COMPRESSIBILITY OF WATER, GLASS, AND MERCURY. 



o 



I. General Account of the Investigation. 



I will first give a general account of the subjects treated, of the mode of conducting 

 the experiments, and of the difficulties which I have more or less completely overcome 

 in the course of several years' work. The reader will then be in a position to follow 

 the full details of each branch of the inquiry. 



The experiments were for the most part carried on in the large Fraser gun fully 

 described and figured in my previous Report. But it was found to be impracticable t 

 maintain this huge mass of metal at any steady temperature, except that of the air of 

 the cellar in which it is placed. The great thickness of the College walls, aided by the 

 comparative mildness of recent winters, thus limited till the beginning of the present 

 year the available range of temperature for this instrument to that from 3° C. to about 

 1 2° C. As I did not consider this nearly sufficient, and as comparative experiments at the 

 higher and lower of these temperatures could only be made at intervals of about six 

 months, I procured (in May 1887) a much less unwieldy apparatus. It was made entirely 

 of steel, so as to be of as small mass as possible, with the necessary capacity and 

 strength : and could at pleasure be used at the temperature of the air, or be wholly 

 immersed in a large bath of melting ice. As this apparatus was mounted, not in a 

 cellar, but in a room sixty feet above the ground and facing the south, it enabled me 

 to obtain a temperature range of 0° C. to 19° C, with which I was obliged to content 

 myself. A great drawback to the use of this apparatus was found in the smallness of 

 its capacity. Not only was I limited to the use of two, instead of six or seven, 

 piezometers at a time ; but the pressure could not be got up so slowly and smoothly as 

 with the large apparatus, and (what was still worse) it could not be let off so slowly. 

 In spite of these and other difficulties, to be detailed later, I think it will be found that 

 the observations made with this apparatus are not markedly inferior in value to those 

 made with the great gun. 



In the piezometers I have adhered to the old and somewhat rude method of 

 recording by means of indices containing a small piece of steel, and maintained in their 

 positions (till the mercury reaches them and after it has left them) by means of attached 

 hairs. These indices are liable to two kinds of deceptive displacement, upwards or 

 downwards, by the current produced at each stroke of the pump, or by that produced 



