- 26 



successive sacks were as follows : 5.")-,"). .">.">, 56, 54, 54, 54, 54, 55-5, 54, 

 53-5, 54, 54-5, 53, 53-5, 54, 54, 54, 54, 53 C. 



This represents a variation of 3 in the course of treating nineteen 

 ardebs. 



With the low quality seed the temperatures wore much higher, 

 there being no cause for fear with regard to the germination of the 

 seed, and not so regular, probably owing to the variability of the 

 seed. The temperatures of successive sacks were : 59, 56-5, 60, 63, 

 64, 63, 62-5, 59 C. 



Leaving out of account the first two sacks, which can never 

 be relied upon, this gives a range of 5 < '. 



Examination of the treated seed gave the results shown in 

 Table III. These may be summarized as follows : 



With the good seed, treated at temperatures from 53 to 56 C.. 

 the average mortality of the worms was 99-6 per cent and the average 

 germination of the seed was 85 per cent, that of the untreated seed 

 being 82 per cent. 



With the bad seed, treated at temperatures from 56-5 to 64 C., 

 the mortality of the worms was uniformly 100 per cent, and the 

 average germination of the seed was 49 per cent, that of the untreated 

 seed being 48 per cent. 



These results are sufficiently satisfactory, showing that the 

 machine has a range of at least 12 C., over which it kills practically 

 all the worms without damaging the seed, and as it can be made to 

 run with a variation of not more than 5 C., there should be no difficulty 

 in getting uniformly satisfactory results. 



Experience with this model, however, has already suggested 

 some improvements. In the first place, it is a little difficult at present 

 to get a high enough temperature with a reasonable steam-pressure 

 when treating seed of a good qu'ality. This can easily be remedied 

 by increasing the area of the steam-jackets, which at present only 

 surround the lower half of each cylinder. In the second place, the 

 automatic control would be improved if it were made to work on 

 the steam-pressure in the third cylinder only. It may safely be 

 taken for granted that the seed will not be overheated before leaving 

 the second cylinder. If, therefore, the steam in the third jacket is 

 cut off entirely as soon as the temperature rises above 57 C. and opened 

 again as soon as it falls below that point, the temperature of the seed 



