Intelligence and Miscellaneous Articles. 77 



mometer has been heated to the highest degree of temperature 

 employed in the experiments, 355° for instance ; the position of 

 the zero has been observed, the thermometer submitted to any 

 series of operations ; and, on heating again to 355°, the original 

 position of the zero has been recovered ; it remains to raise this 

 depressed zero to its highest position. A thermometer can be kept 

 indefinitely at the ordinary temperature, or even be heated for 

 several -weeks to 100°, without producing the total reelevation of 

 the zero ; and the study in detail of the means which facilitate that 

 movement has revealed new facts of some importance in regard to 

 the theoretic views considered in some preceding communications. 

 I will briefly repeat the explanation which seems to account 

 best for all the movements of the fixed points of a thermometer. 

 The glass, softened during the process of blowing, retains inde- 

 finitely, at the ordinary temperature, a residual separation of its 

 particles, similar to that observed in an exaggerated degree in 

 Rupert's drops. On such glass being heated, greater mobility is 

 imparted to its particles, and the normal shrinking is induced (the 

 disappearance of the abnormal expansion) ; and this action of heat 

 is the more pronounced the more nearly the temperature is 

 approached which produced the original separation. When a 

 thermometer is in operation, the shrinkage causes a permanent 

 raising of the zero-point, which may reach the considerable figure of 

 26°. It is evident that if a separation of the same kind, although 

 inferior in amplitude, can be produced at will, can be made to 

 persist for an indefinite time at the ordinary temperature, and to 

 disappear by heating, this manner of understanding the phe- 

 nomenon will be notably confirmed. Experiment realizes this 

 prevision. In fact it is sufficient to know the true depression of 

 the zero produced by boiling the mercury in a thermometer to 

 ascertain that it persists indefinitely at the ordinary temperature. 

 The depression is about 2° for French glass containing oxide of 

 lead, the reelevation at the end of the first six months is from 

 o, 4 to o, 5 ; the subsequent elevation during from five to ten 

 years rarely exceeds o- 5 ; and after ten years the zero remains 

 stationary, leaving half of the depression subsisting, of which 

 the total removal, equalling an additional degree, can be effected 

 by suitable means. This conclusion has been confirmed by numerous 

 experiments on the depressions produced at different temperatures. 

 It is especially by heating to very high temperatures that de- 

 pressions are produced which do not disappear. This phenomenon 

 resembles that of the permanent expansion of a body when the 

 limit of its elasticity has been passed. 



The total reelevation after a depression of the zero, is accom- 

 plished more rapidly at high temperatures. Let us compare the 

 periods of time when the intervals of temperature remain nearly 

 equal. Five thermometers, after being heated for 24 hours to 

 306°, were kept at 218° until the position of the zero had become 

 constant ; for which four days were required. Afterwards the 

 zero depressed at 218° was reelevated by heating for 18 days to 

 100°. It requires from six months to two years for the total 



