-93- 
sum , both these species having false leaf-traces in the stem precisely as in 
T. australis. Here we see the danger of generalizing too far. 
T. australis agrees exactly with certain Splachnums in another charac- 
teristic taken from the vegetative parts, namely in the structure of the costa. 
It is Lorenz again who has given some hints on this subject. According to 
him the costa of the Splachnaceae is composed of two large ventral cells 
(“guides”), of a certain number of large dorsal cells, and of a central 
cellular body whose principal element and characteristic is a bundle of thin- 
walled, angular cells of the kind he has called “comites in addition, one 
usually finds between these “comites” and the large dorsal cells a larger 
or smaller number of cells which he designates as “intercalary” cells, 
formed sometimes like the guides, sometimes difficult to distinguish from the 
“comites,” sometimes thin-walled, sometimes rather incrassate. In Tetra- 
plodonshe observes that these intercalary cells are present, at least in the most 
developed part of the costa, and that they are nearly stereid ; as regards 
Splachnum he describes the same structure in S. sphaericum, while in S. 
luteum he finds a bundle of “comites” surrounded by numerous, rather 
large, polygonal, thin- walled cells. In S. vasculosum and in S. ampullaceum 
I have observed a little different structure: the “ comites” form here a band, 
which extends between the ventral and the large dorsal cells along the 
median line of the costa; at each side appear, usually, some large rounded, 
thin-walled intercalary cells, having nearly the aspect of the ventral 
ones. These intercalary cells are wanting, however, in 5 . ampullaceum , 
in the upper part of the leaf where the central cellular body is con- 
sequently represented by the “comites” alone. The same structure is 
found in T. australis , (and likewise in Haplodon) in all parts of the costa 
which is composed here also of “ comitesT surrounded by the large ventral 
and dorsal cells. We have here another characteristic which distin- 
guishes T. australis from true Tetraplodon and at the same time connects 
it with Splachnum ampullaceum. 
That T. australis belongs to the genus Splachnum is thus set beyond 
doubt, and it becomes apparent that it is S. ampullaceum with which it is 
most nearly allied. This relationship is in fact so close that without 
incroaching on its specific claims, we can characterize it as an S. ampulla- 
ceum with hyaline seta, apophysis reduced to a minimum and narrower, 
deeply dentate leaves. 
It goes without saying, that this conception should be expressed in the 
name. In transferring it to the genus Splachnum, however, I take the 
advantage of the occasion to restore to it the oldest specific name, so long 
unnoticed and revived only in 1883 by Lindberg, who was convinced by his 
studies of the herbarium of Dillenius that the type of Phascum caulescens L. 
is the same plant as Tetraplodon australis Sulliv. & Lesq. 
Splachnum caulescens (L.) Dicks, in Trans. Linn. Soc. Ill, p. 239 (1797) 
Synon.: 
Sphagnum foliis tenuibus, gramineis , pellucidis Dill. Hist. Muse. p. 550, 
(i74i). 
