Meiotic Mitoses of Osmunda . 155 
others, has already split into two parallel rows of beads. A nuclear limiting 
membrane makes its appearance (Fig. 99), the skeletons of the quondam 
chromosomes become less and less recognizable as their beaded sides 
separate, and thus inaugurate a reticulum (Fig. 99). The beads gradually 
become distributed throughout the nucleus (Fig. 100), which thus passes 
into the fine reticulate, or resting, stage found in the cells of the spore 
tetrad. 
The predominating feature of the homotype division lies in the pre- 
cocious separation of the daughter chromosomes, suggesting weak cohesion 
between them. Possibly some physical conditions associated with the 
reception into the nucleus of entire premeiotic chromosomes during the 
previous telophase of the heterotype division may be responsible for this 
phenomenon. 
Synangia. 
Five cases of synangia have been found, three in O. palustris , var. 
aurea , and two in O. palustris , var. undulata. Of these one is in arche- 
sporial division (PI. XII, Fig. 102) the others in the first meiotic division 
(Figs. 103, 104, 105, 106). One side of the synangium may have its spore 
mother-nuclei in a considerably more advanced stage than the others 
(Figs. 105 and 106). 
It is interesting to note that Bower ( 1 ) has described synangia as 
occurring not unfrequently in O. regalis (PI. Ill, Fig. 5 1 ) 5 he has also 
found them in Gleichenia (PL I, Fig. 31), and mentions that they may be 
observed occasionally in Todea barbara . 
General Considerations. 
No attempt will be made to touch even the fringe of the vast literature 
which has grown up round the hotly debated question concerning the mode of 
origin of heterotype chromosomes. It is merely proposed (1) briefly to 
recall the main differences in the views held by telosynaptists and para- 
synaptists, (2) to suggest a possible explanation for these different views by 
showing how variously the details of mitosis may be modified, (3) to con- 
sider in particular the evidence afforded by the mitoses of Osmunda. 
I. The Main Differences between the Telosynaptic and Para- 
synaptic Views . 
Farmer (6) has drawn attention to the confusion which has arisen 
round the terms * telosynapsis J and 4 parasynapsis ’, and sums up the posi- 
tion in a few cogent sentences. He shows that by emphasizing a point of 
comparative unimportance, these terms ‘ have led to a misconception on the 
part of many people, as to the really fundamental differences which still 
divide the two schools of investigators \ There is no essential difference 
between a lateral approximation achieved by the twisting together of the 
sides of a loop, ‘ and an approximation produced by the coming together 
