106 THE PRINCIPLES OF FLORICULTURE 



to the glass in large houses than in small houses. This 

 difference in the ratio of glass to cubic contents is easily 

 shown by placing two houses of the same width and roof- 

 pitch together in ridge and furrow fashion, then compar- 

 ing the glass and cubic contents of the single house to 

 the glass and cubic contents of a house of twice its width. 



" There is another fact that upsets an attempt at exact 

 calculation of radiating surfaces in greenhouses. This 

 is, that as the atmosphere in a greenhouse is usually 

 humid, when the outside temperature is in the neighbor- 

 hood of zero, a sheath of ice covers the glass. This sheath 

 stops the openings between the laps and also changes the 

 rate of conduction through the glass so covered. It is not 

 possible to say at just what outside temperature this 

 sheath forms, as the formation of this ice is obviously 

 dependent on the degree of humidity inside the house, 

 the temperature in the house and the proximity of the 

 radiating surface to the glass." 



82. Comparison of amount of radiating surface re- 

 quired for steam and for hot- water. "In all calcula- 

 tions for the radiation of dwelling houses, it is usual to 

 consider the amount of hot-water radiation required, and 

 the amount of low pressure steam radiation required, as 

 being equal to the ratio of 1 to .6, or in other words, where 

 one hundred square feet of hot-water radiation would be 

 used, sixty square feet of low pressure steam radiation 

 would produce the same results. In the heating of green- 

 houses, it has been found that one linear foot of three and 

 one-half inch cast-iron pipe in hot-water heating will pro- 

 duce the same results as one linear foot of one and one- 

 quarter inch low pressure steam pipe, and the ratio of 

 hot-water surface to low pressure steam surface in green- 

 houses is, therefore, 1 to .4, one linear foot of three and 



