filed the following dissenting opinion:

The majority of the Court holds that the legacy to the children of the testator’s nephews and nieces was not a *290class gift, but a gift to contemplated and designated individuals with added description, and I concur. This is equivalent to a gift to George Marion, son of my nephew. The Court decides that if George Marion was not the son of his nephew, misdescription prevails over the designation (in that it nullifies the gift) because it is supposed that testator would want this result. In situations where there must be a choice between a designation and a misdescription, neither course is easy to travel nor is the end of the road a haven of complete satisfaction. Yet, to me it seems that the majority has chosen the wrong road and reached the wrong destination. In the absence of fraud, undue influence or compelling evidence to the contrary, the one absolute or certainty in the problem is that the testator named the individual who is to take and he should take, as the cases cited in the opinion of the majority hold. This Court adopts the conclusion of the chancellor that “it was likely” that the testator did not know George Marion was not the son of his nephew. This is an assumption—it was not proven—and if he did not know, he was not deliberately misled—fraud is not claimed—and the legatee had nothing whatever to do with any belief testator may have had. If it be assumed for the argument that Dr. Davis thought George Marion to be the son of his nephew, how can anyone ever know, or with any degree of certainty believe, that he would not have included him in the description of children or nephews and nieces? If his nephew regarded him, and so held him out as a son, this may well have been enough for Dr. Davis to so regard and treat him. He did not name the great-grandnieces or nephews, nor, as far as can be told, were they distinct personalities to him. Apparently, he thought of them in relation to nieces and nephews who were known personalities to him and whom he named. • He was good to the nephew who, for all practical purposes, had adopted George Marion. He sent him money for support and must have had an affection for him. How, except for speculation, can it be decided that the controlling motive for the gift to George Marion *291was that he was thought to be a blood relative and not because Dr. Davis thought of him in the same way as did the nephew? Significant in this regard, and unfavorable to the reasoning of the Court, is that Dr. Davis wanted known individuals to take and not all of his great nephews and nieces because they were such. If this were not so, he could have made a class gift which closed at his death or at a later named time.

If, for argument’s sake, it be assumed that it is likely that the testator did not know the facts and would have reacted as the Court has reacted for him if he had, is a will to be judicially rewritten because of no more than a belief that a testator would have acted as a court thinks he should have? Against any likelihood as to what would have been done, on any given belief, there stands the one indisputable fact that an intelligent testator left a legacy to George Marion. Why he did so, in what belief, or whether he would not have done so if there had been another belief, can only be a matter, to use the phrase of Smith v. Diggs, 128 Md. 394, 399, of “wildest speculation.”

I would affirm the decree.

delivered the opinion of the Court.

On a bill filed by trustees for the construction of the will of S. Griffith Davis, deceased, it was held that George Marion Sisk is entitled to a share in the income of the trust created therein for certain relatives including the seven great-nephews and great-nieces of the testator, and that Sandra Lee Davis, a great-niece born out of wedlock after the execution of the will but legitimated after the testator’s death was not entitled to a share. Appeals were taken by four of the great-nephews and great-nieces and by Sandra Lee Davis, through their respective guardians ad litem.

By item 4 of his will, Dr. Davis left the entire residue of his estate in trust, and provided that the net income therefrom should be paid to five brothers and sisters and six nephews and nieces named therein and to “the seven children of my nieces and nephews” without naming them. The scheme of this item was to allocate a specific number of shares in the income to each taker in such manner that each brother and sister would-receive four shares and each niece and nephew two shares, except that Steward Griffith Davis, a nephew, would receive only~ one share. This accounted for a total of thirty-one shares. It was then provided that the trustees should pay “To each of the seven children of my nieces and nephews one share thereof, in consequence of which if all of the above named beneficiaries are living at the time of my death the income of my estate will be divided into thirty-eight shares divisible as indicated above. Upon the death of any beneficiary his or her share of the net income shall cease. As a result of this plan the number of shares will continue to decrease as successive deaths occur with a consequent increase in the amount of income payable to each share. Upon the death of my brothers and sisters and my nephews and nieces, the income from my estate will eventually be divisible into seven shares, one of which will be payable to each of the seven children of my said nephews and nieces. When the youngest of said seven children of nieces and nephews *282attains the age of twenty-one years, assuming that my brothers and sisters and my nephews and nieces, have all died,' the trust will terminate and the entire trust estate is to be equally divided among the seven children of my nieces and nephews, the child or children of any deceased child to take the parent’s share per stirpes. Should any of the children of nieces and nephews (great-nieces or great-nephews) die without leaving issue, the share of such great-niece or great-nephew shall be included in the trust share of the survivors.” The will was executed on December 24, 1941, and the testator died on April 21, 1943.

The Chancellor found that when the testator made his will all of the named brothers and sisters and nephews and nieces were living, and there were six living infants, the children of a nephew or niece. These were the four children of Dr. Edward Hollister Davis, by his first marriage, and the two children of Martha D. Jay Galajikian. There was also living at the time of the execution of the will a child, George Marion Sisk, who resided in the household of the testator’s nephew, Steward Griffith Davis.

It was shown that Steward Griffith Davis and his present wife met in Phoenix, Arizona in February, 1939. Mrs. Davis, who was then Esther Lee McCann Sisk, wife of Marion Sisk, had separated from her husband and gone to Phoenix from Washington, D. C., in 1938, with her infant son George Marion Sisk, who had been born on March 14, 1935. Shortly after her meeting with Steward Griffith Davis, they began to live together and to hold themselves out as man and wife. George Marion Sisk was a member of this household and was held out to friends and neighbors as George Marion Davis. Beginning in 1939, both Steward Griffiith Davis'and his ostensible wife corresponded with the testator, who sent them funds from time to time. In this correspondence they each mentioned George Marion and referred to him as their son. Mrs. Davis testified that she saw letters from the testator in which he alluded to George *283Marion. In 1940 they moved to Texas. Subsequently, in November, 1942, a daughter named Sandra Lee was born to the parties. Mr. Davis testified that Sandra Lee is the only child of his own blood he has ever had. In 1945 Mrs. Sisk obtained a divorce from her husband, Sisk, and in 1946, she married Steward Griffith Davis in New Orleans, La.

The guardian ad litem for the four children of Dr. Edward Hollister Davis, a nephew of the testator,- contends that the Chancellor erred in holding that George Marion Sisk was entitled to a share in the accumulated and future income of the fund. It is conceded that the claimant was not in fact related to the testator in any way and was not a child of his nephew, Steward Griffith Davis, although, as the Chancellor found, the testator knew of his existence and “probably believed that George Marion Sisk was the blood child of his nephew.” It is contended that George Marion was not sufficiently identified as an object of the testator’s bounty and in any event cannot satisfy the condition of being a child of a nephew. It is further contended that he is barred from taking by reason of the misrepresentation as to his true status. The guardian ad litem for Sandra Lee Davis contends that there was a class gift to the children of nephews and nieces in which she can share by reason of her birth before the death of the testator. It is further contended that she is not disqualified by the fact that she was then illegitimate, in view of her subsequent legitimization.

On the point of identification, the Chancellor’s finding that the testator was aware of the existence of George Marion is supported by the evidence, and we think it is clear that the testator had him in mind as one of the seven children who would take, although deceived as to his true status. The will not only allocates one share out of the total of thirty-eight shares to “each of the seven children”, without provision for possible enlargement of the number, but allocates the total of thirty-eight shares, necessarily including the seven *284shares, to “all of the above -named beneficiaries”, although in fact the seven children were not named. 'For purposes of identification the objects of a gift are to be determined as of the date of the will and not as of the death of the testator. Darden v. Bright, 173 Md. 563, 572, and cases there cited. The will was evidently drawn by a competent lawyer and there is no reason to suppose that the use of the number seven was inadvertent. It is only by .counting George Marion that the numerical specification can be satisfied.

We see no basis for the contention that this is a ciass gift. In Evans v. Safe Deposit & Trust Co., 190 Md. 332, 340, it'was said: “* * * the Jarman definition of a class gift [citing -1 Jarman, Wills (6th ed.) p. 262] * * * [is] predicated on the intention of the testator to make a gift to ‘a body of persons uncertain in number at the time of the gift, to be ascertained at a future date.’ Without such an intention, there is no class and the lore of class gifts * * * is irrelevant. Of. Restatement, Property, §279, comment b. The ‘antithesis’ of a class gift ‘is a gift to an individual either by name or by some description sufficiently explicit to permit- the donee-to be identified as-the particular individual for whom the gift was intended.’ Boulden v. Dean, 167 Md. 101, 106, 173, A. 26, 28.” See also Miller, Construction of Wills, §68. '

In Jarman, Wills (8th ed.), p. 419, it is said: “If the testator refers to a number of children- in such a way as to show that he has certain individuals in-his mind, they take as personae designatae, and not as a class; as where he gives a sum ‘to be divided between the six children of A:’ ” Again (p. 449) it is said: “And if a testator after a gift to ‘children’, proceeds to name them, or if he specifies their number, as by giving ‘to the five children of A’, this is a designatio personarum, and is a bequest to those who are named, or to the five in existence at the date of the will * * These statements may be compared with the rule announced in the Restatement, Property, sec. 280, comment c: “When a convey*285anee, in describing the takers thereunder, uses a descriptive term [such as children] * * * and also states the number of takers, and the number so stated is the total number of those who fit the descriptive term at the time of the execution of the conveyance, then, unless a contrary intent of the conveyor is found from additional language or circumstances, the conveyance is construed to make a gift to individuals distributively. This form of limitation effectively identifies the intended takers.”

These authorities would seem to effectively dispose of the contention of the guardian ad litem of Sandra Lee that this is a class gift, and it is unnecessary to discuss the contention as to her legitimacy. George Marion, however, is sufficiently identified as one of the designated seven, and the crucial question is whether this designation can prevail over the description as “child” ■ of a nephew or niece. We find no occasion to ascribe an artificial meaning to the word “child”, so as to include a putative step-child within the description. This court refused to do so in the case of Fouke v. Kemp, 5 H. & J. 135, 138, where there were children and step-children living in the same household. The case might be different if there were no children to answer the description and the misdescription could be attributed to mistake. Re Jeans, Upton v. Jeans, 72 L. T. 835 [1895]. It may be noted that this court has always been reluctant to extend the literal meaning of the word “child.” See Sabit v. Safe Deposit & Trust Co., 184 Md. 24, (grandchildren), and cases there cited. See also Gutman v. Safe Deposit & Trust Co., 198 Md. 39, (adopted child).

As indicated in the Restatement, supra, the designation may be sufficiently certain, and yet may yield to a contrary intent shown by additional language or circumstances. We have found no case directly in point. Cases are cited in Jarman, Wills (8th ed.), pp. 416 and 1238, and Page, Wills (Lifetime ed.) §1007, where inaccurate descriptions have been rejected in favor of a person named. Where not named but merely designated, Jarman states that the same principle might apply but *286stronger evidence would be required. These cases usually involved situations where a legacy was left to a named individual described as the wife of another, when in fact the person named was not lawfully married.

In Anderson v. Berkley [1902] 1 Ch. 936, the testator left a legacy to his son’s “wife Letitia”. It was held that she could take, although the son had falsely represented that he was married to Letitia. In Re Lowe, 61 L. J. Ch. 415, it was held that a legacy to “the present wife” of the testator’s brother was good, although it appeared that the claimant and the brother were not married. See also Turner v. Brittain, [1863] 3 New Rep. 21, and Stothers v. Flieger, 13 N. J. Super. 379, 80 A. 2d 583. On the other hand, in In Re Taylor, [1925] 1 Ch. 739, where there was a legacy to the child or children of a deceased great-nephew, it was held that an illegitimate child could not take. In Perkins v. Iglehart, 183 Md. 520, 530, we held that Sally Woods could not take as the “widow” of William James Rucker, although it was shown that the testatrix knew she was engaged to Rucker when the will was executed and married to him twenty-one days later. It was said (p. 531) : “It was the relationship to the son, and not any particular friendship for Sally Woods which motivated the bequest.”

There is also the curious case of Stokeley v. Gordon, 8 Md. 496, where there was a devise “to Anna Maria German, wife of Jonathan German”. Jonathan had a daughter named Anna Maria and a wife named Catharine, both of whom were well known to the testatrix Mason, J., thought the wife should prevail. Eccleston, J., thought the daughter entitled, while Tuck, J., thought the devise was void for uncertainty. One of the chief points of disagreement was as to the extent to which extrinsic evidence was admissible. Judge Mason referred to a maxim attributed to Lord Bacon that “the truth of the names takes away, or controls, the error of the description”, but stated that it was well established that the rule is not inflexible.

*287In the instant case, while the matter is not free from difficulty, we think that George Marion is not entitled to a share. Aside from two bequests to a church and for upkeep of the family burying ground, the entire estate was left to persons related by blood to the decedent, and this scheme was carried out to the permissible extreme. There is no evidence that the testator ever saw George Marion or knew him in any personal capacity. It is of some significance that he was not named, although the testator knew his name. The testator was obviously mistaken as to his true status and we cannot find, under the circumstances, that the testator was indifferent to ties of blood, or had any desire to provide for a putative stepchild whom his nephew was under no legal or moral obligation to support. We think the description in this case is a stronger indication of intent than the numerical designation.

We reach this conclusion the more readily because of the fact that the mistake was due to misrepresentations made to the testator by the nephew and his ostensible wife. In Smith v. Diggs, 128 Md. 394, 398-9, it is stated that “ * * when a legacy is given to a person, under a particular character which he has falsely assumed, and which alone can be supposed the motive of the bounty, the law will not permit him to avail himself of it, and therefore he can not demand his legacy. But before applying this rule, the Court must be satisfied that the assumed character was the motive of the bounty’. The cases illustrating this rule are chiefly dealing with facts where legacies have been left to ones who have falsely created the impression in the mind of the testator that they are either married or unmarried according to the needs of the situation; or legacies left to children under a false impression as to their parentage.” In that case a favorite niece of the testator was alleged to have concealed from him her participation in litigation over her grandfather’s will, of which the testator disapproved. The court said (p. 400) that the testator’s motive was “because he was genuinely attached to her, as abundantly *288appears; and might not have allowed her alleged actions, even if known to him, to have affected his attitude to her.” It was held that there was no legally sufficient evidence from which it could be found that the will-was .procured by fraud. The case is not directly in point, but the statement of the general rule is apposite. See also In Re Carson’s Estate, 184 Cal. 437, 194 P. 5, and note 17 A. L. R. 247.

In 1 Roper, Legacies, p. 171, it is stated: “Wherever a legacy is given to a person under a particular description and character, which he himself had- falsely assumed; or where a testator, induced by the false pepresentations of third versons to regard the legatee in a relationship which claims his bounty, bequeaths him a legacy by a description according with such supposed relationship, and no other motive for such bounty can be supposed, the law will not, in either case,, permit the legatee to avail himself of the description, and therefore he cannot demand his legacy.” (Italics ours). The case relied on for the italicized statement is the case of Ex Parte Wallop, 4 Brown Ch. Cas. 90 [1792], where the disreputable Mrs. Brown palmed off children on the gullible Mr. Fellowes, who were neither his nor hers, resulting in legacies to her and the children by name. On the other hand, there are cases where a misrepresentation is not necessarily fatal, in the absence of a showing that it was made by the legatee to induce the legacy. Turner v. Brittain, supra; Anderson v. Berkley, supra. In Wilkinson v. Joughin, L. R. 2 Eq. Cases 319, 14 L. T. N. S. 394 (1866), it was held that a misrepresentation of widowhood, inducing a bigamous ■ marriage, barred the person, so representing herself, from taking a legacy, but not her child, who was left a legacy by name, with a description as “step-daughter”, by the deceived- spouse. .

In the instant case, while there is no direct evidence that the misrepresentation was designed to induce the legacy, it may be inferred that the references to “our son”, in letters asking for financial assistance which the testator gave, were intended to- arouse his sympathy and *289measure the extent of their needs, both present and future. The representations were not entirely innocent, although they may not have amounted to fraud. The fact that George Marion made no representations would not seem to be controlling under the circumstances. Cf. Davis v. Calvert, 5 G. & J. 269, 312. It is conceded that the testator was in fact deceived, not by strangers, but by persons in a position to benefit indirectly from his bounty.

As a matter of construction, we find error in the Chancellor’s ruling that George Marion is entitled to a share, and the decree must be affirmed in part and reversed in part, and remanded for the entry of a decree in accordance with the views here expressed. The ratification of the settlement agreement between the guardians ad litem of George Marion and Sandra Lee becomes inoperative since neither takes under our construction of the will. As to the disposition of the present and future income from the share set aside for George Marion, there may be a question as to whether it should be divided between the six great-nephews and great-nieces, cf. In Re Sharp [1908] 2 Ch. 190, or between all of the persons entitled to shares. Or there may be an intestacy, as suggested in argument. We leave the question open, as it was not considered below or argued in the briefs. The appellants urge that the Chancellor should have passed a declaratory decree as to the principal as well as a decree as to the present and future income. On this point we note that the bill did not pray such a declaration. In any event, the matter is largely in the discretion of the Chancellor and we find no abuse of discretion.

Decree affirmed in part, reversed in part, costs to be paid out of the trust , estate.