in Dictyota dichotoma. 187 
of chromosomes be known also in the vegetative cells of 
these generations. 
For the solution of these problems a thorough and pro¬ 
longed study at the sea-shore is necessary. 
Summary. 
When we consider the facts presented in the foregoing 
paragraphs concerning the development of the karyokinetic 
figure, it will be seen that in Dictyota the nuclear spindle 
originates in two systems of kinoplasmic radiations or asters 
that lie close to the nuclear membrane and some distance 
from each other, generally on opposite sides of the nucleus. 
The radiations of each system are centered upon a very 
distinct rod-shaped body, the centrosome. For the sake of 
convenience I have spoken of the centrosome together with 
its radiations as a centrosphere. The centrosome is present 
during the two nuclear divisions which take place in the 
tetraspore mother-cell ; it is also found in the germinating 
tetraspores, and its persistence has been observed in the first 
three or four cell-generations of the resulting seedling plant. 
In all vegetative cells of the thallus bearing tetraspores, the 
centrosome is readily recognized during karyokinetic activity. 
In the beginning of mitosis it divides longitudinally, and the 
daughter centrosomes with their radiations, or we may say 
the daughter centrospheres, move apart along the nuclear 
membrane to positions occupied by the poles of the future 
spindle. During the process of karyokinesis, the centrosomes 
may undergo a change in size, appearing smaller in the 
equatorial plate stage. 
The fact of the persistence of the centrosome from one 
cell-generation to another, and of its multiplication by division, 
is certainly strong evidence in favour of its morphological 
unity as an organ of the cell. In the light of what is known 
in the higher plants, and owing to our limited knowledge 
of karyokinetic activity in the lower plants, I do not think we 
