July 20, 1882] 



NATURE 



28- 



very slight taper, but in order that the column may not break 

 the bore must be so small that friction and capil'arity render 

 correct graduation impossible. Whiting (Pogg. Ann. cxvii., 656, 

 1862) proposed in place of that to use two tubes of different 

 diameters joined together with, at lower surface of mercury, an 

 ivory plate with a little glass bulb at ached above it. In the 

 horizontal barometer of Bernoulli and Cassini the longer range is 

 obtained by enlarging the tube at the upper level of the mercury 

 and replacing the cistern by a tube of fine bore bent at a right 

 angle. In Sir Samuel Moreland's diagonal barometer the top 

 part of the tube is bent more or less from the perpendicular. It 

 is said to have been invented by Derham, as he refers to his 

 "former communication about a crooked tube." The instru- 

 ment at South Kensington bears date 1750, and was made by 

 Watkins and Smith, London ; that at Peel Park, Salford, is by 

 T. Whitehurst, Derby, 1772; there is also one in the museum 

 at King's College. Hicks [Prov. Prat. No. "40, 1S62) proposed 

 to increase the range by terminating the tube at the top in a coil. 

 The maximum registering barometer of Traill is the same as 

 the diagonal, but with the addition of a short steel rod in the 

 tube above the mercury ; his minimum is on the same principle 

 as the horizontal barometer, but the bottom open tube forms a 

 smaller angle than a right angle with the vertical tube. The 

 steel rod is placed in this open tube above the mercury. How- 

 son's (Pal. No. 1616, 1S61) is on an entirely different principle, 

 as in it the cistern is supported by the floating power of a sealed 

 gla»s tube which is inserted in the mercurial column, and is 

 attached to the bottom of the cistern. This was reproduced by 

 Vitli (Les Mondis, iii., 25, 1S63) two years later. 



As the absence of air above the mercury renders the tubes 

 very liable to fracture, from the bumping of the column against 

 the top, when barometers are carried or moved, the means of 

 rendering the instruments more or less portable has occupied the 

 attention of several. Nairne long ago overcame this difficulty 

 by making the lower half or two-thirds of the tube of very fine 

 bore. Passemente (175S) adopted the expedient of twisting the 

 middle of the tube so as to form tw o or three coils of a flattened 

 spiral. Spry (Phil. Trans, lv., S3, 1765), who unfortunately 

 gave no illustration, wrote, "The small bowl at the top with 

 beads therein, render it far less liable to break by the mercury's 

 ascent, the bowl giving it an immediate expanse from the column, 

 and the beads counteracting in force as so many springs." 

 Uzielli (Dent. Chan. Gcs. Per. v., 1055, 1S72) makes a some- 

 what similar proposal: "Above Sco mm. a glass valve is in- 

 serted in the tube, so that by inclining the tube the mercury 

 rises above the valve ; then, on bringing the tube upright again, 

 the air is above the valve and the valve is sealed by mercury." 



Recorders. — In Fontana's arrangement the barometer tube 

 and short open tube are cemented into elbows at either end 

 of a short horizontal tube (in this there is a stopcock). A 

 float on the surface of the mercury in the open tube trans- 

 mits its motion to a small section of a horizontal cylinder 

 about 700 mm. in diameter, and covered with paper. Every 

 hour an impression is made on this paper by a steel point moved 

 by a clock. Kreils used a siphon barometer, and connected the 

 float with the shorter arm of an unequal lever. The longer end 

 of the lever carried a pencil, which, being struck every five minutes 

 by a hammer moved by a clock, made a dot on a sheet of paper 

 fixed to a frame drawn horizontally in front of it. Negretti and 

 Zambia attach the float by means of cords attached to a balance 

 with unequal arms to a pencil fastened to a square collar. This 

 collar moves freely up and down a vertical rod of square cross-sec- 

 tion ; the rod is near the vertical cylinder round which the paper is 

 wrapped, and the same clockwork which causes the cylinder to 

 revolve moves the rod (at regular intervals) so as to bring the 

 point of the pencil against the paper. Keith (Encyc. Metrop. 

 1S45) attaches the recorder to the float rigidly by means of a 

 thin steel rod ; but he obtains nearly the whole of the move- 

 ment of the mercurial column in the open limb by attaching 

 to the upper limb a horizontal tube of large diameter. He thus 

 makes the movement of the upper surface of the mercury scarcely 

 perceptible. Redier (Symons' Met. Mag. x., 33, 1S75) connects 

 one pole of a battery to the float, and the other pole to a metal 

 point which is lowered at regular intervals to make contact. 

 The same clockwork which lowers the point draws a pencil 

 along (but not touching) the paper which is wrapped round a 

 horizontal cylinder. At the instant at which electrical contact is 

 made the pencil marks a dot on the paper; it is then drawn 

 back, and remains at rest for a certain time, when the operation 

 is repeated. By this arrangement the ratio betw een the move- 



ment of the pencil and that of the float can be increased to any 

 extent without offering any resistance to the movements of the 

 latter. The same principle is employed by Hough ; but as his 

 instrument is a complete meteorograph, the mechanical details 

 are varied in order that the movements of the barometer may 

 be recorded on the same paper as those of the other instru- 

 ments. Theorell's is a very similar instrument ; but, by means of 

 type, the record is made in printed numerals. It is not easy to 

 decide under what heading Russell's instrument should be de- 

 scribed, but this seems its most appropriate place. The baro- 

 meter tube is fixed, but the cistern (which is a small one) floats 

 in a vessel of mercury. The pen is attached to a rectangular 

 framework which is drawn backwards and forwards once a 

 minute in front of the paper. On electrical contact being made 

 between a lever attached to the cistern and the side of a wire 

 triangle attached to the pen-frame, the pen is pressed against the 

 paper, and thus the position of the cistern is recorded. 



Photographic Recorders. — In Brooke's (1S46) apparatus a lever 

 with arms of very unequal length has its short arm attached to a 

 float, w hich rests on the mercury in the lower end of a large 

 syphon barometer. The long arm carries a screen with a small 

 hole in it ; through this hole the light from a lamp produces a 

 mark on a sensitive paper which is wound round a verticil 

 cylinder moved by clockwork. Ronalds (Brit. Assoc. Rep. 346, 

 1S51), 1S47, °f whose apparatus in the South Kensington 

 Museum the only part visible is the case which, for aught one 

 knows, may contain nothing, made the surface of the mercury 

 trace its own line without the intervention of a movable screer. 

 A lamp is placed behind the barometer tube and a lens is so 

 adjusted that the surface of the mercury may throw its image on 

 a sheet of prepared paper or a daguerrotype plate, which is 

 moved horizontally by clockwork. At the same time correction 

 is made for temperature by means of the metal rods which sup- 

 port the cistern of the barometer. In the improved form (Rep. 

 Met. Comrn. Roy. Soc. 40, 1867) of the instrument the sensitive 

 paper is wound on a cylinder driven by clockwork, and the time 

 is recorded by a stop which intercepts the marking for four 

 minutes every two hours. The temperature-compensating appa- 

 ratus is attached to the vertical slit at the barometer, so that an 

 alteration in the temperature is indicated by a variation in the 

 base line on the sensitive paper. 



Volpicelli (Compt. rout. lxx. 334, 1870 ; Les Maudes, xxii. 365, 

 1S70) constructed a barometer of which the following are the 

 main details : — The millimetre scale is on a sheet of glass, and 

 is photographed with the barometric variations. A solution of 

 alum is interposed between the lamp and the mercurial column 

 to intercept the heat-rays. It is arranged to allow the barometer 

 to be read off without moving any of its parts. The interior of 

 the casing is freely ventilated, and the clock which moves the 

 paper makes a mark every hour, so that the record is 

 uninterrupted. 



Balance Barometers. — The Steelyard barometer (Moreland) is 

 one of the oldest forms of the above. The tube is suspended to 

 the shorter arm, and is balanced by the longer arm, at the end 

 of which is the pointer, w hich moves over a graduated arc. The 

 cistern is but very little wider than the tube ; thus when the 

 atmospheric pressure increase-', the pointer rises, and vice versa. 

 The static barometer (Magellan) is very similar, except that the 

 balance has arms of equal length, and the tube is balanced by a 

 weight ; the pointer is attached under the beam, and the ex- 

 tremity of it moves to and fro along a horizontal scale. Wild, 

 -in his recording-barometer, uses a tube with an enlarged upper 

 extremity, so that the variations in weight produced by the 

 alterations in the pressure of the air may be considerable. A 

 pencil at the end of the pointer, which is fixed below the beam, 

 records on a horizontal cylinder. The balance has a bent beam, 

 the arm from which the tube is suspended being horizontal, and 

 the arm to which the counterpoise is rigidly attached descending 

 obliquely. Secchi (1S67) used a slightly-bent beam with arms of 

 equal length ; the pointer which descended from the beam termi- 

 nated at its lower extremity in a hinge, to which was attached 

 one end of the horizontal rod, which carried the recording- 

 pencil ; parallel motion was obtained by a rod to which the other 

 end of the horizontal rod was hinged. The record was made on 

 a flat descending sheet of paper. Brassart (1872) did not in any 

 way alter the principle of the instrument, but arranged it in a 

 slightly different manner, so as to render more compact the 

 metec rograph of which it formed a part. Schreiber makes his 

 instrument record on a vertical revolving cylinder by means of a 

 pencil attached to a rod suspended from the other arm of the 



