99 
crystals as needle-shaped, hut bifurcate at one extremity— 
which is peculiar. 
However abnormal many of the stems belonging to these 
various orders may become, and however difficult it may be 
to trace their divergency from the normal structure, there 
can be no doubt that the characteristic elements of the di- 
cotyledonous stem are all present during some portion of 
their lives. Their unequal development may be brought 
about either by the vital energy of the growing tissue of the 
bark being in excess of that of the wood, or vice versa, from 
which circumstance will arise the curious outlines presented 
by the relative distribution of each; or else it may be pro- 
duced by a much more copious deposition of woody tissue 
at some points of the circumference than at others, from 
which will result the curious forms presented by the Bau- 
hinias and many of the Malpighiacese. 
The monocotyledonous division of the vegetable kingdom 
has also its representatives amongst these ropes. There are 
two species, perhaps belonging to the grasses, which I have 
met with ; but in neither case is the entire stem used. One 
species is much larger than the other, their diameters being 
about two inches and four inches respectively; both are 
hollow and are divided into strips for use. 
There are many other species found amongst these ropes 
which belong to other natural orders, such as the Menis- 
permacese, Gnetacese, Asclepiadacese, &c., but our knowledge 
of them is too limited to assign them to their respective 
orders. Most of my specimens have come from bales of 
Santos Cotton, and it would be as well to keep a record 
of the localities from whence they are derived. I am very 
anxious to get some from the Pacific coast, where many 
species differing from Brazilian species must be found. 
Gaudichaud mentions the neighbourhood of Guayaquil, in 
Ecuador, as being particularly prolific in these lianas. 
