17 



The reactions with ferric salts, with potassixim bichromate, and vjith copper 

 acetate, exclude the presence of tarjiic or resinous substances. The stain 

 reactions obtained v;ith Sudan III and osmic acid do not malse known the pre*- 

 ^ence of the smallest y.u^ntity of oleose substance or the proteinaceous na- 

 ture of the body, considering the fact that even crystals of protein react 

 to these stains Q.) . 



Conclusions . The body in question, occupying the same position as the 

 Qlsioplasts, of a composition analogous to them, and being found, moreover, 

 in the genus Hippeastrum, where these latter $[the elaioplasts] were found, 

 I believe to be nothijig more than a form of their [the elaioplasts'] develop- 

 ment. 



Eucharis subdenta'^ Benth. 



So far as regards the study of the history of the develorment of the 

 elaiopla-sts in this genus, I have been unable to examine the floral axis 

 in process of development, or the bulbs in geminative activity, ana an 

 unable to make any pronouncement. 



I exciEiined solely the bulbs following the stage of bloom-ing and enccui>- 

 tered the elaioplasts in the external epidemiis of their scales. 



They occur immersed in the cytoplasm and are generally of notable di- 

 mensions. The color varies from yellow to brov.Ta .and in form the elaioplasts 

 are roundish or irregular. 



Often drops of oil may develop from their mass. Examining one by one of 

 the scales, I noted that in all cases they [the elaioplasts] are not of ir- 

 regular form and variable size and that in no case are they whollj' lacking. 



