Really, it is the theory of necessity which violates the uniformity of nature, for it requires us to believe that provided certain of the conditions (viz. all the circumstances except the will) are the same, then the result must be the same, no matter how much the remaining condition (the will) changes. We may, indeed, evade this conclusion by simply denying that the will is one of the conditions of its own modifications, and we may say that the wax contributes nothing to the form which it takes on when impressed by the seal. The truth is that if the will or the wax appears in the result, it must have been present and active as one of the conditions: it contributes to its own determination, and is in part self-determining. If it be in accordance with the uniformity of nature and with our experience of what actually happens, that the circumstances should be the same and the will different on two different occasions, then the theory of necessity breaks down: if we can will and act differently under the same circumstances, we have all the freedom we want. But if--all the circumstances, save the will, being the same--the resulting modification or determination of the will is different, then the difference of result must be due to some difference in the conditions; all the conditions save one were _ex hypothesi_ the same; the remaining condition, therefore, viz. the will, must have changed. What caused the change? Not the circumstances: one attempt to explode a barrel of gunpowder may resemble another in all the circumstances save one (the dampness of the powder), but the circumstances which remain the same (application of the spark, etc.) are not the cause of the difference in the remaining condition. If, then, we do as a matter of fact at times under the same circumstances will different things, and if the circumstances are not the cause of the change of will, then the will changes itself, _i.e._ is self-determining, self-modifying. And, as we all know from experience, it determines itself at the moment of choice, not before. Until all the conditions requisite for the effect are combined, neither physical nor mental science requires us to assume that they must combine in this particular way--that the light must be applied to the powder because an explosion will take place if it is applied, that the motive of gain must be adopted because it will be gratified if it is obeyed.