368 Mr. R. Spruce on the Musci and Hepatice of the Pyrenees. 
malous Polytrichum, which may possibly be a starved form of P. na- 
num. It has the capsule subcernuous, nearly spherical; the colu- 
mella 4-sulcate, or rudimentarily alate (not ¢erete as in typical P. na- 
num); and the calyptra sheathing the whole of the capsule. 
183. P. aloides, Hedw. Muse. Frond. 1. t. 14; M. P. 304; 
Br. Europ. Pogonatum, p. 6. t. 8. 
Hab. Zs 1 humidiusculis, presertim in arena rivulorum, al- 
tius versus alpes ascendens quam P. nanum. 
184. P. urnigerum, L. Sp. Pl. p. 1573; M. P. 305; Br. Europ. 
Pogonatum, p. 7. t. 9. 
Hab. Zo. im arenosis humidiusculis. 
185. P. alpinum, L. Sp. Pl. p. 1573; Br. Europ. Pogonatum, 
p: 3. t. 10: 
Hab. Z,_, P. c. in monte Crabioules. 
Var. caule valde elongato, subsimplici; M. P. 306.—Hab. in 
saxosis umbrosis vallis Lutour prope Cauterets. 
The curious way in which the epiphragm (the tympaniform dila- 
tation of the summit of the columella) is attached to the peristome 
in this and other Polytricha does not seem to have been anywhere 
described. In P. alpinum it is originally placed at the base of the 
teeth, to which it is attached by means of processes equaling them 
in number and exactly covering their internal face. After the fall of 
the lid, these processes are gradually detached and the epiphragm 
rises, probably from the pressure of the full-grown spores beneath it, 
so as to allow the latter to escape through the interstices of the pe- 
ristome. When the epiphragm is quite liberated, either naturally or — 
by art, the processes curve inwards upon its upper surface (see Pl. I. 
fig. 1) so as to be with difficulty seen, unless the light be properly 
regulated or the epiphragm be set up on its edge. ‘The processes 
are composed of only a single layer of cellules, which are so disposed 
that their interstices form vertical lines corresponding to those on 
the teeth. 
The adhesion of the epiphragm to the teeth is so great as to resist 
the action of the columella to draw it down into the capsule, and 
often ultimately to cause the rupture of the columella. The ragged 
portion at the underside of the section (fig. 2) is where the rupture 
takes place between the columella and the epiphragm. 
In P. urnigerum, aloides and nanum the epiphragm is attached to 
the teeth in the same manner, but the processes are very thin and 
tender, and when the epiphragm is detached by force they often re- 
main adhering to the teeth. In P. angustatum (as also probably in 
P. undulatum) the processes are united to each other by an inter- 
vening membrane, which is granulated on the surface, while the pro- 
cesses themselves are smooth and marked by lines similar to those 
on the teeth: in other words, the epiphragm is bordered by an in- 
flexed continuous membrane, by means of which it adheres to the 
peristome, The figure of P. angustatum in ‘ Bryol. Europea’ repre- 
sents this pretty well, but no mention is made of it in the text. 
ATP CS 
