EXPERIMENTAL KNOWLEDGE OF THE PEOPERTIES OF MATTER. 105 



cation with a reservoir of mercary -whicli had been thoroughly heated 

 and dried, while the reservoir could be adjusted vertically so as to alter 

 the level of the mercury in the open limbs of the U-tube and therefore 

 to raise or lower the top of the column of mercury in the barometer-tubes. 

 One of the barometer-tubes was used as a barometer for comparison with 

 the other by a cathetometer ; the other was graduated and served for the 

 introduction of oxygen. It will be noticed that by the arrangement 

 described above the pressure of the same quantity of oxygen could be 

 increased or diminished at will by raising or lowering the reservoir of 

 mercury. 



After the introduction of the oxygen its volume in cc. is read and 

 the pressure in mm. of mercury at a constant temperature. Several 

 readings were taken at intervals of about an hour generally. No pres- 

 sure was finally read off after an interval of less than two hours, and 

 often the readings for a single pressure and volume were taken at inter- 

 vals of from twelve to twenty-four hours. Each result recorded is the 

 mean of several observations extending over some time. Another series 

 is then taken at the same temperature after introduction of a little more 

 oxygen ; and perhaps still another. These give us data for tracing a part 

 of the curve, say fromp (abscissa) = 0'12 mm. to p=7'45 m. ; and ^w 

 (ordinate) =13'21 5, to ^y^l39'9G ; and so by other series of experiments 

 Bohr extends the curve up to p=15'02 ; ^i;=311'83. 



The point which was alluded to above is at 0'70 mm. pressure, at 

 which this volume is (between limits) indeterminate ; so that if the oxy- 

 gen is at the fixed temperature of the experiment submitted to a pressure 

 of very slightly less than 0"70 mm., and allowed to remain five or six 

 hours, the volume (and the p v) will be considerably different from that 

 which would be exerted by the oxygen after standing the same time at 

 the same temperature at a pressure very slightly more than 0'70 mm.; 

 thus on p. 472 {loc. cit.) for 0'70 mm. twenty-three observations gave as 

 mean volume 47 cc, and twenty-five observations at same pressure gave 

 52-41 cc. The peculiarity at this point has been verified by Bohr by re- 

 peated observations specially directed to it. 



As a confirmation of all these conclusions concordant results were 

 obtained by using tubes of different internal diameter — one of 18'5 mm., 

 the other of 32 mm. 



Thus Bohr finds, for representing relation of pv to p, for oxygen at 

 pressures from 300 mm. downwards, a line very slightly convex and nearly 

 parallel to the axis of abscissEe for some distance, till from say 60 mm. 

 to 70 mm. it curves down very considerably ; and again from 0"70 mm. 

 onwards towards mm. another curve which curves rapidly down so as 

 to tend to become nearly vertical ; this shorter part of the whole he calls 

 the small branch, and the part from 07 mm. to some hundreds of mm. 

 the long branch of the whole. The volumes of oxygen varied from 

 about 20 cc. to 200 cc. in different experiments. 



The short branch of the curve he finds can be approximately repre- 

 sented by the formula 



(p + 0-07)v=Z;; 

 and the long branch by 



where Tc (and also Jc') will be different numbers according to the amount 

 •of oxygen present. 



