As an inch of rise of the mercury in the longer leg will cause an inch of descent in the shorter, the bores. being equal; the two surfaces will thus be 2 inches. apart ; and this alteration cannot be effected by a less pressure of the air than that which causes 2 inches of rise in the Torricellian tube. Hence the range of the scale is only half that of the common barometer. But this defect is compensated by converting the per pendicular motion of the mercuryinto a rotatory one, ., and exhibiting it on .a circular dial plate. For this purpose, a piece of ivory of a bell form is made to float on the surfa -e of the mercury in the shorter leg, having a silk thread fastened to its upper end, which, passing over a pulley, is stretched by a weight that is nearly a counterpoise. By this means the mo tion of the mercury is communicated to an index, which turns round a graduated circle, and thus the vertical range is enlarged at pleasure. The motion of the index in the wheel barometer has been render td more sensible by Mr Russel of Falkirk, by the addition of a little wheelwork.
Under the same figure we may refer to the syphon barometer ; for the form of the tube, and the motion of the mercury, are the same in both. This, at least in theory, is one of the most elegaat modifications of the Torricellian tube. It consists of the tube appli ed to a frame of wood, with the addition of a sliding bar of bress nearly of the same length. At the lower end of the bar is an index, which, in observing the mercury, is to be set opposite its surface in the short er leg : the upper end of the bar is divided into a !Tale of inches, whose zero is the index below. The scale is furnished with a sliding hand for pointing to the surface of the mercury in the longer tube, and also with a vernier for distinguishing the smaller divi sions. By this .simple contrivance, the sum of the motions of the mercury in both legs is correctly ex hibited, and measured at top.
The inclined or diagonal barometer (Fig. is another form of this instrument, for augmenting the scale. It differs from the common barometer in having the upper part of the tube, where the scale is applied, bent.at B, so that the range is on AB instead of a B. By this arrangement, when the stands at d in the Torricellian tube, it will stand at D in this instrument, D d being parallel to the horizon. Hence the scale, compared with that of the barometer, will be enlarged in the ratio of AB to a B ; a ratio which admits of unlimited increase, since AB. may be made indefinitely great with respect to a B.
Of all the methods which have been proposed for increasing the range of the barometrical scale, this seems to be the most susceptible of accuracy. The.
only objections.to which it is liable, are, the friction. arising from the increased column of mercury the sides of the tube, and the possibility of the mer cury separating into.detached portions during its de scent 'in the inclined part. The invention,of the in clined barometer is generally ascribed to Sir Samuel. Moreland; but this is doubtful. Derham, who has described it in the Phd. Trans, for 1698, No. 23G, only mentions that it had been communicated to him. by a friend. The invention itself is so obvious in principle, that it scarcely merits enquiry to whom it is due.
The rectangular barometer (Fig. 9:) consists of a tube ABC bent at right angles, and swelled out at AD, which includes the vertical range of the mer cury into a cylindrical cistern. The scale is reckon ed on the horizontal part. of the tube CB, and may be enlarged to an unlimited extent, by making the bore of the tube indefinitely small, in comparison of the cistern AD., The mercury is prevented from flowing out at C, by the pressure of the air acting upon its surface at E. When the weight of the air is increased, so as to produce an ascent of one inch of the mercury in the Torriccll:an tube, it will also pro duce an ascent of one inch in AD; but in order to do this, the mercury must be supplied from CB, and the space through which it moves from E towards B, will be to one inch, as the square of the diameter of AD to the square of the diameter of the bore of C B.
Another contrivance for enlarging the scale of. the barometer, is represented by Fig. 10. ABD is a bent tube of the common diameter, terminated at the upper extremity by the bulb A, in order that the as cents and descents of the mercury may be chiefly ter the leg BD. EG is an index moveable about F as a centre ; DC is a• float of ivory or glass, attached by a slender wire to the extremity E of EG, and which, by its ascent and descent on the surface of the mercury, communicates motion to the index. By this means the scale is enlarged in the ratio of FG to EF ; but the friction is considerable, so that little reliance can be placed on the degrees pointed' out by the index. It may be useful, however, merely to point out changes in the atmospherical pressure, where the real. amount of these is of little importance.
We shall conclude the account of the methods that have been proposed for enlarging the extent. of the barometrical scale, by a description of a barometer in vented for that purpose by Mr Rowning.