Amblystegiaceae Kindb.

The following family description is modified from Kanda (1976) and Hedenäs (2003).

Plants small to robust, mat-forming, mostly in damp to aquatic habitats. Stems creeping or erect and self-supporting, irregularly to pinnately branched, with branches either distichously or radially arranged, in cross-section usually with a central strand and with or without a hyaloderm. Stem leaves erect-spreading, spreading, or occasionally squarrose, inserted in many rows, straight or falcate, lanceolate, ovate-lanceolate, ovate, or nearly round, sometimes cordate or auriculate and often decurrent, rounded, acute, or acuminate at apex, entire to serrate, mostly plane at margins; mid laminal cells oblong to elongate, smooth or sometimes prorate, incrassate or thin-walled, not or variably porose, mostly unistratose or rarely bistratose to multistratose; alar cells mostly differentiated, inflated or small and thick-walled. Costa mostly single and elongate, sometimes forked or short and double, rarely absent. Branch leaves similar but often smaller than stem leaves. Axillary hairs numerous or sparse, cylindric, consisting of c. 4–9 cells, with the basal 1 or 2 weakly pigmented and the upper cells yellow-brown. Paraphyllia absent or present.

Autoicous or dioicous. Perichaetia lateral, scattered on stems, the leaves costate and often plicate. Perigonia mostly gemmiform and scattered. Setae elongate, smooth; capsules oblong or cylindric, mostly curved and inclined or rarely ± erect, smooth, constricted below the mouth when dry; exothecial cells variable; stomata restricted to neck, superficial, long-pored; annulus differentiated or rarely absent; operculum conic or rarely short-rostrate. Peristome hypnoid; exostome teeth yellow-brown and cross-striolate below, ± hyaline and papillose above, narrowly bordered, trabeculate on inner surface; endostome with a high basal membrane, pale, the segments keeled and ± perforate, with cilia in groups of 1–4, elongate, nodose or appendiculate. Calyptra cucullate, naked. Spores spherical and finely papillose.

The Amblystegiaceae are a large, systematically and taxonomically difficult family. The family is best developed in temperate and colder regions of the northern hemisphere, with most species occupying aquatic or moist habitats. The family has significant outliers in the southern hemisphere, including N.Z. The family is treated here in the broad (and conservative) sense advocated by Hedenäs (2003), who has studied the family extensively in the northern hemisphere and in the neotropics. A broad interpretation of the family is preferable until a greater consensus is achieved concerning the family and generic limits in it on a world scale.

The Amblystegiaceae in N.Z. is considered here to include twelve genera: Amblystegium, Calliergon, Calliergonella, Campyliadelphus, Cratoneuron, Cratoneuropsis, Drepanocladus, Leptodictyum, Sanionia, Scorpidium, Straminergon, and Warnstorfia. These genera each have only a single species in N.Z. with the exception of Amblystegium, Campyliadelphus, Scorpidium, and Warnstorfia which each have two species.

An alternative treatment recognising a segregate familiy (Calliergonaceae) in addition to a somewhat reduced Amblystegiaceae has been advocated by Goffinet et al. (2009). They recognised all of the 12 genera cited above, albeit with some variation in family assignment. Additionally, they accepted Orthotheciella (Müll.Hal.) Ochyra (which is discussed briefly under Amblystegium, but not accepted here) and Hypnobartlettia Ochyra (which is here included within Cratoneuropsis).

The proposed placement (excluding Calliergonella, which they placed in the Hypnaceae) of the N.Z. genera by Goffinet et al. (2009) can be summarised as follows:

Amblystegiaceae: Amblystegium (plus Orthotheciella), Campyliadelphus, Cratoneuron, Cratoneuropsis (plus Hypnobartlettia), Drepanocladus, Leptodictyum, Sanionia, and Scorpidium

Calliergonaceae: Calliergon, Straminergon, and Warnstorfia

A different interpretation of the Amblystegiaceae and its allied families was provided by Buck & Goffinet (2000).

Generic limits and placements in the Amblystegiaceae remain confused, despite decades of study and debate. Because the greatest diversity of taxa occurs in the northern hemisphere, the study of the N.Z. taxa can contribute only modestly to an overall understanding of the family. The generic limits employed here rely heavily on the studies of Hedenäs (particularly Hedenäs 1989a, 1989b, 1992, 1993, 1995, 1996, 1997, 2003; Hedenäs & Kooijman 1996). For the most part they are similar to the generic limits employed by Buck (1998). Some of Hedenäs’ generic concepts seem to partly derive from Drepanocladus segregates advocated by Tuomikoski et al. (1973).

Broader generic concepts in this family are presented in several 20th century regional Floras, including those of Crum & Anderson (1981) and Smith (1978). (The second edition of The Moss Flora of Britain and Ireland (Smith 2004) adopts the narrower generic concepts advocated by Hedenäs.) Crum & Anderson’s (1981) discussions highlight many instances where the genera intergrade as well as instances (such as in Drepanocladus s.l.) where the recognition of segregate genera result in ill-defined and unnatural residual groupings of species. Kanda’s (1976, 1977) treatment of the family for Japan generally presents narrow generic concepts (with the exception of Drepanocladus). Kanda’s concept of Camypliadelphus is followed here. Kanda’s (1976, 1977) Japanese treatment and Hedenäs’ (2003) monograph for Central America and tropical South America are particularly useful in understanding N.Z. species.

Species of Amblystegiaceae with a single elongate costa can easily be confused with members of the Brachytheciaceae. However, in the Amblystegiaceae the capsules are more elongate and paler in colour and the exostome teeth are usually yellow-brown rather than dark red-brown. The stomatal pores in Amblystegiaceae are elongate, while those in Brachytheciaceae are short and ± round. The degree of costal development often varies in a single plant in the Amblystegiaceae; this rarely occurs in the Brachytheciaceae. In the Brachytheciaceae the costa often ends in a terminal spine, but terminal costal spines do not occur in the Amblystegiaceae. In the Amblystegiaceae the stem and branch leaves are nearly always similar in shape, whereas there is a strong tendency in the Brachytheciaceae towards dimorphy, both in shape and size. Finally, members of the Amblystegiaceae usually inhabit wetter sites than do the generally more meso- or xerophytic species of the Brachytheciaceae.

Axillary hairs (see Hedenäs 1990) appear to be of little taxonomic value in the Amblystegiaceae. The ease with which they can be observed in different genera varies widely; staining with a proprietary dye such as toluidine blue can be helpful. I am confused by Hedenäs’ illustrations (see Hedenäs 1990, figs. 69–107) of the basal cells of the axillary hairs being more pigmented than the upper (cylindric) cells. In my limited experience the upper cylindric cells are yellow-brown in colour and the basal cell is the least pigmented. I have made very little use of axillary hair characters here.

Acrocladium Mitt. was placed in the Amblystegiaceae by Sainsbury (1955), who recognised a single N.Z. species. Acrocladium auriculatum sensu Sainsbury is considered here to be a synonym of Acrocladium chlamydophyllum (Hook.f. & Wilson) Müll.Hal. & Broth. and will be treated elsewhere in the Flora. The genus was placed in the Lembophyllaceae by Goffinet et al. (2009), but a recent analysis by Tangney et al. (2010) suggested it would be more appropriately placed in a monotypic family with affinities to Lepyrodontaceae.

Calliergidium (Renauld) Grout was recorded from N.Z. by Bartlett (1984). He recorded Calliergidium austro-stramineum (Müll.Hal.) E.B.Bartram from Cape Foulwind (Nelson L.D.) and from near Haast Bridge (Westland L.D), based on determinations by H. Kanda. Bartlett stated that the two localities "represent only two of the known localities as the plant has been collected in places between these extremes". The value of the genus Calliergidium, the taxonomic status of Calliergidium austro-stramineum, and its occurrence in N.Z. are all controversial. The genus Calliergidium was not accepted by Goffinet et al. (2009). Kanda (1977, p. 47) considered it "doubtful that Calliergidium [was] worthy of taxonomic recognition" and implied that he would include it within his broad concept of Drepanocladus, while Crum & Anderson (1981) recognised the genus but considered it to be of "doubtful value". Calliergidium austro-stramineum was included in a checklist of N.Z. mosses by Fife (1995). The type specimen of Hypnum (C.) austro-stramineum is from South Georgia. Ochyra & Matteri (1997) placed it in the synonymy of Warnstorfia laculosa (Müll.Hal.) Ochyra & Matteri from Staten I. Ochyra et al. (2008, p. 544) subsequently placed C. austro-stramineum and W. laculosa in the synonymy of W. fontinaliopsis (Müll.Hal.) Ochyra, with the last based on a Kerguelen type. Only the poor and sterile Cape Foulwind collection is present in N.Z. herbaria; it is in my opinion a Warnstorfia. Additionally, I strongly doubt that the Cape Foulwind material is conspecific with Kerguelen material in part because it is unlikely that a subantarctic species should occur on a low-elevation peninsula in Nelson L.D. Secondly, the habitat described by Bartlett (1984) for the Cape Foulwind material (as "very damp, acidic bogs in old paddocks and ditches") is consistent with its interpretation as aberrant W. fluitans. The Cape Foulwind material (J.K. Bartlett 23375; CHR 348341, WELT M007484) has leaves that are scarcely denticulate at their apices (a feature usually clearly marked in W. fluitans); the quality of the collection precludes making observations on its branching pattern. The Bartlett collection is illustrated in Image: Image\1IH7. After weighing all the available data for this material I consider its interpretation as a growth form of W. fluitans to be the most acceptable conclusion. The genus Calliergidium and its component C. austro-stramineum are thus excluded from further consideration.

Campylium (Sull.) Mitt. was recognised by Sainsbury (1955). Both the single species treated by Sainsbury (as "C. polygamum (Bryol. Eur.) Bryhn") and C. stellatum (Hedw.) C.E.O.Jensen are discussed here under Campyliadelphus.