896 REPORT OF COMMISSIONER OF FISH AND FISHERIES. 

 The following abstract from tables shows the truth of this: 



Thus, when the difference in temperature is less than 10°, "five min- 

 utes are sufficient"; but as the differences increase, so does the time 

 necessary to produce an equilibrium increase. The third experiment in 

 Table 21 points to the necessity of having the mercury perfectly at rest 

 when about to use the instrument. 



An experiment was tried to ascertain the effect, if any, of the sun 

 shining on the stems of the thermometers, the temperature in the sun 

 being 71°; four thermometers with bulbs level in a bath were recorded, 

 and the instruments were entirely immersed, with following results: 



No. 



1, 9 inches exposed, stood .3G°. 



No. 2, 4 inches exposed, stood 36o.5. 



No. 4, inches exjjosed, stood 36°. 



No. 5, G inches exposed, stood 36°.5. 



Entirely immersed and read in two minutes they stood : No. 1, 36°.2; 

 No. 2, 3G°; No. 4, 36°; No. 5, 36°.5. So that no appreciable error arises 

 from this cause. 



