igo SCIENCE PROGRESS 



study of these questions may be said to begin with the work 

 of Hermann Kopp, whose results were published in a series 

 of memoirs contributed to Poggendorff's Annalen der Physik 

 and to Liebig's Annalen der Chemie at intervals extending 

 over fifty years, viz. from 1839 to 1889. Strictly speaking, 

 Kopp was not actually the first to broach the subject, but 

 such tentative observations as were made before his time 

 possess no practical importance and have only a very slight 

 historical interest. This arose partly from paucity of material, 

 but more especially from the absence of any well-defined 

 basis of comparison, whereby the recognition of any possible 

 generalisations became hopelessly obscured. It was of little 

 use, for example, to determine the weight of the unit-volume 

 of the liquids to be compared at some uniform temperature, 

 say at the melting point of ice, or the mean atmospheric tem- 

 perature, the conventional temperatures to which specific 

 gravities are usually referred, since liquids are not necessarily 

 under comparable conditions at these temperatures on account 

 of their variable thermal expansibilities. Kopp imagined 

 that a comparable thermal condition would be found at that 

 temperature at which presumably the internal pressures are 

 the same for all liquids, that is, at the temperature at which 

 the liquid passes wholly into the state of gas or vapour, or, 

 in other words, at its normal boiling-point. Other bases of 

 comparison may be suggested nowadays and possibly some 

 of them may be preferable to that adopted by Kopp, but at 

 the time he wrote no other suitable method was open to him 

 and it is at least doubtful even now whether any other system 

 would serve to reveal any more comprehensive or more definite 

 relations. Be this as it may, practically all subsequent 

 workers have followed Kopp's example in principle, whilst in 

 some cases varying his particular experimental procedure. 

 Kopp's method was to select liquids of assured purity and, of 

 course, of known molecular weights ; to ascertain their boiling- 

 points under standard conditions, the details of which he 

 greatly improved ; to determine their thermal expansibilities 

 over a range sufficiently wide to enable the law of the thermal 

 expansion of each to be ascertained so long as it remained a 

 liquid under normal pressure ; to find its specific gravity at 

 some convenient temperature, and thence to calculate the weight 

 of the unit-volume at the normal boiling-point. This value 



