revived, provided that writing clearly discloses the testator's intent so to do. In such case the effect is to republish the original will as of the date of the last writing. But the mere revocation or destruction of a codicil which itself revokes a will does not resurrect the latter as an effective instrument. It can be made so only by a republication in the presence of its attesting witnesses. If, on the other hand, the codicil only modifies a previous will in certain particulars, and not as a whole, the original will stands except as to the matter so modified. For instance, if the codicil provides for a legacy in addition to those mentioned in the will, the latter is modified to the extent of the new legacy. In such case the revocation of the codicil leaves the will as originally drawn, except as to the amount of the legacy given by the codicil, which passes as intestate property to the next of kin.

It only remains to consider the converse of these propositions. Suppose the will to which the codicil relates is itself revoked by destruction, or otherwise, what becomes of the codicil? The general rule is that, in such case, the revocation. of the will carries with it the codicil also. Of course if the codicil disposes of all the testator's property and is wholly independent of the will, it will not be affected; but such a codicil would

To

hardly be called a codicil at all. It would practically amount to a new will. If the reader has been able to digest these legal refinements, he will probably have come to the conclusion that a codicil is something of a vehicle of trouble. an extent this is true, and we are quite willing to advise him never to execute a codicil; that is, in the sense that it is but a partial disposition of his estate. Beside the complicated questions we have just adverted to, it involves a double probate and perhaps a double contest, all of which may be avoided by the execution of a new will.

17. Papers referred to in wills.

Wills are frequently found in which the testator has made reference to other papers and documents, as, for instance, a personal letter to the executors containing private instructions, or an unattested paper setting forth the items of property or the names of proposed beneficiaries, with the idea that they will be given effect in connection with the will itself. This is a great mistake. No paper can be proved as a will unless it has been duly signed, published and attested by witnesses, as required by statute. Thus, in one case, a will directed the executors to carry out the terms of a certain deed of trust which had been made by the testator, and if this could not be done, to set aside

a certain portion of the income and distribute it among the beneficiaries named in such deed of trust in the proportions therein specified. The court decided the will created no trust, because it disposed of no property, and that it received no support from the deed of trust because that instrument had not been authenticated according to the Statute of Wills. Any instrument or writing, however, which is so executed may be incorporated into a will by reference thereto, and be proved and recorded therewith.

But unattested papers are not wholly without effect. While they cannot be proved as a part of the will, reference to them may be made for purposes of identification and description. For example, a testator may direct his executors to deduct from the shares of one of his children such amount as appears by his books of account to be owing him by that child. In such case the entries in the testator's books may be considered, so as to enable the court to properly construe the will. They form no part of the will, however, and whether they exist or not does not in any way affect the question of proving or rejecting the will.

A will once signed and attested in all respects as the statute commands, remains valid and

effective until the testator's death, unless it be revoked in one of the ways about to be described. Even if it be lost, or destroyed by accident, it may still be proved as we shall see later on (see post § 47).

In the first place, a will may be expressly revoked by the act of the testator in executing a paper declaring such revocation, or by physical cancellation of the instrument. It may be indirectly or impliedly revoked, by a later inconsistent will or codicil, or by a change in the character of the property affected by the will, or of the testator's interest therein. Lastly, a will may be revoked by what is called the operation of law; that is, the law declares that certain circumstances occurring in the life of a person have the effect of revoking any prior will he may have made.

19. Same: by later will or codicil.

We have already seen how a codicil may operate to revoke a prior will. The same result may follow from the execution of a later will containing either a revoking clause, or provisions so inconsistent with the earlier instrument that it can no longer be regarded as in effect. A will may also be expressly revoked by a writing which is not testamentary in its character, but to have that effect the paper must be executed with the same

formalities as the will itself. Thus, where the testator wrote a memorandum on the margin or back of the will, stating that the latter was thereby revoked, but which, though signed by him, was not witnessed, it was held by the courts that the writing was insufficient for any purpose.

And in another case where the will itself provided that it was to be regarded as cancelled the day that the testator entered matrimony, it was held not to be effectually revoked by his marriage, and was admitted to probate.

20. Same: by destruction.

A will may also be revoked by burning, tearing, cancelling or otherwise destroying it, if done with the intention of accomplishing that result. This may be done either by the testator or by a third person in his presence, by his direction and consent. Complete destruction or cancellation is not necessary to constitute revocation. If the will be torn into several fragments it will be sufficient, even though it may be possible to piece them together again. Cases have occurred where a will, known to have been in the testator's possession, has not been found, or if found, the testator's signature appears to have been crossed out. In each of such events there is a presumption of revocation; in the one, by destruction, and in the