546 Evans: HeEpaticAE OF PuERTO RIcOo 
of this cell, which is commonly more or less curved, there is a 
slight depression in which a hyaline papilla is situated, very much 
as in Drepanolejeunea. The apical tooth and the lunulate sinus 
connecting it with the end of the keel assist in forming the open- 
ing into the water-sac. Sometimes the latter widens out abruptly 
just within the opening, but there is commonly a narrow and 
curved passage-way leading from the enlarged part of the water- 
sac to the external air. The cells of the lobes are comparatively 
small and are commonly plane or nearly so on their free surfaces ; 
their vertical walls are either slightly and uniformly thickened or 
show small but distinct trigones. In anumber of species a distinct 
group of four elongated cells may be detected at the base of the 
lobe, similar in many respects to those found in Drepanolejeunea 
subulata and its allies. Usually, however, the two cells of this 
group which are turned toward the antical margin of the lobe 
become modified into bulging ocelli, and sometimes one or both 
of the remaining cells undergo a similar transformation. In other 
species the position of the ocelli is less definite, but their occur- 
rence in some part of the lobe can almost always be demonstrated. 
The underleaves of Harpalejeunea are unique among the Le- 
jeuneae. They are of small size and widen abruptly from a nar- 
row base into an obcordate expansion, the apical sinus being shal- 
low and varying from lunulate to subacute; the lobes of the 
underleaves are more or less divergent and are broad and rounded 
at their extremities.» 
The species of Harpalejeunea are almost always dioicous. 
The branch bearing the female inflorescence varies greatly in length 
but the flower itself is invariably subtended by one or two innova- 
tions. The perichaetial bracts are more bluntly pointed than the 
leaves and are sometimes narrowly winged along the keel. The 
perianth is sharply five-keeled in the upper part, but the keels, 
although distinctly winged in several species, are never produced 
as horns. Here, again, the genus shows a certain approach to 
Drepanolgeunea subulata and D. anoplantha. 
Three species of Harpalejeunea from Puerto Rico have been 
studied by the writer. The first of these is much more abundant 
than the others and was at one time referred by Stephani to Lé- 
jeunca (Harpa-Lej.) stricta Lindenb. & Gottsche, a species orig- 
