underleaves are very well seen in Harpanthus ( Fig . 20), where they attain 
an appreciable size, but they are also to be found in many other species. 
They become more complicated when their margins are irregularly toothed 
or ciliate, as in many species of Lophozia {Fig. 21). 
Variously lobed and divided underleaves are characteristic of many 
genera. In Frullania {Fig. 14) and in Lejeunea the underleaves, which are 
subrotund in form, are bifid, sometimes to beyond the middle. In Lopho- 
colea {Fig. 8), Chiloscyphus, Geocalyx and other genera, the underleaves are 
much narrower, but are also deeply bifid. In Lepidozia {Fig. 11 ) the under- 
leaves are broad and deeply trifid or quadrifid in the larger species. In 
Ptilidiitm ciliare and in Trichocolea the underleaves resemble the leaves in 
being strongly ciliate along the margins of the divisions. In none of our 
northern species, however, do the underleaves produce water-sacs, although 
this phenomenon occurs among certain antarctic genera. 
It will be seen from what has been said so far that the leaves and under- 
leaves afford generic characters of much importance. These are supple- 
mented, in the purely vegetative part of the plant, by characters derived 
from the branching and from the cell-structure ; but, as it is difficult to dis- 
cuss these without entering into considerable detail, we may pass at once to 
characters connected with the reproduction. 
The antheridia and archegonia of the scale-mosses are essentially like 
those of the true mosses. The archegonia are borne singly or in groups on 
the tips of specialized branches, the growth of which is thereby terminated. 
As a rule only one archegonium of a cluster develops a sporophyte. Even in 
the absence of fertilization, the archegonial branches yield important char- 
acters. Instead of being prostrate such branches tend to be ascending or 
erect, and their leaves are oftentimes very different in appearance from the 
ordinary vegetative leaves. These leaves are called “ bracts,” and the cor- 
responding underleaves “ bracteoles.” In many cases the latter are nearly 
or quite as large as the bracts, and this is true of some species which lack 
underleaves on ordinary stems. In a number of species which are destitute 
of underleaves, bracteoles also fail to be devoloped. The various species of 
Radula are striking examples of this condition. As a rule the bracts are 
less highly specialized or less definite in their characters than the leaves. In 
a species with bifid leaves, for example, the bracts tend to be irregularly two-to 
four- lobed {Fig. 22) ; in a species with well-developed water-sacs these struc- 
tures are not developed on the bracts {Fig. 23): in certain species with 
undivided leaves, the bracts are bifid. In still other cases the bracts are 
scarcely to be distinguished from the leaves. Usually the archegonial branch 
shows a gradation between typical leaves and typical bracts. 
The archegonia, the young sporophyte and the calyptra cannot be seen 
as a rule without careful dissection. This is because they are covered over 
and concealed by other parts of the plant, and are apparently thus protected 
from being dried up. In very rare cases, the covering is done by the bracts 
alone. Usually, however, in addition to the bracts, the archegonial branch 
bears a remarkable tubular organ called a “ perianth,” or else itself develops 
into a hollow structure known as a “ perigynium.” 
