HYGROSCOPICITY 1 73 



(mean temperature 18° C, or 64 to 65° F.) there was an 

 increase in weight of 22-5 per cent., and under cool conditions 

 (mean temperature about f C, or 41° F.) an increase only of 

 4-1 per cent., whilst with seeds exposed in turn to both the 

 warm and cool conditions there was an increase of i6*6 per cent. 



One is inclined to think that condensation occurred in the 

 experiment under warm conditions. The seeds were placed 

 in a porcelain vessel standing in water and covered with a 

 glass globe, the range of temperature being 13° to 21° C, or 

 ^^° to 70° F. With a range of about 15" F., condensa- 

 tion is quite possible ; and if it occurred at night it might 

 not be evident in the daytime. We are told that the occur- 

 rence of mildew brought the experiment to a close, and that 

 the seeds were still adding to their weight at the end. The 

 same method was adopted with all the flax-seed experiments, 

 and also with the seeds of the Kohl-rabi (Brassica okracea 

 caulorapa\ which increased their weight 23-5 per cent. It is 

 to be noted that this investigator was not inclined to accept 

 Hoffmann's much smaller estimates for the hygroscopic increase 

 of weight of the seeds of Linum and Brassica^ namely, 4*7 and 

 4*6 per cent, respectively. 



The trend of my own results is all in the direction of 

 Hoffmann's smaller determinations of a seed's hygroscopicity. 

 I should imagine that flax-seeds {Linum)^ containing as they 

 do so much colloid material capable of combining with water The danger 

 in the form of mucilage, would give results far from typical. such°experi- 

 But of more importance still is the fact that mould or mildew "i^nts. 

 interfered with both the experiments on flax-seeds made by 

 Nobbe, where the gain in weight reached i6|- and 11^ per 

 cent. ; and in neither case had the increase ceased when the 

 experiment ended. The danger in all experiments where 

 seeds are exposed to air saturated with moisture is the develop- 

 ment of mould, which, on account of its hygroscopic properties, 

 would go far to vitiate the results. Its onset is insidious ; 

 and, as was often indicated in my drying and similar experi- 

 ments in the tropics, long before the fungus became evident 



