1 12 Salmon. — On the Genus Fissidens . 
of this line indicates, I believe, a different origin, and would 
not occur on the hypothesis that one of the vaginant laminae 
has prolonged itself to form the superior lamina. In dealing 
below with an allied genus, this question of the origin of the 
superior lamina will be mentioned again, and I will only 
state here that I regard the superior and inferior laminae 
as one laminar outgrowth, arising primarily from the back 
of the leaf, quite independently of the nerve. 
There are still two more theories as to the morphology 
of the Fissidens leaf that must be mentioned. The first 
of these was put forward by Spruce ( 15 ) in these words: — 
‘ The leaves of Fissidens were originally 3-lobed, with the 
medial and longer lobe, by a half turn on its axis, placed 
vertically, i. e. at right-angles to the other two lobes and 
to the base of the leaf, which is inserted transversely on 
the stem. As we now usually see them, the lobes have 
become connate ; the two lateral lobes complicate into an 
equitant sheath, and at the keel, but especially at the apex, 
winged with the vertical medial lobe; but are still occasionally 
found more or less dissevered, as in the ancestral type/ 
In support of this theory, Spruce mentions the occurrence 
of perigonial bracts ‘ not complicate, but concave, cloven at 
the apex into three subequal short obtuse lobes ; the middle 
lobe twisted half round so as to set it at right-angles to 
the rest of the leaf,’ and cites Bryol. Europ., fasc. 17, 
F. bryoides , Hedw. t. 2, f. 10, and Sullivant’s leones Mus- * 
corum, F. obtusifolms , Wils., t. 22, f. 20, as figuring similar 
male bracts, more or less distinctly 3-lobed. 
An examination of the floral leaves of Fissidens shows, 
among not only the perigonial, but also among the peri- 
chaetial bracts, certain leaves of the ‘ trilobate ’ form 
mentioned by Spruce (Figs. 36-39). But taking the whole 
series of the leaves of either inflorescence, we can see that 
this ; third lobe ’ is simply the reduced or rudimentary ‘ lamina 
verticalis ; ’ and if we are to take the most simple form of 
these leaves as representing the ‘ancestral type,’ we must 
choose the outermost simple concave leaves, which are quite 
