NATURAL PHILOSOPHY. 183 



vessel in water, and this ves.se! carries a rod with a pencil, which 

 marks the force of the wind on a cylinder which is moved by clock- 

 work. There is another cylinder which is moved by a connection 

 with a wheel in the upright tube, which gives the direction of the 

 wind ; so that the time, force, and direction of the wind are given by 

 the instrument. 



HYDROBAROMETER. 



A NEW instrument with the above name has been constructed by 

 M. Walferdin, of Paris, for ascertaining the depth and temperature of 

 the sea. M. Walferdin shows that the bulb of an ordinary thermome- 

 ter, formed by being blown on the extremity of a tube, is compressible 

 between the finger and thumb so far as to cause the mercury to ascend 

 appreciably, and that consequently the ordinary arrangement for 

 ascertaining the temperature at depths is faulty. At the artesian well 

 of Grennelle, before the jetting of the waters, six maxima thermome- 

 ters (" thermometers a maxima a deversement " of M. Walferdin) 

 protected from pressure and sunk to a depth of 505 meters, indicated 

 as the mean temperature 26 43 C. A seventh thermometer at the 

 same time was sunk, unprotected from the pressure, and it indicated 

 39 50 C. A pressure of 50.5 atmospheres had thus increased the 

 result by 13.07, equivalent as the arbitrary scale of the instru- 

 ment to | an atmosphere. This experiment was performed by 

 Arago and M. Walferdin. 



It hence follows that a minima thermometer, protected from the 

 action of pressure, will indicate the true temperature of depth ; while a 

 maxima inverting thermometer exposed to the pressure, and sunk 

 with the other, may indicate the pressure of the sea. Such is the 

 principle upon which M. Walferdin constructs his new instrument 

 which he calls a liydrobarbmeter. 



The best means of avoiding the effects of pressure on a thermometer, 

 according to careful trials, is by enclosing the instrument in tubes of 

 glass, more or less thick according to the pressure they will have to 

 endure, and hermetically sealing the tubes. 



MEAN DENSITY OF THE EARTH. 



REICH has published the results of a second series of experiments 

 on the mean density of the earth, undertaken principally with a view 

 of determining the cause of the difference between his own previous 

 results and those of Baily. (It will be remembered that Cavendish 

 found 5.48, Reich 5.45, Baily 5.66.) Three series of experiments, 

 making in all 70 single determinations, were made. The first gave as 

 mean 5.5712 with a probable error of 0.0113, the second gave 5.6173 

 with a probable error of 0.0181, the third gave 5.5910 with a proba- 

 ble error of 0.0149. The general result of these three series is 

 5.5832 with a probable error of 0.0149. The third series was made 

 with a bifilar suspension wire, but the results did not correspond better 



