It may be objected to our explanation of the formation of these bright streaks, that so far as our means of observation avail us, we fail to detect any shadows from them or from such marginal edges as might be expected to result from a sideway-spreading outflow of lava from the cracks which afforded it exit in the manner described. Were the edges of these streaks terminated by cliff-like or craggy margins of such height as 30 or 40 feet, we might just be able at low angles of illumination and under the most favourable circumstances of vision, to detect some slight appearance of shadows; but so far as we are aware, no such shadows have been observed. We are led to suppose that the impossibility of detecting them is due not to their absence but to the height of the margins being so moderate as not to cast any cognizable shadow, inasmuch as an abrupt craggy margin of 10 or 15 feet high would, under even the most favourable circumstances, fail to render such visible to us. Reference to our ideal section of one of these bright streaks (Fig. 45), will show how thin their edges may be in relation to their spreading width. The absence of cognizable shadows from the bright streaks has led some observers to conclude that they have no elevation above the surface over which they traverse, and it has therefore been suggested that their existence is due to possible vapours which may have issued through the cracks, and condensed in some sublimated or pulverulent form along their courses, the condensed vapours in question forming a surface of high reflective properties. That metallic or mineral substances of some kinds do deposit on condensation very white powders, or sublimates, we are quite ready to admit, and such explanation of the high luminosity of the bright streaks, and of the craters situated at the foci or centres of their divergence is by no means improbable, so far as concerns their mere brightness. But as we invariably find a crater occupying the centre of divergence, and such craters are possessed of all the characteristic features and details which establish their true volcanic nature as the results of energetic extrusions of lava and scoria, we cannot resist the conclusion that the material of the crater, and that of the bright streaks diverging from it, are not only of a common origin, but are so far identical that the only difference in the structure of the one as compared with the other is due to the more copious egress of the extruded or erupted matter in the case of the crater, while the restricted outflow or ejection of the matter up through the cracks would cause its dispersion to be so comparatively gentle as to flood the sides of the cracks and spread in a thin sheet more or less sideways simultaneously along their courses. There are indeed evidences in the wider of the bright streaks of their being the result of the outflow of lava through _systems of cracks_ running parallel to each other, the