and that if the gifts to illegitimate children had not been upheld, the gifts would have failed altogether. In some of the cases, also, the description of children might have been applied to legitimate children if there had been any, without violating the intention of the testator, and, in such an event, no doubt would have been so applied. But when the intention of the testator to give to illegitimate children is clear, the fact that legitimate children are in existence, who might possibly take under the description of the will, cannot defeat that intention. In all such cases the language used must be interpreted in its ordinary sense, unless, as stated by Ch. Walworth, in the cases above mentioned, there be something upon the will, in conjunction with the circumstances surrounding the testator at the time it was made, to show that another sense was intended. The rule is a reasonable one, and is also a logical consequence of the power of testamentary disposition. A man has a right, qualified only by the restraints of positive law, to give his property to whomsoever he pleases. If he has both legitimate and illegitimate children, he may make either or both the objects of his bounty. Under the description of “ children,” alone, only legitimate children will take. But the will by itself, or in connection with such extraneous evidence as the law permits to be given, may show that the objects intended by the testator include both classes, or one class, to the exclusion *156of the other. The intention of the testator is the pole star of construction, and must in every case determine the effect to be given to the will. (1 R S., 748, § 2.)
In the case before us I think that the will furnishes very cogent evidence that the persons whom the testator intended by the description of “ children ” were those who were born of the woman whom he called his wife. In the case of Hill v. Crook (L. R., 6 H. L. Cas., 285), before cited, Lord Chancellor CairNES said: “ It appears to me that the terms 1 husband’ and ‘ wife,’ ‘ father’ and ‘mother,’ and ‘children’ are all correlative terms. If a father knows that his daughter has children, by a connection which he calls a 1 marriage ’ with a man whom he calls her ‘ husband,’ terming the daughter the ‘ wife ’ of that husband, I am at a loss to understand the meaning of language, if you are not to impute to that same person, when he speaks of the 1 children ’ of his daughter, this meaning, that as he has termed his daughter, and the man with whom she was living, ‘ wife ’ and ‘ husband,’ so also he means to term the offspring, born of that so-called marriage, the children, according to that nomenclature. That is all that your Lordships have to find. If you find that that is the nomenclature used by the testator, taking his will as the dictionary from which you are to find the meaning of the term he has used, that is all which the law, as I understand the cases, requires.” In Barlow v. Orde (supra), Lord Chancellor Westbury held “that the word ‘ children ’ must be taken in that sense that it was used by the testator — that is to say, its extent of meaning in the vocabulary and mind of the testator must be determined in the will itself.” These remarks applied to those cases, and to this case are sensible and just. See, also, prevailing opinion in Cutter v. Doughty (7 Hill, 305), reversing. S. C. (23 Wend., 513). Whether the testator was lawfully married to Catharine Lalor or not, there can be no question that he thought she was his lawful wife, and that he did not regard his connection with Jane Valentine, whatever it was, as one which created a marriage between them. That being so, it follows as a corollary that the children whom he intended to provide for were the children who were born of the woman whom he called his “ wife.” If no dispute existed respecting the validity of the testator’s marriage with said Catharine, those children would *157have taken by force of the meaning which the law gives to the term “children.” Using the will as the testator’s glossary, or dictionary, to ascertain the meaning of that term, the result must be the same. In other words, the children designated in the will are those born of the testator’s wife, and for the purpose of the construction of his will she must be treated as his “wife,” whom the testator called his wife. Additional evidence that such was the intention of the testator also arises from the fact that all the provisions for children are in terms for sons and daughters, in the plural number, which, of course, are not strictly applicable to the one son and one daughter of said Jane; from the appointment of his so-called wife Catharine guardian of the persons of his infant children, whereas only one of his children by said Jane was an infant, and from the extreme improbability that the testator intended to make his so-called wife guardian of the person of the infant son of said Jane. Having thus ascertained that the children of Catharine are those intended, the others must necessarily be excluded, for the reason that they do not fall within the geiieral description of children according to the vocabulary of the testator.
Upon the whole I think that the judgment, with the modifications suggested, should be affirmed, with costs to the respondents, 'to be paid out of the estate.
BarNard, P. J., concurred; DykmaN, J., not sitting.Judgment affirmed, and modified in accordance with opinion.