Endosperm of Hordeum vulgare during Germination. 13 



granule as " sub-alenronic " and " sub-scutellar " respectively; for 

 although occasional instances may occur where one form of attack 

 merges insensibly into the other, yet, looked at generally, they differ 

 so much from each other as to suggest that the transforming agents 

 are essentially different. 



The accompanying photographs (Plate 1) illustrate these differ- 

 ences far better than can any mere description. 



It appears to us that the phenomena which are observed when the 

 endosperms of Hordeum are deprived of their embryos, and are 

 treated in the manner we have described, must be attributable to one 

 or more of the following causes : 



1. They may be the result of micro-organisms originating in the 

 culture-medium, and gradually invading the endosperm- tissues,, 

 which undergo progressive alteration either by the direct action of 

 the organisms or in .virtue of their secreted enzymes projected into, 

 the endosperm. 



2. The phenomena may be due to residual enzymes, cjtohydro- 

 lytic, amylohydrolytic, and proteohydrolytic, left in the endosperm at 

 the time of maturation and desiccation of the grain. 



3. They may be due to the revival of metabolic activity of still 

 living cells of the endosperm when these are placed under favourable 

 conditions of moisture and temperature, and facilities are afforded for 

 the removal of the products of change. If this is the correct solution? 

 the active cells may be those of (a) the " aleurone-layer," or (6) the 

 amyliferous portions of the endosperm. 



We must now consider these three possibilities in detail. 



We have already stated that, no matter how careful we may be in 

 sterilising the apparatus and culture-medium, the appearance of 

 micro-organisms is only a question of time, unless we employ anti- 

 septic methods of so drastic a nature as to seriously imperil the 

 vitality of the endosperm-tissue, a course which would render it 

 impossible to get the answer we require as to the respective parts 

 played by organisms and by autonomous changes in the endosperm- 

 cells themselves. We can, however, arrive at certain conclusions by 

 making a large number of experiments and by confining our observa- 

 tions to the period prior to the appearance of organisms, a period 

 which, under favourable circumstances, may extend to about eight 

 days. When this is done we find that the changes originating in the 

 first place under the "aleurone-layer" of the degermed seeds so far 

 precede in point of time the appearance and multiplication of the 

 Bacteriacece- and moulds as to render it in the highest degree improb- 

 able that the two sets of phenomena are causally related to each 

 other. 



A much more satisfactory proof of the truth of this proposition 

 may be obtained in an entirely different manner. Endosperms of 



