CAMPBELL, C.J., delivered the opinion of the court.

On March 10, 1930, a writing bearing date of October 26, 1927, and purporting to be the last will and testament of J. I. Triplett, a deceased citizen of Shenandoah county, with codicils thereto bearing dates of November 6, 1927, August 6, 1929, and September 17, 1929, was produced before the Circuit Court and, after proof of due execution, was admitted to probate. Herbert Trotter, the executor named in the will, was permitted to qualify as such executor and entered into bond in the penalty of $750,000.

At a subsequent term of the Circuit Court, the appellees, claiming to be heirs and distributees of J. I. Triplett, deceased, under the statute of descents and distribution, and who were not parties to the probate proceedings, filed their bill of complaint, alleging that the paper writings purporting to be the will and codicils of J. I. Triplett, deceased, were not the last will and testament of the decedent, and praying that an issue devisavit vel non be made up and tried by a jury. An answer was duly filed by appellants, and by agreement a jury was waived and all matters of law and fact were submitted to the chancellor.

After hearing proof and inspecting the said paper writings, the court entered a decree reading in part as follows: "* * * the court * * * doth adjudge, order and decree that the certain paper offered for probate by Herbert Trotter, the executor named therein dated October 26, 1927, purporting to be the last will and testament of Dr. J. I. Triplett and purporting to be attested by George R. Geary and Chas. I. Dellinger, is denied probate as an attested will.

"It being proved by George R. Geary and Charles I. Dellinger, two disinterested, credible and competent witnesses, however, that said paper writing dated October 26, 1927, and the paper known as Codicil No. 1, dated November 6, 1927, and the paper marked Codicil No. 2, dated August 6, 1929, and the other certain paper dated September 17, 1929, all offered for probate by said Herbert Trotter, including the interlineations therein appearing, are each in the handwriting of Dr. J. I. Triplett and were signed by him and that at the several dates of the same he was of sound and disposing memory, and the same having been thus proved for probate as a holographic will in the manner required by law, it is adjudged, ordered and decreed that, pursuant to the findings of the court set out in Paper A above, that the said will and said codicils, including the interlineations therein, be admitted to probate, together as the unwitnessed and holographic will of the said Dr. J. I. Triplett and be recorded together as the last will and testament of said Dr. J. I. Triplett, deceased.

"And it is further ordered that the papers tendered by the defendants herein, marked in the record as Exhibits X and Y, being proved by the same witnesses as being entirely in the handwriting of J. I. Triplett, deceased, and that he was of sound mind and disposing memory at the date of the same, be and the same are hereby denied probate."

It is from that decree this appeal was allowed.

In an able opinion which is filed as a part of the record, the chancellor, Honorable Philip Williams, has discussed the pertinent questions raised by the petition for an appeal. The opinion is as follows:

"This is an issue of devisavit vel non which was submitted to the court for determination, a jury having been waived by the parties. No questions of fraud, undue influence or mental competency are in controversy and there are but two issues between the parties. One of these issues arises from interlineations, erasures and additions which appear in the papers offered for probate, and which the contestants contend to be alterations which will prevent probate as an attested will. The other issue involves two papers which contestants assert should be admitted to probate, but which the proponents contend are no part of the will.

"Dr. J. I. Triplett was a resident of Mount Jackson, in Shenandoah county, Virginia, and died there on March 6, 1930, in the eighty-fifth year of his age. He had never married. He was educated a doctor of medicine, but did not practice that profession. He had devoted himself to various business interests during a long and successful life, and had accumulated a large estate, consisting of both personal and real property. He owned real property in Virginia, Pennsylvania and Florida. He wrote his own will, in which, with the exception of bequests to certain of his nephews and nieces, he provided that his estate should be vested in trust for the establishment and maintenance of the 'Triplett Business College' to be located at Mt. Jackson, Virginia.

"Dr. Triplett first wrote a will of thirteen pages, entirely in his own handwriting, signed and published by him in the presence of two witnesses, and dated October 26, 1927. This will is written on folio sheets, with four pages to each sheet, and the pages and text prove that each of the sheets formed a part of the document subscribed by the attesting witnesses. It was written in black ink at first, but there appears changes in blue ink in the handwriting of the testator.

"On November 6, 1927, he wrote a codicil to the will, entirely in his own handwriting and properly signed, but without witnesses. This paper is written on the next page of the folio sheet on which the will was terminated. It, too, is in black ink and shows two changes in the handwriting of the testator in blue ink. This is codicil number one.

"On August 6, 1929, he wrote a second codicil immediately following codicil one and on the same folio sheet. This, too, is entirely in his handwriting and is without witnesses. Black ink was used in writing this codicil, but changes appear therein written by the testator in blue ink. This is codicil number two.

"On September 17, 1929, Dr. Triplett wrote in his own handwriting immediately after codicil two, the writing termed in the record codicil number three, though it was by him given no designation, as were the other codicils. This codicil is likewise signed, but is without witnesses. It was written in black ink and changes appear in blue ink in the handwriting of the testator.

"At some time after the execution of the attested paper the first two codicils, Dr. Triplett began to rewrite his will; and there was found in the same envelope with the will and codicils, above described, an incomplete draft of a will in his handwriting, unsigned. This paper is not dated. It was written entirely in blue ink of, apparently, the same color as that of the changes referred to in the description of the will and codicils. In the record this instrument is designated as 'Exhibit X.'

"On December 11, 1929, Dr. Triplett wrote what he termed 'This memorandum statement,' entirely in his handwriting, signed by him, but without witnesses. This writing is on a single page of small size note paper. It is written in blue ink similar in shade to that of Exhibit X. This paper is called 'Exhibit Y' in the record.

"Prior to the execution of the will dated October 26, 1927, Dr. Triplett had ascertained that two witnesses were required to wills of real estate by the law of Florida. Though he had apparently written his will without assistance in order that its contents might not be known, when he came to execute it, he had it witnessed by the cashier and bookkeeper of the Mt. Jackson National Bank, of which he was the president. These witnesses testify, however, that when he requested them to witness his will, he had it folded in such manner that only about one-third of the last page was visible. They could have read the names of the trustees and two lines of the writing above those names. They saw his signature and the attestation clause. But they did not read any portion of the will, and they could not testify whether or not there appeared any erasure, interlineation, or change of any kind in the original text, at the time they witnessed the will.

"Dr. Triplett lived in an apartment in the hotel at Mt. Jackson. It was his habit to write at two different tables in this apartment; he made use of one of them when he wrote in the day and of the other when writing at night. He also habitually kept several ink bottles on these tables and supplied himself with ink from the office of one of his mills which was near his apartment. Blue ink and black ink, similar to that used in the writing of the will and in the changes appearing there, were in use at that office.

"Dr. Triplett had kept his will in his deposit box in the bank. During his last illness, the cashier, Mr. Geary, having occasion to open this box, noticed that there was no will in it. He informed Dr. Triplett's nephew, Herbert Trotter, of this fact, who requested W. L. Vehrenkamp to look in Dr. Triplett's rooms for the will. A few days before, Mr. Vehrenkamp had gathered together all the papers in Dr. Triplett's room and had placed them in a box there. Upon the request of Mr. Trotter, Mr. Vehrenkamp and Mr. Geary went to Dr. Triplett's room and there found an envelope which contained all of the papers which have been described. No other papers of a testamentary character have been found.

"An examination of the attested will shows changes made in the original text in black ink and in blue ink. Those in black ink appear to have been made at the same time the will was written, but they are important in that they show that when they were made the writer was using the same color ink as that in which the will was written, and that the blue ink changes, inferentially, were written in a later revision of the will.

"Until the twelfth page is reached the changes in blue ink are of no consequence; they do not change the meaning of the will and might be rejected without in manner affecting it. On that page a bequest was originally number '4th' in black ink. This has been erased and '5th' written in blue ink both in the marginal numeration and in the body of the bequest. In the same bequest a blank left in the original text has been filled in in blue ink with the name 'P. P. Hanson.' It is also apparent that the amount of the bequest as originally written has been erased, and the words 'five hundred' inserted in blue ink. This bequest reads as follows, the blue ink changes being indicated by parentheses:

" '(5th) Bequest, being min e ful of two nieces, whom I p ur posed remembering in making bequests. Therefore will number this one in order, making it my (5th) bequest, I bequeath Mrs. Ethel Bursey and Mrs. Loretta (P. P. Hanson.) children of my dec 'sister Mrs. Malinda B. Craig of Washington, D.C. (five hundred) dollars each, payable in twelve months after my demise.'

"On page thirteen, three changes of text appear in blue ink. The first is an interlineation ' power,' correcting the name of Woodstock Mills and Electric Light and Power Company, Inc. The second is an erasure of the text and the insertion of 'two thousand.' The third is also an erasure of the original writing and the insertion of the name of 'W. L. Vehrenkamp' as one of the trustees. The part of the will where these changes occur reads as follows, the parentheses indicating the changes:

" 'I desire my nephew Herbert Trotter who has been with me from childhood to continue in the management of my Woodstock Mills and Electric Light ( Power) Company, Inc., and I appoint him my executor without bond at a salary of 'five hundred dollars annually his salary as manager of the Woodstock Millsc will be continued at (two thousand) dollars annually, as executor he will be in charge of my general business till all improvements have been completed, which will be built by contract and let to the highest bidder the standing and reputation of the firm being considered —, and all construction will be subject to inspection, whilst in course of construction by a competent party, at a reasonable cost. I appoint Herbert Trotter, J. I. Triplett, Jr., F. S. Tavenner, Jr., and (W. L. Vehrenkamp) trustees.' The attested will ends with the clause quoted above.

"Changes in blue ink appear in codicils one and two, which were written in black ink, but they need not be detailed, it being noted, however, as a circumstance later to be discussed.

"In codicil three, originally written entirely in black ink, appears an addition in blue ink. It is apparent that this addition was inserted after this codicil had been completed. The period which originally terminated the sentence has been changed to a comma and the blue ink addition inserted between the signature in black ink and the last line of the original text. This codicil reads as follows, the addition being enclosed in a parentheses:

" 'My will and testament with the several codicils made Oct. 26th, 1927, and legally closed — shall not be affected by my undertaking herewith to rewrite the will and fail to complete it, (and that which I have written is herewith made subject to acceptance by the trustees.)

" 'J. I. TRIPLETT, Sept. 17, 29.'

"There is no denial, and it is proven, that all of these papers, and the changes in them, are in the handwriting of Dr. Triplett. Furthermore, it is admitted that the paper purporting to be an attested will may be probated as a holographic will even if it cannot be proven an attested will.

[3, 4] "It would seem that this will, including its three codicils, is a valid holographic will under the Virginia statute (Code 1930, section 5229). Under the statute in West Virginia, which is similar to the statute in Virginia, it has been held that erasures and other alterations made by the testator in a holographic will, even though they are made after the will has been executed, do not invalidate it if his name still remains in such manner as to manifest that it was intended as a signature. The will then becomes re-executed with all the changes as valid and subsisting parts of the new will. LaRue v. Lee, 63 W. Va. 388, 60 S.E. 388, 14 L.R.A.(N.S.) 968, 129 Am. St. Rep. 978. Moreover, even if this were not regarded the will and codicils would have become validated by codicil three, by which they would have become republished. For even though the addition to the said codicil be held to have been made after its execution, that part would be rejected without affecting the remainder. In Gooch v. Gooch, 134 Va. 21, 113 S.E. 873, it was held that where the testator in a codicil refers to a will and gives sufficient demonstration that when making the codicil, he considered the will as his will, a republication of the will may be implied. The paper known as Exhibit Y would also affect a republication. However, it does not seem necessary to resort to the principle of republication in this case, because the holding of the case of LaRue v. Lee, supra, is believed to be sound.

"But may this will, or that part of it which is attested, be admitted to probate as a valid attested will?

"The answer to this question depends upon when the changes were made which appear in the will which purports to have been attested. If they were made before attestion, the answer would be 'yes;' if they were made afterward it would be 'no.'

"The evidence extrinsic to the will, and that of the papers themselves must be relied upon to determine the question; and lacking any testimony as to when these changes were made, the reasoning must be by inference from the circumstances disclosed.

"The burden of proving that the will has been executed and attested in the manner required by the statute (Code 1930, section 5229) is upon the proponents; but is there a presumption that alterations in a will are presumed to have been made before execution, or is there a presumption that they were made after execution?

"There has been found no case in Virginia which decides this question with relation to an attested will. Wilkes' Adm'r v. Wilkes, 115 Va. 886, 80 S.E. 745, held that where a holograph will shows a line drawn through a certain clause, there being no evidence to the contrary, it will be presumed that this was done by the testator, and the conclusion was to admit the will to probate with that clause stricken out. Another Virginia case is Harris v. Wyatt, 113 Va. 254, 74 S.E. 189, where an attested will showed an erasure and the question was whether or not this was a revocation. It was shown in evidence that this erasure had been made after the execution of the will, but there was no evidence to show who made the erasure. The court stated that there was no presumption that the erasure had been made by the testator, and that the erased clause was properly admitted to probate as part of the will.

"That these decisions established a rule that alterations appearing on the face of an attested will are presumed to have been made before execution, appears doubtful. They do not seem to be in conflict with the principles stated in City National Bank v. Slocum (1921, C.C.A.6th), 272 Fed. 11 (certiorari denied in [1921] 257 U.S. 637, 42 S.Ct. 49, 66 L.Ed. 409), as follows:

" 'When a will is presented for probate and discloses interlineations and erasures, it is obvious that, if they were made before the will was executed, they form a part of it, and the document, as so changed and altered, is the one which should be received; while, if they were made after execution, they form no part of the will and are of no effect whatever, unless they sufficiently support an inference of cancellation.

[8, 9] " 'It is claimed, on one side, that there is a presumption that such changes were made before execution, and, on the other, that the contrary presumption prevails, and it is said that there is a conflict of authority on this subject. We have made as thorough an examination of the cases cited and others as we can, and, while there are decisions indicating that the judges supposed there was a presumption one way or the other, we are satisfied that there is no substantial conflict, and that the true rule is that there is no presumption of law; that the burden of proof is on the proponent to show that any alteration which he wishes to be considered effective was made before execution; but the face of the paper and the obvious circumstances may amply meet that burden, and the inference to be drawn is always one of fact. If the will is drawn and executed with full formality and under careful advice, it is customary to note such changes in the attestation clause. If so noted, this supports the resulting inference of fact that they are part of the will; if not, absence of the notation may or may not be suspicious, depending on the extent of formality and care that seems to have been observed; and the changes may be, in themselves, suspicious enough to indicate that they were made later, or so natural and probable as to support the contrary inference.'

"See, also, upon this question Martin v. Martin (1929), 334 Ill. 115, 165 N.E. 644, 67 A.L.R. 1127, and an annotation to this case where numerous authorities are collected.

"Here changes appear on the face of the paper in ink of a different color and otherwise of a character to indicate that they may have been made after execution, the proponents must, therefore, sustain the burden of proving that these changes were made prior to execution.

"The evidence shows that the will was written in black ink and was executed October 26, 1927. Thereafter, two codicils were written in black ink, one of them dated November 6, 1927, and the other August 6, 1929. The third codicil is dated September 17, 1929, and was written in black ink. That codicil states an intention upon the part of the testator to rewrite the will. In March of 1930 the will was found in the testator's room, and with it an incomplete draft of a rewritten will, together with a paper dated December 11, 1929, stating that he was engaged in rewriting his will. The two papers, last referred to, are written in blue ink. In the will appear erasures with insertions in the handwriting of the testator in blue ink similar to that in which those papers are written. Alterations in similar blue ink also appear in the three codicils, in the last of which is clearly an addition made after this codicil has been completed, and this addition is in the similar blue ink. This addition to this codicil is significant; it shows obviously that it was made not earlier in time than the date of that codicil, and very probably was made at a later date. The blue ink alterations in the other codicils also show, obviously, that they could not have been made prior to the date of those codicils, though they might have been made later. The blue ink was not used in writing the will and the three codicils. More than two years after the will was written, blue ink was being used by the testator in rewriting his will, and that ink is similar to that in which all of the alterations to the will and the codicils were written. This evidence and the appearance of the papers themselves, which words cannot photograph, seem to sustain the inference that the erasures and insertions in the will of October 26, 1927, were made after the execution of that instrument; and this inference appears more reasonable and of more weight than that to be drawn from the evidence tending to show the contrary.

"This conclusion having been reached, it follows that no part of the will may be admitted to probate as a will attested by witnesses. It will be seen that some of the alterations are material, and are properly effective as parts of the holograph will because they were made in the handwriting of the testator. But these alterations cannot form any part of the will which was attested by the two witnesses. That will does not now exist, it has been changed, though being holograph it is entitled to probate in its altered form.

"Not without the most serious deliberation and some reluctance, has this conclusion been reached. There is nothing in this evidence to cast any suspicion upon the genuineness of the alterations or to raise any doubt that the testator intended them to be effective. But, of course, the protective requirements of the statute, designed to insure testamentary disposition of property against fraud and imposition, must be of general application, and must be proven to have been complied with.

"The other issue for determination shows the contestants assuming the position of proponents in offering the papers designated as Exhibit X and Exhibit Y for probate; while the position of contestants is taken by the parties who were proponents of the other papers.

"Briefly, it is asserted, on the one side, that Exhibit Y has the requisites of holographic will; that Exhibit X is incorporated by reference into that paper so far as the third clause is concerned; and that, therefore, these two papers should be probated as a part of Dr. Triplett's will; this is controverted, on the other side, it being contended that Exhibit X is not properly identified to be incorporated by reference.

"Exhibit Y is as follows:

" 'Mt. Jackson, Va., Dec. 11th, 1929.

" 'This memorandum statement is intended to show that my will written October 26th, 1927, and witnesseth by George R. Geary C. S. Dellinger is not to be affected by my undertaking to rewrite it and failing to complete it. The change made in my third request I prefer being substituted for the original.

" 'Witness my hand and seal,

" 'J. I. TRIPLETT, (Seal)."

"Exhibit X is the unfinished will which Dr. Triplett was engaged in writing and which was found in the same envelope with Exhibit Y and the will and codicils. This paper contains the preface of a will followed by four enumerated clauses providing for a testamentary disposition of property. The third clause reads as follows:

" '3rd. I bequeath unto my nephew J. I. Triplett, Jr., fifteen thousand dollars, the same to be credited on his indebtedness for advanced payments made in the Melville acreage with the accrued interest thereon since the date of payment. He has materially contributed toward my success in real estate investments in the city of Jacksonville and rendered valuable services professionally.

" 'He has long since been given my father's secretary and book-case, which is in the library in my brick flat. This is a family relic, came from my grand-father Dr. Wm. H. Triplett of Front Royal, being a gift to my father. I also will to my nephew the 8 day clock, which also antique relic of the family, which I have in use in the Mt. Jackson National Bank keeping accurate time of the day also of the time of the moon. Its works all being hand made.

" 'My nephew legally heirs it, being the only living representative of the family name.'

[14, 15] "There are three requisites in order that an extrinsic paper may be incorporated into a will by reference. It must be made a paper in actual existence at the date of the execution of the will; it must appear from the face of the will that it is a paper in actual existence at that time; and it must be identified and described with reasonable certainty in the will. Harrison on Wills and Administration, section 102(7); William J. Bryan's Appeal from Probate, 77 Conn. 240, 58 A. 748, 68 L.R.A. 353, 107 Am. St. Rep. 34, 1 Ann. Cas. 393, and note; Pollock v. Glassell, 2 Gratt. (43 Va.) 439; Selden v. Coalter, 2 Va. Cas. (4 Va.) 553.

"The paper in question fails to measure up to these requisites, and therefore should not be admitted to probate.

"Exhibit Y states that the testator was 'under-taking to rewrite' his will, and was dated December 11, 1929. He did not die until March 6, 1930. How much of the paper sought to be incorporated by reference was written after Exhibit Y was written; or how much had been written at the date of that exhibit? We do not know. How, then, is it proven that the paper was in existence at the time it was claimed to have been incorporated by reference? If it be answered that the '3rd bequest' appears complete in Exhibit X, and therefore must have been in existence because the reference is to that bequest, the question at once arises: Was the reference to the 3rd bequest in the will or was it to the 3rd bequest in Exhibit X? The testator did not write: 'The 3rd bequest I prefer being substituted for the original.' What he did write was: 'The change made in my 3rd bequest I prefer being substituted for the original.' When Exhibit X is examined and compared with the will of October 26, 1927, it will be found that a number of changes from that paper were made in Exhibit X; and that some of these changes appear in the 3rd bequest and some in the 4th bequest of the exhibit. The numeration is not the same in the two papers. Therefore the reference to the '3rd bequest' in Exhibit Y does not prove how much of Exhibit X was in existence at the requisite time, even if it be assumed that the reference to that paper fulfills the requirement; which assumption also is doubtful. The testator appears to have been referring to the 3rd bequest in the will; rather than that bequest in the paper he was then engaged in writing; and the evidence does not show how much of the changes in the 3rd bequest of the will was in existence when Exhibit Y was executed.

"It may be said, too, that Exhibit X was found in the envelope with the paper called Exhibit Y; and that this is evidence to show that it was in existence when that paper was executed; and that this fact is also evidence to show it was the paper referred to. But this fact still does not prove the requisite existence at the date of the execution of Exhibit Y; nor does it identify it as the paper to which reference was made in that exhibit. It may or it may not have been the same paper; it may or may not have been in existence at the time Exhibit Y was executed; or it may have been written in part of that time and partly written thereafter. The proof is not sufficient.

"Therefore, the decree shall be to admit to probate as a holographic will, but not as an attested will, the will dated October 26, 1927, together with the three codicils thereto, with the erasures, interlineations and additions appearing in these papers; but to deny probate to the papers designated as Exhibit X and Exhibit Y."

Upon mature consideration of the matters brought under review, we concur in the conclusions reached by the trial court, and the decree complained of will be affirmed.

Affirmed.