Evans: NOTES ON GENUS HERBERTA 195 
to either leaves or underleaves. It is of course possible that their 
aberrant position is due to secondary displacements caused by in- 
equalities of growth, but this could not be definitely determined 
from the material studied. The secondary stems are sometimes 
direct extensions of the creeping rhizomes, but it is usually possible 
‘to demonstrate their origin as branches. Even under ‘these 
circumstances they usually bear small leaves at the base and 
only gradually give rise: to the normal leaves. Sometimes the 
change is more abrupt, and all gradations may be present between 
the small-leaved prostrate rhizomes and the typical secondary 
_ stems with large leaves. 
The secondary stems differ greatly in length in different species 
and even in the same species under varying conditions. In some. 
of the robust tropical species a length of 25 cm. may be attained. 
The stems are usually ascending but are often pendulous and 
sometimes form conspicuous festoons hanging from the branches 
of trees. The branches are always intercalary, just as in the case 
of the rhizomes. The normal vegetative branches take their 
origin in the axils of underleaves and are rarely subdivided. In 
most cases they spread widely and show a flagelliform appearance, 
their leaves being distinctly smaller than those of the stem. 
When the growing point of.a stem is injured, however, a ventral 
branch sometimes changes its direction of growth and assumes 
that of the stem, its leaves at the same time increasing in size. 
The same thing is true of the-subfloral innovations, which may be 
either ventral or lateral in position. The growth of the stem is 
here brought to an end by the development of the archegonia. 
Rhizoids are often absent altogether from the secondary stems. 
When they occur they are usually restricted to the underleaves 
and leaves of the ventral branches, although the normal leaves 
and underleaves have the power of developing them. 
The leaves (TEXT FIGS. I, 14, 15, 21-25) and underleaves of 
the secondary stems present a great deal of uniformity throughout 
the genus. They are more or less deeply bifid with a narrow sinus 
and sharp divisions. In rare cases a leaf may be trifid instead of 
bifid, a peculiarity to which certain species seem to be more subject 
than others. The divisions are sometimes distinctly grooved but 
show little tendency to be complicate, both lying approximately in 
