Revisionary studies of the Chandonanthoideae (Jungermanniales, Jungermanniaceae)
Schuster, Rudolf M.
The Chandonanthoideae Inoue are regarded as a rather ''weak'' segregate from the Lophozioideae (Jungermanniaceae). No taxon of the subfamily has heretofore been studied from sporophyte-bearing plants. Such studies, and studies of branching patterns, show that the current two-genus classification is not defensible. The group is regarded as formed of three genera which, basically, correspond to the three subgenera I recognized (Schuster 1961) under the base genus, Chandonanthus Mitt. Of the three elements, formerly united as Chandonanthus, the most isolated is C. squarrosus (Hook.) Mitt. The other two elements include Tetralophozia (Schust.) Schljakov and Plicanthus Schust., gen. n. The species of Plicanthus were heretofore included in Chandonanthus but are much more closely allied to Tetralophozia (in, i.a., the almost vertical and transverse, 3-4-lobed leaves, divided for 0.75-0.85 their length). Chandonanthus s. str. is shown to include a single species, C. squarrosus, isolated phytogeographically (it is the only purely Australasian element) and morphologically (it is the only element with consistently bilobed leaves, with a strongly succubous insertion and orientation; it is unique in the 7-8-layered capsule wall). Chandonanthus, s. str. is cool/cold Australasian in range; the genera Tetralophoziai and Plicanthus appear to be Laurasian in origin but show secondary invasion into tropical/subtropical sectors of Gondwanaland. The present revision has been limited by (a) lack of sporophytes of a number of taxa; (b) lack of androecial plants of many taxa; (c) lack of copious gatherings of many taxa, so that ramification patterns could not be thoroughly studied for most of them. A formal taxonomic revision, therefore, is premature. Until recent years, the Chandonanthoideae were held to include a single genus, Chandonanthus. The type species, C. squarrosus, with uniformly bilobed leaves and a coarsely papillose cuticle, is totally isolated from all other taxa in the group. Subg. Tetralophozia Schust. (1961) was elevated to generic rank by Schljakov (1976) and the entity has been generally adopted at this rank subsequently (as by Urmi 1983). However, Tetralophoziai is much closer in its basic criteria to the Chandonanthus hirtellus-hamatus element than the latter is to Chandonanthus s.str. The currently accepted two-genus classification is demonstrably artificial: either Tetralophozia is accepted as a subgenus of a diversified genus Chandonanthus, or the ''rump'' genus Chandonanthus is split in two. In essence, we adopt either a three-genus classification or a one-genus one. Although I had opted for a one genus classification (Schuster 1969), with general adoption of Tetralophozia as an autonomous genus a procedure I continue to have conceptual problems with two choices are open to us: (a) move all taxa not placed in Chandonanthus into Tetralophozia, except for the generic type; or (b) recognize three genera. I have, with misgivings, opted for this last solution. If Tetralophozia is recognized for the T. setiformis element, then Chandonanthus piliferus should be assigned to it (as T.pilifera), but this element is quite isolated by its leaf cells and by the formation of incipient water-sacs; it perhaps should be recognized as an autonomous subgenus. If Tetralophozia is recognized for, basically, taxa with ± symmetrically quadrifid leaves, then but only then the other taxa with more than 2-lobed leaves, the C. hirtellus element must be regarded as forming a third genus, Plicanthus gen. n. All taxa fitting here have, basically asymmetrically trifid leaves vs. subequally quadrifid ones in Tetralophozia vs. consistently bifid ones in Chandonanthus. Although such a crude sorting of taxa has the merit of simplicity, it only imperfectly reflects reality: Tetralophozia has sporadic trifid leaves.