522 



NATURE 



{April 27, 1876 



the point directed downwards. (This point serves the 

 purpose of fixing with certainty the surface of the mercury 

 contained in E, for the point is reflected in the surface of 

 the meniscus.) E is connected by means of caoutchouc 

 tubing with a wide vessel D, which is fixed to a moveable 

 carrier, and can be easily moved up and down. The tube 

 C, which forms the vertical prolongation of the vessel D 

 upwards, moves before a scale with millimetre divisions. 



y - - - -— - K- 



When the apparatus is to be used, a portion of D and E 

 is first filled with mercury. Then the cock h above is 

 opened and D pushed up to the height of the cock ; where- 

 upon the rising mercury fills the measuring vessel B, while 

 the vessel D gets emptied. When the rising mercury 

 comes into view in the small funnel s;, you shut the cock 

 and bring the vessel D down again, whereupon the mer- 

 cury runs out of the measuring vessel and leaves there an 

 empty space. If B be now connected with a the air is 

 sucked over from the absorption jar into the measuring 

 vessel, while the mercury of the pneumatic trough rises 



in the absorption jar. This air is also removed from 

 the apparatus in the way above described. The air- 

 specimen to be examined is now brought into the absorption 

 jar, and allowed to pass over into the measuring vessel, 

 which is surrounded with snow or pounded ice.^ When 

 the gas is cooled, the mercury in E is so placed that the 

 tongue point touches the meniscus. Since the scale num- 

 bers begin at the height of the tongue point, by reading oflf 

 the position of the mercury in D, the height of the mercury 

 column is obtained, which taken together with the ten- 

 sion of the included air-specimen, makes equilibrium with 

 the external air-pressure. On deducting from the imme- 

 diately observed barometer height the observed column of 

 mercury, the pressure of the enclosed air at o'^ is obtained. 

 Next, by suitable position of the vessel D and turning of 

 the cock, the measured air-specimen is allowed to flow 

 over again into the absortion jar. Afterwards, some small 

 pieces of fused pyrogallic acid are introduced with a pin- 

 cette under the mercury into the absorption jar. When 

 the resulting absorption of the oxygen is finished, you 

 measure, in the same way, in the measuring vessel, the 

 tension of the remaining nitrogen. (The tension of the 

 saturated water vapour at 0° = 4"6 mm. is deducted in 

 each calculation from the tension obtained in the measur- 

 ing vessel.) 



Let V be the volume of the measuring vessel, P the 

 pressure of the air-quantity contained in it, then, accord- 

 ing to Mariotte's law — 



V P = 760 j>', 

 if y denotes the volume of this air-quantity at 0° C and 

 760 mm. Let P' be the pressure of the remaining nitrogen, 

 then — 



VP' = 76ojr, 



where x denotes the volume of this quantity of nitrogen 



at 0° C and 760 mm. Consequently — 

 p/ pr 



X — y — — 100 — 

 -^ p p 



'\{y = 100. 



When it is wished to analyse breathed air with this 

 apparatus, the carbonic acid is first removed with potash 

 lye, and the oxygen with pyrogallic acid. The oxygen 

 may also be removed with a piece of phosphorus, by in- 

 flaming it with an induction-spark in the absorption jar. 



It is known that the experiments which have been 

 made by Maxwell, O. E. Meyer, Obermeyer, and Puluj, 

 on^e dependence of friction of gases on temperature, 

 have not led to concordant results. M. Puluj has there- 

 fore lately made a large number of experiments with air, 

 hydrogen, and carbonic acid, in order definitely to deter- 

 mine the relation between friction of gases and the tem- 

 perature. This investigation was carried out in the 

 laboratory of Prof. Kundt, in Strassburg. The apparatus 

 used for the purpose was the friction apparatus of Kundt 

 and Warburg, which is essentially not very different from 

 Maxwell's, and consists of a glass disc oscillating between 

 two fixed discs. 



According to the dynamical theory of gases, the con- 

 stant of friction should be proportional to the square root 

 of the absolute temperature, i.e. — 



« = «o (i + « ^)i 

 where 11^ denotes the constant of friction of the gas at 

 temperature 6 = 0° C, and a the coefficient of expansion 

 of the gas. If we make, generally, 



n = ;/o (i -f a 5)» 

 we have from the experiments of Puluj — 

 For air, «=072i96± •01825 between — 3° C. and + 25° -6 

 For H, «=o-693i2± 0-01088 „ - 1° -5 C. and + 30^ C. 

 For CO2, «=o-9i6S4± 0-01394 „ +1° C.and + 29° C. 



The exponent for air is smaller than Meyer's (£), 



I To apply the ice a vertical sheet-iron cylinder is used, separable into 

 halves ; it surrounds the measuring vessel at an interval of about 3 cm. In 

 the figure the outline of the cylinder is shovvn by dotted hnes. 



