In order to make out a justification of deception, on the ground of expediency in any case, all the possible results, direct and indirect, must be taken into the account. But this is impossible except to omniscience itself. Even in those cases which appear the most clear to us, there may be consequences of the most grave character utterly hidden from our view. In the instance just related, the captive was very certain, from some circumstances, that he directed his captors to a spot where there were no white men. The uncertainty of our knowledge of the circumstances of each case prevents then our defining any limits, within which deception shall be bounded. We can make no accurate distinctions, which will enable us to say, that it can be beneficially employed in one case, while in another it will be inexpedient. I have now finished the examination of the various considerations which have been suggested to my mind in relation to this subject. And I think that they settle the question as to the expediency of deception beyond all doubt. I think it perfectly evident, that the good, which may be done by deception in a _few_ cases, is almost as nothing, compared with the evil which it does in _many_ cases, when the prospect of its doing good was just as promising as it was in those in which it succeeded. And when we add to this the evil which would result from a _general_ adoption of a system of deception, the importance of a strict adherence to truth in our intercourse with the sick, even on the ground of expediency, becomes incalculably great. In the passage, which I quoted in the beginning of this article from Percival’s Medical Ethics, the writer makes, I conceive, a false issue on the question under consideration. He assumes that the injury, which results from a sacrifice of the truth for the good of the sick, comes upon him who practices the deception, and that in doing it, “he generously relinquishes every consideration referable only to himself.” But the considerations that I have presented show, that the injury is very far from being thus confined. Often the very person intended to be benefited is injured, perhaps deeply, in some cases even fatally. And then the indirect effects can not be estimated.