of sea waters , in different parts of the ocean , &c. 1 65 
apparatus was sunk into the sea, the valves would, of course, 
be kept open by the current of water passing freely through 
the machine so long as it descended; and when drawn up 
again, the valves would be kept closed by the water acting in 
an opposite direction. As however, in heaving a machine of 
this kind from a great depth, it is almost impossible that some 
oscillations should not take place in it, either from the motion 
of the boat, or from some accidental jerks in pulling the line, 
it is evident that these oscillations being necessarily communi- 
cated to the valves, partial changes of the water contained in 
the vessel are almost unavoidable. It was principally with a 
view to remove this objection, that I modified the principle 
of the apparatus in the following manner. I procured a strong 
cylindrical brass vessel, capable of holding about half a pint of 
water, and having, like Dr. Irving's machine, an aperture both 
at the top and bottom, with a valve at each aperture opening 
upwards. But these valves, instead of being allowed to move 
backwards and forwards freely upon their hinges, were, when 
closed, firmly locked by springs ; and when open, could only 
be kept in that state by the application of a certain degree of 
force. This force consisted in a weight of several pounds, 
suspended to the extremity of a cord, the other end of which was 
fastened to the valves ; and the machine was so constructed 
(as will be readily understood by referring to the annexed 
sketch, Plate XI. ), that the moment the valves were relieved 
of the weight which kept them open, they closed instantly, and 
were forcibly locked by the above mentioned contrivance. Now 
the machine being open, with the weight suspended to it at a 
few feet distance, and the whole being let down into the sea, 
it is evident that the apparatus must continue perfectly open 
