SPECIAL MOHPHOLOGY: CLUB-MOSSES. 189 



can be explained), yet manifest a tendency towards this form." Sad in- 

 deed, that men of intellect should seek science amid such a confusion of 

 words ! By others, Robert Brown among the rest, stomates have been 

 considered as channels for the evacuation of the spores ; this, however, 

 they cannot be, since they never open a communication between the 

 spore-cavity itself and the exterior. The spongy cellular tissue below 

 them always passes gradually into a close cellular tissue, the internal 

 membrane, as it approaches towards the spore- cavity. They do not deviate 

 in the most trifling degree in Polytrichum alpinum from the ordinary 

 structure of the stomata, although such would appear to be the case from 

 the. incorrect delineation of Treviranus (see as above, fig. 18.). I have 

 myself been unable to examine the pores of the capsule of Lyellia ; but 

 if the delineation of Treviranus (fig. 17.) be correct, they have nothing 

 to do with the stomata, and are organs of an entirely special kind. I 

 would remark, that, although I am not confident of the fact, I believe I 

 have seen spiral fibres in the cells of the peristome ; for instance, in 

 Hypnum triquetrum. 



b. Agamic Plants having Roots. 

 VI. CLUB-MOSSES (LYCOPODIACEJS). 



104. A perfect history of the development of the Lycopodiacece 

 remains still a desideratum. Only so much is certain, that in the 

 germination of the larger spores, which will be mentioned, a true 

 root appears. In the perfect plant, the stem, which is almost always 

 recumbent, develops roots on the lower side, along its whole length, 

 and dies off from below upward. The leaves always follow one 

 another closely around the stem, and are sometimes twisted in such 

 a manner as to appear as if they were ranged in one plane on op- 

 posite sides of the stem. The branches which are developed from 

 the axillary buds often stand similarly, in such a manner that the 

 ramification is pinnatifid, or the bifurcating branches are erect and 

 arranged in pyramidal forms ; in rare cases the stem is flat, and the 

 leaves far asunder, as in Bernhardia complanata. The leaves are 

 almost always narrow, lanceolate, similar to the leaves of Mosses, 

 but bearing more resemblance on the recumbent stems (where they 

 evidently stand in two rows) to the leaves of the Liverworts, and 

 are likewise smaller and of different form on the under side of the 

 stem. All are provided with a simple mid-nerve. The greatest 

 deviation is in the stem of the Isoetes, which is shortened into a 

 thick disc, and has long, narrow, grass-like leaves, that enclose 

 one another below in a sheath-like manner. In a few of the Ly- 

 copodia, the axillary buds are developed into a somewhat more 

 fleshy condition in all their parts, and spontaneously (?) separate from 

 the stem to constitute bulbels (bulbilli). 



