ON DRAWING, WRITING, AND MEASURING. 101 



a right line, besides that it is guided by the flatness of its lower surface. We 

 judge of the straightness of a line, by means of the well known property 

 of light, which moves only in right lines, so that if we look along the edge 

 of a ruler, we easily discover its irregularities ; and this may be done with 

 still greater accuracy,, if we look through a small hole made with a pin in a 

 card. Rulers of silver, bniss) or ivory, have a material advantage over those 

 of wood, as they are not liable to be spoilt by warping, A pen filled with 

 ink cannot be applied close to the edge of a ruler without inconvenience ; it 

 is therefore best, for diagrams which require great accuracy, to draAV the 

 lines first with a steel point, or a very hard black lead pencil, and to finish 

 them with ink if necessary. The paper should also be fixed on a drawing 

 board; and plates of lead or copper may be employed, instead of paper, for 

 very delicate purposes. The carpenter's chalk line is a useful instrument for 

 supplying the place of a very long ruler; it becomes straight when it is 

 stretched, because a right line is the shortest distance between any twa- 

 points.^ 



For drawing a circle of a gfven radius, we use compasses, with one pofnt 

 generally of metal, the other of various descriptions. Compasses are some- 

 times made with a spring, instead of a joint, and opened or shut by a screw: 

 sometimes a graduated arc is fixed in one leg, and passes through the other; 

 and when great accuracy is required, hair compasses may be employed, having 

 ajoint with a spring in one of the legs, which is bent a little by means of a 

 fine screw. Beam compasses are useful for drawing circles of larger radii : 

 they have also the advantage of being steadier than the common compasses,," 

 and of admitting readily the application of a graduated scale, so as to indi- 

 cate the measure of the radius of the circle which is described. Sometimes, 

 for drawing portions of very large circles, two wheels, differing a little in dia- 

 meter, are fixed on a common axis, and thus made to revolve round a point, 

 which is more or less distant, accordingly as the wheels are set at a greater or 

 less distance on the axis, the surface of the wheels tracing the circles on the 

 paper; or two rulers joined together, so as to form an angle, are made tojslide 

 against two points, or edges, projecting from a third ruler, so that the angular 

 point remains always in the arc of a circle. The same effect may be produced, 

 somewhat more commodiously, by means of a thin piece of elastic wood, 

 which is made to assume any required curvature, by the action of screws, appli^ 



