With all this in mind, let us return to the question, “Are the spots, these gigantic areas of disturbance, comparable to whirlpools or to volcanoes?” It may seem unphilosophical to assume that they are one or the other, and in fact they may possibly be neither; but it is certain that the surface of the sun would soon cool from its enormous temperature, if it were not supplied with fresh heat, and it is almost certain that this heat is drawn from the interior. As M. Faye has pointed out,[1] there _must_ be a circulation up and down, the cooled products being carried within, heated and brought out again, or the sun would, however hot, grow cold outside; and, what is of interest to us, the earth would grow cold also, and we should all die. No one, I believe, who has studied the subject, will contradict the statement that if the sun’s surface were absolutely cut off from any heat supply from the interior, organic life in general upon the earth (and our own life in particular) would cease much within a month. This solar circulation, then, is of nearly as much consequence to us as that of our own bodies, if we but knew it; and now let us look at the spots again with this in mind. [1] To Mr. Herbert Spencer must be assigned the earliest suggestion of the necessity of such a circulation. [Illustration: FIG. 25.--TYPICAL ILLUSTRATION OF FAYE’S THEORY.] Fig. 21 shows a drawing by Father Secchi of a spot in 1854; and it is, if unexaggerated, quite the most remarkable case of distinct cyclonic action recorded. I say “if unexaggerated” because there is a strong tendency in most designers to select what is striking in a spot, and to emphasize that unduly, even when there is no conscious disposition to alter. Every one who sketches may see a similar unconscious tendency in himself or herself, shown in a disposition to draw all the mountains and hills too high,--a tendency on which Ruskin, I think, has remarked. In drawings of the sun there is a strong temptation to exaggerate these circular forms, and we must not forget this in making up the evidence. There is great need of caution, then, in receiving such representations; but there certainly are forms which seem to be clearly due to cyclonic action. They are usually scattered, however, through larger spots, and I have never, in all my study of the sun, seen one such complete type of the cyclone spot as that first given from Secchi. Instances where spots break up into numerous subdivisions by a process of “segmentation” under the apparent action of separate whirlwinds are much more common. I have noticed, as an apparent effect of this segmentation, what I may call the “honeycomb structure” from its appearance with low powers, but which with higher ones turns out to be made up of filamentary masses disposed in circular and ovoid curves, often apparently overlying one another, and frequently presenting a most curious resemblance to vegetable forms, though we appear to see