In Re the Estate of Burke

48 N.J. 50 (1966) 222 A.2d 273

IN THE MATTER OF THE ESTATE OF ELIZABETH WEST BURKE, DECEASED.
TRUST FOR MADELEINE FORREST APPONYI (FORMERLY BURKE).

The Supreme Court of New Jersey.

Argued December 21, 1965. Decided August 29, 1966.

Mr. James C. Pitney argued the cause for defendants-appellants and cross-respondents Edith L. Wilkinson, Lawrence Wilkinson, J. Burke Wilkinson, Patricia Wilkinson Howard, Kenneth S. Walker, Mabel Burke Walker Lewis and Elizabeth Lee Walker Stoddard. (Messrs. Pitney, Hardin & Kipp, attorneys; Mr. Pitney, on the brief).

Mr. Donald B. Jones argued the cause for defendants-respondents and cross-appellants William Goldberg, et al., sub-assignees of Randolph Forrest Burke.

Mr. Paul E. Doherty argued the cause for defendant-respondent and cross-respondent The Chase Manhattan Bank, executor of the estate of Madeleine Forrest Apponyi, deceased.

Mr. Paul M. Hanlon argued the cause for defendant-respondent and cross-respondent Manufacturers Hanover Trust Company, surviving and successor trustee of the Trust created under Item V of the last will and testament of Elizabeth West Burke, deceased, for Madeleine Forrest Apponyi (formerly Burke). (Messrs. O'Mara, Schumann, Davis & Lynch, attorneys; Mr. Frederic W. Schumann and Mr. Hanlon, on the brief).

The opinion of the court was delivered by HALL, J.

The question on this appeal is whether the remainder interest in one of three residuary trusts created by the will of Elizabeth West Burke pours over into the other two trusts or is to be disposed of as intestate property of the testatrix. It arises on the death of the life beneficiary by reason of the prior demise of all designated remaindermen in that trust. The will makes no provision whatever for devolution in this contingency. There is therefore a case of complete gap, compare In re Estate of Morton, 48 N.J. 42 (1966), bringing into play the matter of "probable intent." See Fidelity Union Trust Co. v. Robert, 36 N.J. 561 (1962).

The Chancery Division, in the trustee's action for instructions, found that the probable intent of the testatrix in the circumstances could not be ascertained and consequently directed that the trust remainder be distributed by intestacy among the next of kin of the testatrix determined as of the date of her death. The principal appeal is by the beneficiaries of the other two trusts. We certified it on application prior to argument in the Appellate Division. R.R. 1:10-1A.

The circumstantial setting of the problem is evidentially confined to the will itself and family history. The will was executed in 1916. Mrs. Burke, apparently then a widow, had had three children, a son, Edward F. Burke, and two daughters. Mabel Burke Walker and Edith L. Wilkinson. Mabel had died in 1912, leaving her husband, John Y.G. Walker, and three children. Kenneth S. Walker, Mabel Burke Walker (now Lewis) and Elizabeth Lee Walker (now Stoddard). Edward had passed away in 1915, shortly before his mother's will was executed, leaving his wife, Madeleine Forrest Burke (she was later remarried to a man named Apponyi) and three children. Randolph Forrest Burke, Edward W. Burke and Madeleine (who later married Charles G. King, III). All of these grandchildren must have been quite young. We know only that Randolph was then 15 years old. Edith was living in 1916, and still is. She presently has three living adult children. Lawrence Wilkinson, J. Burke Wilkinson and Patricia Wilkinson Howard. Whether any had been born by 1916 we are not advised.

At the time of the execution of the will, the testatrix was a resident of Essex County, New Jersey, and continued to be so until her death in 1924. The will was prepared, however, by a Baltimore attorney, who seemingly had been her son's lawyer, the son being a resident of Maryland when he died. There were no deaths in the family between the 1916 will and Mrs. Burke's passing in 1924. Thereafter, in the Mabel Burke Walker branch, Mabel's husband, John Y.G. Walker, died in 1940, leaving the three previously mentioned children of that marriage, who still survive. In the Edward F. Burke branch, the daughter Madeleine King died in 1929, a resident of Rhode Island, intestate and without issue, but leaving her husband (who subsequently died in 1952). The son Edward W. died intestate in 1936, a resident of Texas. He had never married. The son Randolph died a resident of Florida in 1961, at the age of 60, testate, but without leaving issue or a surviving spouse. Edward F. Burke's widow, Madeleine Forrest Burke Apponyi, survived until 1963 when, at the age of 87, she died testate. Thus, all her children by her marriage to Edward F. Burke had predeceased her and none of them left issue. This fact gives rise to the question before us.

Mrs. Burke's main testamentary scheme was a division of the residue of her estate, "real, personal and mixed, of whatsoever kind and description and wheresoever located, whether in possession or remainder," into three equal parts, each to be held in separate trust for the benefit of one of the three branches of her family. Preliminarily, she made some small bequests and set up two other trusts. The first of these trusts comprised her real property and was to extend for a maximum period of five years to permit orderly sale within that time and management in the interim. The proceeds were directed to be divided into three equal parts, to pour over into each of the residuary trusts. The second consisted of $25,000 in bonds, the income from which was to be paid to her sister during the latter's life, with the corpus thereafter also to be divided into three equal parts and similarly paid into the residuary trusts.

The residuary trusts we will designate as the "Apponyi trust" (Item V of the will, with which we are particularly concerned), the "Walker trust" (Item VI of the will), and the "Wilkinson trust" (Item VII of the will). The basic pattern of each was payment of the income to her surviving child (Wilkinson trust) or her deceased children's surviving spouses (Apponyi and Walker trusts) for life, with each remainder to be divided, on the death of the life beneficiary, equally among her then surviving grandchildren in that branch, per stirpes. The specific language of the Apponyi trust read: "upon the death of the said Madeleine Forrest Burke, I direct that my said trustees shall divide this trust fund into as many equal parts as there are children of the said Madeleine Forrest Burke and my son, Edward F. Burke, at that time surviving, the children of any deceased child to take by right of representation the share of the parent * * *." The Walker trust language was to the same effect. The identity of the grandchildren who could possibly take in each of these two trusts was fixed when the will was executed for the Burke family parent had already died in each instance. This was not true of the Wilkinson trust since Mrs. Wilkinson was then living and could have further children.

The time and method of payment of the share of each surviving grandchild or his representative were thoroughly spelled out in each trust. The basic pattern was essentially the same, with some individual variations. In the Apponyi and Walker trusts, distribution of the share of principal was to be made when each grandchild attained age 25. Each Apponyi grandchild was also to receive a $30,000 advance on his share when he reached that age, though his mother be still living. In the Wilkinson trust, the age was 21. In all three cases, each share was to continue in trust until the distribution age was reached, with the income paid to or for the benefit of the ultimate beneficiary in the interim.

The contingency distribution provisions were detailed and complex. As has been said, in the event any grandchild predeceased his life tenant leaving a child or children, such issue took his remainder interest by representation. The share of principal to which he thereby became entitled was to be paid to him when his parent would, in the case of the Apponyi and Walker trusts, have become 25 years old had he lived. It was further provided that if a grandchild should survive the life tenant, but die before reaching 21 years of age without leaving a child or children, his share would devolve by survivorship upon the remaining grandchildren interested in that trust. In addition, the testatrix gave the Apponyi and Walker grandchildren surviving the life tenant the power between age 21 and age 25 to dispose of their prospective shares by will and, in default of such appointment, the share was ultimately to pass in the event of the beneficiary's death during that period as his intestate property pursuant to, in the case of the Apponyi grandchildren, the laws of Maryland and, with respect to the Walker grandchildren, the laws of New Jersey.

When the entire will is surveyed, it is apparent, however, that several contingencies common to all the trusts are not covered. For present purposes, the following may be noted. There is no express provision as to what should occur in the event a life beneficiary predeceased the testatrix. There is a complete gap — the situation confronting us — in the event the life tenant survives the testatrix, but dies leaving no issue. Another unprovided for situation is the death of the life tenant subsequent to that of the testatrix, with children then living, and the subsequent death of all such children before reaching age 21 without issue. In appraising the unmet contingencies, it can fairly be concluded that the testatrix provided in great specificity for occurrences in the usually expected chronology of death as between generations, i.e. children surviving their parents, but made little provision for the eventualities of deaths happening outside of that course.

With further reference to the Apponyi trust, by 1949 Madeleine King and Edward W. Burke had died without issue, leaving Randolph, then 48 years old, as the sole remaining grandchild who would take the entire corpus of that trust upon his mother's death if he survived her. The corpus was then valued at just short of $200,000. (Its present worth is about $270,000.) In that year Randolph sold $150,000 of his expectancy in the corpus to one Leonard P. Levy for $25,000, then paid to him in cash, and assigned his interest accordingly. The assignee also paid a $5,000 finder's fee to some third party. Levy thereafter made sub-assignments of the $150,000 interest to some 28 others in varying percentages. These 28 comprise the sub-assignees who are among the defendants-respondents and are the cross-appellants here. Mention is made of the transaction to identify these parties and their interest. The alienation of the expectancy has no bearing on the resolution of the question before us. It was not prohibited by the will (the testatrix probably never even thought of the possibility), the sub-assignees' rights rise no higher than those of the original assignor at the time of the sale, and we fail to see how the transaction in any way aids in ascertaining the testatrix' intention with respect to the failure of issue on the subsequent death of Mrs. Apponyi.

The positions of the interested parties in the trial court, which are renewed here, and the substance of the outcome there should be briefly outlined. By reason of the death of the life tenant of the Walker trust in 1940 leaving three children of the marriage each over 25 years of age, the principal thereof was then distributed equally among those children and the trust terminated. They, along with the beneficiaries of the Wilkinson trust (the life tenant and her adult children), took the position that the Apponyi trust corpus should pour over one-half into the Wilkinson trust and one-sixth going to each of the three Walker remaindermen. The Chancery Division, in holding that the Apponyi trust remainder passed as intestate property of the testatrix, decided that each of the three family branches was to receive one-third thereof. This gave Mrs. Wilkinson one-third outright and each of the Walker children one-ninth. They are the appellants here.

The other parties in interest appearing, including the trustee, the sub-assignees and Mrs. Apponyi's executor, claimed devolution should be by intestacy. The trial court agreeing, the judgment directed that the other third of the Apponyi trust pass to the testatrix' next of kin in that branch, with each of these sub-branches representing the three deceased grandchildren receiving one-ninth. Madeleine King's one-ninth (her estate is not represented) passed, according to the intestacy laws of the state of her residence at death, half to her husband's intestate estate (thus passing out of the Burke family) and half to her mother's (Mrs. Apponyi) estate (thus presumably also passing out of the Burke family by the Apponyi will). Edward W. Burke's one-ninth (his estate is not represented) passed, according to the intestacy laws of the state of his residence at death, half to his mother's estate and half to the estate of Randolph, who had survived him. Randolph's one-ninth, plus the one-eighteenth coming from Edward's share, passed to the sub-assignees to the extent of their interest, with any excess going to Randolph's executor. (Randolph's executor appeared at the trial level, but not on appeal.) Since, under the trial court's judgment, Randolph's total participation amounted only to about $45,000 against the sub-assignees' $150,000 interest, no excess remained for Randolph's executor.^[1]

The sub-assignees took an alternative position that the contingent remainder of the Apponyi trust became vested entirely in Randolph after the death of his brother and sister, by reason of certain language used in the will, and thus the entire trust corpus passed to Randolph's executor on his mother's death, subject to the $150,000 assignment. The trial court rejected this thesis and the sub-assignees have cross-appealed from this portion of the judgment. The same view is asserted here as well as by Mrs. Apponyi's executor. (We fail to see the latter's interest in this theory unless Randolph's will, the terms of which are not disclosed, bequeathed something to his mother.) We likewise can detect no merit in the contention. As the trial judge aptly put it: "Clearly the contingency that the testatrix provided for was that the child or children survive their mother, Madeleine Forrest Burke (Apponyi), the life tenant, not their fellow remaindermen." The rule of construction is clear, in the absence of a contrary intention, that in such a contingency, survivorship constitutes a condition precedent to a right to the gift. 6 New Jersey Practice (Clapp, Wills and Administration), § 572, 573 and 584 (3d ed. 1962). We can find no evidence whatever of any opposite intent here.

Before reaching the main point in appellants' argument for reversal, we should dispose of an alternative proposition they advance. It is suggested that the inclusion in the earlier quoted opening of the will's residuary clause of all property "whether in possession or remainder" was intended to encompass in the residue the remainder, which had failed, of one of the trusts created by the residuary clause itself. We thoroughly agree with the trial court that the view is not well taken. The phrase in question constitutes ancient words of art, referring only to the nature and extent of the testatrix' property owned at the time of her death, inserted to make certain all was included in the residue and would pass under the will. See Gardner's Executor v. Gardner, 37 N.J. Eq. 487 (Ch. 1883), affirmed o.b. 39 N.J. Eq. 337 (E. & A. 1884).

As has been indicated, the trial court turned the result on inability to ascertain the intent of the testatrix upon the happening of the unprovided for contingency. It therefore felt compelled to apply the old common law rule that when a gift of a portion of the residue fails, the share passes to the heirs and next of kin of the testator by intestacy rather than to the remaining residuary beneficiaries, absent a contrary intention expressly or impliedly found in the will. 4 Bowe-Parker: Page on Wills § 33.56 (1961); 5 Clapp op. cit. supra. § 402; Annot: Devolution of lapsed portion of residuary estate, 36 A.L.R.2d. 1117 (1954). The rule has been changed by statute in New Jersey, N.J.S. 3A:3-14 enacted by L. 1947, c. 380 (amended as to title only by L. 1948, c. 139), applicable only, however, to wills of testators dying after July 3, 1947. The trial judge expressed the view that but for this limitation, the statute would dictate the opposite result here.^[2]

Appellants urge that the common law rule be retroactively overruled to conform to the policy of N.J.S. 3A:3-14, as has recently been done in Illinois, (Schroeder v. Benz, 9 Ill.2d 589, 138 N.E.2d 496 (1956); cf. In re Arens' Trust, 41 N.J. 364 (1964)), or that in any event, there is sufficient indication of Mrs. Burke's probable intention to require the division of the Apponyi trust remainder between the other two trusts.

There is no doubt that the "no residue of the residue" rule has long been followed in New Jersey (e.g., Rippel v. King, 126 N.J. Eq. 297 (Ch. 1939), affirmed o.b. 128 N.J. Eq. 179 (E. & A. 1940); Lawes v. Lynch, 6 N.J. 1 (1950)) in common with the majority of other jurisdictions. Annot., supra, 36 A.L.R.2d at p. 1118. Though a basis for distinction is conceivable, it has been considered generally applicable not only to outright residuary dispositions but also to those in trust, here (e.g. Brown v. Fidelity Union Trust Co., 134 N.J. Eq. 217 (Ch. 1943), affirmed o.b. 135 N.J. Eq. 461 (E. & A. 1944)) and elsewhere (e.g. Davis v. Mercantile-Safe Deposit & Trust Co., 235 Md. 266, 201 A.2d 373 (Ct. App. 1964)). The rule seems to have rested on the technical ground that residuary gifts which failed were already part of the residue and on the supposed intent of the testator that he would not wish surviving residuary beneficiaries to receive larger shares than the will gave them.

Criticism of the rule has been heavy. Some states have rejected it by decision (e.g. Commerce National Bank of Toledo v. Browning, 158 Ohio St. 54, 107 N.E.2d 120 (Sup. Ct. 1952)) or, like New Jersey, have abrogated it by statute, to some extent at least. The bases of the criticisms are well set forth in two decisions in this State. In re Moloney's Estate, 15 N.J. Super. 583 (Cty. Ct. 1951); Skinner v. McCormick, 62 N.J. Super. 256 (App. Div. 1960). The substance of them is that the presence of a complete residuary clause rather conclusively evidences the testator's intent not to die intestate as to any of his property and that his intent is ordinarily better served by giving the share which failed to the remaining residuary beneficiary or beneficiaries. Consequently, even before the days of the modern doctrine of probable intent, it was not unusual to find the use of exceptions or other methods of avoidance of the rule on rather tenuous grounds where the court sought to effectuate its view of the testator's desire in the contingency. See Aitken v. Sharp, 93 N.J. Eq. 336 (Ch. 1921); Brown v. Fidelity Union Trust Co., supra (134 N.J. Eq. 217). The suggestion of some of the respondents that the Maryland lawyer who drew this will advised his client, correctly, that Maryland followed the common law rule, and therefore the omission of contingent provisions was deliberate to bring about an intestate result if there were a failure of all remaindermen in one of the trusts, seems unrealistic. Further it may be noted that Maryland courts today strain to find a non-intestate intent in analogous situations. See Davis v. Mercantile-Safe Deposit & Trust Co., supra, (201 A.2d 373).

Any mechanical rule for the disposition of a lapse in the residue — whether it be by the common law in favor of intestacy or, oppositely, leading to survivorship, or by statutory provision — is, standing alone, likely to be quite arbitrary. Failure to provide for the contingency may be an oversight, intentional in the sense that in the expected and normal order of deaths the named beneficiaries would survive and take and the testator thought it unnecessary even to consider provision for contingencies (see Skinner v. McCormick, supra (62 N.J. Super., at p. 263)) or, in rare instances, deliberate in the sense that the testator knew the possible consequences and was entirely willing that intestate succession then take over. The point is illustrated by the case of a hypothetical testator, an elderly widower, who leaves his residuary estate in two equal shares, one half to his two children, per stirpes, with survivorship, and the other half to a charity. Both children unexpectedly are killed in an accident, leaving no issue, and the testator himself dies within a day or two thereafter before he has an opportunity to change his will. In the absence of some evidence of his intent in the situation, it might or might not be true that the testator would desire the charity to take the whole residue to the exclusion of his other relatives. Since the common law rule and most statutes indicate they are to be applicable only in the absence of a contrary intent, solution of the question in most cases should come down to the matter of the testator's intent in the circumstances. Under the present concept of probable intent in this State, the old common law rule, and in fact N.J.S. 3A:3-14, have lost most of their former force and effect. We feel it unnecessary to consider further the question of retroactive overruling of our common law rule with respect to testators dying before 1947 since, in disagreement with the trial judge, we are satisfied that Mrs. Burke's probable intent in the premises can be properly ascertained and that such intent dictates the pour over of the Apponyi trust remainder into the other two trusts in equal shares.

The current view of probable intent to which this court is now committed was crystallized in Justice Jacobs' opinion in Fidelity Union Trust Co. v. Robert, supra (36 N.J. 561), following the ground-breaking decisions in In re Klein's Estate, 36 N.J. Super. 407, 415 (concurring opinion of Judge Clapp) (App. Div. 1955) and Bank of New York v. Black, 26 N.J. 276 (1958). The essence of this broader and more liberal approach to will construction, is found in certain key expressions in Robert. "* * * `when we say we are determining the testator's intent, we mean his probable intent.'" "* * * [I]n ascertaining the subjective intent of the testator, courts will give primary emphasis to his dominant plan and purpose as they appear from the entirety of his will when read and considered in the light of the surrounding facts and circumstances * * *. So far as the situation fairly permits, courts will ascribe to the testator, `those impulses which are common to human nature, and will construe the will so as to effectuate those impulses.'" "* * * [T]he court's endeavor is to put itself in the testator's position insofar as possible in the effort to accomplish what he would have done had he `envisioned the present inquiry.'" (36 N.J., at pp. 564-566.) "Since the goal is the ascertainment of the testator's probable intent in this particular will construction case, precedents involving the construction of other wills have no great force * * *. Similarly, canons of construction which are largely designed to serve as aids in ascertaining the testator's probable intent have no controlling force." (36 N.J., at p. 568.) As was very recently said in In re Estate of Morton, 48 N.J. 42 (1966): "The inquiry is always peculiar to the individual circumstances." See also 5 Clapp, op. cit. supra, § 196 (1965 supp.), 416. Among cases since Robert applying this approach may be noted In re Cook's Estate, 44 N.J. 1 (1965), and Darpino v. D'Arpino, 73 N.J. Super. 262 (App. Div. 1962). See also In re Coes Estate, 42 N.J. 485 (1964).^[3] While a court may not, of course, conjure up an interpretation or derive a missing testamentary provision out of the whole cloth, it may, on the basis of the entire will, competent extrinsic evidence and common human impulses strive reasonably to ascertain and carry out what the testator probably intended should be the disposition if the present situation developed.

This positive approach has particular significance in the case, as here, of an unprovided for contingency causing a failure in the complete disposition of the residue. Formerly the judicial viewpoint was quite negative. When the problem arose in Lawes v. Lynch, supra, the court looked at it this way in arriving at a conclusion of partial intestacy:

"Since the testator is presumed to know the law, it must be presumed that he knew how to provide for the right of survivorship as well as how to provide for substitution. If the testator had intended survivorship as an attribute of the estate given he could have expressly so provided. Manifestly he did not do so." (6 N.J., at p. 9.)

To be contrasted is the opposite result reached in analogous situations in Robert and Cook where judicially ascertained probable intention carried the day. In such situations, of paramount importance is the great strength of the so-called presumption against intestacy, cogently expressed in the Robert opinion by a quotation from a New York case: "`A testator, by the act of the making of a will, casts grave doubt on any assumption that he expressly intends to chance dying intestate as to any portion of his property * * *. "The idea of anyone deliberately purposing to die testate as to a portion of his estate and intestate as to another portion is so unusual in the history of testamentary disposition as to justify almost any construction to escape it."'" (36 N.J., at p. 572.)

Consideration under these principles of the Burke will and the surrounding circumstances earlier detailed leads to a pour over and not to intestacy. First, looking at the will as a whole, one gains the very definite impression from the generally complete and detailed provisions that the testatrix intended all of her property to pass by the will. The several common unprovided for contingencies, when compared with the great specificity otherwise, show that, as previously pointed out, she was generally proceeding on the assumption that family deaths would occur in the expected chronology. It seems certain that she thought it unnecessary to make provision otherwise and that this is not a case of a knowing decision for intestate distribution to prevail in the event of the occurrence of deaths out of the normal order. Indeed, an intended intestate result at any stage of an estate plan is so unusual that, where that is the actual desire, one almost invariably finds an express provision in a will to that effect, especially in a situation like this one where the estate was a large one, the scrivener was obviously a careful practitioner and the will was otherwise so detailed. The absence thereof is a persuasive indication that Mrs. Burke had no such intent.

Secondly, the over-all testamentary plan gleaned from the provisions previously sketched, strongly dictates the conclusion we reach. The primary purpose was to benefit only members of the family, both as to income and principal, in equal shares between each of the three branches. The first, but limited, objects of Mrs. Burke's bounty were her living daughter and the surviving spouses of her two deceased children. She wished them to receive the income for the remainder of their lives. But her intention is clear that such was to be the extent of their interest. They were given no right whatever in the corpus and it appears certain that she did not intend them or their estates to receive any under any circumstances. The opposite would be the case as to Mrs. Apponyi's estate if the corpus of that trust were to pass by intestacy. It is equally clear that the main beneficiaries were to be the grandchildren surviving the income beneficiary, or their issue by representation. The testatrix chose to have outright distribution at that level of her descendants at the designated ages of the distributees, each thereafter, of course, to have complete freedom to deal with his inheritance as he wished. The point is that Mrs. Burke's intent is clear that the corpus of the estate was to stay in the family until the respective distributions had been so made. To allow the unexpected contingency of the death of all grandchildren without issue before that of their parent-life tenant to mandate distribution measured by intestacy of the testatrix would do violence to that basic intent. It would permit, as is the fact here, corpus to leave the family, at least in part, before the testatrix wanted that possibility to exist.

The trial court appears to have been largely actuated to reach an intestate result by the provisions in the Apponyi and Walker trusts authorizing a grandchild who survived the life tenant and attained the age of 21, to appoint by will the recipient of his prospective share of the corpus should he die before reaching the distribution age of 25 and, in the event of such death without the power of appointment having been exercised, directing the share to be ultimately distributed as the grandchild's intestate property. We think the court's reliance was misplaced and that this single provision does not require, or even indicate, a probable intent favoring distribution of the entire corpus as intestate property of the testatrix in the event of the contingency with which we are dealing. The court suggested that the testatrix' use of possible intestate distribution in this isolated situation showed that the concept of intestate succession was not repugnant to her. She was, however, providing for such distribution as the intestate property of the grandchild and not of hers. Moreover, the provision is clear evidence that where she wished intestacy to enter into devolution in any respect, she specifically said so.

Stress is also laid in support of the intestacy result on the independent, tripartite division of the estate without express provision for any interrelation between the three trusts. In our opinion, this really proves nothing with respect to the question before us, when it is considered, as has been pointed out, that we have here a case of a testatrix relying on the normal order of family deaths and so finding it unnecessary to provide otherwise.

A persuasive indication of Mrs. Burke's probable intent in the contingency which here occurred is to consider, in the light of this will pattern, what she would likely have done if, when the will was executed, her daughter-in-law had had no children by the testatrix' deceased son or all children born of that marriage had already passed away without issue. It seems clear to us that she would then have expressly provided for a pour over of the corpus of that third of her estate equally between the other two trusts upon the death of her daughter-in-law. We are convinced she would have similarly provided if she had envisioned the occurring contingency of the death of all her son's children without issue after her own death but before the demise of their mother. We conclude that the trial judge erred and that the Apponyi trust remainder should be declared to pass one-half to the Wilkinson trust and the other half by equal division among the three Walker trust remaindermen.

The sub-assignees suggest in the conclusion of their brief (although without any supporting discussion in the body thereof) that, if this court should reach the result it has, equitable terms ought to be imposed upon the successful appellants by directing that they take the Apponyi trust remainder subject to repayment of the sum paid to Randolph in 1949 on the sale of his expectancy, plus a reasonable increment on that amount. We cannot perceive any merit whatever in such a claim. It is most clear that the purchaser of the expectancy and his sub-assignees were simply gambling on Randolph's surviving his mother. They lost their wager and have no equities in their favor.

The judgment of the Chancery Division is modified in accordance with this opinion.

For modification — Chief Justice WEINTRAUB and Justices JACOBS, FRANCIS, PROCTOR and HALL — 5.

Opposed — None.

NOTES

[1] The last paragraph of the trial court's judgment granted leave to the "attorneys for all the parties herein * * * to make application for additional counsel fees upon the determination of any appeal which may be taken from this Judgment." Since the judgment allowed counsel fees for services in the trial court, we presume the quoted portion refers to fees for appellate services and authorizes application therefor to be made to the trial judge. Although no objection is asserted, we must comment that any such practice is erroneous. Applications relative to costs and allowance of counsel fees in all cases must be made in the court in which the costs accrued or the services claimed for were rendered. R.R. 1:9-3, applicable in the Appellate Division by virtue of R.R. 2:9-2, as construed in United States Pipe and Foundry Co. v. United Steel-workers of America, 37 N.J. 343, 357 n. 1 (1962); Sarner v. Sarner, 38 N.J. 463, 471 (1962).

[2] N.J.S. 3A:3-14 reads:

"When a residuary devise or bequest shall be made to 2 or more persons by the will of any testator dying after July 3, 1947, unless a contrary intention shall appear by the will, the share of any such residuary devisees or legatees dying before the testator and not saved from lapse by section 3A:3-13 of this title, or not capable of taking effect because of any other circumstance or cause, shall go to and be vested in the remaining residuary devisee or legatee, if any there be, and if more than 1, then to the remaining residuary devisees or legatees in proportion to their respective shares in said residue."

The statutory phrase "unless a contrary intention shall appear by the will" means "by the will as construed in the light of the surrounding circumstances," i.e., by application of the concept of probable intent. In re Cook's Estate, 44 N.J. 1, 10 (1965).

[3] The division in the court in Coe (dissenting opinion, 42 N.J., at p. 495) and Cook (dissenting opinion, 44 N.J., at p. 11) was not by reason of disagreement with the probable intent concept. Rather, in Coe it concerned the view that a long-standing positive rule of law ought not to be overturned retrospectively where there was nothing to show that the testator's intent was contrary thereto. In Cook, the differences involved primarily the matter of the nature and quality of the evidence relied upon to establish the intent found.