184 Alfred J. Ewart: 



germ ina don. This is, however, incorrect, and according to 

 Becquerel, only those seeds can preserve their vitality for long 

 periods of time which are protected by thick coats impe<rmeable 

 to water and oxygen, and which have feebly oxidizable reserve 

 materiails. This is, however, only a partial truth. It is a fact 

 that " starchy " seeds last longer than oily ones, on the whole, but 

 reference to the lists will show that many seeds containing appre- 

 ciable quantities of oil are comparatively long lived (Myrtaceae, 

 etc.) The truth is thait long-lived (macrobiotic) seeds have 

 mainly been developed among the Leguminosae, an order charac- 

 terised by its starchy reserve food materials. Many more or 

 less endospermio Leguminous seeds (Melilotus, Cassia, Cytisus, 

 Lotus tetragonolohus, etc.), are macrobiotic, though incapable of 

 the same dura/tion as the non-endospermic seeds of Gondia and 

 Acacia. 



Longevity depends not on the food materials or seed coats, 

 but upon how long the inert proteid molecules into which the 

 living protoplasm disintegrates when drying, retain the molecular 

 grouping which permits of their recombinaition to form the active 

 protoplasmic molecule when the seed is moistened and supi>lied 

 with oxygen. Chemical changes in the food materials might 

 hasten the irrevocable disintegration which the slow process of 

 time brings about, and in the same secondary manner the 

 character and structure of the integuments will be of importance. 

 The property of longevity is an hereditary peculiarity inherent 

 in the protoplasm of certain seeds, and developed by natural 

 selection as an adaptation to special conditions of life. Accord- 

 ing to Acton, samples of old wheat contained more soluble 

 matter and less water than fresh wheat, while the diastatic and 

 proteolytic ferments present in fresh wheat had entirely disap- 

 peared. This is, however, the result rather than the cause of 

 death, since as long as the power of reconstituting the living 

 protoplasmic molecule is present, the i>ower of reproducing fresh 



ferments is retained. 



Maci'obiotic seeds are all seeds which show no special adapta- 

 tions for dispersal. None are wind or water borne, and although 

 some are more or less accidentally distributed by animals, 



1 A full investisjation of this question is in progress. 



