Holomitrium trichopodum (Mitt.) Klazenga

Plants medium-sized, bright or yellow-green, often turning brown when dry, forming cushions. Stems often branched by forking, commonly c. 10–50 mm, lacking microphyllous axillary shoots, in cross-section with a few thick-walled outer cell layers and a distinct central strand, tomentose with pale brown, smooth, and strongly contorted rhizoids. Leaves loosely to strongly secund when moist, not contorted when dry, narrowly ovate-lanceolate, evenly tapered to a long subulate and ± tubulose acumen, entire except for a few ± spinose teeth at apex, (3–)5–10(–16) × c. 0.4–0.7 mm (under cover slip), scarcely sheathing below, strongly concave to tubulose throughout, with lamina at mid leaf narrow and indistinct; mid laminal cells mostly rounded-oblong, mostly c. 8–15 × 4–6 µm and c. (1–)2–3:1, never oblate, often obscure in surface view, firm-walled or incrassate, smooth or weakly mammillate abaxially, non-porose, transitioning gradually to the cells of the leaf base, those in upper subula obscure and apparently not bistratose; cells of leaf base narrowly oblong to ± linear, incrassate, becoming strongly porose and ± pigmented near insertion; alar cells strongly pigmented, yellow-brown, enlarged, oblong or subquadrate, incrassate, forming a large group extending nearly to the costa and usually extending >8 cells up the margin. Costa filling c. ⅓ the leaf base width, filling the upper half of leaf, cross-section in mid leaf with a single layer of median guide cells and large abaxial and adaxial stereid groups with cells of the exposed adaxial layer (at mid leaf and below) similar to the enclosed cells and lacking a distinct lumen; in adaxial surface view (at mid leaf) cells elongate.

Pseudautoicous. Perichaetial leaves sheathing the setae, strongly shouldered, abruptly tapered from an oblong and sheathing base to a slender and spreading subula, weakly costate, mostly 7–8 mm. Dwarf males rarely observed, c. 2 mm, sometimes innovated and with multiple (–4 or possibly more) perigonia. Setae mostly (17–)20–40(–50) mm, slender and flexuose, pale brown or yellow; capsules erect, cylindric, scarcely constricted at the transverse mouth when dry, c. 2.0–2.5(–3.0) mm, smooth, straw-coloured; exothecial cells irregular in outline, not in distinct columns, with moderately and uniformly thickened walls; stomata few at extreme base of capsule, superficial; operculum long rostrate, nearly equal the capsule in length. Peristome teeth inserted near the rim, paired and often partially conjoined by extensions of lateral trabeculae near their base, red-brown, unequally divided from c. ⅔ their length or ± to the base (the segments sometimes anastomosing), c. 180–200 × 40 µm, strongly papillose-baculate throughout (apparently on both surfaces). Calyptra c. 4 mm. Spores c. 18 µm, spherical, smooth or nearly so.

Klazenga 2003, fig. 29; Klazenga 2003, fig. 2 a–b (both as Dicranoloma trichopodum).

In the field Holomitrium trichopodum is unlikely to be confused with H. perichaetiale, although the two often grow intermixed.

Confusion has arisen concerning the identity of aberrant H. trichopodum from the Stockton and Denniston Plateaux (Nelson L.D.). Some specimens have been reported (Bartlett & Frahm 1983) as belonging to the South American Chorisodontium aciphyllum (Hook.f. & Wilson) Broth., and a modest number of specimens have been subsequently placed in N.Z herbaria as this species by other workers, including me. However, numerous collections (in CHR), made by D. Glenny & K.A. Ford from those plateaux, as part of a survey specifically aimed at clarifying the N.Z. status of C. aciphyllum, demonstrate the variability of epilithic populations of H. trichopodum there. These collections, together with Bartlett’s original collections (in AK and CHR) in some instances have some vegetative leaves extremely short (c. 3–4 mm) for H. trichopodum. However, in such populations leaves from the upper shoots that are longer and more slenderly lanceolate (and hence more representative of the species) have in all instances the areolation and costal structure representative of H. trichopodum. Several collections from the plateaux also have either setae or intact sporophytes representative of H. trichopodum. Chorisodontium aciphyllum is rejected here as a member of the N.Z. flora. Two syntypes of the basionym of C. aciphyllum (Dicranum aciphyllum Hook.f. & Wilson, London J. Bot. 3: 541 [1844]) have been seen in the Mitten herbarium (NY 267989! & NY 267990!).

When sterile, H. trichopodum can be difficult to distinguish from Dicranoloma menziesii without microscopic examination, but H. trichopodum normally has a brighter green coloration. While similar in shape, the leaves in H. trichopodum are narrower (usually 0.4–0.7 mm vs. 0.7–0.9 mm wide under a cover slip). In H. trichopodum the spine-like marginal teeth are strongly restricted to the apex and project at varying angles from the leaf axis. By contrast, D. menziesii is sparsely serrate (not spinose), with the teeth smaller, more numerous, and sometimes extending for as much as ½ the leaf length. In H. trichopodum the mid leaf laminal cells are mostly 2–3:1, while those of D. menziesii are mostly isodiametric. The abaxial surface of the costa in H. trichopodum is smooth or weakly mammillate, while that of D. menziesii is spinose by projecting cell ends. In cross-section the adaxial surface layer of costal cells (at mid leaf & below) of H. trichopodum are stereid-like and lack a distinct lumen, while the adaxial surface layer of costal cells of D. menziesii have a distinct lumen and are clearly differentiated from the adjacent stereids. The elongate nature of the adaxial surface costal cells in H. trichopodum can usually be seen in surface view. When fruiting, these two species are distinguished by numerous sporophytic features, including seta length and capsule orientation, and they are unlikely to be confused.

Holomitrium trichopodum can also be confused with Kiaeria pumila (=Dicranum aucklandicum Dixon). Compared to the latter species, H. trichopodum is a larger plant with longer stems, longer leaves (usually c. 6–10 mm), broader costae (c. ⅓ the width of the leaf base) with two stereid bands, shorter upper laminal cells (c. 3–5:1 and ± oblong), pseudautoicous sexuality, very long and fine setae (mostly 20–30 mm) and erect, elongate (2.0–3.0) mm), and cylindric capsules. By contrast, Kiaeria pumila is a smaller plant with shorter stems (c. 6–13 mm), shorter (c. 3 mm) leaves, narrower costae (c. ⅙ the width of the leaf base), which lack stereid bands, elongate upper laminal cells, autoicous sexuality, shorter setae (c. 8–11 mm), and shorter (c. 1.0–1.3 mm), slightly curved, oblong-cylindric capsules.

Mostly on tree trunks and larger branches, often those that are sloping or nearly horizontal. It is most common on various species of southern beech (Fuscospora and Lophozonia), but also occurs on Beilschmiedia tawa, Elaeocarpus hookerianus, Ixerba brexioides, Weinmannia sylvicola, Halocarpus biformis, and Libocedrus bidwillii, as well as on tree ferns. It is often found on fallen branches in beech forest, suggesting that this species is well developed in the forest canopy. At Lake Paringa (Westland L.D.) material fallen from the forest canopy has been seen persisting as an ellipsoid "moss-ball" on the forest floor, in mixture with Cladomnion ericoides, Mesotus celatus, and Papillaria flavolimbata. Holomitrium trichopodum also occurs on rock (granite, gneiss, sandstone, and probably greywacke) at higher elevations, and rarely occurs on leaf duff. This species is common on rock at the Stockton and Denniston Plateaux (Nelson L.D.), where it forms loose cushions up to c. 8 cm diam.

Holomitrium trichopodum is largely confined to high rainfall areas. On the South I. it occurs primarily near or west of the Main Divide; it also appears to be absent from drier parts of the North I. Its N.Z. distribution mirrors its predominantly western distribution in Tasmania (Klazenga 2003).

Occurring to at least 975 m (but recorded from 1400 m at Mt Manuoha, Gisborne L.D.) on the North I. and from near sea level to at least 1190 m (Mt Euclid, Nelson L.D.) on the South I. This species can be associated with a wide variety of epiphytic/epilithic species, with Dicnemon spp. and Holomitrium perichaetiale, and the hepatics Herbertus alpinus and Lepicolea scolopendra, the most frequent. Cladomnion ericoides, Dicranoloma menziesii, Ditrichum punctulatum, Leptostomum inclinans, Pyrrhobryum mnioides, P. pennatum, Wijkia extenuata, and Chandonanthus squarrosus are less frequent associates.

As in H. perichaetiale, the ♂ plants are normally nanandrous (dwarfed). However, I have seen fully developed (large stature) ♂ plants in a herbarium specimen from Pegasus Bay (Stewart I., B.A. Fineran 211, CHR 609464), as well as what appeared to be a single, weakly dwarfed (5 mm long) ♂ plant in material from Kellys Creek (Westland L.D., W. Martin 155.17, CHR 609453).

The generic placement of this species is problematic. Sainsbury (1955) retained this species in Dicranum, where it was originally described by Mitten; Fife (1995) followed this placement. Brotherus (1924) made a new combination in Dicranoloma, the genus in which it was retained by Klazenga (2003). Klazenga noted a gametophytic resemblance to D. menziesii but also (p. 467) expressed doubt that the species should remain in Dicranoloma. More recently Klazenga (2006), using both nuclear and chloroplast spacer sequence data and peristome morphology, argued for placement of this species in Holomitrium. Although neither of his molecular-data-derived cladograms showed the Holomitrium clade to be strongly supported (and in one case critical parts are unresolved), Klazenga argued that the congruence of results from two molecular data sets added weight to this conclusion. Klazenga (2003) discussed the H. trichopodum peristome in detail, and subsequently (Klazenga 2006) he discussed the variation in peristome structure in both the New World and Old World members of the Holomitrium complex. While conceding that the divided peristome teeth of H. trichopodum are aberrant in Holomitrium, Klazenga argued that divided teeth have been recorded from two other species (including H. perichaetiale) in this genus. Klazenga’s generic assignment is reservedly followed here. Klazenga (2006, p. 301) stated, "because of its apparently isolated position in the Holomitrium-clade, its lack of many morphological features that characterised the rest of the clade, and its sharing of other morphological features with the related Dicranum- and Dicranoloma-clades, Holomitrium trichopodum may have an important role to play in the resolution of the phylogenetic relationships and the study of character evolution within the Holomitrium clade."