REPORT ON THE COMPOSITION OF OCEAN-WATER. 
155 
Notes to Section I. 
1. When in the above table V 0 and are omitted, this means, in the majority of 
cases, that they were “ not determined because the necessary data were wanting.” In a 
few cases, it is true, the omission was caused by mishaps in the laboratory. 
2. With respect to the samples No. 823 and No. 1696. Having repeatedly observed 
that the gases from deep-sea waters more especially possessed a peculiar nauseous smell, 
strong enough to be perceptible even in the minute bell of gas which failed to find its way 
into the eudiometer, I made an attempt, in the case of gas No. 823, to determine the organic 
matter, and also the sulphuretted hydrogen presumably present. For this purpose the 
gas was treated first of all with solution of acetate of lead. This reagent produced a 
considerable contraction, but no coloration ; hence the contraction was certainly owing 
to carbonic acid absorbed by the acetate. It was accordingly added on to the amount of 
carbonic acid found in the residue by means of caustic potash. 
The gas free from carbonic acid was exploded with very carefully prepared fulmi- 
nating gas (2H 2 + 0 2 ), the change of volume noted, and the residue treated with caustic 
potash, to determine any carbonic acid produced by the combustion. In this manner two 
analyses were made (except that in the second the acetate of lead treatment was 
omitted). In both cases the contraction involved and the carbonic acid apparently 
produced in the combustion had small positive values, which were separately reduced to 
the equivalent quantity of marsh gas. The mean of the two results served for the cal- 
culation of what in the following report figures as “per cent, of marsh gas : ” — 
I. 
II. 
Carbonic acid, 
18-60 
18-97 
Marsh gas, . 
0-56 
0-33 
Oxygen, . 
20-36 
20-43 
Nitrogen, . 
60-48 
60-27 
100-00 
100-00 
Considering that these analyses (like all the rest) had to be carried out on a very small 
scale, I felt very diffident as to the real existence of the combustible matter reported 
above as “ marsh gas ” ; and this impression was confirmed by the result of another 
attempt at determining the organic matter, which was made with gas No. 1696. In this 
analysis the contraction observed in the explosion of the gas free from carbonic acid 
with fulminating gas amounted to only OT of a volume in 1 S2 ‘8 volumes, which is within 
the limits of unavoidable errors. I therefore, in the case of No. 823 also, put down the 
assumed marsh gas as imaginary, and re-calculated the analyses on this basis, to obtain 
the results reported on pages 153 and 150. 
