Carbonic Acid by Saline Solutions. 491 



Synthetical Experiments. In order to check the above experiments, it 

 appeared to me to be of importance to determine the absorption-coeffi- 

 cients of one or more of the solutions for carbonic acid. Before knowing 

 any thing of the retention of carbonic acid by sea-water, I had determined, 

 if possible before the sailing of the ' Challenger,' to investigate the solu- 

 tions of some of the salts occurring in sea-water, with reference to their 

 power of absorbing the atmospheric gases. It was well known that 

 sea-water, in common with most other salt-solutions, absorbed a smaller 

 quantity of air than distilled water would do under the same circum- 

 stances ; but it had never, to my knowledge, been attempted to find out 

 whether this diminution of absorptive power was distributed equally 

 over the three gases, or was exhibited more strongly in the case of one 

 gas than in that of another. From a preliminary experiment in which a 

 2 per cent, solution of Na Cl in distilled water was saturated with air and 

 the air then expelled and analyzed, it appeared that the oxygen was 

 present in slightly greater quantity relatively to the nitrogen than would 

 have been the case if the liquid had been distilled water. Of course, this 

 being the result of only one experiment, no conclusion can be drawn from 

 it ; but it shows the necessity of the investigations which I had proposed 

 to myself. Unfortunately, the time at my disposal was too short to allow 

 of any thing being done, except a few experiments with carbonic acid and 

 solutions of sulphate of magnesia and of sulphate of lime. 



Eor this purpose I made use of a Bunsen's absorptiometer, and fol- 

 lowed his method, with the modifications rendered necessary by having 

 to do, not with a simple liquid, but with a saline solution. The carbonic 

 acid was introduced into the absorption-tube and measured, not in the 

 mercurial trough, but in the absorptiometer itself, the lid being left open. 

 This is a much more expeditious way, inasmuch as the gas quickly 

 assumes the temperature of the water of the absorptiometer ; and as the 

 readings after absorption are all clone in this way, there can be no 

 object in reading the gas alone in another way. After absorption, the 

 instrument is always read with the lid shut, so that what corresponds 

 to the height of mercury in the trough is given by the height of it in the 

 outside graduated leg of the absorptiometer. In all of the determinations 

 this height is given, not in the reading on the leg itself, but in the cor- 

 responding reading on the absorption-tube, which can be directly 

 observed with sufficient accuracy with the ordinary telescope used in gas- 

 analysis. As after shaking the instrument and opening the stopcock 

 connecting the leg with the body of the instrument some of the water 

 frequently passed into the former, we have generally a reading marked 

 " water in outer leg," which forms a factor in estimating the tension of 

 the gas. 



The solutions experimented on were, one containing 1-23 per cent, 

 crystallized sulphate of magnesia and one containing 0*205 per cent, 

 gypsum. It was necessary that these solutions, before being introduced 



