176 AGBICULTgRA.L EXPERIMENT STATION, ItHACA, N. Y. 



TABLE II — Average Pipe Temperatures. 



Riser, a b, average inside temperature, 211 degrees. 

 Riser, A, average outside temperature, 193 degrees. 

 Riser B, average outside temperature, 183 degrees, plus. 

 Return a b, average inside temperature, IGO degrees, plus. 

 Riser c d, average inside temperature, 212 degrees. 

 Riser C, average outside temperature, 19G degrees, plus. 

 Riser D, average outside temperature, 192 degrees, minus. 

 Return c d, average inside temperature, 131 degrees. 

 Riser f, average inside temperature, 155 degrees, plus. 

 Riser E, average outside temperature, 137 degrees. 

 Riser F, average outside temperature, 124 degrees, plus. 

 Upper return f, average inside temperature, 122 degrees. 

 Lower return f, average inside temperature, 111 degrees. 



By a comparison of these averages it is seen that the loss of 

 heat in passing the whole length — some sixty feet — of tlie 

 steam series A B averages ten degrees, and owing to the fact 

 tliat the thermometer at the farther end showed an error, the 

 difference is probably even less than this. Frequently when there 

 was steam pressure of three or four pounds the entire circuit of 

 about 150 f(3et was made with a loss of only ten or eleven degrees. 

 The average loss between the opposite ends of series C and D — 

 which is a wai-mer run — was only slightly more than four degrees. 

 Allowing fifteen to eighteen degrees for the difference between 

 outside and inside measurements at the last end, it is seen that in 

 the returns there is an average loss of about forty degrees between 

 the farther extremity of the house and the heater, although tliis' 

 difference is not at all uniform. 



In the tomato and cucumber houses, C and D, heated by steam, 

 the figures used as before stated, are those obtained by disregard- 

 ing the observations on the mornings of February twenty-third, 

 twenty-fourth, twenty-fifth, twenty-sixth and twenty-ninth, when 

 there was practically no circulation in this run of pipes. Here 

 the average difference between ^the registering of the ther- 

 mometers inserted in the riser, and the first one outside, is only 



