202 Evans: NOTES ON GENUS HERBERTA 
in much the same way. The entire cell-cavity is eventually lined 
by a uniformly thick layer of deposit, the only thin places left 
being the pits in the vertical walls (TEXT FIGs 5,6). Even some 
of the pits may be obliterated with age. When this takes place 
the deposit either fills the pit-chamber altogether or bridges it 
across, leaving a minute vestige of the chamber next to the closing 
membrane (TEXT FIG. 7). When the cells are mature it is some- 
times possible to detect evidences of the original bands of thicken- 
ing and of the scallops along the vertical walls, but the free walls 
present an absolutely uniform appearance. Although the thick- 
‘ enings look very much as if they were formed by the coalescence of 
trigones and intermediate thickenings, it is clear from their 
method of development that they are not directly comparable 
with the trigones and intermediate thickenings of other Hepaticae. 
Although the description just drawn from H. juniperina prob- 
ably applies in its essential features to the entire genus, the 
details are not always as clear in some of the other species. In 
H. adunca, for example, it is difficult to demonstrate bands of 
thickening except in the basal portion of the vitta, and it is doubtful 
if they ever form complete rings. The scalloped appearance of 
the vertical walls is likewise much less evident. Possibly the more 
indefinite conditions found are associated with the fact that the 
thickening of the walls in H. adunca is usually much less marked 
than in H. juniperina. 
The leaf surface in Herberia, the so-called aide of authors, is 
striolate or verruculose. The roughness is sometimes very ap- 
parent and sometimes made out with difficulty. Whether differ- 
ences in the degree of roughness yield constant differential char- 
acters in separating species is perhaps doubtful, although such 
differences seem very striking when certain species are compared. 
The androecia in Herberta are relatively short and usually 
include from four to eight pairs of bracts, those at the upper and 
lower ends being somewhat transitional in character. The male 
plants tend to produce smaller leaves than female plants or vigor- 
ous sterile plants, and although the bracts are considerably larger 
than the leaves borne between the successive androecia, they 
scarcely equal in size the normal leaves on female individuals. 
Several years ago Schiffner made the remarkable discovery that 
