This is an appeal by seventeen appellants from an order granting partial distribution to the seven respondents.

Appellants are ten first cousins and seven second cousins of the deceased while respondents are seven second cousins of the deceased. The seventeen appellants were likewise granted partial distribution of the estate by other orders which are not involved on this appeal. The question presented here is whether respondents were persons entitled to take under the provisions of paragraph three of the will of the deceased.

Preliminarily it may be stated that deceased left surviving him his wife but no other relatives closer than first and second cousins. The residence of the deceased was in San Francisco but the residences of the first and second cousins were scattered throughout the world. It appears probable that deceased had not been in close touch with his first or second cousins as he made provision for some who had died long prior to the execution of the will and also made provision for the children of some who had died without leaving children, long prior to the execution of the will.

Appellants and respondents represent four branches of the families of the parents of the deceased, being children or grandchildren of deceased’s aunt and uncle on his father’s side and the children or grandchildren of the deceased’s aunt and uncle on his mother’s side. The evidence does not definitely show that the four branches which were mentioned in the will and which are represented by appellants and respondents were the only branches of the families of the parents of deceased, but there is nothing to show that they were not. It is a fair assumption that the testator made provision for all branches of which he had knowledge and there is no suggestion that there were other branches. Under the interpretation placed upon paragraph three of the will by the trial court, persons in all four branches have been held to be entitled to take under the provisions thereof. . Under the interpretation for which appellants contend, no person in one of the four branches, that is the branch represented by the descendants of David Craig who was the uncle of the deceased on his father’s side, would be entitled to take.

In paragraph four of the will, the deceased stated “All of my property is separate property, being the property and avails thereof acquired by me through inheritance from my deceased father,” and in one of the concluding clauses of paragraph three of the will, deceased stated “. . . I am not advised as to whether all of the above mentioned parties are now alive, or as to their whereabouts. ...” Paragraphs two and three made disposition of the entire estate of the deceased.

By paragraph two of the will, deceased left one-half of his estate in trust for his wife, who is not a party to the present controversy. Paragraph three then followed, the pertinent portions of which are:

The last quoted clause of paragraph three, commencing with the words “It is my desire,” has been termed in different ways by the parties to this appeal and, for the sake of convenience in this discussion, said clause will be termed the “concluding clause.” The first quoted clause of paragraph three, reading “the other one-half of my said estate, I give devise and bequeath to the following persons in equal proportions, to-wit:” will be termed the “opening clause.” The clause lettered (n) in paragraph three is the other principal clause for discussion.

The clause lettered (n) uses the words “The children or child living at the time of my death of David Craig (now deceased) my father’s brother. ...” Taking the word “children” in its primary sense, there were no “children” of David Craig living at the time of the death of the deceased or at the time of the execution of the will. The will was executed in 1939 and both John Craig and Herbert Craig, who were first cousins of the deceased and were the only “children” of David Craig, deceased, had died in 1933. Five of the respondents are the children of said John Craig, deceased, and the remaining two respondents are the children of said Herbert Craig, deceased. It is therefore conceded by all that if the word “children,” as used in clause (n), is to be interpreted in its primary or strict sense, then there were no persons answering the description of “children . . . living at the time of my death of David Craig. ...” And it is further conceded that under such interpretation of clause (n), respondents would not be entitled to take by virtue of the provisions of section 92 of the Probate Code as said clause (n) referred only to those persons answering the description of “children,” as so interpreted, “living at the time of my death.” (Estate of Bounds, 180 Cal. 386 [ 181 P. 638 ].)

Respondents contend, however, that when the will is read as a whole and clause (n) is interpreted in the light of the other provisions of the will and particularly of the concluding clause of paragraph three, or in the light of the extrinsic evidence introduced, or in the light of both such other provisions of the will and such extrinsic evidence, the only reasonable conclusion to be drawn is that the deceased intended the word “children,” as found in clause (n), to be used in its enlarged sense so as to include grandchildren. In our opinion respondents’ contention must be sustained and under this view, it is clear that the trial court correctly held that respondents, as grandchildren of David Craig living at the death of the testator, were entitled to take under said paragraph three of the will.

There are many authorities holding that while the word “children” is ordinarily presumed to be used by a testator in its primary or strict sense, other provisions of the will or extrinsic evidence may show that the word “children” should be interpreted in its enlarged sense so as to include grandchildren. (I n re Schedel, 73 Cal. 594 [ 15 P. 297 ]; Rhoton v. Blown, 99 Cal. 645 [ 34 P. 513 ] ; Estate of Eelerle, 155 Cal. 723 [ 102 P. 935 ]; see, also, numerous authorities collected in 104 A.L.R. 282 .) There are two cases in this jurisdiction in which the claim that the word “children” should be interpreted in its enlarged sense was denied but each of said cases is distinguishable and each recognizes that said word may be interpreted under certain circumstances as including grandchildren. In Estate of Willson, 171 Cal. 449 [ 153 P. 927 ], the bequest was to “the children of my said brothers and sisters.” Six of said “children” survived the testator as did also a number of grandchildren whose parents had died after the execution of the will. The court there said at page 456, “The word ‘children,’ in a will, has been held to include grandchildren in only two classes of eases. The first class is where there is an ambiguity in the will itself which leaves the meaning of the testator in respect to the word ‘children’ uncertain. . . . The other class consists of cases where there is a latent ambiguity. . . . Where the will purports to make a gift to ‘children’ and it appears that none were living at the execution of the will, or afterward, extrinsic evidence may be given tending to prove that a gift to grandchildren was intended. ... As there were a number of children of the brothers and sisters living, both at the time of the execution of the will and at the time of the death of the testatrix, this case does not come within the second class. . . . There is nothing in the context of this will which throws doubt upon the mean ing of the word ‘children’ as used in the residuary clauses, nothing to suggest that it was not used in the ordinary sense to describe the immediate offspring of the brothers and sisters named.” In Estate of Vizelich, 129 Cal.App. 347 [ 18 P.2d 773 ], property was left “to my surviving children, that is to say, my sons and daughters herein named. ’ ’ The court said on page 352, “The word ‘children’ in its primary significance means descendants in the first degree, and can receive no other construction, in the absence of satisfactory evidence that the testator intended it to include grandchildren. ... No such intention is apparent in the testator’s will. ... It is evident that Mrs. Vizelich, the testatrix herein, desired the share of any of her children or grandchildren who might die without issue, to go to her surviving children, and as a precaution against uncertainty in that classification she added, ‘that is to say, to my sons and daughters herein named. ’ ”

We therefore turn to the circumstances presented in the present case to determine whether the word “children,” as used by the testator in clause (n) of paragraph three, should be interpreted in its primary or strict sense so as to exclude grandchildren or in its enlarged sense so as to include grandchildren. When the evidence showed that there were no sons or daughters of David Craig, deceased, living either at the time of the death of the testator or at the time of the execution of the will, a latent ambiguity arose with respect to said clause (n) which called for interpretation by resort to the other provisions of the will or by resort to extrinsic evidence, or by resort to both. (Prob. Code, § 105 and authorities cited above.) In our opinion a reading of the will as a whole without resort to extrinsic evidence, calls for the interpretation of the word “children” in the enlarged sense as determined by the trial court.

Respondents lay particular stress on the concluding clause of paragraph three reading “It is my desire that each of the children of my deceased first cousins living at the time of my death shall share equally with each of my first cousins then living, and if any of the children of my said first cousins have predeceased me leaving lawful issue then living, such issue shall take by right of representation the share in said estate which said decedent would have received if living.” Respondents were children, living at the time of the testator’s death, of two of the testator’s deceased first cousins. In passing, we may state that it does not appear that any of the children of the testator’s deceased first cousins had predeceased the testator leaving issue who might take by right of representation under the last portion of said concluding clause.

The concluding clause is termed by respondents as the “key” or “master” clause expressing the desires and intentions of the testator and we believe that it may properly be so termed. It is significant to note that it would be sufficient, standing alone and in the absence of any of the lettered clauses of paragraph three, to dispose of the one-half of the estate of the testator mentioned in the opening clause of paragraph three. (Estate of Tooley, 170 Cal. 164 [ 149 P. 574 , Ann.Cas. 1917B 516]; Estate of Briggs, 186 Cal. 351 [ 199 P. 322 ]; Estate of Lawrence, 17 Cal.2d 1 [ 108 P.2d 893 ].) Clause (n) must be interpreted in relation to this concluding clause so as “to form one consistent whole,” if such is possible, and if the concluding clause should be found to be “absolutely irreconcilable” with clause (n), the concluding clause should prevail. (Prob. Code, § 103.) We believe, however, that these clauses are not absolutely irreconcilable but on the contrary may be interpreted as entirely consistent with each other if the word “children,” as used in clause (n) is interpreted in its enlarged sense so as to include grandchildren.

The entire plan of the testator seems perfectly evident from reading the will as a whole. He declared in his will that his entire estate had come to him by inheritance from his deceased father. He manifested throughout his will his desire that said estate should be shared at his death by the descendants of the several branches of the families of his parents. There were four such branches represented by the descendants of his two uncles and two aunts and, as above indicated, it does not appear that there were any other branches. The testator stated in his will his uncertainty as to whether all his first cousins and second cousins, who were mentioned therein, were alive and if alive, as to where they might be located. He apparently did not know whether some of his first cousins had ever married and if so, whether they had had children and if so, the number and names of said children. It is entirely probable that he did not know who were the living children or grandchildren of Ms deceased uncle, David Craig. He was, of course, uncertain as to which, of his first and second cousins might predecease him.

With the limited knowledge which the testator had concerning the existence and the names and addresses of the descendants of his deceased uncles and aunts, it seems clear that the testator did all that his knowledge permitted to designate such descendants as beneficiaries by name or by description and to make reference to addresses which might aid in locating such descendants; and it further seems clear that he believed it necessary and proper to give definite expression to his general plan by adding the concluding clause of paragraph three. That plan as therein expressed, which no doubt appeared just and proper to him, was to have each living first cousin and each living child of a deceased first cousin share equally and, in the event that any child of a deceased first cousin predeceased the testator leaving lawful issue, to have such issue share but only by right of representation. There is nothing in the will to show that the testator intended to discriminate between the four branches of the families of his parents. Every indication is to the contrary.

In referring to a will bearing some similarity to the one before us, the court referred to the strong family spirit exemplified saying that the will was “patriarchal in design” and holding that the word “children” should be interpreted in its expanded sense (Davis Trust Co. v. Elkins, 114 W.Va. 742 [ 175 S.E. 611 ]). The expanded sense was also adopted where there was no person living coming within the description of a “child,” in the primary and restricted sense of the word, but there were grandchildren living, the court pointing out that there as here “the beneficiaries were cousins and lived at a distance, and were much scattered, and it did not appear that the testator was acquainted with the situation of their families, and the number or names of their children.” (Dunn v. Cory, 56 N.J.Eq. 507 [ 39 A. 368 ].) And in another case, where the wording of the will was similar to the one before us in that it read “to such children as should be living at the time of her death,” it was held that the word “children” should be interpreted so as to include grandchildren (Prowitt v. Rodman, 37 N.Y. 42 ). While it is difficult, if not impossible, to find cases on all fours to aid in the in-' terpretation of a particular will, the foregoing authorities indicate the trend of the decisions and to these may be added the following. (Pfender v. Depew, 136 App.Div. 636 [121 N.Y.S. 285]; Matter of Paton, 111 N.Y. 480 [ 18 N.E. 625 ] ; Matter of Brown, 93 N.Y. 295; In re Marshall’s Will, 18 App. Div. 211 [ 45 N.Y.S. 956 ] ; In re Faust’s Estate, 83 Misc. 250 [ 145 N.Y.S. 842 ] ; In re Scholl’s Estate, 100 Wis. 650 [ 76 N.W. 616 ]; Spencer v. Title Guarantee Loan & Trust Co., 222 Ala. 221 [ 132 So. 32 ]; Douglas v. James, 66 Vt. 21 [ 28 A. 319 , 44 Am.St.Rep. 817 ] ; Bowker v. Bowker, 148 Mass. 198 [ 19 N.E. 213 ].) As was said in Pfender v. Depew, 121 N.Y.S. 285 at page 287, “Where two interpretations of the word ‘children’ are possible, the courts have been much disposed to adopt that one which will not serve to disinherit the heirs of the testator. ’

Appellants argue for an interpretation which would disinherit the seven respondents whose relationship as second cousins of the testator was the same as that of seven of the appellants. To adopt this interpretation would result in disinheriting the only heirs shown by the record to bear the name of Craig, which was the family name of the testator and of the deceased father of the testator from whom the estate had been inherited. Such interpretation seems entirely out of harmony with the general plan of the testator as expressed in his will and entirely out of harmony with the trend of the authorities above cited.

We are not impressed with appellants’ argument that the concluding clause must be interpreted as a clause inserted by the testator solely for the purpose of fixing the proportions which the legatees, under appellants’ interpretation of the legatees included in the lettered clauses, were to take and for no other purpose. Such proportions, under appellants’ interpretation of the legatees included in the lettered clauses, had been previously fixed by the words “to the following persons in equal proportions” found in the opening clause. Even if it be conceded that appellants’ argument is based upon a possible interpretation of the concluding clause, the interpretation placed upon said clause by respondents is at least equally reasonable and the latter should be preferred as it renders clause (n) operative (Prob. Code, § 102) and prevents the disinheriting of the heirs who were descendants of David Craig.

Appellants further argue that under respondent’s interpretation “any other first cousins of the testator or children of deceased first cousins would be entitled to inherit along with the expressly mentioned first cousins and children of first cousins expressly referred to as deceased. ’ ’ While this is a possible interpretation of the concluding clause, it does not appear that there were, at the time of the death of the testator or at the time of the execution of the will, other branches of the families of the parents of the deceased or that, in addition to appellants and respondents, there were other living first cousins or other living children of deceased first cousins within the four branches of said families mentioned in the will. We are therefore not called upon to determine whether the concluding clause, which standing alone would be sufficient as a dispositive clause, should be interpreted as such when read with the lettered clauses. We are merely called upon to hold that the lettered clauses must be read in the light of the concluding clause and that said concluding clause was properly interpreted by the trial court as expressing the general plan and intention of the testator with respect to the several branches of the families which were mentioned in the lettered clauses and as indicating that the word “children” was used in its enlarged sense in the clause lettered (n).

Appellants further contend that the trial court committed prejudicial error in admitting certain extrinsic evidence. The evidence to which particular reference is made is certain evidence given by Mr. Randolph V. Whiting, the attorney who drew the will, and by the testator’s surviving wife.

We have heretofore indicated our belief that it was unnecessary for the trial court to have resorted to anything other than the remaining provisions of the will to clarify the latent ambiguity which arose with respect to clause (n) when it appeared that there were no sons or daughters of David Craig living at either the time of the death of the testator or at the time of the execution of the will. However, if it may be said, after a reading of the entire will, that the intent of the testator still remained in doubt, the testimony of Mr. Whiting, covering the specific instructions given to him by the testator for the purpose of drawing the will, left no doubt concerning such intent. It is well settled that when a latent ambiguity exists in a will, testimony concerning such specific instructions is admissible. (Estate of Dominici, 151 Cal. 181 [ 90 P. 448 ]; Estate of Little, 170 Cal. 52 [ 148 P. 194 ]; Estate of Greenwald, 19 Cal.App.2d 291 [ 65 P.2d 70 ].) The testimony of the testator’s surviving wife was at best merely cumulative in nature. Under these circumstances we find no prejudicial error in the admission of any of said testimony.

The order appealed from is affirmed.

Nourse, P. J., and Sturtevant, J., concurred.

A petition for a rehearing was denied May 27, 1944, and the following opinion was thereupon rendered:

In the petition for rehearing, appellants call our attention to the fact that toward the end of the opinion heretofore filed respondents were referred to as “heirs” of the testator. While appellants concede that “appellants and respondents were the testator’s closest blood relatives,” it appears that neither appellants nor respondents were heirs of the testator as the testator left a surviving wife. (Prob. Code, § 224.) We are satisfied, however, that the conclusions reached in the opinion heretofore filed were correct regardless of whether respondents were or were not heirs of the testator.

The petition for a rehearing is denied.

Appellants’ petition for a hearing by the Supreme Court was denied June 26, 1944.