Nay, an eddy of wind will not stay quiet for three minutes, as that cloud did to be drawn; as all the others did, each in his place. You see there is perfect harmony among the curves. They all flow into each other as the currents of a stream do. If you throw dust that will float on the surface of a slow river, it will arrange itself in lines somewhat like these. To a certain extent, indeed, it is true that there are gentle currents of change in the atmosphere, which move slowly enough to permit in the clouds that follow them some appearance of stability. But how to obtain change so complex in an infinite number of consecutive spaces;--fifty thousand separate groups of current in half of a morning sky, with quiet invisible vapor between, or none--and yet all obedient to one ruling law, gone forth through their companies;--each marshalled to their white standards, in great unity of warlike march, unarrested, unconfused? "One shall not thrust another, they shall walk every one in his own path." [Illustration: FIG. 81.] § 8. These questions occur, at first sight, respecting every group of cirrus cloud. Whatever the form may be, whether branched, as in this instance, or merely rippled, or thrown into shield-like segments, as in Fig. 81--a frequent arrangement--there is still the same difficulty in accounting satisfactorily for the individual forces which regulate the similar shape of each mass, while all are moved by a general force that has apparently no influence on the divided structure. Thus the mass of clouds disposed as in Fig. 81, will probably move, mutually, in the direction of the arrow; that is to say, sideways, as far as their separate curvature is concerned. I suppose it probable that as the science of electricity is more perfectly systematized, the explanation of many circumstances of cloud-form will be rendered by it. At present I see no use in troubling the reader or myself with conjectures which a year's progress in science might either effectively contradict or supersede. All that I want is, that we should have our questions ready to put clearly to the electricians when the electricians are ready to answer us. § 9. It is possible that some of the loveliest conditions of these parallel clouds may be owing to a structure which I forgot to explain, when it occurred in rocks, in the course of the last volume. [Illustration: FIG. 82.] When they are finely stratified, and their surfaces abraded by broad, shallow furrows, the edges of the beds, of course, are thrown into undulations, and at some distance, where the furrows disappear, the surface looks as if the rock had flowed over it in successive waves. Such a condition is seen on the left at the top in Fig. 17, in Vol. IV. Supposing a series of beds of vapor cut across by a straight sloping current of air, and so placed as to catch the light on their edges, we should have a series of curved lights, looking like independent clouds.