tion of the word, since the person named may have children. born after the date of the will, or even after the death of the testator, or where the time of vesting is deferred, before that time arrives, all of whom, in such case, would be entitled to claim the bequest.28 In the last case, the question arose in regard to illegitimate children taking by the general description of "children," the person named having no other, either at the date of the will or the death of the testator.29 We shall have occasion to consider this point more hereafter, as regards illegitimate children.

11. We will here indicate what seems to us to be the latest determination of the English courts of equity upon this and all the analogous questions.

(1.) The terms "children," grandchildren," "nephew," "niece," "cousin," or any similar term, must ordinarily be presumed to have been used in its primary and natural sense; and this sense cannot be enlarged or qualified, so as to admit any person not coming within such primary meaning, unless there be something upon the face of the will clearly indicating such intent in the mind of the testator, at the time of making his will, or unless such qualified or enlarged sense of the term may be gathered, with reasonable certainty, to have been the intention of the testator, either from the necessity of so construing it, in order to give it any operation, or from some other equally satisfactory probability, arising out of the state of facts existing at the date of the will, and which are of a character to be properly admissible in aid of the construction of the will.30

* Harris v. Lloyd, Turn. & Russ. 310.

* In Harris v. Lloyd, supra, the learned judge said: "I have not the least doubt this testator meant illegitimate children, but I am clearly of opinion that there is not enough, upon the face of this will, to authorize me to carry that intention into effect." Cartwright v. Vawdry, 5 Vesey, 530; Swaine v. Kennerley, 1 Ves. & B. 469. The proposition of the learned judge is scarcely maintainable, as we may show hereafter.

* 2 Story, Eq. Jur. § 1065 c, where we took occasion thus to classify the more

(2.) It is apparent, from a careful examination of the more recent decisions of the English courts of equity, that they adhere with great strictness and pertinacity to any rule of construction, when once established, although it may lead them quite one side of the obvious intention of the testator. In the American courts, probably, no such persistent adherence to the strict and literal application of the recognized canons of construction, to every case, has yet been arrived at, as a general

recent decisions upon this somewhat perplexing question, thus: "In a recent case before the Lord Chancellor, after a good deal of examination and discussion at the bar, it was held that a bequest to cousins,' simpliciter, includes first cousins only, in the absence of anything to explain the meaning of the testator. The Lord Chancellor said: 'I think that if a testator says no more than that he gives to "cousins," he must be taken to mean first cousins. That will be a practical construction, and one by which the parties entitled will be easily ascertained; it coincides, too, with ordinary experience, for when a person speaks of cousins, he generally means first cousins, the children of an uncle or aunt; — and I think that in the present case, there being first cousins, (and nothing to indicate that any others were intended,) this is the proper construction to adopt.' It seems to us this view, as a general exposition of the difficulty, is extremely satisfactory. Stoddart v. Nelson, 6 DeG., M. & G. 68; Stanger v. Nelson, ib. The counsel, in arguing the extension of the term 'cousins,' so as to include all persons coming within the general import of the term, referred to the following cases: Caldecott v. Harrison, 9 Sim. 457; Sanderson v. Bayley, 4 My. & Cr. 56; and Williams on Exrs. vol. 2, p. 885, 3d ed. See also Thompson v. Robinson, 5 Jur. N. s. 1196. The same construction was given to the word 'niece.' Crook v. Whitley, 7 DeG., M. & G. 490. See also Pride v. Fooks, 3 DeGex & Jones, 252; Jenkins v. Lord Clinton, 26 Beavan, 108; Smith v. Liddiard, 3 Kay & J. 252. And in order to enable illegitimate children to take under a bequest to 'daughters,' it would seem to be requisite to show that there were no other persons who could answer the description, and that their reputed character did answer it, and that this was understood by the testator, which last fact will not be inferred. Herbert in re, 6 Jur. N. s. 1027. How far illegitimate children, or their children, shall take, by general description, is matter of intention. Allen v. Webster, 6 Jur. N. s. 574. But illegitimate children, born after the date of the will, cannot take by general description, as 'such other child that may be born of my housekeeper,' &c. Medworth v. Pope, 5 Jur. N. s. 996."

thing. Particular instances of that kind may occur. But, in general, it will be found, that the American courts depart, with some degree of reluctance, from what is the obvious and unquestionable meaning of the testator, where so expressed as to be intelligible. In the case of Edmunds v. Fessey, already referred to, where Sir John Romilly, M. R., a very learned and experienced equity judge, felt bound to exclude illegitimate children wherever there were legitimate children answering the description. The result of the application of this rule to the particular case was, to admit an illegitimate daughter, there being no other daughter, and to exclude the illegitimate son, there being more than one legitimate son, thus meeting the words of the will, "sons and daughters" of A. B., the learned judge thus apologizing for the "anomalous result": "The result is necessarily somewhat anomalous, for I admit one of the illegitimate children and exclude the other. It is to be observed that there are two legitimate sons, sufficient to satisfy the word 'sons' in the plural. I regret the decision to which I feel myself obliged to come, because it is evident that the testator intended to include all the children of his cousin, whether legitimate or illegitimate.'

(3.) We believe it is safe to conclude that an American court would, under such circumstances, without hesitation, have included all the illegitimate children, and, as it seems to us, upon the best-established and most unquestionable rules of construction, as applicable to wills, that where the testator has used the word in question, or its equivalent or correlative, in the course of the same instrument, in a particular sense, it is admissible to give it that construction throughout the instrument, where that seems best to conform to the probable intent of the testator, in the particular place in question. The fair application of this rule, which is as well established, and of as controlling a character as any other, would have enabled the learned judge to escape the "anomalous result" which he seems so much to deplore, and at the same time to meet the acknowledged justice

and truth of the case, which is always a most gratifying result upon all subjects; and one, in regard to the determination of causes in courts of justice, which, if it is not reached, must always operate, in some degree, either as an impeachment of the law, or of the competency of the court in its administration, either of which we should always feel reluctant to admit.31

(4.) We think, too, that it is safe to affirm, that no court would be required to go into all possible conjectures, in regard to the occurrence of future facts, which might give the words of the will an intelligible signification, and to be able to exclude all such conjectural states of fact, before any secondary sense of the words of the will can be admitted. We should therefore dissent from the view, that, in order to extend the term "children" to include those of another generation, or those not legitimate, it must be shown, that, at the time the testator used the term "children" of A. B., the father had deceased, and had left no children legitimate, but only grandchildren, or illegitimate children, as has been urged by some writers of great authority upon this subject.32 All that reason, and the generally received rule of law upon the subject, would seem to require, in order to

$1 We do not intend here to imply, that many cases do not occur, where, from some mistake or defect in the will, it may become impracticable, without the violation of acknowledged principles of law, to reach the intent of the testator, and the justice of the case. As in Drake v. Drake, 8 Ho. Lds. Cas. 172, where the name and the description of the person in the will cannot be made to apply to the same person, but really identify two persons, with equal certainty. It not being a case of equivocation, direct evidence of intention is not admissible. And unless some preponderance of probability of intent can be reached by way of construction, the name must prevail over the description, without regard to the probability of which would be most likely to be mistaken in the particular

case.

2 2 Jarman, (ed. 1861,) 136. It is here said that the extended construction of these terms will only be allowable, when "otherwise the gift never could have an object." It is obvious that even the case of Edmunds v. Fessey, already alluded to, and where the construction adhered to was inconveniently nice, fell far short of the strict rule for which Mr. Jarman here contends.

give the words a secondary import, is, that it should appear that the testator was aware, at the date of his will, that the term children, or any similar term, used by him, could have no strict, literal, and technical application to the existing facts, in order to give it a secondary signification. And this knowledge it has been held must be proved, and cannot be presumed.33

(5.) And it seems that in general, where the word "children" is so construed as to include other descendants, it will be held to include all the descendants, as being synonymous with "issue." 34 But there will occur many cases where the testator will give evidence, by the language used in his will, or in other ways coming legitimately within the limits of aiding the construction of the words used, that while he intends "children" to include "grandchildren," he does not intend it shall have a wider extension.35 And this same indefiniteness of extension, when once it is allowed to go beyond its literal import, will apply, to some extent, to other relative terms, such as "nephew," "niece," "cousin," &c.

(6.) It seems to be well settled, as a general rule, that the construction of the words of a will cannot be affected by the occurrence of contingencies, not in the mind of the testator, either at the time of making his will, or before his death.36 For as no one can know, with certainty, what the testator might have been disposed to do, in a state of facts not presented to his

"Crook v. Whitley, 7 DeG., M. & G. 490, 496, by Lord Cranworth.

* Pride v. Fooks, 3 DeG. & J. 252, 275; Lord Cranworth in Crook v. Whitley, 7 DeG., M. & G. 490, 496.

of

"Fenn v. Death, 23 Beavan, 73.

"Pride v. Fooks, 3 DeG. & J. 252, 275. This rule of construction is one

very general application to wills. Hence, in aid of the construction of the will in general, we can only introduce extrinsic evidence to show the state of facts and of the testator's knowledge, at the time of making the will. But as for most purposes the words of the will speak from the death of the testator, and the instrument is merely inchoate and ambulatory, until that time, it may often afford some aid in conjecturing the sense in which the testator expected his will to be received, by showing the surrounding facts and the state of the testator's mind and knowledge up to the time of his death.