Relief will be found only in the way mentioned - by the publication of synopses of the genera, with full synonymy. Many of the genera present other and serious difficulties which need not be enumerated here. The lobes of the PiDDlen are rounded, yet appear more or IC8B acute and distant in the dried specimen, owing to the usual bending of the margins.
MUT all to be of the same species, regardless of certain slight variations in size and shape. Your epcciea belongs to the Coastal flora. and is not rare on the side of the Coast Mountains. Specimens of the type coHection of this species have not been seen by the author.
The Qriginal description mentions the margin of the pinnm as a whole, which is by no means always true. The fertile leaves are long and fewer in number and stand erect from the center of the rosette. The edges vary from serrated to slightly ereuate.aerrate, depending on the fertility of the pinnre.
From A. melanorachis it differe conspicuously in the charactere enumerated in the
THE NORTH Al[EllICAlif TREE FERNS OF THE GENUS DICKSONIA
In the Dicksonieae, the species are much more difficult to distinguish than in the tribe Cyatheae, partly because there is a smaller one. Furthermore, the branching of the veins is more variable and more directly correlated with degrees of fertility. This is especially true of the larger, fully fertile segments of the lower and middle parts of the pin, in which the fertile veins in the proximal .eide of the segment are often ODce forked, while thoae on the distal side are almost invariably simple.
The branches of the veins are delicate, however, and do not stand out, their development is accompanied only by a slight creation of the fertile lobule!!. Rika specimens collected in the vicinity of Colibla.nco, on the elopes 01 Tunialbll volcano, altitude about 1,950 meters. Mo:ron 325), which agree fairly well with Karaten's illustration and with fragmentary South American and Costa Rican material in the Underwood Fern Herbarium at the New York Botanical Garden.
It is clearly related to Dicbonia ghiabrtghtii, from which it differs in the characters listed under that species. 8B short spur immediately below; eori mostly 4 or 5 pairs, occupying the lower two-thirds of the segment, about 1 mm. wide; plll'8physes numerous, rather long, lax, light brownish. 676144, collected near the top of the Divide, above Camp I, Holcomb Trail, above El Boquete, Chiriqui, Panama, elevation about 1900 meters, by William R.
Cert,g. and individuals at this location, which are supposed to belong to the same species, had trunks as high as 6 meters. The genus Odontosoria, as recognized by Diels in the PHanzenfamjljen of Engler and Prantl, comprises two sections or subgenera, the first (Eu-Odontosoria) containing rather small species of erect or climbing habit and definite growth, the second (Stenoloma) containing three species indefinite growth spurts. These two groups are entitled to recognition as separate genera. If we accept this view, the name Odontosoria should be applied to the second group, the large climbing species, instead of the first. for this consideration are introduced in the following brief account of the taxonomic history of the genus with a review of the American peculiarities of the true Odontosoria. Presl appears to have been the first to use the name Odontosoria, applying it to a group of 17 species comprising his fourth section of the genus Davallia. Among these are included species of the two species mentioned above. Of course, the simpler, smaller plants of erect, vigorous growth are listed first, and one of these (Davallia tenuifolia Swartz) is shown, As this is the only species of this subgenus illustrated by Presl, it might re.
Most of the species of Presl's section OdontoBoria were placed together by F6e under his new genus Stenoloma, and no distinction was made regarding the extremely different growth habits of the species thus included. The systematic relationship of the many species listed by Kuhn under Lindsaya (43 species), Schizoloma (25 species), and Odontosoria (3 species) must be carefully considered; but it is evident that the plants called Odonto8oMa chimnsis and O.
Pinnules of the fourth order rhombic to irregularly obcordate, simple, or the larger Olles with 1 or 2 large ones (ree obcordate lateral segmenta, the terminal segment subrhombic, irregular). Collection type specimens of this species are not Other factors that influence the form of the individual to an unusual degree are the age of the plant, the degree of fertility, and the seasonal condition of the plant when collected.
Jenman'a description ia inaCCllI"&te in one aspect: and of particular of !K'g1Ilcnts, Odontoson'a fltxuO&a is well dis. Surprisingly the secondary and tertiary rachises are not at all flexu08C, as in other species of the genus, and usually the primary ra.chis is also straight.
In the shape and size of the ultimate segments the similarity of this species to. 8Chkchtendahlii is 80 great that a specimen of the above collection (Lehma.nn XXXIV) was 80 determined by Hieronymus.1 But O. 88 OdontoBO'fia jenmanii, a species not represented by any of the :listed figures and until the present never given a valid name.
With abundant material, there can be no possible confusion of these species, the only difficulty is found in explaining historical errors due to too much material, misidentification of plates and a connection. Beyond the poor figure and the declaration of Ute of the West Indies as the type locality, there appears to be no means of identifying this species, which technically muat be considered unpublished. Apparently studied by Kuhn, Bince is one of the three species listed by him under Lindayopsis, in his work on this group.
This, which is one of the species described by Kunze in a critical commentary on Hookerta's treatment of this group in the fine volume of Species Filicum t, is listed by Christensen (under Dav&llia) as an unidentified or doubtful member of Odontos. >ria.. specimen collected by Poitea.u and presented to Kunze by Bory.. lfOTES UFOlf BO)[J(ERTA AND LATE GENERA. Of the previously known Opecies, two, Bommeria pedata and B. hupida, are completely free-veined; while the third, B. dirmh Most of the species are soft-haired, and their agreement in other general characters is so close 88 as to suggest a common 8.Ilcestry. 172 DISTRIBUTIONS FROM THE NATIONAL HERBARIUM. arranged in two airways, the 8Dlaller rOBUlate, horizontal, mostly tlterial, 1.5 to 2.5 em. long, the stalk from half to a third as long as the small suborbicular-cordate em. long, the stalk from half to a third as long as the small suborbicular heart-shaped . dense piloee lamina; larger fronds two or three, 4.5 to 8 em. long, stiffly erect or somewhat arched, the stem 2.5 to 5.5 em. long, pillless, slightly pallid at the .. baM, the lamina 2 to 5 cm. broad and long, BUborbicular or very obtuse rounded- .. triangular from a cordate or reniform base, the SinUB usually very deep; leaf tissue. The name Gymnopteris is not used by Diels in the sense of Bernhardi and Underwood, but is rather erroneously applied to. National Herbarium shows that it is rather a member of the genus currently known as Coniogramme and that it is closely related to Coniogram1M japonica, as Eaton suggested. In fact, the satisfactory arrangement of the species of this entire group must await a careful analysis and comparison of very numerous and widely divergent fOl"IIIs, many of them from distant regions. Lack of complete southern material prevented a complete treat- ment of the tropical American species by UndeI'wood and Lloyd in their paper' of 1906, although the North American members of the genus, with few exceptions, are now tolerably well understood. Stem very slender (about 0.5 mm in diameter) ), entirely hidden by the a.ppr&!ud imbricate Bases of the very numerous close set leaves,; leaVOll at the time in 8 Rows, not twisted at B. A deviant member of the group of L. par- I.on·cen&t UDderw. de: Lloyd,l which is known only from Puerto Rico. brachiatum a.ppeaJ"l!J to be judicious, which is unusual for members of the verticillatum group. portoricro.!e mostly on strongly divaricate branches. thoi of L. portoricenu diverging at an angle of SOo to 45° ) and fewer, shorter, more. Only a few South American members of the 8elago subgroup, to which !hiB 8p~iea belongs, have leaves arranged in 10 rows, and of these L. Of this i the last is io suppressed with reed from thruebout ita eteDl8 to reddish, ita Ieeer size and its lever much darker and broader8. Although a lax plant, it is very Ieee 10 as L. delicate species of the whole genus. epiphytic, pendulous, 20 to SO cm. long, 5 to 8 times dichotomous, the branching. luly scattered, discontinuous sporangiate in the apical part. 676072, collected from the fallen branch of a large tree in dense humid forest along the upper Caldera. NATIONAL IlERBABIUII:. to linear.lanceolate), less conapicuou.ely COI!Itate, and of a very thmel texture. Stems elender (0.6 to 0.8 mm, in diameter), partly obscured by the rather dense covering of leaves near baae, less 80 above, coarsely bulbous; leavetl dull yellowish greed, rather stiff membranous herbs, strongly ascending. There have recently been described by Brause 1 five new species of Cyathea from the West Indies, as follows: Oyathea tenuis from Cuba and Oyathea urbani, O.
