Opinion by
Me. Justice Jones,The issue (narrow in scope but complex in nature) presented on these appeals is: under the will and two codicils thereto of Daniel Grimm (testator), should distribution of the principal of a testamentary trust be made per stirpes or per capita among the several classes of testator’s descendants. The court below—the Court of Common Pleas, Orphans’ Court Division, of Venango County—decreed distribution per capita among testator’s grandchildren and per stirpes among testator’s great-grandchildren.
Testator, a Venango County resident, died June 1, 1928, survived by his wife and seven children.1 Testator left a will dated May 2, 1901, a codicil dated September 15, 1922 (first codicil) and a codicil dated February, 1924 (second codicil), all of which testamentary writings were probated.
Insofar as presently pertinent, the will provided: (1) a bequest to testator’s wife of household goods; (2) a gift of a fund—consisting of testator’s homestead, certain securities and the income arising from the proceeds of certain insurance policies—in trust, to provide a $5,000 annual income to testator’s wife; (3) a direction to the trustees to distribute any of the wife’s unused annual income as well as the income from the residue of the estate2 among all testator’s liv*130ing children and all the children of any deceased child or children, per capita, until the death of testator’s last child; (4) a direction to the trustees, upon the death of testator’s wife, to divide equally the fund which had generated the income for the wife between all his then living children and all the then living children of any deceased child or children per capita, “so that the portion of each individual distributee . . . shall be the same, whether the distributee be child or grandchild” of testator; (5) the trustees were further directed that, upon the death of testator’s last child and after his last grandchild, living when the last child died, reached the age of twenty-one years, to distribute the principal equally among the then living grandchildren per capita. The tenth paragraph of the will provided: “No one shall take anything by representation under this will, except the law make such provision in contravention of the intent of this will.”
In 1901, at the time testator executed his will, he had a wife, seven children and one grandchild. It is clear beyond question that, when testator made his will, he contemplated a per capita, not a per stirpes, distribution of both income and principal.
On September 15, 1922, testator executed the first codicil to the will:
“First. Paragraph Six (6)3of my said Will and the Distribution therein made, I now revoke, and instead thereof, I direct that the Distribution therein made shall be per stirpes and not per capita, that is that 6. After the Trustee shall have made sure annually that my said wife, Carrie F. Grimm, has had her *131annual income, they shall annually distribute the residue of the income from the Estate equally among all my children who shall be living at the time of my decease and the child or children of any children of mine who may then be deceased, but the child or children of any child or [sic] mine who may be then deceased, only to take and have such part or share as his, her or their deceased parent would have taken if he or she had been living at the time of my decease.
“Second. Paragraph Seven (7)4 of my said Will and the Distribution therein made, I now revoke, and instead thereof, I direct that the Distribution therein made shall be per stirpes and not per capita, that is that 7. Upon the death of my said wife, Carrie F. Grimm, the homestead and funds set apart for her use shall thereupon forthwith be divided equally among all my children who shall be living at the time of my decease and the child or children of any children of mine who may be then deceased, but the child or children of any child or [sic] mine who may be then deceased only to take and have such part or share as is, [sic] her or their deceased parent would have taken if he or she had been living at the time of my decease.
“Third. Paragraph Eight (8)5 of my said Will and Distribution therein made, I now revoke, and instead thereof, I direct that the distribution therein *132made- shall be per stirpes and not per capita, that is that 8. The income of the residue of the Estate shall be distributed quarterly among all my children who shall be living at the time of my decease and the child or children of any children of mine who may be then deceased, but the child or children of any child of mine who may be then deceased, only to take and have such part or share as his, her, or their deceased parent would have taken if he or she had been living at the time of my decease.
“Fourth. Paragraph Ten (10)6 of my said Will and the provision therein made, I now revoke, and instead thereof, I direct that all my Grandchildren shall take by representation, that is per stirpes and not per capita.”
When this first codicil was executed, testator’s wife and seven children were still living and there were eleven grandchildren with a twelfth soon to be born.
On an unspecified date in February, 1924, testator executed the second codicil:
“First. Paragraph Six (6) of my said will and distribution therein made, I now revoke, and instead thereof, I direct that the distribution therein made shall be per stirpes and not per capita, that is to say: 0. After the Trustees shall have made sure annually that my said wife, Carrie F. Grimm, has had her annual income, they shall annually distribute the residue of the income among all my then living children and the then living descendants of any children of mine who may then be deceased, but the descendants of any child of mine who may then be deceased,, only to take and have such part or share as his, her or their deceased parent would have taken if he or she were living.
“Second. Paragraph Seven (7) of my said will and the distribution therein made, I now revoke, and *133instead thereof, I direct that the distribution therein made shall be per stirpes and not per capita, that is to say: 7. Upon the death of my said wife, Carrie F. Grimm, the homestead and funds set apart for her use shall thereupon be divided forthwith among all my then living children and all the then living descendants of any children of mine who may then be deceased, but the descendants of any child of mine who may then be deceased, only to take and have such part or share as his, her or their deceased parent would have taken if he or she were living.
“Third. Paragraph Eight (8) of my said will and the distribution therein made, I now revoke, and instead thereof, I direct that the distribution therein made shall be per stirpes and not per capita, that is to say: 8. The income of the residue of the estate shall be distributed quarterly among all my then living children and all the then living descendants of any children of mine who may then be deceased, but the descendants of any child of mine who may then be deceased, only to take and have such part or share as his, her or their deceased parent would have taken if he or she were living: such distribution to continue until the distribution of the principal of my estate as set forth in Paragraph Nine (9) of this my will.
“Fourth. Paragraph Nine (9)7 of my said will and the distribution therein made, I now revoke, and instead thereof, I direct that the distribution of the principal of my estate shall be made per stirpes and not per capita, that is to say: 9. After the death of my last surviving child, and as soon as the youngest child of *134any of my children born at the time of the death of my last surviving child shall have attained the age of twenty-one years, then the whole principal of the estate shall be. divided among the then living descendants of my children, but the descendants of any of my grandchildren who shall then be deceased shall only take and have such part or share as his, her or their deceased parent would have taken if he or she were then living.
“Fifth. .Paragraph Ten (10) of my said will and the provision therein made, I now revoke, and instead thereof I direct that all my children and their descendants shall take by representation, that is to say, per stirpes and not per capita.”
At the time testator executed this second codicil, he had a wife, seven children and twelve grandchildren, one of whom had married.
Testator’s wife survived testator by five years, dying on. June 21, 1933, and the last of testator’s children died on December 8, 1967. The death of testator’s last surviving child terminated the trust and the trustees filed the necessary accounting for the purpose of effecting distribution of principal to those persons entitled under the terms of the will and codicils.
The court below, on petition of the trustees, appointed an auditor to recommend to the court the manner ; of distribution of the balance in the trustees’ hands.8 The sole question involved the manner of distribution, either per capita or per stirpes, of the principal. of the residue of the trust. The auditor concluded, that'testator’s, overall testamentary intent was to treat his grandchildren, upon termination of the trust, as a class for per capita distribution and his great-grandchildren as a class for per stirpes distribution and *135recommended distribution on this basis.8 9 The court below adopted the auditor’s recommendation as to distribution.
The first codicil drastically changed testator’s testamentary scheme outlined in the will. While per capita distribution of the principal of the residuary trust remained unchanged (paragraph 9 of the will), the residuary trust income (paragraph 8 of the will) and the unused income and principal of the separate trust for testator’s widow (paragraphs 6 and 7 of the will) were to be distributed among testator’s children and grandchildren, representing a deceased child or children of testator, per stirpes and not per capita. This change was probably dictated by the fact that in 1922 testator, in addition to seven children, had eleven grandchildren and one grandchild about to be bom, spread in varying numbers among the family units of his seven children.
The interpretation of the second codicil’s language, viewed in the context of the language of the will and first codicil, is the problem we must determine.
Appellants urge that paragraphs 3, 4 and 5 of the second codicil conclusively reveal the testator’s intent to provide a stirpital distribution of the principal of *136the trust among both grandchildren and great-grandchildren and that, even if the language does not reveal testator’s intent, which appellants contend it does, resort to canons of construction would compel the same result.
Appellees, arguing that the second codicil requires a distribution of principal per capita among testator’s surviving grandchildren and per stirpes among his great-grandchildren, rely upon certain factors: (a) the circumstances existing at the time testator executed the several writings; (b) the employment in the testamentary writings of identical stirpital phrases—on occasions when testator’s intent is patently evident—indicates an intent to restrict per stirpital distribution to substituted beneficiaries who would share, if at all, by reason of his, her or their parent’s death; and (3) the testamentary language is so clear and unambiguous that resort to rules of construction is unnecessary but, even if it were otherwise, the rules of construction would support their position.
The real thrust of appellees’ argument rests upon what is termed the “but” argument. The second codicil’s fourth paragraph expressly revokes paragraph 9 of the will and, in lieu thereof, directs that “distribution of the principal” shall be made “per stirpes and not per capita.” In the new paragraph 9, testator seeks to interpret what he means by stating that, upon occurrence of the events which require a distribution of principal, the principal shall be divided “among the then living descendants of my children, but the descendants of any of my grandchildren who shall then be deceased shall only take and have such part or share as [the] deceased parent would have taken if . . . living.” (Emphasis added) Appellees argue this “but” clause clearly reveals an intent to mandate per capita distribution to the children of testator’s children, i.e., living grand*137children, and limits stirpital distribution to the descendants of such of testator’s grandchildren then deceased. This argument would interpret the word “descendants” in two ways: when it comes to “living descendants” of testator’s children, “descendants” would mean “children,” i.e., grandchildren, but when it comes to “living descendants” of grandchildren, “descendants” would mean those persons in the line of inheritance from the deceased grandchild or grandchildren. See: Strunk Estate, 369 Pa. 478, 480, 87 A. 2d 485, 486 (1952).
The ascertainment of testator’s intent, to be garnered from the language of the testamentary writings, is not without difficulty. The result reached by the court below is that the testamentary writings clearly evidence that a per capita distribution was intended to be made within one class, i.e., living grandchildren, and a stirpital distribution to be made within another class, i.e., living children of deceased grandchildren. Examining the changes made in paragraph 10, the result reached in the court below is difficult to reconcile. Paragraph 10 in the will clearly stated: “No one,” regardless of class, should take by representation or per stirpes; the first codicil, revoking paragraph 10, clearly dictates that “all” [testator’s] children and their descendants” shall take per stirpes and not per capita.
Initially, we respond to one of appellants’ arguments that, as a result of the decree below, one of the six family units will receive six-thirteenths (6/13) of the principal whereas the mother of the children comprising this family unit received only a one-sixth (1/6) interest in the income. It is not at all unusual for a testator to give income to his children until the death of the last surviving child and principal per capita to the succeeding generation or generations. In Rosengarten Estate, 349 Pa. 32, 37, 36 A. 2d 310, 313 (1944), *138this Court said: “A study of the philosophy underlying this principle will disclose that when a testator makes a gift to a group of individuals of the same class, he is not mentally dividing them stirpitally, and increasing or decreasing their shares depending upon the fecundity of their parent. He is thinking of each member of the class as an individual.” Regardless of whether one believes the result of the decree is unfair, if that is what the testator intended, we are bound to carry out his expressed intent.
The key question is: what did the instant testator intend? To ascertain such intent, we must look to the language of the testamentary writings; rules of construction are of no consequence unless and until we find that the language of the testamentary writings fails to reveal, with clarity, testator’s intent. As this Court said, in England Estate, 414 Pa. 115, 118, 200 A. 2d 897, 898 (1964): “If the language employed by the testator in disposing of his estate is plain and clearly discloses his intention, the will interprets itself, and no rules of construction are necessary to aid in its interpretation. ... It is a matter of common sense, as well as of law, not to attempt to construe that which needs no construction. . .
Although repetitious, we cannot overemphasize the fact that testator completely changed the scheme of his testamentary disposition from per capita in the will to per stirpes under the first and second codicils. The Tin-equivocal testamentary mandate, contained in paragraph 10 of the will, that “no one,” i.e., of any generation, should take by representation was completely changed by the first codicil. Paragraphs First, Second and Third of the first codicil, read in conjunction with paragraph Fourth thereof, direct distribution per stirpes and not per capita of the income unused by testator’s widow, of the corpus of the widow’s trust and of the income from the residue of the estate insofar as *139testator’s grandchildren are concerned. Ignoring, momentarily, the “but” clause contained in paragraphs First, Second and Third of the first codicil, the intent of testator as to distribution among the grandchildren is clear beyond question because of the direction “that all my Grandchildren shall take by representation, that is per stirpes and not per capita.” Moreover, even though paragraph 9 of the will is not mentioned in the first codicil, the language of the Fourth paragraph of that codicil, revoking paragraph 10 of the will, containing the direction of per stirpital distribution to all grandchildren, would indicate that, even as to the distribution of the principal of the trust, distribution to the grandchildren would have been per stirpes. Had testator died in 1922—subsequent to the first and prior to the second codicil—subject, of course, to the testamentary provisions for his widow, distribution of income would have resulted in a per capita distribution to his living children10 and a per stirpital distribution to all the children of any deceased child or children. This change would seem to have been brought about by the fact that the testator realized that, with a large increase in the number of grandchildren and the fact that his own children were tweuty-one years older than when he made his will, the possibility of inheritance by his grandchildren dictated a method of distribution which would have insured equality among the family units rather than a distribution based on the fecundity of his children. One fact is most evident, i.e.} that under the first codicil, grandchildren would take only per stirpes and not per capita.11
*140The prefatory clauses to paragraphs First, Second, and Third of the second codicil, revoking, respectively, paragraphs 6, 7 and 8 of the will, state: “that the distribution therein made shall be per stirpes and not per capita.” These prefaces are followed by the phrase “that is to say,” which phrase, if it is to have any meaning, introduces a clause descriptive or explanatory of the prefatory clauses. The second codicil differs from the first codicil in the following respects: (1) paragraph 9 of the will is specifically referred to and revoked and the direction of a per capita distribution of the principal of the trust is changed to a stirpital distribution; (2) while the “but” clauses are retained in the second codicil, unlike the first codicil which directed distribution “equally among all my children who shall be living at the time of [testator’s] decease and the child or children of any children of mine who may be then deceased,” the second codicil directs distribution “among all my then living children and the then living descendants12 of any children of mine who may then be deceased;” (3) whereas the Fourth paragraph of the first codicil directs a stirpital distribution to “all” testator’s grandchildren, the Fifth paragraph of the second codicil directs that “all my [testator’s] children and their descendants13 shall take by representation, that is to say per stirpes and not per capita.” Fully cognizant that “every word in a will must be given effect if it is reasonably possible so to do” (see, e.g., Ludwick’s Estate, 269 Pa. 365, 371, 112 Atl. 543, 544 (1921))., we have endeavored to ascertain what testator intended by the use of the words “my living children.” Testator certainly could not have intended to refer to his own children because, having a common ancestor, they could not take by representation or per *141stirpes and, moreover, the event upon which the principal was to be distributed could not take place until all testator’s childi'en were deceased. Unless we translate the word “children” to mean “grandchildren,” the direction of distribution by representation in the Fifth paragraph of the second codicil would be senseless and the only way to render meaning to the words “my living children” is to find that testator intended to use “children” in the more extensive sense of “grandchildren.” Under our case law, we have authority to make such an interpretation. Taylor Estate, 384 Pa. 550, 557, 121 A. 2d 119, 123 (1956); Clark Estate, 359 Pa. 411, 420, 59 A. 2d 109, 114 (1948) ; Joyce’s Estate, 273 Pa. 404, 408-09, 117 Atl. 90, 91 (1922) ; Puterbaugh’s Estate, 261 Pa. 235, 104 Atl. 601 (1918). See, also, Pemberton v. Parke, 5 Binn. 601 (1813).
The “but” clause in the Fourth paragraph of the second codicil, considered with the “but” clauses in other paragraphs, is the keystone of appellees’ position. We cannot, as appellants urge, treat this clause as unnecessary, as a superfluity or as lacking impact on the language which precedes it. In our examination of the “but” clause, in juxtaposition to that which precedes it in the Fourth paragraph, we note that: (a) when the events occur upon which distribution of principal is to take place, i.e., death of testator’s last surviving child and the attainment of twenty-one years by the youngest grandchild living when testator’s last child dies, none of testator’s children then would be living and “descendants” could include only grandchildren, great-grandchildren and other more remote generations; and (b) that the “but” clause provides that only “descendants” of deceased grandchildren shall take by representation and have such part or share as the parent would have taken. The “but” clause, read literally, refers to a per stirpes distribution only among deceased *142grandchildren’s descendants living, inferring a different distribution among the grandchildren. When the second codicil was written, testator’s children were twenty-three years older than when he had drawn his will, he had twelve grandchildren and one of the grandchildren had married, raising the expectancy of a great-grandchild or great-grandchildren. The “but” clause would indicate that the Fourth paragraph contemplated a distribution to living grandchildren—encompassed in the term “living descendants of my children”—per capita and to great-grandehildren—encompassed in the term “descendants” of deceased grandchildren—per stirpes, so that the “but” clause, standing alone and isolated, would indicate an intent to differentiate between the two classes of descendants. We do not deem controlling the prefatory clause of the Fourth paragraph in view of the diversification between classes by way of distribution so evident in the “but” clause.
However, we deem the impact of the “but” clause to have been nullified by the language of the Fifth paragraph of the second codicil. The event which compels the distribution of principal is not only the attainment by testator’s youngest grandchild, living when testator’s last surviving child dies, of the age of twenty-one years but also the death of testator’s last surviving child. Until the class comprised of testator’s children is extinguished by death, paragraph 9, set forth in the Fourth paragraph of the second codicil, does not become operative and distribution thereunder could not take place. Unless we are to completely ignore the phrase “all my living children” as meaningless, we cannot reach any other result than to treat the word “children” as “grandchildren” and find testator intended distribution per stirpes not only among his great-grandchildren but his grandchildren. Reading paragraph Fifth, in the context of paragraph 10 *143of the will and. paragraph Fourth of the first codicil, as a statement of testator’s general intent as to the manner of distribution of his entire estate, we conclude that he intended a stirpital distribution among his grandchildren as well as his great-grandchildren. Such an interpretation squares with the Fourth paragraph of the first codicil, renders intelligible and meaningful the Fifth paragraph of the second codicil and brings about equality between the several classes of descendants.
To read this will and two codicils14 as a progression from a complete per capita to only a partial per stirpes distribution would torture testator’s expressed intent and place testator’s twelve grandchildren on a parity of distribution with testator’s children under the First, Second and Third paragraphs of the second codicil— containing the same “but” clause in identical language ■—as to income and would give a preferential treatment to grandchildren living over the children of grandchildren dead and place a premium upon the fecundity of testator’s children rather than equality among the several family units.15
Ordinarily, the words “per stirpes” are used with respect to substitutional gifts to substituted legatees in the event of the death of a primary legatee or legatees (Re Title Guarantee and Trust Co., 159 App. Div. 803, 144 N.Y. Supp. 889 (1913), aff'd, 212 N.Y. 551, 106 N.E. 1043 (1914), and authorities therein cited), yet “[t]he expression ‘per stirpes’ may be used in two different senses; it may refer, first, to a ‘taking by right *144of representation/ and second, to a taking ‘collectively by families and not equally as individuals’: Hogg’s Estate, 329 Pa. 163, 167.” Corr’s Estate, 338 Pa. 337, 341-2, 12 A. 2d 76 (1940).16
In Mayhew’s Estate, 307 Pa. 84, 160 Atl. 724 (1932), the testatrix, after providing a life estate for her daughter, directed that, upon the daughter’s death, the property should go to the daughter’s issue and, in default thereof, to a named charity. There was no direction as to whether distribution should be per stirpes or per capita. The daughter had six children, none deceased, and twelve grandchildren, each family unit having either no children or varying numbers. Appellants urged a per capita distribution among children and grandchildren since all were “issue” of testatrix. In denying appellants’ position, this Court said: “We therefore agree with the growing majority of states which hold that where the word ‘descendants’ or ‘issue’ is unexplained in the context of the instrument, children do not take concurrently or per capita with their parents, but take per stirpes.” 307 Pa. at 91-2, 160 Atl. at 727.
In Wanamaker Estate, 399 Pa. 274, 281, 159 A. 2d 201, 204 (1960), we said: “It is unusual for a testator to intend and direct a per stirpes distribution of income among a specified class, to-wit grandchildren, and a pér capita distribution of principal among the same class, to-wit grandchildren, with the result that certain grandchildren would receive a per stirpes share of income on Tuesday, but on Wednesday (the next day when the youngest heir or grandchild reached 21 *145years of age) would receive a per capita share of principal. If that was the testator’s intent, it should he clearly expressed. . . .” (Emphasis supplied)
England Estate, 414 Pa. 115, 200 A. 2d 897 (1964), and Rosengarten Estate, 349 Pa. 32, 36 A. 2d 310 (1944), relied upon by appellees, are clearly distinguishable.
Our examination of all the language of the three testamentary writings convinces us of the clarity of expression of testator’s intent and that such intent was that, upon the happening of the events which would trigger the distribution of principal, a stirpital distribution was to take place among living grandchildren and the child or children of a deceased grandchild or grandchildren. Therefore, the need for an application of rules of construction does not arise. We note, however, that if resort was necessary to rules of construction, one rule would certainly be applicable: first, when a will contains two clauses totally inconsistent and incapable of reconciliation, the latter shall have preference: see, e.g., Richley Estate, 394 Pa. 188, 194, 146 A. 2d 281, 284 (1958) ; Hunter, Pa. Orphans’ Court Commonplace Book, Vol. 6, p. 218, Wills 2(e). Application of this rule of construction to the factual posture of the case at bar would compel the same result we have reached without resort to such rule.
Marshall Estate, 377 Pa. 41, 103 A. 2d 420 (1954), appears somewhat apposite. In Marshall Estate, testator gave a life estate to his wife and provided that, upon her death, certain property be given to such of his children unmarried when their mother died to hold so long as any remained unmarried; upon the death or marriage of the last survivor of such children, testator gave the property to such of his married children then living for their lives; on the death of the last surviving child who had a life estate in the property, testator *146directéd that the property he sold, “and the net pro-, ceeds divided among the descendents [sic] of my married children, per stirpes.” When the testator made his will he had seven children, three of whom were then married and six grandchildren. When his last surviving child died, two of his grandchildren had also died. Of the four grandchildren living, three were from one daughter and one from another. Appellants therein contended that each grandchild took per capita or a one-fourth (1/4) interest and that only great-grandchildren, if any, took per stirpes; appellees contended that grandchildren took per stirpes, i.e., one-half (1/2) to the one grandchild and one-sixth (1/6) to each of the other three grandchildren.17 This Court stated: “Instead of making a bequest for each grandchild to share equally or per capita, the testator provided that the net proceeds should be divided ‘among the descendants of my married children, per stirpes.’ ” 377 Pa. at 45, 103 A. 2d at 422. .We upheld stirpital distribution. The sole difference between Marshall and the instant case is that in Marshall “per stirpes” immediately followed the phrase “among the descendants of my married children” whereas in the case at bar “per stirpes” appeared in the prefatory clause. Moreover, unlike Marshall, we have present herein the strong language of the Fifth paragraph of the second codicil.18 We are satisfied of testator’s intent garnered from the language of his testamentary writings and we conclude that tes*147tator intended, that not only his great-grandchildren but his grandchildren were to receive distribution of the principal of his estate on a stirpital basis. In concluding otherwise, the court below was in error.
Decree reversed. Estate to pay costs.
Mr. Justice Cohen took no part in the decision of this case.Pearl Moorliead died October 20, 1934, survived by one child; Uyda Thompson died April 11, 1942, survived by six children; Eugene Grimm died July 8, 1944, survived by one child; Stella Blair died December 19, 1946, survived by one child; Louise Rowland died December 21, 1950, survived by one child; Mayme Koos, whose only child predeceased her, died June 5, 1960, survived by one granddaughter; Evelyn Jenkins died December 8, 1967, survived by three children.
Testator essentially set up two trusts, one consisting of the fund to provide income for his wife, and the other consisting of the residue of his estate.
“6. After the Trustees shall have made sure annually that the said Oarrie F. Grimm has had her said annual income, they shall annually distribute the residue of the income from the estate equally among all my children who are living and all the children of any child of mine that may be deceased. This distribution shall be per capita.”
“7. Upon the death of the said Carrie F. Grimm the homestead and funds set apart for her use shall thereupon forthwith he divided equally between all my then living children and all the then living children of any child of mine that may be deceased. This distribution shall also be per capita, so that the portion of each individual distributee of the property mentioned in this paragraph shall be the same, whether the distributee be child or grandchild of mine.”
“8. The income of the residue of the estate shall be distributed quarterly amongst my children and any children of any of my children who may then be deceased, per capita, during the lifetime of my child who longest survives.”
Paragraph 10 is set forth in.the text of the opinion.
"9. After the death of my last surviving child, and as soon as the youngest child of any of my children born at the time of the death of my last surviving child shall have attained the age of twenty-one years, then the whole principal of the estate shall be divided equally amongst the then living children of my children, per capita.”
The balance in the estate for distribution approximated $350,-000.
This recommended distribution would give one-thirteenth (1/13) to each of the parties in interest, which would give to one of the six applicable family units—that of testator’s daughter, Lyda Thompson—six-thirteenths (6/13) of the principal although Lyda Thompson, during her lifetime, received only one-sixth (1/6) of the income. It is appellants’ position that the three children of Mrs. Jenkins should now receive in equal shares the one-sixth (1/6) interest of their mother, that the six children of Lyda Thompson should each receive a one-thirty-sixth (1/36) interest, that the three other remaining grandchildren—one child of Louise Howland, one child of Eugene Grimm and one child of Stella Blair—should each receive a one-sixth (1/6) interest, and the one great-grandchild should receive the interest which her father would have had had he lived, to wit, a one-sixth (1/6) interest.
Despite the prefatory language of the First, Second and Third paragraphs of the first codicil, we reach this result because, testator being the parent, testator’s children could not take by representation or per stirpes. The use of the word “equally” supports this view.
We reach this result regardless of the “but” clause.
Italics supplied.
Italics supplied.
Where a will and codicil are in conflict, the codicil provisions prevail. See Elkins’ Estate, 339 Pa. 193, 12 A. 2d 83 (1940).
Where the language of a testamentary writing is equivocal, a construction inuring to the benefit of remote lineal descendants is preferred to that which favors immediate issue exclusively. See Clark Estate, 359 Pa. 411, 420, 423 n.l, 59 A. 2d 109, 111, 114 n.l (1948).
In Corr's Estate, 338 Pa. 337, 341, 12 A. 2d 76 (1940), where the gift over was “to be divided amongst such children or descendants”—of the life tenant’s sisters, “said fund to he divided amongst such children or descendants per stirpe, upon the principle of representation,” we held that, under the circumstances, testator intended distribution per stirpes, among the children as well as the next generation.
These contentions were similar to those of the respective parties in the case at bar.
In Marshall, we said that testator “could have provided a per capita distribution for his grandchildren, if he so desired, by merely dividing the net proceeds ‘among the descendants of my married children.’ ” Appellees’ reliance on such statement is misplaced. First, such statement was not decisional and, second, appellees would ignore the direction in the case at bar that distribution be made per stirpes.