DR HOPE ON THE MAXIMUM DENSITY OF SEA- WATER. 247 



apparatus was immersed in a frigorific mixture of temperature 20°, the fluid pro- 

 gressively descended in the stem till it became stationary, without halt or retro- 

 cession in its course. The instrument was then withdrawn from the cold mix- 

 ture and supended in the air ; the fluid immediately began to ascend, and conti- 

 nued to do so steadily, as it gradually increased in its temperature. 



Experiment, No. 6. The same experiment was repeated, and afforded a similar 

 result. These experiments lead to the conclusion that sea- water contracts as it 

 is cooled to its freezing point, and expands from that point when it is heated. 



I conceive that perfect confidence may be reposed in these conclusions, though 

 drawn from trials made in the thermometer-like apparatus, the changes of whose 

 capacity by heat and cold necessarily affect the apparent movement of the included 

 fluid, whether in ascent or descent. When during refrigeration the diminishing 

 capacity of the instrument proceeding from the contraction of the glass necessa- 

 rily causes the fluid to ascend in the stem, and the fluid notwithstanding con- 

 tinues to descend, the descent unequivocally proves that the fluid is undergoing 

 contraction. The case, however, is very different when the fluid during cooling 

 ceases to descend, and then begins to rise in the stem, and continues to do so as 

 it is further cooled. Here the indication is equivocal, as the occurrence may arise 

 either from the diminution of the capacity of the instrument, or the expan- 

 sion of the fluid, and it is no easy matter to decide from which of the causes it 

 proceeds. 



It was in consequence of this circumstance that I discarded this description 

 of apparatus formerly when investigating, in 1803 and 1804, the singular anoma- 

 lous fact of water at low temperatures expanding by cold, and contracting by 

 heat. 



Though convinced by these experiments that Sir Charles Blagden had fallen 

 into an error, I did not rest satisfied with them, but instituted a second series, 

 conducted upon the same plan which I introduced in 1803. But before proceed- 

 ing to describe these experiments, I shall call attention to the observations of Mr 

 Erman. 



He did not make any experiments upon sea- water, but upon a solution of 

 common salt, of the specific gravity of 1.027, which he considered as equivalent 

 to the water of the ocean, and upon two weaker solutions, and conducted the 

 inquiry in four different methods. 



He first tried the question by taking the specific gravity of the representative 

 of sea- water at different temperatures by the weighing-bottle, and then by Nichol- 

 son's hydrometer ; but he put little reliance upon the results, though favourable 

 to the idea of the fluid having no maximum density at some degrees above its 

 freezing point as water has, for the reason that the phenomena and apparent re- 

 sults may depend in either of these modes, as much upon a change of the dimen- 



