TWO:Machinery of transmission being generally a part of the fixed plant of an establishment, experiments cannot be made to institute comparisons, as in the case of machines; besides, there are special or local considerationssuch as noise, danger, freezing, and distanceto be taken into account, which prevent any rules of general application. Yet in every case it may be assumed that some particular plan of transmitting power is better than any other, and that plan can best be determined by studying, first, the principles of different kinds of mechanism and its adaptation to the special conditions that exist; and secondly, precedents or examples.The modern doctrine of evolution, while relying largely on the fertility of multiplied chances, is not obliged to assume such an enormous number of simultaneous coincidences as Epicurus. The ascription of certain definite attractions and repulsions to the ultimate particles of matter would alone restrict their possible modes of aggregation within comparatively narrow limits. Then, again, the world seems to have been built up by successive stages, at each of which some new force or combination of forces came into play, a firm basis having been already secured for whatever variations they were capable of producing. Thus the solar system is a state of equilibrium resulting from the action of two very simple forces, gravitation and heat. On the surface of the earth, cohesion and chemical affinity have been superadded. When a fresh equilibrium had resulted from their joint energy, the more complex conditions of life found free scope for their exercise. The transformations of living species were similarly effected by variation on variation. And, finally, in one species, the satisfaction of its animal wants set free those more refined impulses by which, after many experiments, civilisation has been built up. Obviously the total sum of adaptations necessary to constitute our actual world will have the probabilities of its occurrence enormously increased if we suppose the more general conditions to be established prior to, and in complete independence of, the less general, instead of limiting ourselves, like the ancient atomists, to one vast simultaneous shuffle of all the material and dynamical elements involved.
