It is, however, equally clear to our minds that in the present case the rule in Hancock v. Watson 1 is expressly excluded by the plain words of cl. 18 of the will. It seems necessary to quote again the oft-quoted words of Lord Davey. His Lordship said :- For, in my opinion, it is settled law that if you find an absolute gift to a legatee in the first instance, and trusts are engrafted or imposed on that absolute interest which fail, either from lapse or invalidity or any other reason, then the absolute gift takes effect SO far as the trusts have failed to the exclusion of the residuary legatee or next of kin as the case may be 2". It is obviously involved in this state- ment that there can be no occasion for the application of the rule in Hancock v. Watson (1) unless trusts are engrafted on an antecedent gift to a donee, and there is a failure of those trusts. But this is the very event with which, in the present case, cl. 18 in terms deals. Trusts are engrafted on Phyllis's share in favour of Phyllis's children. When Phyllis died without having any children, there is a failure of those trusts, and cl. 18 provides that her share shall accrue to the other shares. Clause 18 is, as Mr. Bradshaw said, really part of the settlement of the shares.
It is to be noted that the learned judge from whom this appeal comes, after observing that the words used in cl. 18 are fail or determine said We may put aside the question of failure' of the trust. No one contends that the trust has failed". He then proceeded to consider whether there had been a "determination" of a trust. Mr. Bradshaw assured us that he had not intended to convey to his Honour that he did not contend that there had been a failure of engrafted trusts, and we feel sure that there must have been some misapprehension or misunderstanding. As we have pointed out, the rule in Hancock v. Watson (1) could not apply if there had not been such a failure, and his Honour had already held that there had been such a failure. In other words, the very same event which, if cll. 13 and 14 had stood alone, would have attracted Hancock v. Watson (1), attracts, when the whole will is read, cl. 18.
It is, of course, to be remembered that, as Fullagar J. pointed out in Russell v. Perpetual Trustee Co. (Ltd.) 3, the rule in Hancock V. Watson (1) is no more than a rule of construction. In Lassence V. Tierney 4, itself Lord Cottenham said: In every case the question must
be one of construction " 5. It is always described as a rule of construction. Being no more than a rule of construction, it can always be excluded by a settlor or testator, and one of the ways in
1(1902) A.C. 14.
2(1902) A.C., at p. 22.
3(1956) 95 C.L.R. 389, at p. 402.
4(1849) 1 Mac. &G. 551 [41 E.R.
5(1849) 1 Mac. &G, at p. 562 [41
E.R., at p. 1383].