SOMATIC CHROMOSOMES IN TRADESCANTIA 
349 
It may here be questioned whether the vacuoles and open spaces which 
appear and spHt the thin prophasic thread may not be at least partly 
retentions from the preceding telophase, the split therefore being initiated 
in the telophase after all. This is a question which it has not been found 
possible to answer in many cases, since the changes in question occur in 
very minute structures which cannot be interpreted at all in any but the 
most favorable preparations. There can be no doubt, however, that the 
threads for considerable distances are actually single so far as the microscope 
will allow us to determine, and that many new vacuoles and open spaces 
develop where none were visible before. On the other hand it seems very 
probable that a few spaces of the earlier prophase, and hence of the preceding 
telophase, may occasionally persist in the heavier portions of the threads 
as they develop from the reticulate bands, such spaces if properly situated 
being incorporated in the later true split. But only in a very strained 
sense could such occasional vacuoles or spaces be said to constitute the 
initial stages of the split; their relation to it appears to be fortuitous 
rather than determinative. 
It may also be questioned whether the single thread stage of the pro- 
phase, upon the importance of which the writer has insisted, is a phenomenon 
of general occurrence or is a special process peculiar to a few types of cells. 
To this question also no full answer can be given at present. It surely 
occurs in the root cells of Vicia and Tradescantia in spite of the fact that 
it does not appear in the descriptions given by other investigators of mitosis 
in these plants. It is also represented in Miiller's (1911) figure 9 of Najas 
marina, and in the "spiral threads" figured by Bonnevie (1908, 1911, 1913) 
for Allium, Amphiuma, and Ascaris; by Wilson (1912) for certain insects; 
and by several other investigators. How much more widespread it may 
be cannot be stated, especially since the prophasic changes have been fol- 
lowed with sufficient care in so relatively few cases. 
On the contrary, it is not impossible that in some forms the split spirem 
may develop directly from the alveolar-reticulate bands of the earlier 
prophase by a rearrangement of the vacuoles and openings to form a single 
median series as the structures become more slender, but the writer is not 
convinced that such a process has been demonstrated in any instance. 
Even if, for the sake of argument, it were assumed to occur, it would not 
necessarily follow that the telophasic vacuolation should be regarded as a 
splitting or that the chromosomes of the late telophase, resting stage, and 
early prophase should be regarded as double when the vacuoles and openings 
have such an arrangement as that described for Vicia and Tradescantia. 
Chromosomes in this condition are not double in any sense of the word, 
even though they contain open spaces which may later join with others 
to form a split. The chromosome can be said to be ''double" or "split" 
only when its substance has been separated into two distinguishable portions, 
either by the rearrangement of the vacuoles and spaces as provisionally 
