65 
1892-93.] Prof. Tait on the Com^pressibility of Liquids. 
seems almost perfect throughout, for the differences have regular 
alternations of sign. But it is to be noticed that simultaneous 
increase, or diminution, of A and B by as much as 2 per cent, does 
not seriously affect the agreement of the formula with the results 
of experiment. 
I have been for some time preparing to undertake an extended 
series of experiments on the compressibility of various aqueous 
solutions, with the view of finding (although by an exceedingly 
indirect and possibly questionable process) how the addition of a 
salt to water affects its internal pressure. But the recent publication 
of the final results of Amagat’s experiments on the compression of 
water by pressures rising to 3000 atmospheres (more than six-fold 
the range attained in my own work) has led me to make a new 
series of calculations with the view of testing how far the above 
speculations, suggested by the results of pressures limited to some 
three tons’ weight per square inch, are borne out by the results of 
pressures of twenty tons. The agreement, as will be seen, seems on 
the whole highly satisfactory ; though, for a reason already given, 
and presently to be even more forcibly illustrated, the calculations 
are necessarily of a somewhat precarious character. 
Thus we obtain from Amagat’s paper {Comptes Rendus, 9/1/93) 
the following determinations of the volume of water at 0° C., for 
additional pressures of 400 and 800 atmospheres : — 
Pressure. 
Table, No. 1. 
Table, No. 2. 
1 
1*00000 
1*00000 
401 
•98067 
•98071 
801 
•96371 
•96371 
The pressures in Table 1 extend to 1000 atm. only, those in 
Table 2 to 3000 atm. 
These give, respectively, for the average compressibility of water 
per atmosphere for the first p additional atmospheres, p ranging 
from 0 to 800 ; 
0*296 0*3057 
5725 - 1-^9 5939+^ 
whence the compressibility at ordinary pressure may be either 
0*0000517 or 0*00005147. 
VOL. XX. 17/5/93 E 
