REPORT ON MINERAL AND THERMAL WATERS. 5 



that the Indian Ocean, wliich unites the two, is salter on the 

 west where it approaches the Atlantic, than on the east where 



joins the South Sea. 



2nd. That in each of these oceans there exists a maximum 

 point of saltness towards the north, and another towards the 

 south, the first heing further from the equator than the second. 

 The minimum between these two points in the Atlantic is found 

 to be a few degrees south of the equator ; in the Pacific it still 

 remains to be determined. 



3rd. In the Atlantic the western portion is more salt than the 

 eastern ; in the Pacific, the saltness does not appear to vary. 



4th. In proceeding northwards from the point at which the 

 saltness is at its maximum, the specific gravity of the water di- 

 minishes constantly as the latitude increases. 



5th. From the equator to 45° north latitude, the sea-water, 

 from the surface to the depth of 1000 fathoms, continues uni- 

 form in saltness. 



This last conclusion, however, must not be looked upon as instru- 

 fuUy established, since the instruments, by means of which sea- "lents for 

 water has hitherto been drawn from great depths, are considered Wate'"from 

 by the best judges very faulty in their construction, and inca- depths. 

 pable of affording trustworthy results. Such was the opinion 

 expressed by Dr. Marcet*, after examining those that had been 

 invented up to the period at which he wrote ; and such more 

 recently was the impression of M. Aragofj who hence was led 

 to I'ecommend, to the navigators of the French discovery ship, 

 the Bonite, an instrument for the same purpose of M. Blot's in- 

 vention, which is on a different plan from those hitherto em- 

 ployed. 



The description given of this contrivance by M. Arago is in 

 itself very brief, and is unaccompanied by a plate. Possibly, 

 therefore, I maj'^ not have understood every part of its con- 

 struction, but upon the best consideration I was able to give 

 to the subject, it appeared to me that some parts of the instru- 

 ment might admit of improvement. 



I consequently designed an apparatus framed on a similar 

 principle to that of M. Biot, but provided neither with a spring 

 to exclude the external watei-, nor with a stopcock and bladder 

 to receive the compressed gas, both which objects were fulfilled 

 by means of a small hollow cap of brass, which being attached to 

 a conical stopper, accurately ground to the hole in the bottom of 

 the instrument through which the water was admitted, dropped 

 down upon this aperture when the vessel was inverted, and 

 thus at the same time would cut off all communication with the 



* Philosophical Tranc-aclions, 1819. t Jnnv.airc, 1836. 



