Lemaneaceae of the United States . 225 
With the large series of forms before me it is impossible to 
select any important stable character which will differentiate 
them into good species, otherwise one could make a species 
for specimens from nearly each different locality. The varia- 
tions of the antherid-zone and papillae are explained by the 
variations of the branching of the ends of the generative fila- 
ments, which we find varies in the same specimen. I am led 
then only to enlarge the specific characters of the Chantransia- 
form of rigida Sird’t., and in view of the unity of essential 
characters in the large series of forms, to follow Kiitzing and 
Rabenhorst in uniting them into one variable species, and to 
give priority to Tory’s name. 
Among all these variations there are some forms about 
which many may be more or less definitely grouped : — 
Var. a , mamillosa, stout, procarp-zones cylindrical or nearly 
so, papillae very prominent at time of fertilization, sometimes 
more so in age by hypertrophy of the tissue at base of an- 
theridia, sometimes confluent ; young sexual shoots of an 
obscure violet colour, reddening the water in which they stand 
for several days, confined to the basal-cells of the Chantransia - 
form. In age drying yellowish. 
Habitat. Morgan’s and Bolan’s Creek, Chapel Hill, N. C, 
collected by the author. Alabama, T. M. Peters, Coll. (Herb. 
Boston Soc. of Nat. Hist). 
Var. ( 3 , subtilis, very capillary, profusely branched, papillae 
3-5, rarely confluent, procarp-zones cylindrical, or sloping 
from the antherid-zone ; sexual shoots at various heights 
in the Chantransia-i orm. In age drying yellowish. 
Habitat. ( Cascades,’ Melrose, Mass., F. S. Collins and W. A. 
Setchell. 
Var. y, rigida, simple or branched at various lengths in 
the Chantransia-ioxm ; procarp-zones nearly cylindrical or 
strongly constricted ; papillae nearly or quite plain, 3-7, 
often confluent. In age usually drying yellowish, some- 
times obscured by the greenish or dark color of the 
Q 
