386 Agricultural Experiment Station, Ithaca, N. Y. 



which is 10x17 and 12 feet higli. This was connected with the city 

 gas system. Including the connections in the cellar (40 ft.), this 

 little heater was expected to heat 220 linear feet of inch pipe with 

 hot water. A steam heater is not practicable for such a small 

 area. 



The gas was first lighted one December day when the pipes were 

 cold. An hour was required to thoroughly warm up the system. 

 In ordinary snug winter weather (thermometer outside 10° to 15°), 

 the apparatus consumed one-half cubic foot of gas per minute to 

 keep the house at a proper temperature for conservatory plants. 

 The system worked perfectly. jS^ot one thing more could be de- 

 sired — except cheaper gas. A very slight increase in the amount of 

 gas — supplied by simply turning a valve — was sufficient to make a 

 very rapid change in the temperatures of the pipes, so perfectly 

 was the system under the control of the heater. So long as the 

 weather was running nearly uniform, the heater demanded no 

 thought or attention. It is the perfection of a lazy man's machine. 

 The readiness with which the system responded to more or less gas 

 may be indicated by the following test. When the experiment 

 began the apparatus was consuming one cubic foot of gas every 2|- 

 minutes. Thermometer No. 1, on the outside of the riser at its 

 highest point 58 feet from the heater, registered 94° and ther- 

 mometor No. 2 on a return 70 feet from the heater, registered 68°. 

 Gas was turned on until a cubic foot was consumed every 1^ min- 

 utes. The temperatures went up as follows : 



Thermometer 2, 

 degrees. 



7 minutes 

 10 minutes 

 14 minutes 

 20 minutes 

 25 minutes 

 31 minutes 

 40 minutes 

 45 minutes 

 60 minutes 

 75 minutes 

 85 minutes 

 97 minutes 



