78 Intelligence and Miscellaneous Articles. 



reelevation of the zero depressed at 100° when the thermometer is 

 lift at the ordinary temperature. The termination of the reele- 

 vation being the limit of the abatement of a movement, it is 

 evident that it cannot be stated with exact precision ; and it is 

 especially by comparing the course at its different phases for each 

 interval that one gets an exact idea of the variation of rapidity of 

 the reelevation according to the temperature at which one works. 

 "When the intervals are smaller the reelevation is still more rapid : 

 thus, from 0*8 to 0-9 of the total reelevation of the zero depressed 

 at 100° can be produced by heating the thermometer during three 

 days to 80°, four days to 60°, and five days to 40°. 



It will be seen that, the greater the interval between the tem- 

 perature which has produced a depression and that at which the 

 thermometer is maintained to accomplish the reelevation, the 

 slower is the movement ; and it may not be complete if the 

 interval notably exceeds 100°. These data have been utilized in 

 the following manner to prepare some thermometers for the study 

 of the real depressions of the zero. In order to cause the 

 abnormal tension produced during the blowing of the glass to 

 disappear, the instruments nos. 1 and 5 were heated 11 days to 

 3-55° : nos. 13 and 15 were heated three years to 206° and 

 nos. 31 and 32, before being filled with mercury, were heated 1 00 

 hours to 440""', and cooled as gradually as possible during 100 hours ; 

 and finally all were heated for one day to 306° and the zero-points 

 observed. In order to reelevate completely the zero from its 

 position depressed by the heating to 300°, all the thermometers 

 were heated and kept for four days at 218°, eighteen days at 100°, 

 five days at 80", seven days at 60°, six days at 4u°, nine days at 

 20°, three days at 10°, and. two days at zero. 



Here only one series of experiments can be cited ; they are 

 summed up in the following Table : the first column contains the 

 numbers of the thermometers ; those which follow give the de- 

 pressions corresponding to the temperatures indicated at the top 

 of the columns. 



40° 60° 80° 100° 160° 218° 260° 306° 



1 0-00 0-06 019 0-31 0-74 1-12 1-33 1-63 249 



5 0-04 0-08 0-18 0-29 0-56 0-76 0-91 1-14 1*51 



13 0-02 0-03 0-17 0-31 0-69 0-87 1-09 1-30 245 



15 0-01 0-05 0-18 0-31 075 0-97 142 1-40 2-05 



31 0-02 0-06 0-22 0-37 0-84 1-15 1-46 1-77 



32 0-28 0-69 0-98 1-21 1-56 2-06 



Neglecting thermometer no. 5, which is of German soda glass, the 

 numbers for all the others, which are of glass containing oxide of 

 lead, are sufficiently concordant to permit the table to be employed 

 for estimating, with but a very slight error, the depressions which 

 will be produced by heating thermometers manufactured at Paris. 

 The relation between the temperatures and the depressions might 

 be expressed by a general formula ; but simple interpolation 

 suffices for practical needs. — Comptes Rendus de VAcademie des 

 Sciences, May 8, 1882, t. xciv. pp. 1298-1301. 



