542 ANEURA. 
terra humida umbrosa et ad saxa rivulorum vigent. Ubique terra- 
rum inventa sunt. In sylvis Amazonicis et Andinis ad truncos pro- 
stratos putrescentes, seepe muscis irrepentes, valde abundant ; species 
autem caracteribus difficillime eruende, propterea nullam Amazoni- 
cam cum Antillanis, etc. identificare potui. 
The existence of a costa, or nerve, in Anewra is undeniable, and in the 
robuster species the primary costa, or stem, is almost terete, and consists of 
many concentric strata, whose structure is not sensibly different from what 
obtains in other hepaticee. But the costa in the ultimate, and often in the 
penultimate, divisions of the frond usually becomes thin and broad, and 
(contrary to the normal proportions in other genera) the inner layers consist 
of much larger and more pellucid cells than those of the superficial layers, 
whose lateral extension—that of the upper surface especially—constitutes 
the wing or pagina which exists in nearly every species.—The transition 
from species usually (but erroneously) regarded as quite destitute of a 
nerve to those where its existence is undisputed is so gradual that the 
supposed genus Acrostolia Dum. (Pseudonewra Gottsch.), framed for the 
reception of the strongly-nerved species, is inadmissible even as a sub- 
genus. 
The ¢ and @ receptacles are abbreviated pinnules, or branchlets, 
having the same lateral origin as the efloriferous pinnules. Both anthe- 
ridia and pistillidia are solitary, and inserted in two rows, one row on 
each side of the upper face of the costa of the pinnules; the antheridia in 
little pits (alveoles) on the very margin, the pistillidia sessile a little 
within the margin of the costa; and they are partially veiled (involucrate) 
by the incurved, and often deeply laciniate, wings of the pinnule. I 
have not seen the two sexes united on the same receptacle, although that 
is said to sometimes occur in the European A. palmata ; but it is singular 
that, while the involucre of the antheridia is single, that of the pistil- 
lidia is usually double, or bilamellate, each of the twin lamelle being 
besides more deeply cut, with more numerous and narrower laciniz, than 
the ¢ involucre. Possibly, in some anterior synoicous condition of the 
Aneure, the outer of the two lamellae may have subtended the antheridia. 
The involucra that contain only sterile Q flowers are often very little 
shortened, and show the distichous arrangement of the pistillidia very 
clearly ; but where one of the latter has been fertilised, and has grown 
out into a calyptra, the short involucre surrounds its base like a cup, and 
although its duplicate character may be detected, the bilateral insertion 
of the sterile pistillidia that still adhere to its costa becomes indistinct. 
The antheridia vary from 2 to 12 pairs, rarely as many as 20 pairs, except 
in A. latissima n. sp.—the only species that has the antheridia mostly in 
a double row on each side of the costa—where they are sometimes very 
numerous. The pistillidia are usually fewer, from 2 to 8 pairs; but I 
have once found as many as 20 pairs in A. hymenophytoide n. sp. 
J. B. Jack* has given the name of “ Elater-holders” (Schleudertrager, 
or Elateren-triéger) to the short stout elaviform elaters that persist in the 
apices of the capsule-valves, and hold each a pencil of twisted intricate 
slender-fusiform free elaters in mechanical (not organic) attachment ; 
although all finally fall away, and leave the valves quite naked within. 
* In his Review of Dumortier’s ‘ Hepatic Europe’: Bot. Zeit. 1877. 
