124 Description of Sletches of a Self-Registering Barometer. [Jan. 



The indications may be magnified to any extent, by the relative pro- 

 portion of area of float in C and of the diameter of tube, as described in 

 a former communication. It is obvious that this insti-ument is not in- 

 tended for the ordinary use of the traveller ; nevertheless it is desira- 

 ble that it be made as portable as the principle of its construction will 

 admit of. With this object in view the individual parts are made to screw 

 together as will be observed in the sketch. The next consideration is 

 how to dispose of the tube, so that when detaclied from the cistern no 

 air will enter. This object is easily accomplished by screwing on a 

 small cistern, as represented {fg.^) on the principle of the Englefield 

 barometer, after the tube has been inclined to fill it with mercury, but 

 before the end is brought out of the large cistern D. The whole ought 

 to be packed in a case constructed to receive the individual parts. 



Fig. 2 represents that variety of the instrument in which the cistern 

 is 7)ioveable and the tube fixed. The whole is attached to one pillar A ; 

 on the top of which at B is the expanded top of the tube : the narrow 

 part descends inside of the pillar, comes out at C, and then dips in the 

 cistern D. E is a cistern of similar construction to those in fg. 1. In 

 this is a float which, through the medium of rods as represented, buoys 

 up the cistern D. Between the two cisterns the register apparatus F 

 is placed. GG are friction rollers and guides. The pedestal H (of 

 which a top view is given in fg. 4) is furnished with screws and levels 

 as in fg. 1. 



The mode of operation of this variety will readily be comprehend- 

 ed. Suppose an increase of atmospheric pressure to occur : mercury 

 leaves the cistern D to rise into B, and being thus lightened, its float in 

 E rises and carries the cistern with it. The register apparatus records 

 the amount of the movement. 



Fig. 5 represents the apparatus for reading oflf the pencil trace on 

 the rollers. AA is a spindle, on which at u is a disc, and at & a circular 

 nut which is screwed on after the roller has been put on the spindle, to 

 keep the latter firm up against the disc. The spindle goes through C, 

 in which it is cone-sliaped like the ?pindlc of a lathe. After the roller 

 and spindle have been placed in proper position, the pointed screw D 

 keeps them so. B. a circular nut, is then put on, by means of which 

 tlic roller is made to revolve to bring the hour lines successively to the 

 vernier V. The vernier is moved by a tangent screw E, as in the com- 

 mon barometer, which runs along the back of the bar to which it is at- 



