Himanto phyllxun miniatxim Groenl., H. Cyr tanthiflorum Groenl. 

 In these species I had occasion to study, in addition to the vegetative 

 orgaais, also floral organs. With the aid of methods indicated by WaKker, that 

 is, plasmolyzation of the cell, I was easily able to demonstrate that the 

 elaioplast could be foiond in the cytopla.sm. In this case a better reagent is 

 a 15 per cent aqueous solution of potassium nitrate, with the addition of a 

 small quantity of eosine. with such a solution, which has a marlssd power of 

 disintegration, I brought ab-at the plasmolysis and the contraction of the 

 interior protoplasm from .vhich the elaioplasts separated, which remained for 

 a long time visible and maintained their prir.iitive fonm. In the cells of 

 the ovary, in process of development, the elaioplasts in the younger stage 

 appeared as little aggregates of rotund or oval bodies endovved with strong 

 refractility and frecuently were close to the cell nucleus. Each group did 

 not exceed the size of the cell nucleus. 



Later, however, when the flowers have reached aLnost their ultimate 

 size and begin to open in the epidsrrnai csllo of the cvary, which ere 

 preferably chosen for the study of the development of the elaioplasts, these 

 will wax in volijme and in number and each group conse:^r.3:itly becomes greater 

 but never attains to the dimensions cf th3 cell nucleus. In this stage, 

 which may be the culminating point of their development, the elaioplasts 

 begin to lose thsar original fcrm and prirsitive structvjre and show phenomena 

 which, I believe, are to be rsgarded as de^eneratix'e. 



On. the contrary, all the smallest oval or rotund bodies appear at 

 first to flow together, fom'-ins a single mass. The lusicr. takes place 

 graidually in a ir.araier -.hich may fcllo.v all the nirrjits phases of the total 

 disappearance of the charactfristic str-ac:ure of the elaicpls-st. Tnis mass, 



