at \ 
OS ee 
1903] MITOSIS IN PELLIA 41 
(fig. 19) like those of the two preceding mitoses are often 
present, but they are frequently absent, and the caps appear 
with only a few radiations (fig. 78) or even none at all. 
There are no radiations in the metaphase (fig. 7). In short, 
it is possible to select from the third mitosis a series of 
stages identical with a typical series from the apical region of the 
thallus. At the fourth and succeeding mitoses the resemblance 
to the usual vegetative divisions becomes more and more pro- 
nounced, while asters and centrospheres become correspondingly 
rare. 
THE CENTROSOME PROBLEM. 
The centrosome? problem is one of extreme difficulty, and 
perhaps the difficulty is greater for the botanist than for the 
zoologist. At least, the difficulties are different in the two cases. 
That there are in animals well-defined centrosomes which func- 
tion as organs of nuclear division, all investigators agree, and 
animals or tissues in which centrosomes do not occur are regarded 
as exceptions. The existence of the organ is not a serious 
problem ; rather, the more recent investigations have sought to 
establish the permanent or transitory character of an organ which 
all admit to be present during mitosis. In plants, on the other 
hand, even the existence of a centrosome is a problem which 
must be considered separately for the different groups. 
It is of interest to note that centrosomes in plants were first 
observed in diatoms in 1886 by H. L. Smith (31). When 
Guignard in 1891 published his classic paper on fertilization, 
botanists at once accepted the results and confirmatory accounts 
appeared. Strasburger (33) found centrosomes in Larix, Hum- 
Phrey (18) in Psilotum, Mottier (22) in Delphinium, Schaffner 
(26) in Alisma and Sagittaria, Campbell (3) in Equisetum, Lauter- 
born (21) and Karsten (20) in diatoms, and other investigators 
reported centrosomes in various forms ranging from the algae up 
to the flowering plants. In fact, the centrosome seemed to be 
as universally present in plants as in animals. Belajeff (1) and 
*In referring to flowering plants no attempt has been made to distinguish between 
centrosomes and centrospheres. In describing mitosis in liverworts some writers have 
used these terms indiscriminately. 
