12 THE COMPRESSIBILITY OF LITHIUM, SODIUM, 



that at which the connection was broken, since there is sometimes a slight 

 adhesion between the point and the mercury under the last-named circum- 

 stances. Often, however, the making and breaking occurred within an 

 atmosphere's pressure of one another, and with the most compressible sub- 

 stances no difference at all could be observed. 



To prevent leakage of kerosene, a fixed amount (3 mg.) of rubber lubri- 

 cant was always used on the glass stopper. Because this occurred in every 

 case, any slight effect which it might have had in the readings was the 

 same in every case, and hence was wholly eliminated from the final results. 



The special precautions needed for manipulating this apparatus are suffi- 

 ciently discussed in the preceding paper, to which the reader is referred 

 for further details. (Carnegie Institution of Washington Publication 7.) 



THE STANDARD OF PRESSURE. 



The accurate determination of pressure is the most difficult part of these 

 measurements, because all the other uncertainties are eliminated by the 

 method of operation of the apparatus. It has already been pointed out 

 that if the relation of the compressibility of water to that of mercury were 

 accurately known, this could be used as a standard, but unfortunately it j,s 

 not accurately known. The work of Amagat upon the compressibility of 

 water* is indeed in many respects admirable. His apparatus for determin- 

 ing high pressure --a double hydraulic press with rotary pistons seems 

 to have given consistent results, and to have been more satisfactory than a 

 gas manometer would have been. Nevertheless, it is a complex apparatus, 

 and certainly less desirable than a mercury column of constant tempera- 

 ture and known height. Among other defects, it is probable that the 

 diameters of the pistons are neither perfectly constant or easily determin- 

 able with the accuracy required ; these diameters must vary with the press- 

 ures. In another respect, too, Amagat's determinations of the compressi- 

 bility of water are uncertain, namely, because of the fact that in his experi- 

 ments the water was compressed in a kind of inverted thermometer tube, 

 and no allowance was made for the water adhering to the sides of the 

 stem. As the column was pressed upward by the mercury, this adhering 

 water would tend to make the compressibilities at high pressure appear 

 larger than they really were. 



Accordingly, for the present, it was deemed best to refer in this paper, 

 as in our former one, all the pressures to the gauges which we actually pos- 

 sess ; and because these results are all determined in the same way, they all 

 may be corrected when the true compressibility of water is known, by 

 reference to our own figures for water. We hope that before very long 

 this value may be obtained by Dr. W. Ekman at the Laboratory for the 

 Investigation of the Sea, Christiania, Norway. 



*Ann. de Chim. et de Phys. (6), 29, pp. GS and 505 (1893). 



