This is not their only canal connection. When the canals in the dark regions came to be discovered, each of them was found by me, as nearly as difficult observations would permit, to be associated with the caret upon its other side. Thus the lacing of the Mare Cimmerium and Sirenum used them as its reeving-points. Similarly those at the mouths of the Phison and Euphrates did duty likewise to the Maesolus and the Ion. In such manner the carets stood in dual relation to canals; subserving a purpose to the light-region canals on the one hand and to the dark-region ones on the other. In a way the caret, then, holds the same position toward the canals that do the spots in the light or dark regions. Like them it is a canal-distribution point. Unlike them, however, in shape it is triangular instead of round, and we are piqued to inquire to what cause it can owe its different contour. The answer seems to lie in the character of the locality, not simply in its complexion. For the spots in both the northern and the southern dark patches are as circular as those standing in the light, whether they lie in the centre or upon the edges of them. The edges of the northern patches, however, and the other sides of the southern ones do not present the clear-cut character of the northern coast of the diaphragm. Where they seem to be definitely bounded they are so by darker canals. This hints that their contours are not defined by antithesis of level, while that of the northern coast of the great diaphragm is. Difference of altitude is then concerned in their constitution; the canal system here falls to a lower level, and these triangular spots instead of round ones are the result. Topographic only, such explanation leads the way to a more teleologic one, and serves even on first acquaintance to stir curiosity to some satisfying cause.