168 Agricultural Experiment Station, Ithaca, N. T. 



average temperature is eleven degrees higher in the upper one of 

 these pipes, showing that the return water preferred that pipe. 

 The riser for house G passes under the benches of house F. Com- 

 mon wrought-U'on pipes are used throughout the establishment. 



The various character of these houses necessitates many elbows 

 and fittings in the piping and the fall is slight; these are con- 

 ditions to which water is not adapted. But as these are the con- 

 ditions which are usually present in commercial establishments, 

 our results must have an intensely practical bearing. Steam ovei'- 

 comes difficulties much more readily tlian hot water. Our hot- 

 water system is put in after the most approved fashion, and the 

 following tests concern the actual merits of the two systems under 

 practical conditions. In perfectly straight and simple runs, the' 

 results might be very different. 



The observations were made by inseTting thermometers near the 

 boiler inside the riser and return of each series of houses, and 

 placing others on the outside of the risers at the extreme ends of 

 each series. Table I shows the temperatures at all these points, 

 moiTiing and night, from January 18 to February 29, 1892, 

 together with the steam pressure, if any, the outdoor temperature, 

 and the temperature of each of the houses. It should be borne in 

 mind that houses A and B are purposely kept cool, for growing 

 plants requiring a low temperature. The ventilators were often 

 open, as noted by the letter V in the table, while the dooi-s 

 between the cool workroom and both steam series were generally 

 open. These conditions would lower the avei-age temperatui-e 

 shown by these steam-houses. The thennometers used for the 

 inside temperatures were let into the interior of the steam pipes 

 through a fourth-inch tap or hole, and were made tight by a tilling 

 of putty; they remained permiinently in this position during the 

 whole experiment. These were glass thermometers of very deli- 

 cate make. Those used on the outside of the pipes were placed so 

 that the bulb rested on the pipe, the bulb being imbedded in a 

 ball of putty to offset the effect of air temperatures as far as pos- 

 sible. These were placed on the risers at opi)osite ends of the 

 houses, and are the ones corresponding to letters A, B, C, D, E, F, 

 in the following table. These thermometers stood in the houses 



