Their precise width is not precisable. They show no measurable breadth and their size, therefore, admits for certain only of an outside limit. They cannot be wider than a determinable maximum, but they may be much less than this. The sole method of estimating their width is by comparison of effect with a wire of known caliber at a known distance. For this purpose a telegraph wire was stretched against the sky at Flagstaff, and the observers, going back upon the mesa, observed and recorded its appearance as their stations grew remote. It proved surprising at what great distances a slender wire could be made out when thus projected against the sky. The wire in the experiment was but 0.0726 of an inch in diameter and yet could be seen with certainty at a distance of 1800 feet, at which point its diameter subtended only 0.69 of a second of arc. How small this quantity is may be appreciated from its taking more than ninety such lines laid side by side to make a width divisible by the eye. Such slenderness at the then distance of Mars would correspond, under the magnification commonly used, only to three quarters of a mile. Theoretically, then, a line three quarters of a mile wide there should be visible to us. Practically, both light and definition is lost in the telescope, and it would be nearer the mark to consider in such case two miles as the limit of the perceptible. With the planet nearer than this, as is often the case, the width which could be seen would be proportionally lessened. Perhaps we shall not be far astray if we put one mile as the limiting width which could be perceived on Mars at present, with distance at its least and definition at its best. That so minute a quantity should be visible at all is due to the line having a sensible length and by summation of sensations causing to rise into consciousness what would otherwise be lost. A stimulus too feeble to produce an effect upon a single retinal rod becomes recognizable when many in a row are similarly excited.