Proceedings of the Polytechnic Association. 995 



receives water from a separate division in the front chamber, formed 

 bj a diaphragm, into which the circulating tubes are screwed, and 

 which extends nearly to the water level. The Martin form of boiler 

 is probably the best that has yet been devised. The water tube system 

 seems to give the best results. The most recent form of boiler is 

 that in which the cliaraber of the water level is through a tube com- 

 ing from the inside to the diaphragm, it is much the same as the 

 Field boiler, with some modifications. 



Mr. John B. Root remarked that if circulation is to be carried out 

 to its legitimate results, we should obtain some satisfactory theory or 

 data from which the requisite degree and direction of the circulation, 

 the ?ize of water channels, &c., can be computed. The circulation 

 should be sufficient to prevent a steam bubble from forming on the 

 surface of the water. The size of the tube should be large enough 

 to carry a volume of water that would contain all the heat. His 

 theory was that the circulation should be in all cases from the heat- 

 ing surfaces of the boiler to or near to the water line, and back to the 

 heating surface ; that the water line should be far enough above the 

 heating surface to enable the amount of circulation to prevent entirely 

 the formation of steam bubbles upon the heating surface. The water 

 in contact with the heating surface being under a greater pressure 

 than that at the w^ater line, owing to the head of water above it, will 

 take a correspondingly greater temperature, and will begin to part 

 with its excess of heat, by forming steam, when it reaches that point 

 in the ascending circulating column where its temperature corres])onds 

 with the pressure ; the steam being formed throughout the body of 

 the water in suspension. If the steam be formed upon the heating 

 surface, tlie sediment will be there deposited, and the efficiency of the 

 surface diminished by the steam bubbles interposing between the 

 heating surface and the water. Kow supposing the steam to be 

 formed, about two feet below the water line, the temperature at that 

 point should be one degree higher than at the water line, and the 

 pressure one pound greater. Consequently, the amount of water 

 circulated over'the heating surface in a given time should be suffi- 

 cient, when heated one degree, to contain all latent heat of the steam 

 taken from the boiler in the same time. For instance, should the 

 boiler be working under sixty-four pounds pressure, temperature at 

 water line 300°, the latent heat of steam would be QIS'', the total 

 heat contained in the steam 1,212°, evaporation of water being one 

 pound per minute. The circulation necessary to convey the heat 



