.382 Prof. J. R. Green. On the 



soda solution, and these were dialysed for two days in freshly testel 

 dialysers. I expected to find that, instead of dialysing intact, there 

 would be a decomposition, and that the fatty acid would be left 

 behind, while the alkali escaped. Contrary to this expectation, the 

 -dialysates in all cases gave a marked opalescence, or a curdy preci- 

 pitate, with HC1, the ricinoleic acid being liberated thereby from 

 soap which had passed the dialyser. Subsequent careful examination 

 of the dialysing tube proved it to be intact. The experiment, though 

 not without interest, does not throw any light on the mode of 

 .absorption of the acid, for, on neutralising or making slightly acid, 

 the soap is decomposed, and the fatty acid liberated. As the reaction 

 of both endosperm and cotyledon is acid, it is clear that the fatty 

 acid does not pass from the one to the other in the form of a soap. 



The occurrence of the starch must be similarly explained. It 

 results, as in other plants, from the transformation of the sugar 

 which has been absorbed. In sections of the hypocotyledonary 

 portion of the axis I found some roundish bodies, which when heated 

 with iodine stained brown. With a ^ objective these brownish 

 .bodies were seen to contain small crescent shaped bodies which 

 were dark blue. Some contained t\vo, others three or four, of these. 

 There is little doubt that these were amyloplasts, containing starch 

 grains of very small size, but in course of formation. I did not succeed 

 in identifying these in the cotyledons, though starch appeared there in 

 small grains. Gris* has figured bodies exactly corresponding to 

 these, and he says he finds them in the cells of the epidermis of the 

 cotyledon. 



An examination of the relative quantities of these different 

 conditions of the fatty reserve materials present at different stages 

 of the germination confirmed the view given above. The disappear- 

 ance of oil and coincident increase of sugar have already been 

 commented on. The relations between the oil and the fatty and 

 crystallisable acids were separately determined. Some seeds were 

 germinated in an incubator, and samples were examined at intervals 

 of twenty-four hours from their being sown. Care was taken to have 

 all the seeds of about the same size, and as much alike as possible. 

 Two sets of experiments were made, in one the quantities of the 

 different constituents of the whole seed and resulting plant being 

 examined in the other those in the endosperms alone. 



Each sample was crushed and extracted by ether, the extract 

 being evaporated to dryness, and the amount of residue roughly 

 estimated. This residue was then stirred well with water, which was 

 syphoned off, and added for twenty-four hours to the remains of the 

 crushed endosperms now freed from the ether. The extract so 

 obtained was examined for crystallisable acid and for sugar. The 

 * ' Ann. des Sci. Nat.,' Ser. 5, Sot., vol. 2,1364. 



