Now, in both these respects the supposed world-machine differs absolutely from the real machines upon analogy with which the mechanical theory is in the last resort based. Every real machine is, to begin with, the incarnation of the internal purpose of a sentient being. It is something which has been fashioned for the express object of attaining a certain result, and the more perfect its structure the greater is the impossibility of understanding the principle of construction without comprehension of the result it is devised to effect. Why the various parts have precisely the shape, size, strength, and other qualities they have, you can only tell when you know what is the work the maker of the machine intended it to do. In so far as this is not the case, and the structure of the machine can be explained, _apart_ from its specific purpose, by consideration of the properties of the material, the patterns of construction consecrated by tradition, and so forth, it must be regarded as an imperfect realisation of its type. In a perfect machine the character and behaviour of every part would be absolutely determined by the demands made on that part by the purpose to be fulfilled by the working of the whole; our inability ever to produce such a perfect mechanical structure causes all our actual machines to be imperfect and inadequate representations of the ideal we have before us in their construction. Thus a true machine, so far from being purposeless, is a typical embodiment of conscious purpose. It is true that the machine, once set going, will continue to work according to the lines embodied in its construction irrespective of the adequacy with which they effect the realisation of the maker’s purpose. A watch, once wound up, will continue to go, though the indication of the lapse of time may, under fresh circumstances, cease to meet the interests of its maker or owner; and again, if the construction of the watch was faulty, it will not properly execute the purpose for which it was made. The machine has in itself no power of fresh purposive adaptability by which to modify the purpose it reflects, or to remedy an initial defect in its execution. But this merely shows that the purpose exhibited in the machine’s construction originated outside the machine itself, and that the originator had not the power to carry out his purpose with complete consistency. It does not in the least detract from the essentially teleological and purposive character of the machine _quà_ machine.