Branner.) i 470 [April 18, 
and when there, was a point on the surface of the phylophore. Now it 
has become a point on the surface of the trunk, the distance CD be- 
yond its original position. so that the whole bundle must have grown in 
this part. ; 
The theory of downward growth from the frond to the base, as held 
by M. Charles Gaudichaud, appears to me, as it did to Von Mohl,* un- 
worthy of serious consideration. Mirbel’s theory of upward growth of 
the fibro-vascular bundles was denounced by M. Gaudichaud as a physio- 
logical impossibility ; and it might be said of M. Gaudichaud’s theory, 
that downward growth, as held by him, is a mechanical impossibility. 
From the course of the bundles, as already explained, it is seen that to 
grow downward, they would have to pass through the bony outer layer of 
the trunk twice: once on entering it, and again on approaching the sur- 
face lower down. Moreover the bundles at the base of the trunk would 
either have to be extremely small, or the base itself very large, neither of 
which is the case. It will be seen later, also, that the lower extremity of 
a bundle division hardens first, thus precluding all possible growth. 
The reason that the fronds are always developed in connection with the 
central bundles, is because the central bundles are the ones there present, 
and the ones in the most advanced stage of development. It has already 
been noticed that when a frond is developed and a bundle branches, the 
part in connection with the frond is developed rapidly and at the expense 
of the part leading upwards, Having its development thus retarded, and 
being carried to one side by the growth of the trunk, this ascending di- 
vision is attenuated at its lower extremity. Other bundles gain upon it 
in point of development, and take its place at the centre of the growing 
part. But in the course of time this division regains its vigor, and its 
place at the centre of the phylophore, where it is again the most advanced 
in development, and again has a frond or spadix formed in connection 
with its branches, and is again curved outward. 
At the time of branching the formation of the frond is the immediate 
object of the bundle, but provision is at the sarhe time made for other 
fronds higher up, 
The branching goes on from the base of the trunk to its apex, varying 
only as the tree becomes old, and its vital powers diminish, the result ot 
which is shorter internodes, and consequently shorter divisions of the! 
fibro-vascular bundles, 
The number of bundles may be said to be the same in all parts of. the 
stem, and’ it is to this fact, taken in connection with the average even 
size of the bundles themselves, that the equal size of the palm trunk must 
be attributed. In specimens of which I have estimated the number of 
bundles at different altitudes, there has generally been a difference in| 
favor of the base of the stem, but this difference is so slight that I believe 
the decreased vitality of the plant is sufficient to account for it. It is also 
' 
* Ray Society, 1849, p. 62. 
