The sense of touch confines its operations to a very narrow sphere around us, but those of sight extend far beyond; this sense is therefore liable to be mistaken. With a single glance a man takes in half his own horizon, and in these myriad impressions, and judgments resulting from them, how is it credible that there should be no mistakes? Sight, therefore, is the most defective of all our senses, precisely because it is most far-reaching, and because its operations, by far preceding all others, are too immediate and too vast to receive correction from them. Besides, the very illusions of perspective are needed to make us understand extension, and to help us in comparing its parts. If there were no false appearances, we could see nothing at a distance; if there were no gradations in size, we could form no estimate of distance, or rather there would be no distance at all. If of two trees the one a hundred paces away seemed as large and distinct as the other, ten paces distant, we should place them side by side. If we saw all objects in their true dimensions, we should see no space whatever; everything would appear to be directly beneath our eye. For judging of the size and distance of objects, sight has only one measure, and that is the angle they form with our eye. As this is the simple effect of a compound cause, the judgment we form from it leaves each particular case undecided or is necessarily imperfect. For how can I by the sight alone tell whether the angle which makes one object appear smaller than another is caused by the really lesser magnitude of the object or by its greater distance from me? An opposite method must therefore be pursued. Instead of relying on one sensation only, we must repeat it, verify it by others, subordinate sight to touch, repressing the impetuosity of the first by the steady, even pace of the second. For lack of this caution we measure very inaccurately by the eye, in determining height, length, depth, and distance. That this is not due to organic defect, but to careless use, is proved by the fact that engineers, surveyors, architects, masons, and painters generally have a far more accurate eye than we, and estimate measures of extension more correctly. Their business gives them experience that we neglect to acquire, and thus they correct the ambiguity of the angle by means of appearances associated with it, which enable them to determine more exactly the relation of the two things producing the angle.