1912] GRIGGS—RHODOCHYTRIUM 155 
The formation of the chromosomes, though it presents certain 
striking peculiarities, conforms in its essential features to the 
process usually found in dividing nuclei in other organisms. As is 
not unusual, spindle formation and chromosome formation, being 
in a sense unconnected processes, may go on side by side with a 
certain degree of independence, so that in two nuclei of the same 
age one may have the more mature spindle, while the other has 
advanced further in chromosome formation (figs. 52, 53). 
Spirem formation will be understood by a glance at fig. 51. 
Between those chromatin spherules which lie in the equatorial 
region of the nascent spindle there arise connecting bands of linin, 
forming an irregular spirem. At first the stains differentiate the 
chromatin and the linin elements, but in later stages the spirem 
stains homogeneously like other spirems. In the beginning its 
position may not be so definite, but as it contracts it comes to lie 
wholly within the spindle (fig. 52). After some further contraction 
it segments into chromosomes in the usual way (figs. 53, 54). 
Only a small portion of the chromatin of the primary nucleus 
is utilized in the formation of this spirem. On the dissolution of 
the nuclear membrane the remainder is cast out into the cytoplasm. 
There is no indication of any difference between those chromatin 
masses which are cast out and those which enter into the spirem, 
nor of any principle of selection other than that occasioned by the 
mere position of the masses which are utilized. Sometimes the 
masses of discarded chromatin persist for some time as deeply 
staining globules in the cytoplasm (fig. 22), but more often they 
lose their affinity for stains before the nuclear membrane breaks 
down and cannot be followed in later stages. 
During metaphase the spindle, which previously may have 
been shorter than the diameter of the nuclear cavity in which it 
lay (fig. 53), begins to elongate, piercing the membrane (fig. 54), 
and later, as the membrane weakens preparatory to dissolution, 
distorting the nucleus (fig. 55). The only anaphases seen were of 
the first type, occuring in the fourth mitosis. Apparently the 
chromosomes are drawn away from the equator in the usual way 
(fig. 56). No stages showing the formation of the membranes of 
the daughter nuclei were seen in spindles of the first type, but two 
