10. TRICHOSTOMUM          Plates 11, 12– 13.

Trichostomum Bruch, Flora 12: 396, 1829, nom. cons. non Hedw., 1801. Lectotype: Trichostomum brachydontium Bruch.

Subg. Trichostomum (Hedw.) Turn., Musc. Hib. Spic. 35, 1804 (as autonym).

Trichostomum subg. Trichostomum Lor., Bryol. Notizb. 20: 1865, nom. illeg.

Didymodon subg. Trichostomum Kindb., Eur. N. Amer. Bryin. 2: 272, 1897. Type: Trichostomum Hedw., nom. rej., p.p.

Bryum sect. Trichostomum (Hedw.) Relh., Fl. Cantabr. ed 2: 422, 1802, nom. illeg.

Trichostomum sect. Lancifolia B.&S. in BSG, Bryol. Eur. 2: 119 [fasc. 18–20 (Trichost.): 5], 1843.

Trichostomum sect. Pycnophyllum C. Müll., Syn. 1: 567, 1849.

Tortula sect. Trichostomum Mitt., J. Linn. Soc. Bot. 12: 142, 146, 1869, nom. illeg.

Subg. Crispuliformes (Kindb.) Zander, comb. et stat. nov. see below.

Didymodon sect. Crispuliformes Kindb., Eur. N. Amer. Bryin. 2: 272, 1897. Type: Didymodon crispulus (Bruch) Wils.

Subg. Laminanchium Zander, subg. nov. see below. Type: Trichostomum tortelloides (Broth. & Dix.) Zand.

Subg. Oxystegus Limpr., Laubm. Deutschl. 1: 569, 1888. Type: Trichostomum cylindricum (Bruch ex Brid.) C. Müll. fide Saito, J. Hattori Bot. Lab. 39: 436, 1975, hom. illeg.

Oxystegus (Limpr.) Hilp., Beih. Bot. Centralbl. 50: 666, 1933.

Stephanodictyon Dix., J. Linn. Soc. Bot. 50: 86, 1935. Type: Stephanodictyon borneense Dix.

Paraleptodontium Long, J. Bryol. 12: 181, 1982. Type: Paraleptodontium recurvifolium (Tayl.) Long.

Didymodon subg. Oxystegus (Limpr.) Roth, Eur. Laubm. 1: 304, 1904.

Sect. Campylopus Arnott, Mém. Soc. Linn. Paris 5: 244, 1827.

Sect. Leptomitrium Wallr., Fl. Crypt. Germ. 1: 170, 1831.

            A large genus found on soil, rock (often calcareous) or organic material on all continents except Antarctica.

     Trichostomum has been a large, “wastebasket” genus of species that cannot be easily assigned to other genera. It differs from similar Barbula (Merceyoideae) species with plane or erect upper leaf margins (B. sect. Convolutae) by the cells of the outer layer of the stem usually with large lumens (Pl. 11, f. 14; 13, f. 11), lower leaf margins plane, upper leaf margins seldom dentate (but commonly minutely serrulate by projecting cell walls), costal hydroid strand seldom present (Pl. 11, f. 12), dorsal costal epidermal cells sometimes not differentiated (Pl. 12, f. 21), ventral stereid band generally large, almost or about the thickness of the dorsal band, perichaetial leaves generally not differentiated from the cauline, and propagula (Pl. 11, f. 5) very rarely present. The genus differs from Hyophila by the mostly narrower leaves that are broadest at the base, laminal cells usually somewhat bulging on both exposed surfaces (rather than just one surface though this is variable in Hyophila), propagula usually lacking, and peristome usually present (among other characters, see Hyophila); however, the thick-walled exothecial cells of Hyophila (forming a series of concatenated semicircles as seen in transverse section of the theca) are matched in some eperistomate species of Trichostomum.

     I have found no good distinctions between Trichostomum and Oxystegus at the generic level. There are no characters of the sporophyte that acceptably distinguish the two; for instance, Trichostomum (subg. Trichostomum) brachydontium may have the spiral peristome tooth ornamentation typical of T. (subg. Oxystegus) tenuirostre). Gametophytically, subg. Oxystegus differs only weakly from subg. Trichostomum in (1) generally plane (occasionally tubulose when dry) upper leaf laminae, (2) a tendency toward broadly sheathing leaf bases with the differentiated marginal cells somewhat running up the margins as in Tortella, (3) medial upper basal cells thick-walled and rectangular (forming a group sensibly different from the upper cells and lower basal cells), and (4) the upper laminal cells often distinctly enlarged and rounded rectangular, 1–2:1. The first two of these characters are, however, also present to some degree in Trichostomum sect. Trichostomum, and the last is neither constant nor easily gauged. Crum and Anderson (1981) likewise proposed that Trichostomum include Oxystegus. Saito (1975a) recognized Oxystegus but did not discuss it as a genus. His key to genera of the Trichostomoideae separated Oxystegus and Pseudosymblepharis from Trichostomum by the presence of a central strand in the last; Eddy (1990) used the same character to separate the two. This distinction does not, however, hold outside of Japan or, apparently, Malesia. Stoneburner (1985) expressed the opinion that differences between Trichostomum and Oxystegus were probably artificial. Norris and Koponen (1989) distinguished between Oxystegus and Trichostomum by “a suite of differing, although somewhat overlapping characters”: Trichostomum having leaves with lengths less than 6:1, blunt papillae grouped over the lumens, and laminal marginal cells thick-walled near the shoulders; Oxystegus having leaves longer than 6:1, sharp papillae grouped around the lumens, and cells of the shoulders thin-walled. These differences are recognized here at the subgeneric level. They also indicated that although Oxystegus and Pseudosymblepharis are morphologically close gametophytically, the red peristome of the former contrasts with the whitish peristome of the latter.

     There are trends in subg. Oxystegus towards three morphotypes of robust stature: a lingulate leaf as in Trichostomum recurvifolium (= Paraleptodontium Long 1982b, a synonym), a lanceolate leaf (e.g. Trichostomum hibernicum), and a linear leaf (e.g. Trichostomum borneense, originally published as the type of Stephanodictyon). Such tendencies are duplicated in infraspecific variation in the central species of the Oxystegus complex—T. tenuirostre. Eddy (1990) made Stephanodictyon a synonym of Pseudosymblepharis (along with Chionoloma, here recognized as a good genus).

            Of species common in North Temperate Zone climates, Trichostomum (subg. Oxystegus) tenuirostre is rather different from T. crispulum and T. brachydontium (both of subg. Trichostomum) in its plane margins, vaguely vee-shaped area of differentiated basal cells, and less well-developed peristome; as a genus, however, Oxystegus cannot be maintained at the world level because of considerable variation in expression and combination of characters among the species. For instance, the difference between the peristomes of T. crispulum (often 32 filamentous rami like of those of Tortella, but sometimes reduced) and T. brachydontium (16 variously cleft or perforated lanceolate teeth) is greater than that between the peristomes of T. brachydontium (generitype of Trichostomum) and T. (subg. Oxystegus) tenuirostre.

            Tortella can be separated from Trichostomum on the basis of a single relatively constant character, the vee-shaped basal cell region, with the usually long and twisted peristome and generally larger upper laminal cells being associated trends; see Cladogram 15 for analysis of phylogenetically important character state changes. Species of Trichostomum may also have basal cells differentiated in a (less distinctive) vee-shape, and those with flattened autoicous buds (e.g. Trichostomum fragilifolium, Pl. 12, f. 14) differ gametophytically from certain Tortella species (like Tortella humilis) with similar buds only in the basal cells being differentiated evenly across the leaf rather than running higher up the margins. Quite probably, Trichostomum, Pseudosymblepharis and possibly also Tortella represent a complex of several transformation series. This needs to be studied. Future revision of the Trichostomum complex may turn up a satisfactory generic classification recognizing several separate phyletic series, with perhaps Trichostomum brachydontium, T. crispulum, Weissia jamaicensis or even species of Tortella variously at the unreduced end of each.

            Species of Weissia (and of Tetrapterum) with clearly plane or erect-incurved upper laminal margins (including subg. Hymenostomum and subg. Astomum) examined during this study are here transferred to Trichostomum s. lat. Groups of Trichostomum species very similar to each other in areolation and certain other gametophytic characters may be construed (until analysis demonstrates otherwise) as separate reduction series toward smaller stature of gametophyte, shorter seta or capsule (the latter occasionally almost spherical), short or absent peristome, and monoicy, and are recognized below as subgenera.

Trichostomum subg. Trichostomum

Leaves ovate to long-lanceolate, apex usually rounded, plane; upper margins plane; costa thick and excurrent in a stout, sharp mucro; basal cells moderately differentiated across leaf base.

   Occasionally the upper margins in some leaves are rather narrowly incurved, like Weissia species. A reduction series (large to small stature with concomitant reduction of morphological details) might include: Trichostomum brachydontium, T. planifolium, T. urceolare, T. sinaloense (Pl. 13, f. 16–19), T. termitarum, T. williamsii, T. unguiculatum and T. austrocrispum. Also belonging here is T. incertum. Trichostomum subg. Crispuliformes (Kindb.) Zander, comb. and stat. nov. Basionym:  Didymodon sect. Crispuliformes Kindb., Eur. N. Amer. Bryin. 2: 272, 1897. Type: Trichostomum crispulum Bruch ex F. A. Müller.

Leaves elliptical to long-lanceolate; apex rounded-acute, often somewhat cucullate; upper margins broadly incurved; costa evenly tapering and generally ending before the apex or percurrent; basal cells moderately differentiated across leaf base.

   This taxon includes the series: Tortella inflexa, Trichostomum crispulum (Pl. 11, f. 18–21), T. castaneum, T. caespitosum, T. brittonianum, T. subangustifolium, T. atrocaule, T. perligulatum, and possibly Weissia crispa. Also belonging here are T. connivens and T. pulicare.

Trichostomum subg. Laminanchium Zander, subg. nov. Type: Trichostomum tortelloides (Broth. & Dix.) Zand.

Folia longiligulata, et plerumque in regione supra basem subvaginantem angustata et regione sub apice, apice acuto, plano, marginibus supernis et interdum infernis planis, costa decrescenti excurrentique, mucronem formanti, cellulis basalibus valde in regione mediana distinctis et in marginali cellulissubangustatis praedita.

     Leaves long-ligulate, often pinched just above the somewhat sheathing base and also just below the apex; apex acute, plane; upper margins and sometimes the lower margins plane; costa tapering and excurrent as a mucro; basal cells strongly differentiated medially with narrower cells marginally.

     The upper medial superficial cell walls of most species of this subgenus are ventrally bulging and dorsally nearly flat, and are also commonly papillose. Few sporophytes have been seen, but these appear to uniformly lack peristomes and sometimes have rather short setae. Species included here are similar to Barbula sect. Convolutae by the mainly medially differentiated basal cells, but the basal cells of species of sect. Convolutae are rather thick-walled and not strongly differentiated. Species of subg. Laminachium are: Trichostomum bombayense (Pl. 12, f. 1–3, and see discussion of Townsend 1983), T. contractum (Pl. 11, f. 13–16), T. criotum (autoicous, with flattened axillary perigonia), and T. tortelloides (Pl. 13, f. 20–23). See also discussion of Calymperastrum.

Trichostomum subg. Oxystegus Limpr. Type: Trichostomum tenuirostre (Hook. & Tayl.) Lindb.

Leaves lanceolate to linear-lanceolate, occasionally with a sharply dilated base; apex rounded to narrowly acute, plane; upper margins plane; costa tapering and excurrent as a short mucro; basal cells moderately differentiated, often running up the margins somewhat as in Tortella, and bordered distally by a region of thick-walled rectangular cells.

     Species of this subgenus could include the following series: Trichostomum tenuirostre (Pl. 11, f. 1–6), T. duidense (Pl. 12, f. 4–8), T. melanostomum, T. spirale and T. abyssinicum. There are certainly many other names that might go here.

Trichostomum caespitosum, having leaves with small upper laminal cells and broadly incurved margins, and being autoicous, was recognized in Pottia recently by both Smith (1978) and Corley et al. (1981), but clearly fits in a reduction series involving T. crispulum. Its distinctly differentiated perichaetial leaves are, however, unusual in the genus.

     The idea that Trichostomum might include taxa with much reduced or absent peristomes has been entertained recently by several bryologists (pers. comm.) with interest in the group. Magill (1981) pointed out that Trichostomum brachydontium has rudimentary peristomes or gymnostomous capsules both in South Africa and South America. He also noted that the gametophyte of the gymnostomous and rhexolytically operculate species Phasconica tisserantii (= T. unguiculatum cf. Crundwell & Nyholm 1974) resembles that of T. brachydontium. Brotherus (1924–25) avoided the problem of the relationship of species with Trichostomum-like gametophytes and cleistocarpous capsules by referring all these to Tetrapterum.

     Tuerckheimia is quite like Trichostomum in general appearance but the leaf bases of the former genus are not differentiated in shape, the basal cells are merely short-rectangular in a small area near the insertion, the upper laminal cells have a massive, multiplex papilla over the center of each lumen (but not covering the lumen as in Oxystegus), and the leaves (often widest near the middle) are pale yellow in KOH solution.

     Non-type specimens from India collected by Ramskuhl (BM) and identified as Desmatodon longirostris (Griff.) Mitt. (= Merceyopsis longirostris (Griff.) Broth. & Dix.) proved to be the superficially similar Scopelophila cataractae; the correct name for D. longirostris is now Trichostomum contractum, nom. nov. (see below).