[KIRSCH] CERTAIN STRUCTURES IN THE PTERIDOPHYTES 393 
the xylem in many respects, the main difference lying in the much greater 
length of the latter elements. 
Vessels or sacs with epithelium, as described for the stipe, were not 
observed in the rhizome at any place. 
From the above account it is seen that, contrary to Terletzki (21, 
486), the structures which he termed “ Stumpfzellen-Stringe ” are pre- 
sent in the rhizome of Pteris aquilina. : 
As noted before, the main rhizome is divided into a large number 
of unequal parts, and an account will now be given of what was observed 
in the short portions. A great many of these were examined and all 
presented the same phenomena, so that it will be sufficient to describe 
one which measured about + em. in length. ‘The bundles throughont all 
the sections showed the protoxylem to be practically intact (Photo 12), 
although in some places the protoxylem elements showed signs of con- 
siderable compression. In one or two instances, however, there was the 
slightest indication of canal formation in the presence of a minute slit 
between two protoxylem cells, where the wall had begun to split. But 
in no case was this carried far enough to produce a definite cavity. As 
would be expected from this, there was no sign of thyloses in these 
bundles. 
The above is a further proof of the fact that in regions of slow 
growth the canal is not formed at all, or only slightly, proving that the 
formation of the canal is a function of growth. 
This feature is further illustrated in the bundles of the branches 
of the rhizome which bear the stipes. These branches are always short, 
never growing to more than 2 em. in length, and the protoxylem in their 
bundles was generally found intact. In some of the longest ones, how- 
ever, indications of canals were evident, the protoxylem elements having 
separated considerably. However, the action had not proceeded rapidly 
enough to rupture the cells, so that the small cavities were of purely 
schizogenous origin. 
The growing points of the rhizome will now be dealt with. These 
differ greatly in length, from about 0.5 em. to 6 or 7 inches. At the 
anterior end all the tissues are in a formative state, whilst at the pos- 
terior end they are fairly mature. In the smallest specimens very little 
evidence of the canal was to be seen, and since these represent regions 
of slow growth, the canal is probably never formed in them. 
In those of vigorous growth, i.e. which measured about 6 cm. or 
longer, breaks in the protoxylem were evident at the very tip. Proceed- 
ing from the anterior to the posterior end these breaks became larger 
and larger and assumed the character of canals. In some of the largest 
