Life-history of Tetraclinis articulcita. 581 
Tetraclinis , but a spindle becomes obvious about the time when the spireme 
segments, and on it the chromosomes arrange themselves, and, breaking up, 
begin to move towards the poles (Figs. 10, 11, 12). The chromosomes 
apparently become split at this time, since more than twelve may be counted 
in polar view (Figs. 13 and 15) passing to the pole, at about the stage of 
Figs. 12 and 14. This point will be discussed below. The chromosomes 
on arrival at the pole first form a rather ragged and deeply staining lump 
(Fig. 1 6), which soon opens out to form a well-defined daughter nucleus 
(Figs. 17 and 18). No trace of a cell-wall, either permanent or transitory, is 
seen at any time between the two daughter nuclei. The next figure (Fig. 19) 
shows a stage which does not appear to agree with that described for other 
plants. A spireme is organized and at once splits up, before any spindle is 
recognizable, into about twenty-four chromosomes (i. e. the 2 # number) and 
these become oriented on the spindle, when the latter becomes visible, and, 
without any division or split in individual chromosomes, separate into two 
groups of about twelve V-shaped rods, which pass to the poles in the usual 
way (Figs. 20 and 2t). It is worthy of mention that the two spindles of 
the ‘ homotype ’ division are invariably at right angles to one another, never 
parallel, as is said to be the case sometimes in other genera. 
As the four resulting nuclei reorganize, the spindles of the homotype 
division disappear entirely, giving the structure shown in Fig. 22. From 
this stage to the formation of the spores has been followed very carefully, 
in view of conflicting statements in other genera as to whether or not the 
mother-cell becomes chambered, and the stages obtained here suggest 
a possible explanation of those statements. In Tetraclinis no trace whatever 
of spindle fibres is found between the stages shown in Figs. 22 and 23 
respectively, but the cytoplasm cleaves into four equal parts, leaving between 
them a trace of residual cytoplasm. This residual cytoplasm is not always 
seen, but would probably be very quickly absorbed in the formation of the 
wall which is now laid down on each of the four microspores. 
In Juniper US' Nor6n ( 37 ) states that the four spores are formed in an 
unpartitioned mother-cell, but does not mention seeing any residual proto- 
plasm. Nichols ( 36 ), working on the same genus, maintains that the 
mother-cell becomes partitioned, as in Pinus , though the partition walls are 
not easy to demonstrate satisfactorily. It appears possible that what was 
seen by Nichols was the residual protoplasm. In preparations showing 
less contraction from the mother-cell wall than that seen in Fig. 23, it would 
be quite possible to interpret the appearance as due to very faintly staining 
partition walls, but if these were really walls one would certainly expect the 
cytoplasm to contract away from them, which is obviously not the case 
here. It is perhaps worth stating that although the phenomena just 
described can be successfully demonstrated in a good triple-stained Canada- 
balsam mount, yet they can be seen much more readily after the violet and 
