A Note on the Genus Gymnocolea Dum.

Abstract

Two species referred by Stephani to Lophozia Dum. are found to belong to Gymnocolea Dum. and the new combinations G. multiflora (Steph.) Schust. and G. cylindriformis (Mitt.) Schust. are proposed. The former is the type of a new subgenus, Gymnocoleopsis, which differs from subgen. Glymnocolea in lacking terminal branches, the 2-3-stratose perianths, the reduced seta, and other features. The generic and subgeneric placements of many taxa of the Lophoziaceae remain unresolved. In the course of completing the treatment of the genus Gymnocolea for Volume II of The Hepaticae and Anthocerotae of North America, I have tried to resolve various problems dealing with the delimitation of Gymnocolea as opposed to Lophozia. Two antipodal taxa have been studied which, although previously referred to Lophozia, cannot remain there if a genus Gymnocolea is to be recognized. Although they are clearly closer to Gymnocolea, they form a foreign element in that genus. In my opinion, a division of Gymnocolea into two subgenera is necessary. The holarctic species all belong to the subg. Gymnocolea; in the Andes of South America, a totally deviant taxon occurs: Gymnocolea (Gymnocoleopsis) multiflora (Steph.) Schust., comb. nov. (Basionym: Lophozia multiflora Steph., Spec. Hep. 6: 113. 1917). This species should perhaps be divorced from Gymnocolea entirely, and is surely distinguishable at the subgeneric level. It may be separated from subgen. Gymnocolea by the following key: KEY TO SUBGENERA OF GYMNOCOLEA 1. Terminal branching of the Frullania-type common; postical-intercalary branching frequent; perianth 1-stratose, usually caducous in absence of fertilization; dioecious; seta with numerous cell rows; capsule wall 2-stratose, its cells little elongated; gynoecium with bracteole absent or vestigial ------------------Subgen. GYMNOCOLEA Dum. 1. Terminal branching of the Frullania-type lacking; postical-intercalary branching lacking; all branches lateral-intercalary; perianth 2-3-stratose to middle or higher, persistent; autoecious; seta with 8 epidermal, 4(5) internal cell rows; capsule wall 2-3-stratose, of strongly elongate cells; gynoecium with a large, ovate to bilobed bracteole united with both bracts. -------------------Subgen. GYMNOCOLEOPSIS Schust.2 The peculiar southern hemisphere plant from Kerguelen Island (Royal Sound, Eaton) described by Mitten as Jungermannia cylindriformis Mitt. and placed by Stephani (Spec. Hep. 2: 143. 1901.) in Lophozia, as L. cylindriformis also appears to belong here. It has the brownish coloration, terete and clavate perianths lobulate and crenulate at the mouth, of Gymnocolea. Also, as in Gymnocolea, the perianths are feebly plicate above only; the leaf cells are thick-walled, with yellow-brown to brown 1 Department of Botany, University of Massachusetts, Amherst, Massachusetts. 2 Gymnocolea subgen. Gymnocoleopsis Schust., subgen. nov. Monoica; rami intercalares laterales; folia concava, lobis triangulatis, apice rotundatis; perianthia 2-3-stratosa, apice truncato, inciso-lobata, minuteque crenato vel integra; androecia longe spicata, bracteis 6-12 jugis. Type: G. multiflora (Steph.) Schust. This content downloaded from 157.55.39.120 on Sun, 04 Dec 2016 04:52:08 UTC All use subject to http://about.jstor.org/terms 112 THE BRYOLOGIST [Volume 70 walls; the underleaves are absent to minute on gynoecial stems; and the leaves are bilobed with blunt lobes. Although, in the lack of adequate data on axial anatomy, sporophyte anatomy, and other criteria, generic placement is hardly definitively possible, I agree with both the original annotations of Mitten and Stephani that the plant resembles Gymnocolea inflata, and suggest it be placed in Gymnocolea as G. cylindriformis (Mitt.) Schust., comb. nov.3 3Basionym: Jungermannia cylindriformis Mitt., Jour. & Proc. Linn. Soc. 15: 196. 1877. I have studied fragments of the type (c); the plants no longer regain their form in water. The Discovery of Tortula vectensis in North America