Opinion
MASTERSON, J.
In 1982, Evelyn I. Anderson executed a will that exercised a testamentary power of appointment over a portion of a trust [238]*238created by her deceased husband. In 1993, Anderson executed a second will that expressly revoked all prior wills and that inadvertently failed to exercise the power of appointment.
After Anderson died in 1995, the executor of her estate petitioned the trial court to admit to probate the 1993 will and that portion of the 1982 will exercising the power of appointment. One of Anderson’s stepgrandchildren initiated a will contest and objected to the admission of any portion of the 1982 will.
After conducting an evidentiary hearing, the trial court found that Anderson intended to exercise the power of appointment and that the 1993 will did not revoke the provision in the 1982 will exercising that power. Applying the doctrine of dependent relative revocation, the trial court admitted to probate the 1993 will and the portion of the 1982 will exercising the power of appointment.
The contestant challenges the trial court’s order to the extent it admits to probate the provision in the 1982 will. We affirm.
Background
In 1948, Anderson married Wilbur R. Irwin (Irwin). For both, it was their second marriage. Irwin had five sons from his first marriage. Anderson had one daughter from her first marriage, Carole De Paul (De Paul), whom Irwin adopted.
Irwin died in 1963. His will created a trust which paid all of its income to Anderson during her life and gave her a testamentary general power of appointment over 50 percent of the trust assets. The will expressly required that Anderson exercise the power of appointment by specific reference in her own will. Irwin’s will further provided that if Anderson did not exercise the power of appointment, her 50 percent share of the trust assets would be divided equally among Irwin’s living issue by right of representation. (See Prob. Code, § 246.)
In 1982, while residing in Omaha, Nebraska, Anderson executed a will. She left various items of personal property to her husband (E.L. Anderson), daughter (De Paul), and a niece. The will also made monetary gifts in specified amounts to Anderson’s brother, sisters, nieces, and nephew, and established a trust for her two grandchildren (i.e., De Paul’s children). Article eighth of the will stated: “All of the rest, residue, and remainder of my estate, real, personal and mixed, of whatsoever kind and wheresoever [239]*239situated, including specifically that portion of the trust estate over which I possess a power of appointment which was created in and by the Last Will and Testament of my former husband, Wilbur Riseby Irwin, Deceased, ... I devise to (and I hereby exercise said power of appointment in favor of) my daughter, Carole Irwin DePaul . . . -”1
In 1993, after moving to Pasadena, California, Anderson retained a local law firm to rework her estate plan. On May 4, 1993, Anderson met with a partner of the firm who was a trust and estate lawyer and gave him a copy of her 1982 will. Anderson instructed the partner to use the 1982 will as a model and to carry forward its provisions into the new estate planning documents except as she directed otherwise. Anderson requested that the partner convert her existing estate plan from a will into a living trust with a pour-over will. She also indicated that she wanted to make a gift of $1,000 to each grandchild of her former husband (Irwin). The partner valued Anderson’s assets to be worth approximately $1.3 million, including $500,000 that he attributed to her 50 percent share of the Irwin trust.
During the meeting, the partner took two pages of notes on the changes Anderson wished to make in her estate plan. He also made handwritten comments on his copy of the 1982 will. The notes mentioned the creation of a living trust with a pour-over will and a $1,000 bequest to each of Irwin’s grandchildren. The notes also stated, “see will for other changes.” The partner’s comments on the 1982 will indicated, among other things, that Anderson’s third husband (E.L. Anderson) and one of her sisters had died. The partner did not make any comments on the will about article eighth. He testified that the lack of notations beside a particular article meant that Anderson “instructed me to leave the documents as is.”2 Anderson did not specifically refer to article eighth or to the power of appointment in telling the partner what to include in the new documents. Nor did she ever tell the partner that she wanted to revoke the exercise of the power of appointment.
After the meeting, the partner gave his notes and the annotated copy of the 1982 will to an associate attorney at the firm and asked her to draft the new estate planning documents. The partner told the associate that Anderson’s 1982 will would be the basis for the new documents and that, other than the changes mentioned in his notes and on the copy of the will, the basic plan was to remain intact. The partner did not give the associate any specific [240]*240instructions about the power of appointment—whether to exercise it or mention it.
The associate drafted several estate planning documents for Anderson, including a new will, a trust, and two durable powers of attorney. The will appointed De Paul as the executor and stated that Anderson “revoke[d] all former Wills.” None of the new documents exercised Anderson’s power of appointment over her portion of the Irwin trust, nor did they specifically refer to the power of appointment.
In May 1993, the partner sent draft copies of the new estate planning documents to Anderson for review. In response, Anderson asked that her son-in-law be removed as the alternate attorney-in-fact on the durable powers of attorney.
On June 24, 1993, Anderson reviewed the final version of the new documents at the law firm. None of the documents exercised Anderson’s power of appointment with respect to the Irwin trust. In the presence of the partner and the associate, Anderson signed the documents. The two lawyers then signed an attestation clause at the end of the will.
On August 18,1994, Anderson wrote a letter to the First National Bank of Omaha, the trustee of the Irwin trust, seeking answers to questions about the operation of the trust. The bank responded by letter dated August 23, 1994, stating in part: “At the time of your death, you can control where 50 percent of the assets in this trust are to be distributed by specifying in your Last Will and Testament. The other 50 percent of the assets are to be transferred to the children of Wilbur Irwin. [<][] Because you have a ‘general power of appointment’ over 50 percent of the assets we hold in trust, those assets are subject to Federal Estate Taxes. . . .”
In September 1994, Anderson met again with the partner at the law firm to discuss her estate plan. They discussed the bank’s letter but did not talk about whether Anderson’s new will exercised her power of appointment over the Irwin trust. The partner recommended that Anderson consider amending her own trust to place a portion of the trust assets into a generation-skipping trust. Anderson agreed and executed an amendment to that effect in December 1994.3 Throughout the discussions about amending Anderson’s trust, the partner assumed that Anderson had exercised her power of appointment over the Irwin trust in favor of De Paul, i.e., that De Paul would receive [241]*241Anderson’s 50 percent share of the Irwin trust. Only later did the partner realize that, in his words, “[i]t was an inadvertent omission to reconfirm that exercise” in the new estate planning documents.
Anderson died on March 2,1995. As executor, De Paul petitioned the trial court to admit to probate Anderson’s 1993 will and article eighth of the 1982 will. One of Irwin’s grandchildren (who is one of Anderson’s stepgrandchildren) appeared as a contestant and objected to the admission of article eighth. The contestant argued that the 1993 will had revoked the 1982 will, including the exercise of the power of appointment. Under the contestant’s theory, because the 1993 will did not exercise the power of appointment, Anderson’s 50 percent share of the Irwin trust had to be distributed to the remainder beneficiaries designated in Irwin’s will, as follows: one-sixth to De Paul, one-sixth to Irwin’s sole surviving son, and four-sixths to seventeen of Irwin’s grandchildren (including the contestant).4
On November 3, 1995, the trial court held an evidentiary hearing on the contestant’s objections to De Paul’s petition. By order dated November 29, 1995, the trial court denied the contestant’s objections, issued letters testamentary to De Paul, and admitted to probate the 1993 will and article eighth of the 1982 will. The trial court ruled that, “pursuant to the Doctrine of Dependent Relative Revocation, the general revocation clause contained in [Anderson’s] . . . 1993 Will was not effective to revoke that portion of Article Eighth of [her] . . . 1982 Will that exercised the power of appointment in favor of Carole J. De Paul.” The contestant filed a timely appeal from the trial court’s order.5
Discussion
Because the record in this case discloses no conflicts in the evidence and no issues of credibility, we independently determine whether the 1993 will revoked article eighth of the 1982 will. (See Estate of Dodge (1971) 6 Cal.3d 311, 318 [98 Cal.Rptr. 801, 491 P.2d 385].)
The contestant argues that the trial court’s order should be reversed because (1) the revocation clause in the 1993 will encompassed article [242]*242eighth of the 1982 will, (2) the elements of the doctrine of dependent relative revocation were not satisfied in this case as a matter of law, and (3) sufficient evidence does not support the application of the doctrine of dependent relative revocation. We discuss each of these contentions in turn.
A. The Revocation Clause in the 1993 Will
As stated, Irwin’s will required that Anderson exercise the power of appointment over her share of the Irwin trust by specific reference in her will. She exercised that power in favor of De Paul in article eighth of the 1982 will. The revocation clause in the 1993 will—which “revoke[d] all former Wills”—did not specifically mention Anderson’s prior exercise of the power of appointment. The question therefore arises as to whether a general revocation clause in a last will can encompass a power of appointment exercised in a prior will. We hold that it can.
“A will or any part thereof is revoked by . . . [U . . . [a] subsequent will which revokes the prior will or part expressly or by inconsistency.” (Prob. Code, § 6120, subd. (a).) “ ‘As a general rule, the presence of a revoking clause in a later writing is accepted as conclusive proof of an intention on the part of decedent to nullify the provisions of an earlier instrument.’” (Estate of Rose (1928) 95 Cal.App. 580, 584 [273 P. 92]; accord, Estate of Holloway (1925) 195 Cal. 711, 713, 735 [235 P. 1012].)6 In construing a general revocation clause, an exercise of a power of appointment in an earlier will is treated in the same manner as any other provision of that will: “A will does not become legally operative until the death of the testator. Hence, the terms of the exercise of a power that are set forth in the will of the donee of the power may be revoked or amended by the donee to the same extent as any other provisions of the will.” (Rest.2d Property, § 15.2, com. b, p. 172.)
Thus, although Anderson had to specifically mention the power of appointment in order to exercise it in the first will, she did not have to make a specific reference to that power in order to revoke it in the second will. (Compare Prob. Code, § 632 with id.., § 695, subd. (b).) The revocation clause in the second will, which revoked all former wills, was sufficient to revoke article eighth of the first will, unless the doctrine of dependent relative revocation applies.
B. The Doctrine of Dependent Relative Revocation
“Under the doctrine of dependent relative revocation, an earlier will, revoked only to give effect to a later one on the supposition that the [243]*243later one will become effective, remains in effect to the extent that the latter proves ineffective. . . . The doctrine is designed to carry out the probable intention of the testator when there is no reason to suppose that he intended to revoke his earlier will if the later will became inoperative." (Estate of Kaufman, supra, 25 Cal.2d at pp. 858-859, citations omitted.) Put another way, where a “question arises as to the effectiveness of the second instrument, ... so that upon its failure to be operative for want of proper execution or other cause, the testator will be presumed to have intended the original instrument to stand to the extent that the later proves ineffective.” (Estate of Salmonski (1951) 38 Cal.2d 199, 212 [238 P.2d 966].) “The doctrine thus requires that, in revoking a prior and executing a subsequent will, it be specifically intended that certain provisions in the former testament have a continuing effect, either through similar provisions in the new will or because it is intended to make only conditional changes which subsequently do not become effective for the reason that the conditions on which [they are] predicated fail to come into being.” (Estate of Cuneo (1963) 60 Cal.2d 196, 202 [32 Cal.Rptr. 409, 384 P.2d 1, 7 A.L.R.3d 1132].)
As a leading authority has explained: “The testator doubtless has power to make an expressly conditional revocation. The cases, however, go further and hold that if he revokes under some mistake, the condition may be implied or presumed. With respect to certain situations, this has become crystallized into the doctrine of dependent relative revocation. Thus, if the testator destroys a first will in the mistaken belief that a second will is valid, the law presumes that he intended to revoke only if the second were valid', i.e., the revocation is not absolute, but is relative, and dependent upon the validity of the second will. Similarly, if the testator, in a new will or a codicil, revokes gifts to certain relatives, stating that they are dead, the doctrine may be applied to nullify the revocation if the belief turns out to be untrue." (12 Witkin, Summary of Cal. Law (9th ed. 1990) Wills and Probate, § 240, p. 274, italics in original.)
In Estate of Robertson (1968) 266 Cal.App.2d 866 [72 Cal.Rptr. 396], the court applied the doctrine of dependent relative revocation in a situation similar to that before us. There, the decedent had executed a will in 1953 leaving her entire estate to her niece and nephew in equal shares. In 1963, she executed a codicil naming a new executrix. In 1965, she executed a new will that (1) expressly revoked all prior wills, (2) left her “personal effects” to a friend, who was named as the new executrix, and (3) placed all cash, securities, and “like assets” in the residue of the estate. The 1965 will mentioned the decedent’s relationship to her niece and nephew, but did not leave them anything. Nor did it dispose of the residue of the estate. However, the new will did recite, “I have other relatives to whom I purposely leave nothing.” (Id. at p. 868.)
[244]*244The decedent’s niece and nephew sought to admit the 1953 will and the 1963 codicil to probate. In opposition, the heirs at law of the decedent’s predeceased husband argued that the distribution of the estate should be governed solely by the 1965 will. At an evidentiary hearing, the attorney who drafted the 1965 will testified that the decedent had asked him to change the executor named in the 1963 codicil and to leave her personal effects to the new executrix but otherwise to retain the same will. He drafted the will in longhand, leaving the residue of the estate to the decedent’s niece and nephew, but his secretary failed to include the residuary clause in the typed will. The attorney did not review the final will before taking it to the decedent for signature. Because the decedent suffered from impaired vision, the attorney summarized the will for her, stating that it left her personal effects to the friend, named the friend as the new executor, and left everything else to the decedent’s niece and nephew. The decedent then signed the will.
The trial court entered judgment in favor of the decedent’s niece and nephew. The Court of Appeal affirmed on the ground that “[u]pon all the evidence, the conclusion is inescapable that testatrix intended to revoke her earlier will only if the new one left all of [the] real value in her estate to [her niece and nephew]. Particularly impressive is the recital of the 1965 will naming respondents as her nephew and niece, and stating that T have other relatives to whom I purposely leave nothing.’ FID Courts are enjoined to give effect to every expression of a will and to avoid an interpretation which will lead to intestacy .... Here, all property left by decedent had been community property of herself and of her predeceased husband. Thus if any part of her estate passes by intestate succession, appellants will share in it as heirs of the husband. . . . Appellants’ construction not only would result in intestacy as to the great bulk of the estate under the law of succession, but would distribute half of it to the ‘other relatives’ to whom [the decedent’s] 1965 will said T purposely leave nothing,’. . . . FJD We are well aware that it is not our function to draw a new will for the testatrix. Yet the overwhelming evidence here assures against that danger, and clearly brings the case within the established rule of dependent relative revocation . . . .” (266 Cal.App.2d at pp. 868-869, citations omitted.)
The contestant seeks to distinguish Estate of Robertson, supra, 266 Cal.App.2d 866, by arguing that, under California law, the doctrine of dependent relative revocation applies only where (1) the decedent’s last will is wholly or partially invalid, resulting in intestacy and (2) the last will shows on its face that the revocation of the prior will was conditioned on the effectiveness of the last will. We reject this argument.
[245]*2451. Intestacy Under the Last Will
The doctrine of dependent relative revocation applies where a “question arises as to the effectiveness of the second instrument, ... so that upon its failure to be operative for want of proper execution or other cause, the testator will be presumed to have intended the original instrument to stand to the extent that the later proves ineffective.” (Estate of Salmonski, supra, 38 Cal.2d at p. 212, italics added; accord, Estate of Kaufman, supra, 25 Cal.2d at pp. 858-859.) Plainly, a will may prove “ineffective” if it results in total or partial intestacy, but that is not the only situation in which a will may “fail[] to be operative." (Estate of Salmonski, supra, 38 Cal.2d at p. 212.) A will may also be ineffective if, as a result of a mistake on the testator’s part, it purports to revoke a prior will and fails in a material way to distribute the estate in accordance with the testator’s intent. (See 2 Page on Wills (3d ed. 1960) Revocation, §21.57, pp. 449-450 [doctrine of dependent relative revocation should apply where testator revokes prior will based on material mistake].)
“In the construction of wills the paramount rule, to which all others must yield, is that a will is to be construed according to the intention of the testator, as expressed therein, and this intention must be given effect as far as possible.” (Estate of Lawrence (1941) 17 Cal.2d 1, 6 [108 P.2d 893].) The doctrine of dependent relative revocation is simply one means of implementing this paramount rule: “[T]he sole justification for the use of dependent relative revocation is to effectuate the decedent’s intent as nearly as possible.” (Palmer, Dependent Relative Revocation and Its Relation to Relief for Mistake (1971) 69 Mich. L.Rev. 989, 997; see also Prob. Code, §§21101-21139 [codifying rules of construction that give effect to testator’s presumed intent].) No doubt, the purpose of the doctrine is often served by applying it where intestacy would otherwise result. (Cf. 2 Page on Wills, supra, Revocation, § 21.57, pp. 449-450 [in situations where intestacy would better serve testator’s intent, doctrine should not be applied].) By the same token, the doctrine should apply where the last will purports to revoke a prior exercise of a power of appointment, resulting in substantial gifts to the decedent’s stepgrandchildren contrary to her express intention to exercise that power solely for the benefit of her natural daughter.7
As explained by one commentator: “The second instrument may be executed validly and apparently may be a good and sufficient disposition of [246]*246the estate of the testator, yet operate neither as a disposition of the property nor as a revocation of a prior will. This result obtains if the testator executes the apparent will. . . under the mistaken notion that some fact or law does or does not exist .... [H ... [H ... In each case the problem is to discover the true intent when the second document is executed. Legal bars in the form of hard-and-fast rules not based on reason should never be permitted to interfere. Courts should be free to hold the revocation conditional on future events if the testator so intends it or to find [the intent to revoke] lacking if not present.” (Cornish, Dependent Relative Revocation (1932) 5 So.Cal.L.Rev. 273, 300, 306.)
The facts of this case amply demonstrate why the doctrine of dependent relative revocation should not be limited to invalid final wills that would otherwise result in intestacy, Here, the trial court found that the second will was ineffective because it failed to carry out Anderson’s intention to exercise the power of appointment in favor of De Paul. Under article eighth of the first will, De Paul was entitled to Anderson’s entire 50 percent share of the Irwin trust, i.e., approximately $500,000. However, if the doctrine of dependent relative revocation does not save article eighth, then, under the default provisions of Irwin’s will, De Paul will receive only one-sixth of Anderson’s share—$83,333 instead of $500,000. The remaining five-sixths of Anderson’s share ($416,667) will go to individuals who were to receive nothing under her first will (i.e., a stepson and several step-grandchildren).
Moreover, one of Anderson’s primary reasons for executing the second will was to make gifts of $1,000 to each of Irwin’s grandchildren (i.e., her stepgrandchildren). The second will expressly stated that they were to receive that sum. Yet, if article eighth of the first will were deemed to be revoked, the effect of the second will would be to give 17 of Irwin’s 22 grandchildren bequests ranging from approximately $14,900 to $28,800— gifts substantially greater than the $1,000 that Anderson wanted each of them to receive under the second will.8 The paramount goal in interpreting wills—to effectuate the testator’s intent—would hardly be served if Anderson’s desire to give each of Irwin’s grandchildren a gift of $1,000 ultimately [247]*247resulted in an average gift of more than $20,000 apiece, all at the expense of her only natural daughter. Indeed, we think it fair to assume that Anderson would have preferred that her share of the Irwin trust be distributed under the intestacy laws rather than in accordance with the default provisions of Irwin’s will. If the intestacy laws were applicable, Anderson’s entire 50 percent share of the trust would go to De Paul—the same disposition that Anderson made in her first will and the one she intended in the second. (See Prob. Code, §§ 6400-6402, 6454.)
Accordingly, we conclude that the doctrine of dependent relative revocation may be used to effectuate a testator’s intent even if the last will would not otherwise result in total or partial intestacy.9
2. Consideration of Extrinsic Evidence
The contestant argues that the doctrine of dependent relative revocation does not apply unless the testator’s last will shows on its face that the testator intended the revocation of the prior will to be conditioned on the effectiveness of the later will. We conclude that extrinsic evidence may be considered in determining whether Anderson intended to revoke article eighth of the 1982 will.
“[A] will is to be construed according to the intention of the testator, and not his imperfect attempt to express it.” (Estate of Kime (1983) 144 Cal.App.3d 246, 264 [193 Cal.Rptr. 718].) “Any conclusion as to the testator’s intention must be considered in the light of his knowledge at the time he executed the will.” (Estate of Kaufman, supra, 25 Cal.2d at p. 861.) “To constitute a valid revocation, acts of cancellation or interlineation must be done with sufficient present intent and purpose of revocation . . . .” (Estate of Uhl (1969) 1 Cal.App.3d 138, 143 [81 Cal.Rptr. 436], citations omitted.) In determining whether a will has been revoked, courts may consider extrinsic evidence of the testator’s intent. (See id. at pp. 142-144; Estate of Kaufman, supra, 25 Cal.2d at pp. 860-861; Estate of Robertson, supra, 266 Cal.App.2d at pp. 868-869; see also Estate of Verdisson (1992) 4 Cal.App.4th 1127, 1135-1136 [6 Cal.Rptr.2d 363] [court may consider circumstances surrounding creation of will, including statements of testator]; [248]*24812 Witkin, Summary of Cal. Law, supra, Wills and Probate, §§ 245-246, pp. 280-283 [same]; see generally, Estate of Kime, supra, 144 Cal.App.3d at pp. 261-265 [discussing use of extrinsic evidence in construing wills].)
In Estate of Kaufman, supra, 25 Cal.2d 854, the Supreme Court relied in part on extrinsic evidence in applying the doctrine of dependent relative revocation. The trial court had found that the testator intended to revoke his first will unconditionally. The high court rejected that finding as not supported by the evidence, stating: “All the testimony, including the testimony of [the trust officer] and of [the testator’s attorney] and his secretary, shows that the testator wanted no change in his will except for the naming of a California executor.” (Id. at p. 861.) Similarly, in Estate of Robertson, supra, 266 Cal.App.2d 866, the heirs at law of the decedent’s predeceased spouse argued that the court could not consider the testimony of the testator’s attorney in deciding whether to apply the doctrine of dependent relative revocation. The Court of Appeal disagreed, relying on Kaufman, supra, 25 Cal.2d at pages 860-861, and Estate of Cuneo, supra, 60 Cal.2d at pages 202-204. (Estate of Robertson, supra, 266 Cal.App.2d at p. 869.)
As our Supreme Court has stated with regard to interpreting wills in general: “[E]xtrinsic evidence is admissible ‘to explain any ambiguity arising on the face of a will, or to resolve a latent ambiguity which does not so appear.' ... A latent ambiguity is one which is not apparent on the face of the will but is disclosed by some fact collateral to it. ... ffl] ... [ID ‘The court must determine the true meaning of the instrument in the light of the evidence available. It can neither exclude extrinsic evidence relevant to that determination nor invoke such evidence to write a new or different instrument.’ . . . ffl] . . . ffl] . . . [W]e think it is self-evident that in the interpretation of a will, a court cannot determine whether the terms of the will are clear and definite in the first place until it considers the circumstances under which the will was made so that the judge may be placed in the position of the testator whose language he is interpreting.” (Estate of Russell (1968) 69 Cal.2d 200, 206-210 [70 Cal.Rptr. 561, 444 P.2d 353], italics added, citations omitted.)10
Applying these principles, we conclude that extrinsic evidence may be considered in determining whether Anderson conditioned the revocation of the first will on the exercise of the power of appointment in De Paul’s [249]*249favor. Even if we limit our focus to the written instruments—Irwin’s will, Anderson’s two wills, and the document creating Anderson’s trust—we are left with an incomplete picture requiring the use of extrinsic evidence. The documentary evidence alone indicates that one of Anderson’s primary reasons for executing the second will (and for purporting to revoke the prior will) was to leave a gift of $1,000 to each of Irwin’s grandchildren. Yet the consequence of executing the second will, absent an exercise of the power of appointment, would be to give 17 of Irwin’s 22 grandchildren an individual gift ranging from approximately $14,900 to $28,800. Thus, the effect of the second will appears to be inconsistent with the express provision leaving $1,000 to each of the grandchildren. Moreover, the circumstances surrounding the execution of the second will leave no doubt that Anderson conditioned the revocation of the first will on the exercise of the power of appointment in favor of De Paul. (See pt. C, post.) It follows that we may consider extrinsic evidence of Anderson’s intention to revoke the first will.11
C. Sufficiency of the Evidence
The contestant argues that there is insufficient evidence to support a determination that the revocation of the first will was conditioned on the exercise of the power of appointment. In particular, the contestant emphasizes that Anderson never specifically referred to the power of appointment in discussing the drafting or content of the second will. While that assertion is correct, it does not resolve the matter.
As made clear in Estate of Robertson, supra, 266 Cal.App.2d 866, there is more than one way for a testator to indicate that she wants provisions in an earlier will to be included in a later will. In that case, the decedent told her attorney what changes to make in the first will and otherwise to include the same provisions in the second will. (Id. at p. 868.) Although the first will [250]*250contained a residuary clause, the second will did not. That omission was attributable to the inadvertence of the decedent’s attorney and his secretary. (Ibid.) The Court of Appeal held that, under the doctrine of dependent relative revocation, the residuary clause in the first will remained in effect despite the express revocation provision in the second will. (Id. at pp. 868-869.)
Here, the evidence is equally strong. As stated, the dispositive instruments alone raise a serious question about the interpretation of the second will. They indicate that Anderson executed the second will, in part, to provide each of Irwin’s grandchildren with a gift of $1,000. However, the effect of the second will, absent an exercise of the power of appointment, would be to give most of the grandchildren an average gift 20 times that amount. Further, the partner who supervised the drafting of the new estate planning documents testified that Anderson told him to make certain changes in the first will and otherwise to carry forward its existing provisions into the new documents. Anderson did not tell the partner to make any changes with respect to article eighth or the exercise of the power of appointment.12 The partner, in turn, conveyed those same instructions to the associate who drafted the new will and trust documents. As the partner indicated, the failure to explicitly reference the exercise of the power of appointment in the second will was the result of an error in drafting.
Given this evidence, we reject the contestant’s assertion that Anderson never informed her attorneys that she wished to exercise the power of appointment in the second will. By telling her attorneys to carry forward the provisions of the first will except to the extent she directed otherwise, and by not requesting any changes in article eighth or the exercise of the power of appointment, Anderson effectively instructed her attorneys to exercise that power in the second will.
Nor do we doubt that Anderson made a material mistake in executing the second will. Instead of leaving a gift of $1,000 to each of Irwin’s grandchildren, Anderson signed a document that purported to give most of them a gift of at least $14,000. While article eighth of the first will left Anderson’s entire share of the Irwin trust ($500,000) to her only natural daughter, the second will had the effect of giving five-sixths of that amount ($416,667) to steprelatives who, under the first will, were to receive nothing and, under the express terms of the second will, were to receive a collective gift of $20,000 (i.e., $1,000 for each of Irwin’s grandchildren, not including De Paul’s children). In other words, the second will, absent an exercise of the power of [251]*251appointment, will mistakenly dispose of almost one-third of the assets over which Anderson had control.
On these facts, we conclude that a “question [has] arise[n] as to the effectiveness of the second instrument, ... so that upon its failure to be operative for want of proper execution or other cause, the testator will be presumed to have intended the original instrument to stand to the extent that the later proves ineffective” (Estate of Salmonski, supra, 38 Cal.2d at p. 212, italics added.) Because the second will did not effectively exercise the power of appointment, article eighth of the first will remained in effect under the doctrine of dependent relative revocation.
Disposition
The order is affirmed.
Spencer, P. J., and Vogel (Miriam A.), J., concurred.
Appellant’s petition for review by the Supreme Court was denied October 22, 1997. Kennard, J., and Brown, J., were of the opinion that the petition should be granted.