21. PSEUDOCROSSIDIUM       Plate 26– 27.

Pseudocrossidium Williams, Bull. Torrey Bot. Club 42: 396, 1915. Type: Pseudocrossidium chilense Williams.

Barbula sect. Revolutae BSG, Bryol. Eur. 2: 89, 1842 (Fasc. 13–15 Mon. 27). Type: Barbula revoluta Brid. in Schrad.

Tortula sect. Revolutae (BSG) Spruce, Ann. Mag. Nat. Hist. ser. 2, 3: 377, 1849.

Barbula sect. Platyneuron Kindb., Eur. N. Amer. Bryin. 2: 246, 1897. Type: Barbula platyneura C. Müll. & Kindb.

Barbula sect. Pseudocrossidium (Williams) Nyholm, Ill. Fl. Nordic Mo. 2: 102, 1989.

Barbula subsect. Revolutae (BSG) Chen, Hedwigia 80: 209, 1941.

            A large genus largely growing on soil and rock at high elevations; present in mountainous regions of the New World, Europe, the Middle East, Africa, and Australasia.

            Important characters for this genus are the usual small size or absence of the ventral stereid band and the broadly crescent-shaped dorsal stereid band with a clearly differentiated dorsal epidermis of cells often with semicircular lumens through differential thickening of the walls (Pl. 26, f. 7, 21; 27, f. 8, 13, 17). Other important characters are not consistent in appearance through the genus, but are often striking when found: leaves often ending in a short or long awn (Pl. 26, f. 5); differentiation of photosynthetic tissue either as a ventral pad of costal filaments (Pl. 26, f. 20, 21) or within rolled margins (Pl. 27, f. 17) or both (Pl. 27, f. 13); medial cells more papillose and thicker through (the distance between the two superficial walls) than the marginal cells (Pl. 26, f. 7; 27, f. 8); and perichaetial leaves abruptly enlarged and sheathing the seta (Pl. 26, f. 8).

            A few stereid cells of a second stereid band may be found in occasional specimens of what has been called P. aureum in Mexico and southwestern U.S.A. This is synonymized (independently by Sollman 1990) with P. crinitum, which usually has two stereid bands. The North American material is quite like the type of the South American Barbula arenicola, specimens of which have a single stereid band but which is also here placed in the synonymy of P. crinitum. Pseudocrossidium crinitum (Pl. 26, f. 1–8) is an essentially Gondwanaland taxon, being found in southern parts of Africa, South America and Australasia, which may indicate a southern origin for the genus. The fact that P. crinitum also occurs in Mexico and southwestern U.S.A., albeit as a sterile, depauperate (shorter awned and smaller stature) population, shows that there can be considerable northward extension of the ranges of species of ultimately southern derivation. A simple explanation of the evolutionary history of the genus, following the thread of a previous discussion (Zander 1979f) and discounting long-distance dispersal as a factor, would be that ancestors essentially identical to modern P. crinitum spread across Gondwanaland to be later isolated in austral areas through tectonic plate separation. In migrating northward along the Andes, ancestors of P. crinitum developed into the several species now there, these characterized by loss of the ventral stereid band and elaboration of photosynthetic tissue along the leaf margins or ventral costal surface. In North America, descendants of the derived species P. replicatum lost many of the characters of Pseudocrossidium while developing into the essentially high-northern latitude taxa P. revolutum and P. hornschuchianum (Pl. 27, f. 1–5), which may be referred to the genus by the general lack of a ventral stereid band, the highly revolute leaf margins with walls somewhat thinner and more papillose than the medial portion of the leaf, and the presence of prop agula on the ventral surface of the costa (possibly a vestige of the ventral costal elaboration). Thus, P. crinitum and P. revolutum, though once both regarded as Barbula species, are actually at opposite ends of a complex north-south evolutionary series through mid-Andean taxa of Pseudocrossidium. The northernmost taxa subsequently became established in southern Africa and Australia probably through human agency. Evidence for anthropogenic disjunction is the discovery of P. hornschuchianum in North America in Massachusetts in the U.S.A. (Mishler & Miller 1983) and in British Columbia, Canada (Tan et al. 1981), associated with parks or gardens; Arts (1988) has reported rhizoidal tubers in the closely related P. revolutum. In the southern hemisphere, P. crinitum can be separated into two morphotypes, which may rate recognition as taxa (if so, types of synonyms, cf. Catcheside 1980, Magill 1981, Weber 1972, should be consulted for the earliest name, and this would require a careful revision of the complex): one has a recurved leaf margin, coarse upper laminal papillae, dorsal costal epidermis absent or weakly developed, and ventral stereid band strong, while the second has revolute upper laminal margins, delicate upper laminal papillae, dorsal costal epidermis well developed, and ventral stereid band absent or weakly developed. Awned species of Syntrichia may have much the same appearance as P. crinitum (obtusely and broadly short-lanceolate leaves with densely papillose upper laminal cells) but differ in their distinctive red KOH color reaction, P. crinitum reacting deeply yellow.

            Pseudocrossidium leucocalyx (Pl. 27, f. 6–8) is unusual for the genus in its costa ending in a distinctive conical, smooth cell and rough dorsally with both sharp prorulae and solid simple papillae; the upper laminal cells medially strongly bulging on both sides, with high, solid, capitulate and spiculose papillae, but marginal cells smaller in several rows, weakly bulging and smooth; and lamina bright yellow in KOH. It may belong elsewhere, possibly in a monotypic genus of its own. It is superficially similar to Hypodontium species, which differ significantly, however, in their strongly incurved upper laminal margins and two strong stereid bands in the costa. Pseudocrossidium porphyreoneurum is a South African species having a long-mucronate costa and strongly revolute leaf margins, but is unusual in its semicircular (not crescent-shaped) stereid band and basal laminal cells not differentiated from the upper cells; it is placed here only tentatively (cf. Magill 1981, p. 213). The genus requires a thorough revision for adequate evaluation.

            Although P. elatum (Pl. 26, f. 18–22) is similar to Crossidium in its smooth upper laminal cells and rather rounded section of the costa, it is recognized here in Pseudocrossidium (following Delgadillo 1975a) because of its elongate stem, lanceolate leaves,a few stereid cells occasionally differentiated immediately below the ventral costal filaments, and the well-differentiated dorsal costal epidermis. The species remains, however, uncomfortably intermediate in morphology between Pseudocrossidium and Crossidium (especially C. spiralifolium of South Africa), while differing from both by the quite elongate patch of ventral costal filaments and poorly differentiated basal laminal cells. Eventual recognition of this species in a monotypic genus may be the best reflection of relational distances.