CHAP, v.] 



GENERAL CONCLUSIONS. 



305 



the time, it is necessary, in order that the results may be com- 

 parable, to reduce them to their values at one common temper- 

 ature. For this purpose a knowledge of the law of expansion 

 of sea-water with temperature is necessary. This had been de- 

 termined with sufficient accuracy for low temperatures by Des- 

 pretz and others; but as the temperatures at which specific- 

 gravity observations are usually made are comparatively high, 

 their results M'ere of but little use, directed as they were chiefly 

 to the determination of the freezing and maximum -density 

 points. When the late Captain Maury was developing his 

 theory of oceanic circulation, owing to difference of density 

 of the water in its different parts, he found the want of infor- 

 mation on this important subject. At his request the late Pro- 

 fessor Hubbard, of the National Observatory, United States, 

 instituted a series of experiments, from which he was enabled 

 to lay down a curve of the volumes of sea-water at all tempera- 

 tures from considerably below the freezing-point to much above 

 what obtains even in the hottest seas. The results are published 

 in Maury's " Sailing Directions," 1858, vol. i., p. 237, and have 

 evidently been carried out with great care. The composition 

 of different oceanic waters varies, even in extreme cases, within 

 such close limits, that the law of thermal expansion is sensibly 

 the same for all of them : of this Hubbard's experiments afford 

 satisfactory proof. In the table which gives the results of all his 

 experiments, he takes the volume of water at 60° F, as his unit. 

 In the following table the volumes for every centigrade de- 

 gree from —1° C. to +30° C. are given: 



