260 REPORT OF COMMISSIONER OF FISH AND FISHERIES. 
lower end of the net being properly secured, the ends of the lashing 
are carried down to the sinker and made fast in order to keep the net 
in place while going down. 
Four small brass rings are secured to the bag at equal distances, a 
few inches below the upper edge of the silk-gauze lining, and through 
them is rove a soft white tie line, which makes a complete round turn, 
the ends being passed through the same ring, then rove through small 
metal blocks on the lower bridle, and finally secured to leads weighing 
14 pounds each. Two tripping lines, with eyes in their upper extremi- 
ties, are hooked over a friction clamp on the tow rope, then rove through 
small eyes on the rim of the net, and through brass rings on the lower 
bridle, above the metal blocks before mentioned; the ends being hitched 
to the leads, support their weight, allowing the tie, or draw string, to 
hang loosely and the net to retain its natural form while sinking and 
being towed. 
To use the apparatus, prepare it as in figure 1, plate 1, lower it 
vertically to the proper point, and tow it slowly through the water, 
veering and heaving in on the tow line in order to maintain the desired 
depth, which can be determined within a few fathoms by the dredging 
quadrant, an instrument in constant use on board of the Albatross. To 
recover it, stop and back until the tow rope is vertical, heaving in 
sufficient line during the operation to keep the net at the proper depth; 
then send the messenger (fig. 2, m) down to act on the friction clamp 
(fig. 4), release the tripping lines (/), and close the lower part of the net 
as Shown in fig. 6. The net may berun up to the surface at any desired 
speed, the upper portion taking in anything it encounters en route, 
while the lower part remains closed against even the most minute 
forms. The messenger is intwo parts, which, having been placed around 
the tow line, are seized together with marline(s). It sinks at the rate of 
about 650 feet per minute, and the impact can usually be distinctly felt 
by taking hold of the tow line. 
To Cocos Island.—The course was continued towards Cocos Island 
during the night, and two hauls of the trawl made on the 26th in 1,175 
and 978 fathoms. Rocky bottom was encountered in the first haul, 
which wrecked the net, but many valuable specimens were found in the 
remnants. Serial temperatures were taken at one station, but the 
results were not entirely satisfactory ; it becomes more evident from day 
to day that our practice of using the dredgerope for a temperature line 
in the tropics, with the Negretti and Zambra thermometers, must be 
changed. The high temperature near the surface fills the bulbs so full 
that even the moderate jarring caused by the surging of the rope on 
the drum#f the hoisting engine is liable to shake the mercury down 
into the catch reservoir, which, becoming filled, overflows into the tube, 
making it necessary to repeat many of the observations. The weather 
was overcast, with passing rain squalls during the day, and, toward 
evening, frequent flashes of distant lightning were observed, particu- 
