By Mr. Henry Stothert. 



203 



convenient spot for draining, the condensed water is taken away by 

 a small syphon, about three or four inches deep, as shewn in the 

 end of Fig. 13. A steam pipe of one inch diameter is sufficient 

 for a case of fifty feet in length, and if proper attention be paid to 

 the dimensions and distance of the holes, which in this case need 

 not be above one-sixth closer at the farther end than at the com- 

 mencement, scarcely the least difference of temperature will be 

 perceptible at each end of the case, an effect utterly unattainable in 

 the best constructed fire-flue, which, in appearance, it so much re- 

 sembles. There is, however, no particular proportion of the height 

 to the breadth, that depending entirely on convenience. 



Where free-stone cases are used, it is found necessary that they 

 should receive two or three coats of linseed oil, to prevent the 

 escape of steam through them. It is better to give moisture to the 

 house by steam-cocks fixed at the top of the cases, as shewn in the 

 drawing, as humidity can then be regulated at pleasure. 



This mode of heating is extensively applied at Mr. Miller's 

 Nursery, Clifton, where the atmosphere of seventeen houses, con- 

 taining upwards of 12,000 superficial feet of glass are heated by it. 

 A Plan, Plate 8, shewing the dimensions and position of the houses 

 is annexed, describing also the position of the cases, and likewise 

 those beds that receive bottom heat in the manner described at 

 Fig. 3 and 4, and the cistern for aquatic plants at Fig. 6. 



In conclusion, I beg to state that the principles upon which we 

 are induced by daily experience to calculate the size of boilers are 

 as follows : 



1. For forcing houses under ordinary circumstances, we find that 

 one square foot of bottom surface of boiler is equivalent to 150 square 

 feet of glass. 



2. W here bottom-heat is required, one square foot to 135 feet of 

 glass. 



VOL. I. 2nd series. E e 



