These several counts become more expressive of design the farther one looks into them. Straightness upon a sphere means the following of an arc of a great circle. The lines, then, are arcs of great circles. Now, the great circle course is the shortest distance connecting two given points. The canals of Mars, then, practice this economy; they connect their terminals by the shortest, that is, other things equal, by the quickest and least wasteful path. Their preserving a uniform width throughout this distance is an equally unnatural feature for any natural action to exhibit, but a perfectly natural one for an unnatural agent. For means of communication for whatever cause would probably be fashioned of like countenance throughout. Their extreme tenuity is a third trait pointing to artificiality; inasmuch as the narrower they are, the more probable is their construction by local intelligence. Even more inexplicable, except from intent, is their dual character. For them to parallel one another like the twin rails of a railway track, seems quite beyond the powers of natural causation. Enigmatic, indeed, from a natural standpoint, they cease to be so enigmatic viewed from an artificial one; and this the more by reason of what has lately been learnt of the character of their distribution. That they are found most plentifully near the equator, where the latitudinal girth is greatest, and thence diminish in numbers to about latitude 60°, where they disappear,—and this not relatively to the amount of surface but actually,—is very significant. It is quite incapable of natural explanation, and can only be accounted for on some theory of design such as lines of communication, or canals conducting water down the latitudes for distribution. So that this distribution of the doubles is in keeping with the law of development disclosed by the canals _en masse_. Channels and return-channels the two lines of the pair may be, but about this we can at present posit nothing. The relation may be of still greater complexity, and we must carefully distinguish between surmise and deduction. The position of the canals, with regard to the main features of the disk, has a cogency of its own, an argument from time. The places from which the lines start and to which they go are such as to imply a dependence of the latter upon the former chronologically. The lines are logically superposed upon the natural features; not as if they had grown there, but as if they had been placed there for topographic cause. Those termini are used which we should ourselves select for stations of intercommunication. For the lines not only leave important geodetic points, but they travel directly to equally salient ones.