236 
DR. H. T. BARNES ON THE CAPACITY FOR HEAT OF WATER 
cases, on account of the greater opportunity for conduction throughout the water 
column ; but the effect is still shown. It is of interest also to calculate the actual 
temperature of the same heating wire in the two cases. 
Flow 
per second. 
Resistance of 
heating wire. 
Corrected to a mean 
temperature of 26°T. 
First set. 
•60 
•52427 ohm. 
•5243 
•28 
•51740 „ 
■5174 
Second set. 
•60 
•53460 ohm. 
•5331 
•27 
•52383 „ 
•5225 
Correcting to the same mean temperature by the temperature coefficient of the 
platinum, the results show that, as might be expected, the wire held central is 
hotter than when in contact with the glass. This means that the central wire 
is surrounded by a cloak of hot water moving parallel with it, and the more 
completely prevented from diffusing the greater the* velocity of the flow. This is 
shown conclusively by comparing the temperature of the wire, as indicated by its 
resistance, with the temperature of the same wire measured “ cold ” and reduced to 
the same mean temperature. A measurement of this for a current through the wire, 
not sufficient to cause a rise in temperature of more than *1° in the outflowing water, 
gave the value '5100 ohm. 
For the case where the wire is held central in the largest flow and the conditions 
are most perfect for the formation of a moving cloak, the mean temperature of the 
wire, as given by its increment of resistance, over and above the mean temperature 
of the water column in the flow-tube, is of the order of 12° C. This shows that, at 
the very most, only one-quarter of the total quantity of water flowing through the 
3-millim. flow-tube per second was receiving heat from the wire. 
In the case where the wire touches the sides of the tube for its full length, a 
greater area of water is heated by conduction and diffusion throughout the layers 
along the sides of the tube, which do not move at such a high rate of velocity as in 
the centre. The increment of resistance for the large flow given in the first set of 
readings for this case shows that the wire was of the order of 7° hotter than the 
mean temperature of the water, and indicates that about one-half of the total flow 
was employed in carrying off the heat from the wire. When the water is thoroughly 
stirred around the heating wire, and in particular where the flat heating wire, 
twisted into spirals down the flow-tube, is employed, the mean temperature of the 
