240 Proceedings of Royal Society of Eclinhurgli. [sess. 
velocity in seconds per hundred fathoms. I have made many experi- 
ments of this kind, both on sounding line and on wire, and I have 
invariably found the greatest uniformity in the rate for the same 
“ messenger ” and the same line. On board the “ Princess Alice ” the 
messenger used was very heavy ; it consisted of a ring of lead, weigh- 
ing rather over 2 kilogrammes. Its impact on the trigger, at a 
depth of 1000 metres, was felt distinctly at the surface, wire being 
used. It was found, by experiment, to descend 
200 metres in . 
150 seconds. 
500 ,, 
380 ,, 
1000 
752 ,, 
Average 100 ,, 
75 „ 
This rate is very uniform, but slow. In the “ Buccaneer,” where 
hemp line was used with the water-bottle, a messenger of sheet lead, 
rolled into a cylinder fitting loosely over the line, descended at tlie 
rate of 50 fathoms in 26 seconds, or 115 fathoms per minute. The 
slower rate on the wire is due to its cutting deeply into the lead. 
Whatever the rate may be, I have always found it very constant, 
as, in fact, it ought to be ; because, after the maximum velocity has 
been reached, there is no reason why it should suffer alteration. 
The system of sounding line and weights and instruments, after 
the maximum velocity has been reached, descends at a constantly 
decreasing rate, owing to the increased resistance of the continually 
increasing length of line. When wire is used, the speed is more 
uniform. In all cases, however, experience shows that it is possible 
to calculate to a minute or so when the instruments will reach the 
required depth, and how long the messenger will take to reach the 
same depth. With all thermometers it is necessary to allow them 
to remain a certain time in order to take the temperature of the 
medium. In the “ Princess Alice ” the thermometers were allowed 
to remain 10 minutes down, and there was no difficulty in despatch- 
ing the messenger so that it would reach the required depth about 
10 or II minutes after the thermometer. In this way we have all 
the certainty of the method without any appreciable loss of time. 
An objection is sometimes advanced that, when the line “leads” 
out at a great angle, the messenger will not run down and close 
the instruments. This, however, must be looked on as rather an 
advantage than otherwise, because observations made under these 
