DESCRIPTION OF THE EXPEDITION. 13 
amount of brake placed on the drum. A sounding of 2000 fathoms, from the time that 
the order is given until the ship proceeds again upon her course, does not generally take 
more than an hour (PI. 4). 
The weights are connected with the sounding-wire by stout pieces of hemp about 
3 fathoms long, so that an indication of the bottom having been reached is received at 
the machine before the wire can itself touch the bottom. The leads employed by 
us were three in number—the old valved lead so often described; the modification of 
the last regularly employed for deep-sea sounding, which consists of a tube with valves 
running into the middle of the weights, the latter being mechanically detached as 
soon as the bottom is touched; and, lastly, the Snapper Lead, invented by Mr. Lucas. 
For most of our work we found the latter by far the best lead for procuring samples of 
the bottom, particularly for soundings below 1000 fathoms and on more or less hard 
bottom. The form that we employed unfortunately had no detachable weights, and the 
whole of its 28 or 30 lbs. had to be heaved in by the machine, this weight in a seaway 
sometimes throwing an undue strain on the sounding-wire. The apparatus itself consists 
of a steel shaft (ending in a ring for attachment) extending through the centre of 
a conical lead. At its lower end the latter is hollowed out so that a spiral steel spring 
can be inserted around the shaft; this acts upon two brass spoons, which are hinged 
on to the end of the shaft, and which, when closed, fit tightly upon one another. The 
box so formed is kept open by a sear while the lead is sent down, but its sides at once 
snap up tightly when it strikes anything. It seldom fails to bring up a sample of the 
bottom—a small handful of sand, a few pieces of rubble, or fragments detached from 
the rock beneath (see fig. 2). 
The thermometers employed for the ship’s use were of the usual old maximum and 
minimum type, as supplied by Cary, and Negretti and Zambra. When sounding in the 
deep sea, one is attached to the sounding-wire just above the lead. In addition we 
employed them in series at every 50 or 100 fathoms down to 800 fathoms, attaching them 
to a wire of a smaller Lucas machine fixed on a platform on the port side of the poop, 
lowering slowly, and heaving in by hand. Their readings may be relied on to a quarter 
of a degree. In addition, we ourselves employed, generally in pairs, Richter’s reversing- 
thermometers, as approved by the International Bureau for the Exploration of the Sea. 
They reverse by messenger or by propeller. Careful tests showed that the latter, after 
we had readjusted it, released the frame so that it swung over after passing through 
about 7 fathoms of water, and although not so reliable as the messenger form, the error 
of its thermometers from every cause did not, we think, exceed one-tenth of a degree. 
The propeller form has the inestimable advantage that several thermometers may be 
attached as the wire runs out so as to reach different depths. The temperature of ihe 
surface-water of the sea was taken in a narrow canvas bucket by a specially tested 
thermometer. This receptacle was very satisfactory, as we found by experiment that 
its contents, even if hung up in a strong sun, do not vary by a tenth of a degree for the 
first two minutes, long before which the temperature would have been read. 
For the reversing instruments we employed a steel rope half an inch in circumference, 
consisting of six cords each containing nineteen strands. To prevent kinks and to keep 
