Intelligence and Miscellaneous Articles. 325 



p 2 diminishes, the decrement of the temperatnre A£ increases consi- 

 derably ; consequently the heating by the sides ought to increase 

 the deviation more and more if no other cause intervened. We 

 must also remark that this deviation is greater for carbonic acid 

 than for air, although the thermic effect of the sides is less. 



I regard, then, the observed deviation as the result of two distinct 

 causes ; one is the thermic action of the sides, the other is of a dif- 

 ferent nature. 



We have the effect of this latter in the last column of the Table. 

 We see that for the two gases, p" — p' begins by increasing when 

 p 2 diminishes ; this difference reaches a maximum and then decreases. 

 Now there is a mechanical effect which varies in the same manner. 



Let us consider the expansion from 9 atmospheres to 1 in two 

 distinct cases : — 



(1) Without appreciable velocities : the law is that of Laplace and 

 Poisson. 



(2) As takes place in our apparatus : the molecules situated near 

 the orifice are animated with a certain velocity ; there is in the re- 

 servoir A less gas than in the first case ; according as the pressure 

 diminishes, the velocity increases ; but soon it diminishes ; conse- 

 quently it passes through a maximum. According to the period at 

 which the flow is stopped, the difference of the specific gravities 

 which exist in A in the two cases ought to vary in the same manner. 



It is true that my experiments do not exactly realize the second 

 case. Thus in the first series the expansion takes place from 9 to 

 about 5 atmospheres, but the reservoir B is found also under a 

 pressure of 5 atmospheres at the end of the flow ; while in the last 

 series the reservoir A is subjected to 5 atmospheres when the re- 

 servoir B is found subjected to a less pressure. However, we can 

 conceive that this circumstance does not influence the direction of 

 the deviation. 



In fine, the formula of Laplace and Poisson can be applied to a 

 reversible expansion ; but there must be another iaw in the case of 

 an irreversible expansion. The investigation of this law T will be the 

 object of a further study. 



I would also remark that, the difference p" — p being greater for 

 carbonic acid than for air, the impulse of the gas in the irreversible 

 expansion varies in the same direction as the internal work. We 

 meet, in short, with an effect of the gaseous viscosity that M. Reg- 

 nault speaks of in his memoir on the velocity of sound. — Comptes 

 Rendus, August 9, 1869. 



ON THE EMPLOYMENT OF THE SPECTROSCOPE IN ORDER TO DIS- 

 TINGUISH A FEEBLE LIGHT IN A STRONGER ONE. BY M. J. M. 

 SEGU1N. 



To the two poles of a Ruhmkorff-coil of middle size there are 

 attached two fine platinum wires which are kept in a horizontal 

 position, their extremities being separated by an interval of about 

 1 centim. The spark is produced with its usual characteristics, and 

 we especially observe the shell of blue light which envelopes the end 

 of the negative wire. We bring the positive wire gradually nearer 



