384 Agricultural Experiment Station, Ithaca, N. Y. 



It will be seen from table A that three distinct tests, of various 

 duration, were made with the low tank or low pressure, and from 

 B that three short tests were made with high pressure. In drawing 

 conclusions from all the tables, the temperatures of the house should 

 be discarded, and the efficiency of the different trials should be 

 deducted from the temperatures of the pipes as shown by ther- 

 mometers 1, 2, 3, 4 and 5. This is because the temperature of the 

 house was kept as uniform as possible by ventilation, so that as the 

 heat rose in the pipes, when steam or water under greater pressure 

 was used, the ventilators were opened wider. 



By making a general total average of the various pipe tempera- 

 tures in the three systems, A, B, and C, we have the following 

 Hgures : 



D. SUMMAEY OF THE AvERAGE TEMPERATURES IN THE YaRIOUS 



Thermometers. 



It is seen at once that the lowest average efficiency is in the lower 

 pressure water system, the next best is with the high tank, while 

 the highest is with steam. That is, the higher the exj^ansion tank 

 above the heater, within reasonable limits, the hotter the water 

 becomes because it is under greater pressure. This increase of heat 

 was observed in all parts of the system, as shown by the uniformly 

 higher averages in the five several thermometers. The system was 

 also more easy to run, the circulation was moi*e uniform in all the 

 pipes, and its general efficiency was seen to be greater by the work- 

 men who had charge of it. With the greater elasticity and less 



