CLYDE SEA A.REA. 



instrument came to the surface. When the water-bottle was emptied, the thermometers 

 were replaced in the same order as before, sent down to a different depth, and the process 

 repeated again and again, until an observation had been made at every ten fathoms from 

 surface to bottom. Then the figures, recorded by an assistant in the observation-book, 

 were examined, and if there was any sudden difference between two, or any unusually 

 close agreement, readings were made at intermediate depths, and this process of subdividing 

 spaces was carried on until the exact form of the temperature curve had been ascertained. 

 With a crew of three men it was possible, in quarter of an hour, to make and record six 

 observations ; that is to say, two dips of the sounding-line with three thermometers, and 

 also to observe the air temperature, barometer, wind, weather, &c. This may be contrasted 

 with the time required to get the same number of observations with spirit thermometers. 

 The rate of descent of the brass messengers was measured several times. Ther- 

 mometers were fixed on the line at intervals of ten fathoms, and when they had acquired 

 the temperature of the water a messenger was dropped. The time when it struck the 

 surface of the water was noted by the seconds hand of a watch, and an assistant with 

 his hand on the line said " one," " two," or " three," as the respective shocks caused by the 

 impact of successive messengers, which were released by the reversing thermometers, were 

 felt. The instant of hearing these words was also entered. This method was only roughly 

 accurate. The time -intervals between successive shocks included not only the duration 

 of falling of the weight, but the time which each thermometer when struck required 

 to turn through rather more than a right angle (probably about 120°), before the loop 

 holding the messenger slipped, and also the time during which the vibration passed 

 up the line. These intervals were, however, practically constant in the slight depths in 

 which observations were made. The following table (Table I.) summarises the experi- 

 ments made in this way, and shows the rate of fall of the messenger in sea-water 

 (average density, 1"0250), the figures being reduced in the last column to feet per 

 second, the space marked as " ten fathoms " on the sounding-line having, at the time the 

 observations were made, become equal to 62 feet, in consequence of stretching. 



Table I. — Mean Velocity of Fall of Brass Weights in Sea Water. 



Length of Run. 

 Fathoms. 



Average Length 

 of Run. 

 Fathoms. 



Number 



of 



Cases. 



Seconds 



in Falling 



10 Fathoms. 



Average Velocity. 

 Feet per second. 



2 to 5 

 6 to 9 

 10 



13 to 16 

 21 to 34 

 38 to 64 



3 



7 

 10 

 14 

 27 

 50 



15 

 4 



42 

 4 

 5 

 5 



8-7 



8-0 



7-16 



6-9 



6-4 



6-2 



7-1 

 7-8 

 8-6 

 9-0 

 9-7 

 10-0 



This shows that the resistance of the water rapidly causes the acceleration of a falling 

 weight to approach a limiting value ; and from the form of the curve, which expresses the 



