26 
Scott and Brebner. — On the Secondary 
If we include them in the xylem-parenchyma, as Roseler 
appears to have done, we obtain figures which are not very 
different from his We shall return to these numbers later on ; 
it is sufficient now to point out that, leaving the sheath out of 
consideration, more than half the elements in the mature bundle 
as seen in transverse section, are tracheides, while in the xylem 
they form about three-quarters of the whole number. In a tan- 
gential section passing through the middle region of a bundle 
the average number of tracheides cut through is about six. In 
such a section not more than two parenchymatous cells are 
likely to be met with at any one level. The tracheides in 
this species attain an average length of 2-78 mm. (mean of 
twelve measurements). The mean length of a cell of the 
secondary desmogen is -o 75 mm. If therefore the entire 
tracheide is formed by cell-fusion, a series of thirty-seven 
desmogen-cells must, on the average, fuse to form each 
tracheide. If, on the other hand, the tracheides are formed 
by longitudinal growth alone, then each desmogen-cell which 
forms a tracheide must, on the average, grow to thirty-seven 
times its original length. The whole of this elongation would 
be by ‘ sliding-growth,’ as we are speaking of a region in which 
the stem as a whole has long ceased to grow in length. On 
the hypothesis of cell-fusion we should find, in the developing 
bundle, as many rows of fusing desmogen-cells as there are 
tracheides at maturity, i. e. about thirty-six rows in the entire 
strand, or about six in any tangential section passing near the 
middle of the bundle. If, however, the development is by 
sliding-growth, we should find only one cell, on the average, at 
each level, in the whole strand, elongating in order to become 
a tracheide. A thin tangential section could of course only 
occasionally pass through one of these cells at the commence- 
ment of its elongation. 
Now let us consider what we actually find on observing 
a continuous series of tangential sections through the zone of 
secondary increase. 
Proceeding from without inwards, we first come, immediately 
1 1. c. p. 298. 
