56 WOECESTER COUNTY HORTICULTURAL SOCIETY. [1891, 



1,200 feet of 4-inch hot water pipes, heated by one of Smyth & 

 Lynch's largest boilers (No. 7), which is guaranteed to heat 1,600 

 feet of 4-inch pipe. I think that most of the lettuce houses 

 built within the last few years are heated by steam. It is claimed 

 for steam, at least by those who make steam heating apparatus, 

 that you can heat a house quicker, and that one can better con- 

 trol the temperature. 



Some three years ago, Mr. Budlong of Providence, I am told, 

 tore down and rebuilt his four large lettuce houses, varying in 

 length from 150 to 500 feet, and is heating the new houses with 

 steam. Probably the principal reason for displacing his old houses, 

 was to substitute larger glass for his 6x8 light hot-bed sash with 

 which his old houses were covered, and which he could utilize 

 quite as well or better on his hot-bed frames. Whether his new 

 steam heating arrangement is an improvement enough over the 

 water heating, to warrant so expensive a change, I do not know. 

 But I very much doubt if it is. 



I base this opinion upon the results of a careful, comparative 

 test of hot water and steam heating for lettuce growing, made 

 by the Hatch Experiment Station, and published in their Bulle- 

 tins, Nos. 4, 6, and 8, in which every detail of the experiments 

 are minutely described, and the daily and final results are tabu- 

 lated. 



Two houses were constructed during the summer of 1888, 

 75 X 18 feet, as nearly alike as possible in every particular. Two 

 boilers of the same pattern and make were put in, one fitted for 

 steam and one for hot water. Their first published test showed 

 that the hot water gave the best results, and at a saving of cost 

 of about twenty per cent. 



Much discussion having been provoked relative to the results 

 of that experiment, and especially as to the accuracy of those 

 results, last winter, 1889-90, they made a " careful repetition of 

 the experiments to correct any errors that might be found and to 

 verify previous results, the boilers having been run with the 

 greatest care possible from Dec. 1, 1889, to March 17, 1890," 

 and, as before, the temperature was taken five times every twenty- 

 four hours and the coal used in each house daily, weighed. This 

 experiment confirmed the results of their former experiment, as 



