18 



steam-pipe's or other heating agency 'employed, and on the rate at 

 which it passes over them. Of these, the two last factors may fairly 

 easily be kept constant, but the first-named may vary very consider- 

 ably in the course of an ordinary working day. It is obvious, there- 

 fore, that the amount of adjustment required will be much greater 

 in this type of machine than in the first-mentioned -type. Probably 

 the easiest way to make the necessary adjustments is either to regulate 

 the current of air so that the quantity of heat it carries in a given 

 time is always the same, or to keep the current of air constant and 

 regulate its temperature by mixing an adjustable quantity of cold 

 air with the air from the heating chamber, which should be slightly 

 over-heated. The regulation, if possible, should be automatic. 



As regards the cost of running hot-air machines, in those in 

 which steam is used as the heating agent, the cost may be divided 

 under three heads : , 



(1) The steam used in heating the seed. 



(2) The steam lost through radiation and other causes. . 



(3) The power necessary to keep the machine in motion. 



The first of these will vary according to the initial temperature 

 and the quality of the seed, but is quite independent of the type of 

 machine used. Calculation shows that to raise the temperature of 

 one ardeb of fair commercial cotton seed from 15 to 55 C. with 

 steam at 60 Ibs. pressure requires about 8-2 Ibs of steam. 



The quantity of steam lost will probably vary very considerably 

 in different machines. It should, however, always be very much 

 less than the above figure or the machine should be condemned 

 for inefficiency. In a trial with Simon's machine carried out by 

 Mr. Singleton, of the Mechanical Service, the steam consumption 

 was found to be 11-4 Ibs. per ardeb, showing that 3-2 Ibs. or 28 per 

 cent of the total steam used was lost. 



With the last item of expense, if all the different types of machines 

 are considered, the variation will be very great. In those machines 

 in which the seed is heated by ^direct contact with steam-jackets 

 or steam-tubes, however, the power necessary should be fairly uniform, 

 since in all cases it is the power required to keep the seed in a reason- 

 ably rapid state of motion. One horse-power for every four or five 

 ardebs treated per hour is probably an average figure. 



One other point of general interest to prospective manufacturers 

 should be mentioned. Owing to the great undesirability of mixing 



