DEEP-SEA EXPLOKATION. 
327 
however, the higlit of wire between the two jaws of the clamp should be drawn back 
over the head of the clamp screw r, to pi'cvent slipping. The guard m is intended 
for the protection of the propeller against accidental contact with the sounding 
macliine or ship’s side. 
The method of mounting and protecting the thermometer in its case will be readily 
understood by reference to lig. 2, where, securely inclosed in its glass shield, it is ever 
resting in rnbberdined thimbles, which move freely along tlie inner surface of the case 
and guard it laterally, while the delicate spiral springs of phosphor bronze protect it 
longitudinally from the jarring caused by the rapid motion of the reeling engine, which, 
with the old methods of mounting, sometimes affected the reliability of observations 
by shaking the mercury down into the tube. The temperature is read through the 
slot 1c by means of a reading lens. The case is i^ivoted at j in such a manner that it 
swings freely after it is capsized but can not strike the wire while reeling in. 
To take a temperature set the spindle o into the hole in the cap i by screwing it 
down until the ])ropeller blades strike the set screw p; then, by means of the Sigsbee 
clamp q, secure it to the temperati;re rope. The bulb will then be down and the 
mercury in the tube connected with it, the position required to take the temperature. 
The water acting on the propeller during the descent will keep it in ]>osition resting 
against the set screw p, but as soon as the reeling in begins the propeller is set in 
motion, bringing the screw on the ui)per end of the spindle into action, gradually 
raising the propeller until the lower end of the spindle is withdrawn from the hole iu 
the cap i, when the thermometer promptly turns over and registers the tem])erature 
by breaking the column of mercury at the [»oint u, the columu then falling to the 
bottom of the tube. The scale cau be read at any time, i)roviding the thermometer 
has been kept bulb up, as changes of temperature do not affect the reading after the 
column is once broken. 
READING LENS FOR THE TANNER THERMOMETER CASE. 
it is a difficult thing to hold a thermometer vertically and exactly op])osite the 
eye, some observers tipping it forward a little, and some backward, with a conseiiueut 
change in the apparent relative positions of the top of the 
mercury column and of the scale behind it. Dr. Kidder 
tested a number of different observers and found that the 
probable parallax error in reavling, liy those who use the 
thermometers in practice, is not far from 0.3°. While 
this error is of little moment in the ordinary temperature 
observations iu air or at the surface, it assumes greater 
importance in the deep sea, where variations in tempera- 
ture are slight, and, to eliminate errors of parallax as far 
as possible, he introduced a reading lens made by Mr. 
Joseph Zentmayer, of Philadelphia, which he describes as 
follows : 
The lens is about 3 inches focal length, fitted at right 
angles to the center of a brass saddle adapted to the con- 
vex surface of the thermometer case, and provided with 
a sliort draw tube for focusing. The eyepiece opening is made smaller than the ]uipil 
of the eye, and there is therefore no variation in the reading, whatever may be tlie 
inclination to the perpendicular at which the scale is viewed. The magnifying ]>ower 
of the lens makes it much easier than formerly to read the temperature to fractious of 
a degree. 
Cttt 47. — Reading lens for 'I’aii- 
ncr thennoineter case. 
