646 Beer. — Studies in Spore Development. III. 
same relative positions as they did during the telophase of the previous 
division, when their outlines became lost to view owing to the development 
of numerous branches and cross connexions with one another. 
It is especially to be noted that the spireme which develops from the 
reticulum forms at no time a continuous thread. From the beginning it is 
segmented into separate lengths which at fir.st are long and slender, but 
which become thicker, shorter, more deeply staining, and more curved as 
mitosis advances. These facts are shown in Figs. 3, 4, and 5. The last 
of these figures shows that the definite polarization of the chromosomes 
has eventually become entirely lost, and that the curved filaments composing 
the segmented spireme coil in every direction through the nuclear cavity. 
Several conspicuous plasmosomes still remain enclosed within the coils of 
the spireme. 
It will be seen that in the development of this spireme there is no 
concentration of the filaments by the approximation of longitudinal halves 
of previously alveolized chromosomes such as has been described in a number 
of other plants by several writers. 
Indeed, the spireme, both when fully formed and during its development, 
appears in Equisetum to be quite unsplit longitudinally. Each chromosome 
appears in this plant to develop during the prophase solely by the drawing 
in of the anastomoses which had been formed at the close of the preceding 
division and by the concentration of the long, slender, but homogeneous 
filament, which first appears, into a shorter and thicker structure. 
The nuclear membrane then disappears and the coil of chromosomes, 
together with the nucleoli, lies free in the cytoplasm. The chromosomes 
next become drawn upon the equator of the spindle which has developed in 
the usual manner. The chromosomes appear at this time as rather long 
and not very thick bodies which are bent into a longer arm directed towards 
the pole of the spindle and a shorter arm arranged upon the equator of this 
structure. 
Soon after the chromosomes have become regularly arranged upon the 
spindle-equator they can be seen to be longitudinally divided (Fig. 6). As 
the longitudinal halves begin to separate from one another they not in- 
frequently remain adherent or closer together at their distal ends whilst 
they are already quite widely apart at their proximal extremities. In 
these cases loop-like structures are formed such as I have represented in 
Fig. 7. The nucleoli can for some time be seen lying in the cytoplasm in 
the neighbourhood of the chromosomes, either among the spindle fibres or 
free from the mitotic figure, but as the daughter chromosomes begin to move 
apart the nucleoli are lost sight of. 
In Fig. 8 I have represented a stage of the anaphase. It will be seen 
from this that the premeiotic spindle forms a blunt 1 structure, which differs 
1 It is of course possible that the blunt spindle may be due to the removal of the actual spindle 
