POTTIACEAE. 129 
often yellowish, a few rows at margin often remaining quadrate or sub- 
uadrate quite to the base. The capsule is rather long, and nearly always 
either curved or asymmetrical, the peristome long and much twisted, 
rather pale, from a well- developed but not very long tubular basal membrane. 
have examined the type of 7. crispifolia Mitt. at Kew, and am con- 
vinced that it is only one of the longer-leaved forms of B. torquata; the 
nerve is more distinctly excurrent than usual, but in 6 plies chacshen 
it is 1 ae 1 with the more robust states of this spec 
Bellii R. Br. ter. is also this plant. R. ea it may be recalled 
(Trans. 'N. Z. Inst., vol. 30, p. 400), states that he has been unable to identify 
$id a crispifolia, and 7. australasiae 
Ih t from one or two localities in the South Island, but it is 
probably more frequent in the North; it is found also in Tasmania and 
Australia 
3. Barbula rostrata (R. Br. ter.) Dixon comb. nov. 
Syn. Trichostomum rostratum R. Br. ter. in Trans. N.Z. Inst., vol. 29, 
. 485. 
This species is represented by a single tuft without fruit in Brown’s 
herbarium, and I have not seen it elsewhere. In colour and habit it 
resembles Tortula phaea. The leaves are erect and slightly incurved when 
dry, but otherwise little altered ; they are 2-2-5 mm. long, widely oblong- 
lanceolate, scarcely acuminate, broad at the points ; the apex is frequently 
obtuse, but I am not clear whether this is not always due to erosion ; the 
upper, younger leaves being almost always subacute, with somewhat 
incurved points; the nerve, which is very prominent at back, is lost in or 
just below the apex; the margins here and there narrowly recurved or 
reflexed, or subplane ; the re age cells 7-10 » in diameter, angular, more or 
less incrassate, rather opa e chief feature of the plant is the basal 
areolation, which is little differentiated from the upper, the cells being 
shortly rectangular only, and all incrassate, scarcely pellucid, the marginal 
in several rows often quadrate, only a few next the nerve at most being 
elongate-rectangular (3-4 x 1). 
According to R. Brown the perichaetial leaves are erect, linear-lanceolate, 
and acute ; the seta gin. long, the capsule cylindric, with a lid about as 
long. The Hees iee is not described or Gah nd Brown writes of the 
species “(? Tortula),” so that there is no direct palicstion that the peri- 
stome is Trichostomoid, and I do not feel much hesitation in placing it in 
the present Seed the upper cells and areolation generally seem to indicate 
that it is not a Tortula, while the length of ~ lid is very = 
presumptive vides of a long, Barbuloid peristom 
abitat is described by Brown as * aevce of limestone a. 
Broken River, West Coast Road.” 
It. CINCLIDOTEAE. 
{Cinclidotus australis Dixon = Tridontium tasmanicum (v. supra). The 
subfamily Cinclidoteae is therefore not ORI in New Zealand.] 
Ill. POTTIEAE. 
Acauton C. M. 
Cleis tocarpous, annual, phascoid mosses, almost stemless; the lower 
ise ionieas minute, three or four upper much one inflated, concave 
- 
