392 Intelligence and Miscellaneous Articles. 



In order to make with accuracy the correction due to this space, 

 we join to the capillary tube another tube of exactly the same 

 length and the same diameter as the thermometer-stem. This tube, 

 which we have named compensator, is closed at one end; it is 

 attached by means of mastic to a threeway cock, permitting it to 

 be put in communication successively with the outer air and with 

 the Sprengel pump. We close the cock of the compensator at the 

 same time as that of the reservoir of the thermometer ; and after 

 the exhaustion of the nitrogen contained in the latter, following 

 the same order of operations we determine the volume, reduced to 

 0° and 760 millims., of the gas contained iu the compensator, and 

 consequently the volume of the prejudicial space. Deducting this 

 volume from that of the gas extracted from the thermometric reser- 

 voir, we obtain exactly the quantity of nitrogen which remained in 

 the thermometer at the temperature reached. With all these 

 numbers, introduced into a very simple formula, 



1 + ax 760 ' 



we have all that is necessary for calculating, with a high degree of 

 accuracy, the temperature to be determined. 



We will only remark that the determination of the temperature 

 and pressure of the gases successively collected in the graduated 

 tube must be made with the greatest exactness. All the precau- 

 tions necessary for determining these numbers should be taken 

 with minute precision, following the most exact methods indicated 

 by E-egnault. A mode of connexion must also be adopted, between 

 the thermometer and the compensator on the one hand, and the 

 Sprengel pump on the other, such that the volume of the prejudi- 

 cial space in the thermometer shall be precisely equal to the capa- 

 city of the compensator. There are many means of satisfying this 

 condition ; and it is needless to describe them here. 



A drawing of the apparatus, together with the arrangements we 

 have adopted, will be given in a memoir which we shall publish in 

 the Annates de VEcole Normale superieuref. — Convptes Mendus de 

 VAcademie des Sciences, March 29, 1880, t. xc. pp. 727-730. 



* V is the capacity of the reservoir — that is, the volume of the air 

 pumped into the entire apparatus, minus the air pumped into the com- 

 pensator — the whole supposed at 0° and 760 millims. ; V is the volume 

 of the air left in the reservoir at the temperature x and at the exterior 

 pressure H — that is to say, the volume of the air pumped into the reser- 

 voir at the temperature x, minus the volume drawn into the compensator 

 — the whole reduced to 0° and 760 millims. ; k is the sum of the coeffi- 

 cients of the normal dilatation and the permanent dilatation which we 

 have established (Comptes Rendus, t. lix. p. 169). 



t We have merely to add that the Sprengel pump ought to be con- 

 structed without the employment of caoutchouc, and that the vertical 

 tube which serves as pump should, in the part where the impact of mer- 

 cury against mercury takes place, be furnished with a tube of steel or plati- 

 num in order to protect the glass from the effect of the water-hammer. 

 The absolute necessity of this last arrangement will be explained in a 

 memoir by one of us and M. Mascart. 



