346 PHYSICS. 



Corva has also suggested the use of a magnet brought near to the 

 forks, giving Lissajous's figures by projection, for the purpose of varying 

 at will the differences of phase. He uses Mercadier's diapasons mounted 

 on Duboscq's universal support. To vary the period of one of these 

 he employs a supplementarj' electromagnet placed between the arms of 

 the fork, and adjustable by a screw in a plane perpendicular to that of 

 the branches. {J. Phys., June, 1881, x, 253.) 



Koenig has devised an ingenious apparatus for determining with great 

 precision the vibrations of a normal fork. A clock-work movement of 

 great nicety so acts ou a fork making 128 vibrations as to keep it in 

 vibration, while at the same time the fork acts as the escapement of the 

 clock. By comparison with a chronometer the rate of the clock, and 

 hence the error of the fork, is ascertained. If, for example, the error of 

 the clock is i 1 second per hour, then the error of the fork is ± aViru? or 

 0.0355. One of the arms of the fork carries a microscope and the other 

 a steel mirror as counterweight. In this way the movement of the fork 

 can be compared with that of any other vibrating body, by the optical 

 method. The apparatus is regulated for 20° C, and the variation from 19° 

 to 20"^ causes a change of 0.0143 vibration. Within a range of 5° to 30° 

 the mean variation per degree centigrade is 0.059 to 0.054 vibration. 

 Koenig's C3 fork makes at 20° 512.3548 vibrations, and at 2G.2o 512 

 vibrations. The normal French fork, correct at 15°, varies 0.0972 vibra- 

 tion per degree centigrade. The normal fork of the conservatory makes 

 870.9 vibrations at 15"^, and is correct at 870 vibrations at 27.2°. ( Wied. 

 Ann, 1880, ix, 394; J. Phys., May 1881, x, 214.) 



HEAT. 



1. Thermometry and production of heat. 



Russell has given a detailed account of the method of Neumann for 

 calibrating thermometers, which, he says, has very considerable advan- 

 tfiges over the methods in common use, and which combines the great- 

 est simplicity, elegance, and exactness. The columns measured should 

 be as nearly as possible equal in length to a whole number of intervals 

 between the points for which the corrections are required, t. e., for 

 every ten degrees the required columns must be about 10, 20, 30 degrees, 

 etc. The columns obtained are to be measured with their lower ends 

 near all the points for which corrections are required. The method of 

 detaching a suitable column and the details of the method ofcalibrntion 

 are given at length in Eussell's paper. {Am. J. Sci., May, 1881, III, 

 xxi, 373.) 



Pernet has examined thermometers to ascertain whether the distance 

 between the boiling point and the freezing point remains constant at 

 all different stages of secular alteration in volume of the bulbs. He 

 finds that it does remain constant, provided the freezing point is deter- 

 mined immediately after the boiling point. On the other hand, if the 



