30 Scott and Brebner. — On the Secondary 
inner part of the bundle is mainly formed of the long 
elements, with the short cells scattered among them. 
So far the view that cell-fusion is the main factor in the 
development has been, as it seems to us, disproved ; that a 
very large amount of sliding-growth must take part in the 
development of the tracheides, is certain. It remains to 
consider the possibility that a relatively short syncytium, i.e. 
the product of the fusion of a few cells, might grow out to 
form a tracheide. This view might seem to present fewer 
difficulties ; the cell-fusions would be comparatively rare, and 
so much the easier to miss ; there would be fewer nuclei to 
fuse in each element, and thus the apparent absence of fusion- 
stages might be explained away. So far as the evidence 
from transverse sections is concerned, Roseler admitted the 
possibility that the facts might be accounted for if about five 
cells fused to form each tracheide, and the resulting syncytium 
grew out to six or seven times its original length. If this 
happened in the species of Yucca we investigated there would 
be about six parallel rows of fusing cells in each young bundle. 
This view is in reality as untenable as the hypothesis of 
entire development by cell-fusion. The single nucleus of 
the future tracheide is conspicuous throughout, until it is 
absorbed with the rest of the contents. If this nucleus arose 
by the union of several nuclei, some indications of this fusion 
must be found. Some of the growing tracheides observed 
were little longer than the desmogen-cells, so that there was 
no possibility of cell-fusion here. The idea of the outgrowing 
syncytium, though tempting in so far as it might seem to 
reconcile conflicting observations, must be rejected, from the 
entire absence of evidence in its support. 
The development of the tracheides was investigated by the 
same method of serial sections in Dracaena fragrans , Gawl. 
and D. angustifolia , Roxb. In these the secondary bundles 
are concentric, the xylem completely surrounding the phloem. 
The results entirely agree with those obtained in the Yucca. 
The material however was less favourable, as the zone of 
developing secondary tissues was narrower, and consequently 
