164 BOTANICAL GAZETTE [wance 
dle, formed at the center of the cell, always lies as in ordinary . 
cases, 
It is exceedingly difficult to detect the bodies representing 
centrosomes in Limodorum. Multipolar spindles are very fre 
quent, and in general their branches are very numerous and very 
variable in appearance. At the extremity of many of them, 
however, there is seen a granule, or a small mass of substance 
more stained than the rest of the cytoplasm. During the forme 
tion of the bipolar spindle, those granules nearest the pois 
subsequently occupied by the two final poles, points determined 
y the form of the cell, seem to persist, the others disappearing. 
It can be supposed that several isolated bodies sometimes unilt 
to form a single centrosome, for, after the formation of the 
bipolar spindle, bodies are often seen at each pole. A distinct 
sphere was not observed, but time after time very delicate radiae 
tions from each pole were seen. : 
is none the less certain that the higher plants possess differe 
tated elements whose réle is the same as that of those analogots 
bodies observed in the lower plants and in animals. 
Paris, 
