of the Fishery Board for Scotland. 



451 



to my previous reports, I may state that in every case my specific 

 gravity determinations are based on direct weighings of the weights of 

 the equal volumes of pure water and of the respective samples of sea 

 water measured in a modified form of Sprengel's pyknometer at the 

 temperature of melting ice. Every weighing was subjected to a 

 correction depending upon the density of the air at the time of weighing, 

 so that the observational error of any weighing was certainly less than 

 0*00005 grammes. 



The chlorines, or more strictly the total halogens calculated as chlorine, 

 were all estimated by Dittmar's titrimetiic modification of Yolhari's 

 volumetric process. This modification is a happy combination of gravi- 

 metric and volumetric methods, and gives results not much inferior to the 

 most careful gravimetric determinations. 



Dittmar made two determinations of the value for D under the above 

 conditions. The waters he used for this purpose were obtained by mixing 

 several of the ' Challenger ' samples of sea water and diluting portions of 

 this mixture down to different specific gravities. In the one instance 

 referred to, he found D = 1*4593, and in the other D = 1*4619. 



The mean of these two numbers is 1*4606, which is to all intents and 

 purposes identical with the value which I have throughout assumed as 

 Dittmar's value for D, viz., 1*4600. 



The first series of samples I examined were collected during August 

 and September 1883, in the Moray and adjoining Firths. Some of these 

 samples were collected 20 miles out to sea, where the influence of land 

 water would be presumably at a minimum, as compared with the samples 

 collected from the inshore waters, and especially when compared with the 

 samples collected far up in the Inverness Firth near the entrance to the 

 Caledonian Canal, and which had, of course, a very low specific gravity. 



The maximum value for D obtained with the sixteen samples examined 

 was 1*4595, and the minimum 1*4429, the difference therefore from the 

 assumed normal value for ocean water being -0*0171 as a maximum, 

 and - 0*0005 as a minimum. Moreover, thirteen out of the sixteen 

 samples gave values for D falling between the narrow limits of - 1*4595 

 and - 1*4551. In each of these cases the value for D actually found was 

 less than the theoretical value, but the approximation was so very close 

 that my results could not but be regarded as a remarkable confirmation and 

 indeed extension to these Firth waters of Dittmar's proposition. The influ- 

 ence of the river waters in altering the value for D was hardly perceptible. 



To take two extreme cases : — 



In a surface sample collected 20 miles out to sea, the specific gravity 

 at 0° C. (water at 0° C. = 1000), or as Dittmar expresses it, 0 S 0 , was found 

 to be 1028*26, and the chlorine, or rather total halogen calculated as 

 chlorine, 19*363 grammes per kilo. 



Therefore, D = 1000 

 X 



28*26 i , KOK 

 = ^r^w>« = 1'4595. 

 19*363 



In all these calculations x represents the number of grammes of total 

 halogen calculated as chlorine per kilo of sea water. 



In a sample collected at low water near Clachnaharry, at the head of 

 Inverness Firth, 0 S 0 was found to be 1018*56 and the total halogen 

 12*726. 



Therefore, D = o S Q- 1QQQ 

 X 



= = 1*4584. 



12-726 



