re-evaluation of tortella - Missouri Botanical Garden

Bulletin of the Buffalo Society of Natural Sciences 36: 117 - 191. 1998. 

© 1998 The Buffalo Museum of Science 

RE-EVALUATION OF TORTELLA (MUSCI, POTTIACEAE) IN CONTERMINOUS U.S.A. 

AND CANADA WITH A TREATMENT OF THE EUROPEAN SPECIES TORTELLA NITIDA 

Patricia M. Eckel 

Division of Botany, Buffalo Museum of Science, 1020 Humboldt Parkway, Buffalo, New York 14211 

Abstract. The moss genus Tortella (Pottiaceae) is re-examined for continental North America north of Mexico. 

Tortella alpicola is distinguished from T. fragilis as an uncommon but widespread taxon. Tortella japonica is 

considered to be a minor variant of T. humilis. Tortella rigens is new to the United States on the basis of comparison 

of American material with that of Europe, previous reports being erroneous. Tortella nitida is redescribed from 

authentic material and is excluded from the North American flora. The hitherto European Tortella tortuosa var. 

fragilifolia is reported from American plant populations. Tortella inclinata var. densa is reported as new to North 

America as a comb. nov. The bipolar distribution of Tortella fragilis is confirmed. Tortella inclinata var. inclinata is 

excluded from the flora of Arctic North America—specimens on which the species were cited were found to be T. 

inclinata var. densa instead. 

This study of the genus Tortella in continental North 

American north of Mexico was especially undertaken to 

address certain problematic taxa. An attempt was made 

to resolve past ambiguities by focusing on anatomical 

and morphological characters, for example to 

distinguish between the relatively common T. fragilis 

(Drumm.) Limpr. and a hitherto relatively unknown 

element in Tortella, T. alpicola Dix. 

Zander (1994d) added Tortella japonica 

(Besch.) Broth. to the Mexican flora, drawing attention 

to the relationship between the Japanese and Mexican 

floras. This created problems in separation of similar 

specimens with a broad distribution in Canada and the 

United States from the widespread northern hemisphere 

species T. humilis (Hedw.) Jenn. The decision that these 

two taxa are conspecific reveals greater variability in 

Tortella humilis than previously recognized, and 

confirms the widespread distribution of that species. 

One of the consequences of determinations 

made during this study is to relate the Tortella flora of 

North America more closely with that of areas of 

Europe; it also emphasizes the Old-World endemism of 

the species T. nitida by excluding it from the New. 

Two taxa, Tortella tortuosa var. fragilifolia 

and T. alpicola, conflated in previous concepts of T. 

nitida for North America, have been redefined such that 

claims of variability in T. tortuosa in the European 

literature can be related to similar variability in the New 

World. 

Recognition of only two varieties of Tortella 

tortuosa for the entire North American continent north 

of Mexico, a species currently well known for its 

protean nature, may provide a more conservative basis 

for the justification of recognizing some of the 

described European varieties of T. tortuosa as species in 

their own right. Such recognition might have 

biogeographic significance in postulating a higher 

species diversity for the genus in, for example, the 

region of the Alps or southern (Mediterranean) Europe, 

Northern Africa and islands in the Atlantic. 

Tortella alpicola, as T. fragilis var. tortelloides 

S. W. Greene, once claimed to be conspecific to or 

intermediate between T. tortuosa and T. fragilis, has 

maintained its distinctiveness throughout its range as a 

cordilleran species, but with a world-wide distribution 

similar to that of Molendoa sendtneriana (Bruch & 

Schimp.) Limpr. 

The discovery of Tortella rigens in the 

limestone alvars of the Great Lakes region relates the 

biogeography of those areas in North America with that 

of Europe, particularly of the circum-Baltic region, and 

promises to contribute to post-glacial floristic 

reconstructions of these interesting areas. The discovery 

of another taxon largely distributed in the North 

American boreal zone, morphologically close to T. 

inclinata and T. rigens, identified with the European T. 

densa (Lor. & Mol.) Crundw. & Nyh., may be included 

in analyses of northern vegetational histories. It is here 

treated as a variety of Tortella inclinata. 

The present study recognizes seven species of 

the genus present in North America, and three varieties. 

TORTELLA 

Tortella (Lindb.) Limpr., Laubm. Deutschl. 1: 599. 

1888, nom. cons. [Latin tortus, twisted, and - 

ella, a diminutive, in reference to the twisted 

peristome teeth] 

Mollia subg. Tortella Lindb., Musci Scand. 21. 

1879. 

Barbula sect. Tortella C. Müll., Syn. 1: 599. 

1849, nom. illeg. incl. sect. prior. 

Barbula subg. Tortella (Lindb.) Kindb., Eur. 

N. Amer. Bryin. 2: 245. 1897. 

Mollia sect. Tortella (Lindb.) Braithw., Brit. 

Moss Fl. 1: 230. 1885. 

Plants small to medium-sized, in loose or dense tufts or 

mats or expanded in compact sods, fragile, dull green, 

yellowish to dark green above, black, brown or tan 

below with prominent shining costae. Stems erect, often 

branched, in transverse section with hyalodermis,

somewhat weak sclerodermis, central strand absent or 

present, rhizoids dense or few at the base, occasionally 

tomentose; axillary hairs long, 10–20 cells in length, 

hyaline. Stem leaves cirrhate-crispate to incurved when 

dry, spreading to recurved when moist, elongate-oblong 

to linear-lanceolate or linear-subulate, widest at or near 

the base; base hyaline, oblong, erect; margins plane to 

incurved above, generally entire and minutely crenulate 

by projecting papillae, but often slightly or irregularly 

scalloped through indentations at points of laminal 

weakness, occasionally somewhat to strongly undulate, 

rarely with a border of elongate, clear, smooth cells in 

one series above midleaf, gradually tapering upward or 

more or less abruptly narrowed; apex acute or obtuse, 

cucullate or concave, tipped with an apiculus, mucro or 

short subula; costa strong, percurrent to short-excurrent, 

adaxial and abaxial epidermal cells often present, two 

stereid bands, one layer of median guide cells present, 

hydroid strand occasionally present; proximal cells 

enlarged, laxly long-rectangular, thin-walled, hyaline, 

occasionally brown and rather thick-walled, smooth, 

abruptly differentiated from the green cells distally or 

gradual in transition, limit of the proximal region 

usually appears as a V-shape, the limit extending up 

each leaf margin as a short or elongated border; distal 

laminal cells medially and distally rounded-hexagonal, 

chlorophyllose, frequently obscured by numerous, 

dense, C-shaped papillae on both surfaces. Asexual 

reproduction at the stem apex occasional, by deciduous 

or fragile propaguloid leaf tips, or by deterioration of 

fragile leaves along zones of laminal weakness. Sexual 

condition: dioicous, occasionally autoicous. Perigonia 

terminal, short-foliose to gemmate or as stalked buds in 

leaf axils of perichaetiate plants. Perichaetia terminal, 

leaves not or little differentiated, or distinct and longsetaceous. 

Seta 1(–2) per perichaetium, yellow or 

reddish below with age, to 3 cm long, erect, smooth; 

capsule erect and symmetric or slightly inclined, yellow 

to reddish brown with a darker red or brown color to the 

mouth, elliptic to cylindric, more or less wrinkledplicate 

when dry and empty; annulus sometimes present, 

of 1–4 rows of vesiculose cells, persistent; operculum 

conic or long-rostrate, half as long as the urn or longer, 

straight or inclined; peristome orange-red, single, welldeveloped 

from a low basal membrane, of 32 filiform 

rami joined at the base into 16 pairs, twisted 

counterclockwise or merely obliquely inclined, 

branched-spiculose, rarely nearly smooth. Calyptra 

cucullate, smooth. Spores 8–12(–20) µm in diameter, 

yellowish brown, moderately coarsely to finely 

papillose to nearly smooth. 

Species: 53, 7 in the flora, cf. Zander (1993). 

Found on soil, rock or organic substrates in dry 

to variably moist habitats on all continents, generally 

associated with but not confined to calcareous habitats. 

North American species are all smooth on the 

back of the costa except at the extreme ends of 

propaguloid leaf tips. This smoothness gives plants a 

distinctive shining costa in sharp contrast to the green of 

2 

the lamina, which is dulled by papillae on the surface 

of the leaf cells. 

The pottiaceous genus Trichostomum Bruch is 

similar in leaf shape and margin flexion, but has distal 

laminal cells differentiated from the proximal cells in a 

line straight across the leaf base or in a low "U", i.e., 

straight across but with some smooth, hyaline, 

elongated proximal cells extending up the leaf margins 

above the leaf shoulder. The proximal-cell line of 

differentiation usually forms a distinct V-shape in 

Tortella. The peristomes of Trichostomum are erect, 

often short and frequently smooth, whereas those of 

Tortella are long and twisted generally 2–3 times (only 

slightly so in T. flavovirens) and densely spiculose in 

non cleistocarpous species (see Zander 1993). In much 

of the literature the peristomes of Tortella are described 

as papillose, when they are actually spiculose: all 

peristomes of the species treated in this paper have 

spiculose peristome teeth. The term "papillae" may then 

be more usefully employed in the treatment of genera 

with peristome ornamentation that is of tiny, rounded, 

nipple-like or pimple-like elevations such as some 

species, for example, in the genera Trichostomum and 

Pseudosymblepharis Broth. 

Plerurochaete Lindb. does not have the Vshaped 

area of differentiated hyaline proximal 

echlorophyllose cells as in Tortella, but has a median 

area of gradually differentiated proximal cells and a 

strong border of several rows of cells contrasting with 

both laminal and proximal cells in being abruptly 

longer, thinner-walled, smooth and without chlorophyll. 

Plerurochaete also has perichaetia, in addition to 

perigonia, borne laterally on short branches on the main 

axis of the plant. The genus Streptocalypta C. Müll., of 

which two species are endemic to Mexico, also 

resembles Tortella, having abruptly differentiated 

proximal cells extending up the leaf margins in a Vshape. 

It is separated from Tortella by the erect 

peristome teeth and distinctive costal anatomy, there 

being no or rarely a small group of adaxial stereid cells, 

but instead several layers of variously thick-walled 

guide cells adaxial to a well developed abaxial stereid 

band. 

Because the genus Trichostomum has been 

considered closely related to Weissia Hedw., Tortella, 

relatively recently segregated from Trichostomum, has 

been thought to share in this proximity. Species in 

Weissia are similar to Tortella by the incurving leaf 

margins (generally strongly and sharply incurved 

throughout the leaf length) with a tendency toward 

cucullation in the leaf apex. The proximal cells of some 

Weissia species may extend slightly up the margins, as 

in species of Trichostomum, and most especially, 

Weissia jamaicensis (Mitt.) Grout resembles a Tortella 

by a proximal region with a V-shape. After subjecting 

the genera in the Pottiaceae to cladistic analyses, Zander 

(1993) has produced a phylogenetic evaluation of the 

family with novel arrangements of genera, one result 

placing Weissia in a group (Hyophileae) in the 

Pottioideae with genera such as Hyophila,

Weissiodicranum and Weisiopsis, far from its traditional 

grouping in the Trichostomoideae. An important aspect 

of the new arrangement is the grouping of genera, 

including Weissia, with others that have bulging cells on 

the adaxial surface of the leaf relative to the flat abaxial 

surface. 

As for Weissia jamaicensis, were this species 

removed from that genus and placed instead in a group 

of taxa, such as the Chionolomoideae (Zander 1993), 

comprised of Chionoloma, Pseudosymblepharis and 

Pachyneuropsis, genera with many morphological and 

possibly distributional characteristics shared by this 

species, the genus Weissia may display a more natural 

distance from the Trichostomoideae. Unlike many 

species in the genus Weissia, W. jamaicensis is 

dioicous, it has a massive adaxial stereid band creating a 

strongly convex adaxial profile in section, an elongate 

leaf, such as others in the Chionolomoideae, and other 

characteristics that are suggestive of a different 

placement of this species. 

One rather distinctive characteristic of the 

genus Tortella is the flattened, foliose, stalked 

perigonial buds of some species. Only one of the taxa 

here studied possesses them: Tortella humilis. There is, 

however, a group of species of the genus that exhibit 

this character, listed by Brotherus (1924–25). There are 

species in the genus Trichostomum with these, such as 

Trichostomum fragilifolium Dix., and Trichostomum 

spirale Grout, and at least Streptocalypta tortelloides 

(Card.) Zand., of the four species recognized for that 

genus (Zander 1993), also has them. Again, in the genus 

Plerurochaete, where all the species are dioicous, the 

perigonia (and perichaetia) are regularly stalked. These 

genera in the Trichostomoideae may be related by this 

tangential similarity, perhaps through a reduction series 

in the peristome. 

In the following treatment, great emphasis has 

been put on the cross section of the distal region of the 

leaf in delimiting taxa and for discussing relationships. 

Although some bryologists do not care to employ this 

technique, nor are stem and leaf sections often 

considered to be significant in determining taxonomic 

limits at the species level, yet details about these 

characters were employed to assess the concepts of 

earlier students. It was found during the course of this 

study that these characters do in fact aid in identification 

of specimens. 

Some recent floristic treatments of the genus 

have tended to group species with more or less ligulate 

leaves and obtuse apices as segregated from those with 

lanceolate leaves. In the present study, Tortella humilis 

is presumed to be associated with T. flavovirens and T. 

nitida on the basis of the strong stem central strand, the 

variously gradually differentiated proximal cells, with a 

border of elongated, smooth, thin-walled and lax cells 

more or less confined to the border in some or all 

leaves. This characteristic set, together with the stalked 

perigonial bud of T. humilis seems to associate the 

group with the genus Plerurochaete as noted above, but 

also with Trichostomum, some species of which have a 

3 

hint of a marginal border and some have stalked 

perigonial buds. The somewhat reduced peristome with 

erect teeth also suggests that last genus as do the 

undifferentiated perichaetial leaves. Streptocalypta has a 

reduced peristome and stalked perigonial buds in one of 

the three species, and undifferentiated perichaetial 

leaves. 

Tortella alpicola, hitherto discussed (see 

below) as related to or as a variety of T. tortuosa or T. 

fragilis, actually is distinguished by the consistent 

presence of a stem central strand. The absence or poor 

development of an adaxial stereid band in the leaf seems 

to suggest Streptocalpyta; this characteristic, however, 

also applies to the var. fragilifolia of Tortella tortuosa. 

The perichaetial leaves are the same as those of T. 

tortuosa. I shall speculate that Tortella alpicola will 

prove to have a nearly cosmopolitan distribution and in 

fact be relatively ancestral to a T. tortuosa complex. 

One need only refer to the large number of infraspecific 

taxa accorded to T. tortuosa in Europe to posit such a 

complex. The literature of T. alpicola seems to imply 

that it is a relatively recent species when in fact it may 

be one of the more ancient in the Tortella group. 

The fact that Tortella tortuosa in the entire 

North American continent, including Mexico, can be 

accorded only two varieties, one in the extreme north 

(Arctic) or at extremely high altitudes and the other only 

a vaguely defined entity associated with moderately 

high elevations in the Temperate-Boreal zone, might be 

useful in distinguishing species limits in what appears to 

be a welter of intergrading forms in Europe. Tortella 

fragilis may be considered a derived form of Tortella 

tortuosa through the variety fragilifolia along the lines 

conceived by Dixon (1924). 

Another group related to Tortella tortuosa 

through differentiated perichaetial leaves, absence of a 

stem central strand, sharply distinguished V-shape to the 

proximal cell region, and long, spirally twisted 

peristome, may be Tortella inclinata, T. densa (here 

treated as varieties of T. inclinata) and T. rigens. It has 

been speculated that Tortella rigens may share some of 

the genes of T. fragilis (see discussion below). 

Tortella inclinata var. densa has not been 

reported for the North American flora, while T. rigens 

has but based on erroneous and premature 

determinations. The concept of these two taxa and of 

Tortella inclinata employed recently by Crundwell and 

Nyholm in various publications (particularly Crundwell 

& Nyholm 1962; Nyholm 1989) does not fit exactly 

with North American material in some respects, but 

seems to be the most useful to explain the presence and 

distribution of these three taxa and their substrates in the 

region. 

In consideration of the glacial-geological 

implications of the substrates of these three taxa in the 

North American situation, one might speculate that 

plants of Tortella inclinata var. densa that occur in 

upland (rock crevices) situations might be associated 

with glacial refugia (or at least more stable habitats), 

Tortella inclinata var. inclinata with the constantly

shifting but widespread distribution of alluvial deposits 

north of or at the border of glaciation, and T. rigens with 

the alvar substrates created by scouring water 

terminated suddenly (perhaps by stream capture in the 

scenario of retreating continental glaciers) so that their 

4 

beds become subareal with little or no sedimentary 

deposition. 

Herbarium codes are according to Holmgren et 

al. (1990). 

KEY TO THE SPECIES OF TORTELLA IN NORTH AMERICA NORTH OF MEXICO 

1. Leaves oblong-lanceolate or elliptical, apex broadly acute to obtuse, sometimes cucullate . . . . 2 

2. Autoicous, nearly always fruiting; distal margins plane to erect, apex broadly acute; distal leaf cells 

about 6–7 µm; central strand present; cells on adaxial surface of the costa quadrate and papillose 

throughout most of its length; stems short; plants typically rosulate, densely foliose, widespread in the East, 

rare in the West . . . . 1. Tortella humilis 

2. Dioicous, seldom fruiting; distal margins incurved to strongly or variably cucullate at the apex, apex 

obtuse; distal leaf cells 7–11(–14) µm in diameter; central strand present or absent; cells on adaxial surface 

of the costa quadrate and papillose throughout most of their length or mostly or entirely elongate and 

smooth; stems elongate; plants loosely foliose . . . . 3 

3. Stem central strand present; costa with adaxial epidermis: areas on the adaxial surface of the 

costa with quadrate papillose cells; leaf apex variably somewhat cucullate, acute or obtuse with 

apical margins incurved; leaves flat in the leaf middle; fertilized perichaetial leaves not much 

differentiated, mosses of the coast of southeastern United States or Europe . . . . 4 

4. Leaves not fragile, apices of all leaves present; nearly always some leaf apices 

cucullate or subcucullate; distal laminal cells 10–12 µm wide; proximal cells abruptly 

differentiated at least in larger leaves, thin-walled proximal cell region extending up the 

margin into the limb; radicles sparse, confined to the stem base; guide cells never with 

bistratose pairs; mosses exclusively of coastal North Carolina south to Florida and west 

to Texas . . . . 2. Tortella flavovirens 

4. Leaves fragile, apices of nearly all leaves absent; leaf apices infrequently subcucullate, 

sometimes distal margin incurved; distal laminal cells 6–10 µm wide; cells gradually 

differentiated and thick-walled in the proximal cell region in all leaves, except for a 

narrow marginal border of thin-walled cells extending up the margin a short way into the 

limb; tomentose in proximal stem region; guide cells occasionally with bistratose pairs; 

plants of Europe only . . . . 8. Tortella nitida 

3. Stem central strand absent; costa without adaxial epidermis: adaxial surface of the costa with 

smooth, elongate cells throughout the leaf length; leaf apex distinctly cucullate, occasionally 

acute, leaves tubulose; fertilized perichaetial leaves conspicuously differentiated, with subulate 

tips; mosses of the Great Lakes region and north . . . . 5 

5. Leaf cells 11–12 µm or less, stems orange to greenish yellow-brown, leaves deep 

yellow or orange in KOH; leaves irregularly or uniformly twisted on the stem; leaf apices 

usually cucullate to narrowly acute, not deciduous; leaves in section usually keeled at the 

costa, margins incurved; in rock crevices or unconsolidated alluvial sediments near water 

. . . . 6. Tortella inclinata 

5. Leaf cells averaging 14 µm; stems dark-green to brown, leaves green in KOH; leaves 

erect, twisted only at the stem tips; leaf apices variable, usually acute to acuminate, never 

uniformly cucullate, frequently with a narrowed apical deciduous point; leaves in section 

broadly tubulose; mosses of limestone pavements with thin soil cover (alvar) in the Great 

Lakes region . . . . 7. Tortella rigens 

1. Leaves narrowly short to long- to linear-lanceolate, not cucullate or obtuse, apex narrowly acute, tapering to an 

acuminate point, sometimes apex very long, with a long, setaceous point . . . . 6 

6. Leaves without apical propaguloid modifications although some apices may be somewhat fragile . . . . 7 

7. Stems with central strand, leaf cells ca. 14 µm . . . . 3. Tortella alpicola 

7. Stems without central strand, or this rare, leaf cells to 14 µm but often less . . . . 8 

8. Leaves tubulose, margins broadly incurved; most distal laminal cells 14 µm; quadrate, 

papillose adaxial cells on the surface of the costa absent throughout the leaf length to a 

width of two or three cells in section; leaf apices deciduous; known from limestone 

pavements in the Great Lakes region . . . . 7. Tortella rigens 

8. Leaves plane to canaliculate, margins plane to erect; distal laminal cells seldom 

attaining 12 µm, usually less; quadrate, papillose adaxial cells on the surface of the costa 

present in the median leaf region or higher; of general temperate to Arctic distribution . . 

. . 4. Tortella tortuosa

5 

6. Leaves with distinctive apical propaguloid modifications, leaf apices regularly fallen . . . . 9 

9. Stems 1–5 cm, coarsely tomentose; central strand absent; distal leaves to 7 mm, densely 

crowded, rigid, with patches of elongated, non-papillose cells on distal leaf margins of young 

leaves at the stem apex; leaf cells 10–12 µm; lamina below subula bistratose; apical propagula 

falling in a single rigid unit; subulate limb bistratose to multistratose . . . . 5. Tortella fragilis 

9. Stems 0.5 to 1.5 cm, scarcely or not tomentose, central strand present; distal leaves 1.5–2.0 mm, 

sparse, soft, leaf tips without differentiated marginal cells; leaf cells 14 µm; lamina unistratose 

throughout the leaf; propaguloid leaves and apical propagula articulated by periodic constrictions 

falling in several pieces; leaves bistratose only in patches or along costa distally . . . . 3. Tortella 

alpicola 

1. TORTELLA HUMILIS Plate 1 

Tortella humilis (Hedw.) Jenn., Man. Mosses W. Pa. 96. 

1913. 

Barbula humilis Hedw., Sp. Musc. 116. 1801. 

Barbula caespitosa Schwaegr., Suppl. Sp. 

Musc. 1: 120. 1811. 

Tortella caespitosa (Schwaegr.) Limpr., 

Laubm. Deutschl. 1: 600. 1888. 

Trichostomum japonicum Besch., J. de Bot. 12: 

295. 1898, syn. nov. Type: Japan, 

Arima pr‚s de Kob‚, L. Roux, May 10, 

1894 (PC—holotype). 

Tortella japonica (Besch.) Broth. in Engler & 

Prantl, Nat. Pfl. 1(3): 397. 1902, syn. 

nov. 

Plants in loose to dense tufts, usually dull and dark 

green to sordid yellow-green above, brown to black 

below, typically compact and rosulate to more elongate 

and showing annual whorls. Stems: nearly stemless or 

stems to 0.5(–0.7) cm, central strand distinct, densely 

radiculose with red-brown radicles or infrequently 

populations with elongate stems without tomentum. 

Stem leaves densely foliose in rosulate habits, more 

loosely in elongate habits, nearly isomorphic, abruptly 

larger above the whorl-bases consisting of small leaves, 

loosely and variously incurled and contorted, 

occasionally somewhat crisped when dry, widespreading 

to patent when moist, typically oblong or 

oblong-spathulate, in elongate forms more ligulatelanceolate 

to linear-lanceolate, most leaves flat at 

midleaf, somewhat concave and weakly keeled near the 

apex, not cucullate; 1.5–3.5(–4) mm; base sub-clasping 

and often appearing narrower than the limb to somewhat 

broader and elliptic; margins flat to weakly and broadly 

undulate, erect or broadly incurved near the apex; apex 

typically obtuse to broadly or occasionally narrowly 

acute, usually stoutly mucronate, apices usually 

somewhat twisted due to one side of the lamina 

somewhat shorter than the other; costa short-excurrent, 

adaxial surface papillose, adaxial epidermal cells 

typically quadrate to short-rectangular and similar to the 

laminal cells throughout, sometimes smooth and 

narrowly elongate cells (8:1) toward the apex due to 

exposure of the stereid band or, in the extreme apex, 

exposure of the guide cells; in cross section adaxial 

epidermis present, often interrupted in the extreme distal 

portion of the leaf, adaxial and abaxial stereid bands and 

one layer of guide cells present; proximal laminal cells 

rather gradually differentiated from distal cells with an 

zone of cells intermediate in color, cell size, cell wall 

thickness and papillosity, yellow-hyaline, elongate, 

laxly thin-walled, gradually distally papillose, angle of 

hyaline marginal cell differentiation typically shallow 

across the leaf base, then extending up the margins in a 

"U" shape, proximal marginal cells often thicker-walled 

than median and juxtacostal cells near the leaf insertion, 

which are often lax and inflated, the latter often torn on 

removal from stem, marginal proximal cells also longer 

and larger than median cells, occasionally 

approximating a band of differentiated cells; proximal 

cells interior to the marginal ones gradually 

differentiated from the distal laminal cells, those in the 

marginal band more abruptly distinct; distal laminal 

cells small, 6–7(–9) mm, 10–13 µm wide in very 

elongate forms. Asexual reproduction: modifications 

for asexual reproduction none. Sexual condition: 

autoicous. Perigonia small, distinctive, stalked, 

flattened, usually single, rarely geminate, frequently 

yellow or orange, foliose buds of 2–6 leaves in axils of 

the distal leaves alongside and below the perichaetium. 

Perichaetial leaves: inner somewhat smaller than the 

outer, little differentiated or less frequently abruptly 

acuminate-lanceolate, sharply mucronate and sheathing 

the seta. Seta at maturity red near the base to pale 

yellow-green above, about (0.5–)0.7–1.7 cm; one per 

perichaetium, but often three to four per plant from 

different branches. Capsule 1.5–3 mm long; annulus 

deciduous in fragments; operculum 1.2–1.5 mm long; 

peristome teeth orange-red, to over 1 mm long, spirally 

wound 2 or 3 times. Calyptra cucullate. Spores 10–12 

µm, roughened with fine papillae. 

Spores mature in spring. 

This wide-ranging species thrives in various 

conditions: Thuja swamps and bogs, near streams, in 

hard and softwood forests, dry, exposed or moist and 

shaded stations, bark at the base of trees, acid or basic 

substrates, rock crevices and surfaces, sandy or humic 

soil, organic debris, mortar and brick, concrete, from 

maritime and inland forests; from near sea level to 

moderate elevations; B.C., N.S., Ont., Que.; Ala., Ariz., 

Ark., Conn., Del., Fla., Ga., Ill., Ind., Iowa, Kans., Ky., 

La., Md., Mich., Minn., Miss., Mo., Nebr., N.C., N.J., 

N.Y., Ohio, Okla., Penn., S.C., Tenn., Tex., Vt., Va.,

Wash., W.Va., Wis.; Mexico, West Indies, South 

America, Europe, Asia, Africa, Pacific Islands. 

Herbaria examined: BUF, CANM, COLO, 

DUKE, FLAS, MICH, MNA, UBC. 

Although Tortella tortuosa is the most 

common of North American species of the genus in the 

North Temperate and Boreal areas by representation in 

herbaria, certainly T. humilis warrants this distinction in 

eastern middle-temperate and southern regions. With 

few exceptions, all specimens examined were fruiting. 

The species is so far known to be largely absent west of 

the Great Plains, including the Rocky Mountains and 

Basin and Range provinces. There is a disjunctive 

population in British Columbia where it was collected 

once, while the southwestern range ends in Arizona. 

The species appears to be widespread in the 

world but not cosmopolitan. Düll (1984), for example, 

indicated reported locations throughout all of North and 

South America and Africa, Asia Minor and East Asia. It 

has a bipolar distribution and may be widespread in the 

southern hemisphere as well as the northern, however 

some caution is advisable due to the occurrence of 

similar taxa with different sexuality (see discussion 

below). Certain absences of stations, or rarities seem 

curious, such as its absence from the British flora, 

although the climate there dosen't seem to be an 

inhibiting factor when North American stations can be 

found as far north as Quebec. The species is virtually 

absent from Central Europe (Braunmiller et al. 1971; 

Düll & Meinunger 1989). In the former Soviet Union it 

is only reported from European states and the Caucasus 

region, leaving the entire eastern region without a 

representation (Ignatov & Afonina 1992). 

The epithet perhaps refers to the short stem 

resulting in a typically squat, rosulate rather than 

elongated habit that is more characteristic of other 

species in the genus. Tortella humilis is recognized by 

its short stems with leaves in dense, stacked whorls in 

obscure annual innovations subtended by tiny proximal 

leaves, by the relatively broad to occasionally narrowly 

acute, plane, non-cucullate to concave leaf apices, small 

and obscure cells, and autoicous inflorescences. 

In cross section, the costa diminishes in size 

toward the leaf tip: sectioning distally along the leaf, the 

epidermal layer first disappears, leaving up to two 

stereid cells in width exposed. In some instances at the 

extreme apex, even the stereid layer disappears leaving 

the guide cells exposed. The apices of the leaves usually 

seem to twist at the apex so that one lamina lies flat and 

the other extends upwards. This is because one side of 

the lamina is usually somewhat shorter than the other, 

with the shorter side angling toward the vertical. 

The leaves of Tortella humilis, together with 

those of unambiguous specimens of T. arctica, younger 

leaves of T. flavovirens and those of European T. nitida, 

are notable by the gradual rather than abrupt transition 

of proximal cells to the distal laminal ones in addition to 

the marginal thin-walled cells that extend up the leaf 

margins, forming a U-shape to the proximal region, 

rather than the distinctive tortelloid V-shape. This 

6 

characteristic will permit easy confusion of 

infrequently sterile forms of Tortella humilis with 

Trichostomum tenuirostre (Hook. & Tayl.) Lindb. The 

well-developed peristome—elongate and conspicuously 

twisted—will distinguish Tortella humilis from 

specimens of T. flavovirens and Trichostomum species, 

whose peristomes are either not or little twisted, or are 

rudimentary. Any specimen without fruit is unlikely to 

be Tortella humilis: if there is no V-shape in any of the 

leaves on the stem, and if there is the slightest hint of 

marginal scalloping of the lamina and lamina fragility, 

the specimen is probably Trichostomum tenuirostre; if 

the sterile specimen occurs on coastal beaches in the 

southeastern United States, it is most likely Tortella 

flavovirens. 

In areas where the ranges overlap, the leaves of 

fertilized perichatia of the somewhat similar Tortella 

inclinata are stiff, erect, narrowly lanceolate-subulate 

and have long-excurrent costae, whereas those of T. 

humilis are little differentiated from the stem leaves. 

Tortella inclinata has a narrow groove of elongate, 

smooth-walled cells on the adaxial surface of the costa 

throughout the leaf and has no stem central strand, 

whereas T. humilis has quadrate, papillose cells on the 

adaxial surface of the costa in the mid-leaf region and a 

well-developed central strand. 

Because the cells on the adaxial surface of the 

costa are similar in shape and size to the laminal cells, 

and because the leaf base is often narrow (the leaf 

widest in the middle), and because there is a stem 

central strand, there is a superficial resemblance to 

Hyophila involuta (Hook.) Jaeg. & Sauerb.. This species 

typically differs by dentate leaf apices, horned 

propagula in the leaf axils and laminal cells in section 

bulging higher on the adaxial surface than the abaxial. It 

has no peristome nor does it fruit in our area. Plaubelia 

sprengelii (Schwaegr.) Zand., which is dioicous, lacks a 

well-developed adaxial stereid band and has erect 

peristome teeth. 

Plants of Tortella humilis are often confused 

with Barbula unguiculata Hedw., which differs most 

readily in the recurvature of the proximal leaf margin 

and its dioicous condition. Crum and Anderson (1981) 

describe and illustrate the rare species Trichostomum 

spirale Grout from the Great Lakes region (Ontario, 

Wisconsin, Minnesota) that, like Tortella humilis, is a 

nearly stemless plant with gradually differentiated 

proximal cells and a similar leaf shape. It also has 

stalked perigonial buds and is autoicous, unlike the 

usual dioicous condition of the genus Trichostomum. It 

may be distinguished by the peristome teeth which are 

said to be short, erect and smooth or marked with spiral 

lines rather than spiculose papillae, the latter 

characteristic of peristomes of the genus Tortella. 

Two specimens seen from Mexico that 

conform to Tortella humilis (see Zander 1994d) in most 

respects were both autoicous and paroicous on the same 

stem, that is, they had stalked perigonial buds as well as 

robust antheridia in the axils of the distal stem leaves. 

Carl Müller (1878–79: 339–340) originally described

Tortella pseudocespitosa (C. Müll.) Broth. as 

androgynous, with antheridia naked in the axils of the 

distal, fertile (i.e perichaetial) leaves or included in 

small-leaved, very short-stalked gemmulae. This plant, 

known only from Argentina and the southern part of 

South America, has two other varieties that, by their 

names—var. pungens (C. Müll.) Par. probably referring 

to an acute or acuminate-leaved plant, and var. 

brachybasis (C. Müll.) Par., to shorter, broader-leaved 

plants—indicates a plasticity in the leaf shape not unlike 

that attributed to T. humilis (see discussion below) in 

America north of Mexico. These taxa may be 

synonymous, adopting the more complex (paroicous in 

addition to autoicous) condition in the southern 

latitudes. That an Argentinian species might occur in 

Mexico, however, would not be surprising. Cursory 

examination of an isotype of T. pseudocaespitosa (NY) 

shows that there is some promise in this speculation. 

Bryoerythrophyllum recurvirostrum (Hedw.) Chen var. 

recurvirostrum may be similar in that in North America 

north of Mexico the species is synoicous or paroicous. 

In Mexico and Guatemala, however, the var. aeneum 

(C. Müll.) Zand. is "Dioicous or polygamous (with at 

least some gemmiform perigonia in addition to 

synoicous and paroicous inflorescences" (Zander 

1994a). 

After many specimens of Tortella humilis were 

examined, there at first appeared to be two approximate 

facies in the floral area: one with stems short-rosulate, 

tomentose, leaf apices broadly acute to obtuse, 

occasionally broadest above the base to oblong from an 

apparently narrower, sub-clasping base, with proximal 

cells in an indistinct or shallow "U" shape, gradually set 

off from laminal cells, the leaves short, 3–5 (–7):1. Dry 

plant habits were crisped, with leaf tips curved under. 

This is the typical facies nearly universally recognized 

in floristic manuals. 

The other facies, best seen in specimens from 

Texas, British Columbia and Ontario, have elongate 

(not rosulate) stems in obvious annual whorls, 

tomentum generally absent except for coarse radicles at 

the base, leaf apices narrowly acute with leaves broadest 

toward the base, proximal cells higher in a more distinct 

and generically typical V-shape, more abruptly set off 

from the laminal cells, the leaves long-lanceolate, often 

to 4 mm long, in some cases the shape approximating 

that of Tortella tortuosa, 7–9:1. These longer, gracile 

forms have somewhat longer setae and capsules. Dry 

plants of this second morphotype have most leaves 

erect-spreading. This facies was identified by Zander 

(1994d) for Mexico with T. japonica (Besch.) Broth., an 

autoicous species previously known only from Japan 

resembling T. humilis but with lanceolate to narrowlanceolate 

leaves (Saito 1975). The Mexican specimens 

examined by Zander were re-examined for this paper, 

and all specimens fell within the morphologic amplitude 

of the Texas, British Columbia and Ontario specimens, 

and hence can be named T. humilis. 

The variation in facies of T. humilis seems to 

correlate in a bull's-eye fashion with the shorter facies in 

7 

the center of the North American range for the species 

and the elongate ones at the outer boundaries, 

especially, by numbers of herbarium specimens, in the 

south. This variation can be detected in collections from 

northern herbaria, such as MICH, in the center of the 

range where stenotypic plants are typical, and those 

from southern herbaria, such as DUKE, whose 

specimens tend to be more elongate overall. The pattern 

of variation of Trichostomum tenuirostre may be similar 

or identical (Crum & Anderson 1958). 

Tortella humilis was said not to occur in Japan 

(Saito 1975; Noguchi 1988). Thirty-nine specimens 

from TNS determined as Tortella japonica proved to be 

either of two taxa: T. humilis or Trichostomum 

tenuirostre. Material of Tortella japonica, including the 

type specimen borrowed from PC, was indistinguishable 

from the elongate American facies of T. humilis in the 

general elongation of the leaves, becoming narrower 

and more acute in interior perichaetial leaves. Since the 

typical facies of T. humilis grades into this variant in 

North America, especially in Mexico, Texas, British 

Columbia and Ontario, that is, on the margins of the 

range of the species, and since there is no other 

distinguishing characteristic, T. japonica is considered 

here a new synonym of T. humilis. Populations of T. 

humilis in Japan may be interpreted to occur on the 

margin of the range of the more typical facies of the 

species in continental Asia, as described for China, for 

example, by Chen (1941). No short-leaved forms of T. 

humilis were seen from Japan. 

In 1965, the number of Tortella species in 

Japan was eight, and Tortella humilis was considered to 

be one of them (Schofield 1965). Schofield also cited 

the specimen (Schofield and Boas 18192, UBC) 

discussed from British Columbia in the present 

treatment as a form approaching published descriptions 

of Tortella japonica. It was the only collection from 

British Columbia in 1965 and apparently still is to date. 

Saito (1975), however, did not include Tortella humilis 

in his pottiaceous flora of Japan, that is, he excluded it 

from occurring in that country. Saito recognized only 

three species: Tortella tortuosa, T. fragilis and T. 

japonica. Tortella japonica was distinguished from T. 

humilis by the "lanceolate to linear-lanceolate leaves 

with a sharply pointed, acute apex" of the former 

species and indicated that that species had rhizoids 

restricted to the base of the stem, i.e., that the stem was 

not tomentose. The type of T. japonica, however, had 

one of the most robust habits I examined, and it was 

densely tomentose with a rufous tomentum. Further 

specimens examined from Japan were variably 

tomentose and since the nature of the substrate does not 

appear to be a factor, perhaps this variation is due to the 

increased age or size of the plant. 

Although Noguchi (1988) described distinctive 

inner perichaetial leaves for T. japonica, differentiated 

from those of the vegetative leaves as a subulate 

acumen above an oblong base. This character could not 

be well demonstrated in specimens examined from 

Japan, nor in the type specimen. There may be one or

two narrower inner perichaetial leaves whose laminae 

are only somewhat narrower in the distal region, this 

character exaggerated by the presence of a longer mucro 

on most of the leaves of the elongate forms and by 

incurved distal margins. Elongate forms of T. humilis in 

North America share this characteristic as well. 

Occasionally the inner perichaetial leaves may be rather 

abruptly contracted above a more or less sheathing base, 

forming a narrow limb as in some leaves of the type of 

T. japonica. However, this may also be seen in 

specimens of the short-leaved forms of T. humilis in 

North America as well and is considered here to be part 

of the normal variation in the species. The perichaetial 

leaves of T. humilis in either long or short forms are 

considered here to be undifferentiated. Differentiated 

perichaetial leaves in the present paper are different 

almost in kind from the stem leaves: primarily with a 

long, naked extension of the costa into an awn, such as 

occurs in T. tortuosa, T. inclinata var. inclinata, T. 

alpicola and T. fragilis. 

Saito also did not mention that the proximal 

cells interior to the proximal marginal cells gradually 

intergrade with the distal laminal cells, as is 

characteristic of Tortella humilis. In the elongated forms 

of North American material, especially in leaves toward 

the stem apex, the longer leaves have a more 

distinctively V-shaped differentiation of the proximal 

cells. This is true in the longest leaves of intermediate 

American forms, such as in the outer perichaetial leaves, 

whereas the intergradation is less distinct in shorter 

leaves, and merely U-shaped in very short-leaved 

material. Unfortunately, the variable Trichostomum 

tenuirostre also has a U-shaped differentiated proximal 

region very similar to specimens of both Tortella 

japonica and T. humilis, and, as in Saito's illustration, 

the differentiated proximal cells of Trichostomum 

tenuirostre do extend up the margins in various degrees, 

just as in the genus Tortella. While Tortella japonica 

may often be tomentose, Trichostomum tenuirostre is 

never so. 

It is possible that descriptions of Tortella 

japonica by Saito (1975) and Noguchi (1988) include to 

some degree the characteristics of Trichostomum 

tenuirostre (= Oxystegus cylindricus (Brid.) Hilp.), a 

species which may be nearly indistinguishable from 

sterile specimens of the elongate-leaved Tortella 

humilis, and whose ranges and substrates overlap. In 

fertile specimens the best distinction is in the nearly 

always demonstrable autoicous bud in the latter species 

and its long-twisted peristomes in fruiting material. The 

former species is dioicous with comparatively shorterect 

teeth. In fact, the leaf shape of the plant illustrated 

for Trichostomum tenuirostre by Saito (1975) resembles 

closely that of typical Tortella humilis, complete with 

the indication of the "U" shape of differentiated 

proximal cells. Trichostomum tenuirostre also has a 

more elongate stem (to 15 mm), usually more than twice 

the length of the stemless to short-stemmed Tortella 

japonica (to 7 mm). 

8 

Note that, although Saito (1975) indicated 

that Tortella japonica has a stem central strand and that 

Trichostomum tenuirostre does not, the latter species 

does, in many cases, have one, both in the Japanese 

material examined and as attributed to that species in 

the United States (Flowers 1973) and in Europe (Hilpert 

1933). 

When both species are sterile, they may be 

distinguished by the relatively shorter stem of Tortella 

humilis (to 7 mm), its rosulate habit, its variable 

tomentum and its leaf edge firm and orderly (regular), 

as opposed to the more elongate stem of Trichostomum 

tenuirostre (to 15 mm), the leaves more loosely 

disposed on the stem, its lack of tomentum and its leaf 

edge often showing a scalloping or lobing edge 

indicating zones of weakness; there is greater fragility 

of the leaf lamina and occasional obscure and distant 

marginal teeth in the distal leaf region. Also a 

distinctive V-shaped proximal cell region of Tortella 

humilis may often be successfully found on the largest 

leaves, whereas in Trichostomum tenuirostre this will 

not occur. 

The type of Tortella japonica represents a large 

stature that even the ordinary representative specimens 

in Japan do not approach, yet the type material (at PC) 

has the same leaf shape as ordinary North American 

material. Such large stature was duplicated in a 

specimen from Ontario, Canada from the north shore of 

Lake Huron, possessing a proliferation of branches, four 

of which supported fertile perichaetia with a long stem 

to 7 mm, the longest among variants on the margins of 

the North American floral distribution. 

In Zander's (1994d) description of Tortella 

japonica from Mexico, he noted the "shiny upper part of 

the costa, owing to smooth, elongate cells on the ventral 

surface" contributed to the distinctiveness of the 

Mexican specimens examined. However, he also 

described the plants as having a "costa covered on the 

adaxial surface in the middle of the leaf by quadrate, 

papillose cells." In typical short-leaved Tortella humilis, 

the quadrate cells on the adaxial surface of the costa 

disappear toward the apex of the leaf—but this is 

inconspicuous. In leaves significantly more elongate, 

this region of elongate, smooth cells is also longer and 

more conspicuous. The same comparison may be made 

between short and very long leaves of Tortella tortuosa. 

Clearly Tortella humilis, a widespread species 

in the Northern and possibly Southern Hemisphere, 

shows a plasticity not previously recognized. In North 

America, it grades from short, broad plants in the 

northcentral and eastern distribution, becoming more 

gracile on its margins, especially to the south. Indeed, 

upon review of examples at BUF used by Zander for the 

Moss Flora of Mexico, nearly all of the Mexican 

material of Tortella humilis is a "Tortella japonica" 

facies similar to plants from Texas and the southern 

American states. All other material seen in a small 

sample of Mexican specimens at BUF, represents 

minute plants that are sterile, or are possibly not 

satisfactorily named.

The single specimen reported for New 

Brunswick (Albert Co.) by Ireland (1982) is 

Trichostomum tenuirostre, having no central strand, 

fragile leaves, several leaves with thick-walled proximal 

marginal cells, dioicous (perichaetiate with no autoicous 

buds) and is hence excluded from the province. The 

specimen from Nova Scotia (Ireland 1982) reported 

from Inverness County is Hymenostylium 

recurvirostrum (Hedw.) Dix. 

2. TORTELLA FLAVOVIRENS Plate 2 

Tortella flavovirens (Bruch ex F. Müll.) Broth. in E. & 

P., Nat. Pfl. 1(3): 397. 1902. 

Trichostomum flavovirens Bruch ex F. Müll., 

Flora 12(1): 404. 1829. 

Plants in loose or dense tufts, usually pale delicate 

yellow above, yellow-brown below, elongate, not 

rosulate, annual whorls often evident. Stems 0.3–1.0 cm 

high, hyalodermis large, sclerodermis inconspicuous, 

cells of central cylinder thick-walled, central strand 

present, without tomentum, rhizoids thick and sparse at 

the base. Stem leaves loosely foliose, gradually 

somewhat larger and more crowded toward the stem 

tips, loosely to typically tightly incurled-contorted 

around the stem when dry, suberect to spreading when 

moist, oblong-ovate or ovate-lanceolate, strongly 

keeled-concave distally or broadly channeled across the 

leaf proximally, 2–3.5 mm long; base somewhat 

broader, elliptical; margins erect to incurved-involute, 

especially toward the apex, not or somewhat undulate, 

limb gradually or quickly narrowed; apex obtuse, 

subcucullate; costa percurrent or short excurrent, rarely 

disappearing before the apex on some leaves, adaxial 

epidermal cells papillose-quadrate to short-rectangular 

at midleaf, frequently disappearing, showing smooth 

and narrowly elongate (6:1) stereid adaxial cells at the 

extreme apex; in cross section adaxial epidermis of 

quadrate, papillose cells typically present, occasionally 

absent in patches toward the apex, adaxial and abaxial 

stereid bands and one row of guide cells present; 

proximal laminal cells in larger leaves rather abruptly 

differentiated from distal cells, yellow-hyaline and 

nearly concolorous with the distal lamina, longrectangular, 

8(–9):1, laxly thin-walled, gradually 

papillose, in some shorter leaves, however, proximal 

cells gradually differentiated with a marginal border of 

elongate thin-walled cells; marginal angle of 

differentiated proximal cells generally steep; distal 

laminal cells 10–12(–14) µm wide. Asexual 

reproduction: modifications for asexual reproduction 

none. Sexual condition dioicous. Perigonia not seen. 

Perichaetial leaves little differentiated, outer leaves 

somewhat longer, inner shorter than the stem leaves, 

without marginal differtiation. Seta red, sometimes 

vivid below, paler above, 1.1–1.3 cm long. Capsule 

1.8–2.2 mm long, often more or less wrinkled-plicate 

when dry and empty, mouth rimmed with red; annulus 

none; operculum 0.75–1 mm long, nearly subulate when 

9 

young, more broadly long-conic just before 

dehiscence; peristome teeth relatively short, to 0.5 mm 

long, straight or obliquely inclined and slightly twisted, 

especially when newly deoperculate. Calyptra 

cucullate, narrow, the split twisted in conformity with 

the spiralling of its cells. Spores (10–)12(–14) µm, from 

the single capsule seen coarsely papillose, papillae small 

and somewhat irregular in size and distribution, but not 

punctate or fine as in the literature. 

Seldom collected in fruiting condition. Spores 

mature in mid to late spring. 

Infrequently collected, this salt-tolerant plant 

grows near sea level just above the tidal zone in 

saltwater spray on coastal beaches and offshore islands 

of the southeastern United States: exposed sites on 

sandy soil, coquina, among dunes, juniper scrub, lawns 

and other grassy, weedy areas, damp concrete, mortar in 

old forts, and is never found in wooded areas or on 

humic material, such as the bark of trees. Ala., Fla., Ga., 

N.C., S.C., Tex.; Bermuda, in Europe confined to ocean 

and Mediterranean coasts, Macaronesia, Syria, n Africa, 

New Zealand. 


DUKE, FLAS, MICH, MO, NY, UBC. 

Its affinity for sandy stations associated with 

water, together with the cucullate leaves is reminiscent 

of Tortella inclinata of northern regions. Perhaps 

because of the Gulf Stream in the Atlantic Ocean, the 

ranges of T. flavovirens and T. inclinata overlap in 

England and the oceanic coasts of western Europe, 

causing some difficulty in identification there. 

In North America, populations of these two 

species are widely separated, that of Tortella inclinata 

following the southern edge of furthest recent glaciation 

(Miller 1976). The present distribution of that species 

seems to be closely bound with continental glacial 

history. T. flavovirens may be expected to occur on 

beaches associated with warm oceanic currents. Coasts 

associated with cold ocean currents, such as those off 

the northern coast of Pacific North America, might 

preclude the occurrence of the species there. 

The range limits as presently known in the 

United States may reflect more the limits of regional 

collectors than that of the species itself: it should be 

found in the contiguous coast of Mexico and in the 

islands of the Caribbean generally, and possibly also in 

states to the north of North Carolina whose coasts are 

warmed by the Gulf Stream. The species was not 

reported from the beaches of Puerto Rico and the Virgin 

Islands (Crum & Steere 1957), Jamaica (Crum & 

Bartram 1958) or Mexico (Zander 1994d). Many of the 

specimens collected in the United States were 

concentrated around the city of St. Augustine, Florida. 

Sampling of additional populations in the Atlantic- 

Caribbean area may reveal more details about this 

species. 

One possibility to consider is that the species 

has been recently introduced on the American coast and 

is spreading. Rubers (1973) reported that the variety 

glariecola (Christ.) Crundw. & Nyh. is quite common

throughout the seacoast of the Netherlands, especially 

between Voorne and Schoorl, whereas that of the var. 

flavovirens is known only from four localities. In the 

Netherlands, the leaf cell size of var. flavovirens is 8–10 

µm whereas that of var. glariecola is much larger: 10– 

14 µm. Based on a report by Townsend (1965) of 

sporophyte material of the var. glariecola found on the 

island of Cyprus, there is also an apparent different in 

the spore size (Rubers 1973). That American 

populations represent only the var. flavovirens, even 

though the leaf cells are on average larger (to 12 µm 

with some cells reaching 14 µm), may indicate that the 

American populations are depauperate. The issue 

requires further study. 

Braunmiller et al. (1971) state that the center of 

dispersal in Europe is the Mediterranean region where 

the species is broadly distributed and common, with 

scattered outposts along the western European coast to 

Denmark and England. 

The species forms a suite with other 

Mediterranean Tortellas: T. inflexa and T. nitida. These 

authors seem to indicate that only in the northwest of 

Europe does there occur two varieties of Tortella 

flavovirens, the typical one and the var. glariecola. 

They also suggest that the only German reports of T. 

flavovirens are those of the typical variety. 

Stations outside of North America are based on 

published reports. The species has such a large reported 

world distribution, in addition to the Atlantic and the 

Mediterranean, that one might even expect it on tropical 

or subtropical coasts of Pacific North America (A.C. 

Crundwell, pers.comm.) unless cold ocean currents 

preclude this. 

In the United States, all specimens of Tortella 

flavovirens that were inland from coastal habitats were 

redetermined to be some other species: T humilis most 

frequently, Trichostomum crispulum and Weissia spp. 

The fact that Tortella flavovirens is salt-tolerant may be 

helpful in excluding taxa, such as T. inclinata, from 

similar habitats. In Europe, the range of T. flavovirens is 

thought to overlap that of T. inclinata. However, if T. 

flavovirens is an obligate halophyte and T. inclinata is 

intolerant of salty conditions, the ranges may be merely 

contiguous. 

Crum and Anderson (1981: 306) reported that 

"Nyholm [1956: 137] included in the species'[sic] range 

Macaronesia, North Africa, the Middle East, China, and 

Japan." Noguchi (1988), however, did not include it in 

the moss flora of Japan, nor did Redfearn (1993) for 

China. Fife (1995), however, reported it from New 

Zealand where it had been known as T. rubripes (Mitt.) 

Broth., a species previously thought to be endemic to 

New Zealand. It was described by Sainsbury (1955) as 

an "exclusively...maritime plant" which "can properly 

be described as a halophyte, its station being at 

practically the edge of the sea. It is widely distributed 

on the coast in the Auckland District, and appears here 

and there as far south as the Mahina Peninsula, Hawke's 

Bay." 

10 

The characteristic pale yellow leaves (versus 

the more opaque and sordid yellow- or bright green of 

T. humilis and T. inclinata) is indicated in the epithet. 

Dixon (1924) cited Limpricht (1890) as differentiating 

between Tortella flavovirens and T. inclinata, whose 

ranges overlap in Great Britain, by the presence of a 

stem central strand in the former, and its absence in the 

latter. "In undoubted [Tortella] flavovirens from our 

coasts, however, I find the central strand entirely absent, 

and the stem in section exactly similar to that of [T.] 

inclinata" (Dixon 1924). Smith (1978) stated that T. 

flavovirens "often" has a stem central strand. In material 

examined from the southeastern United States, the stems 

regularly had a distinct central strand. 

The leaf cross section of Tortella flavovirens 

exhibits an adaxial layer of quadrate, papillose cells 

across its surface throughout the leaf length excepting 

the proximal region and the extreme apex. In section, 

the leaf is keeled beside the costa, then widens broadly 

out before widely curving inwards in the shape of a 

caliper or tongs. The leaf cross section of T. humilis is 

generally flat with the margins shortly and slightly erect 

or somewhat incurved, becoming more curved or 

slightly keeled in the apical extremities where an 

absence of the adaxial quadrate layer is frequent. 

Tortella flavovirens shows great variability in 

the apex; on the same stem some apices may have 

costae subpercurrent and the apices are round in outline, 

or they may be flat to sharply tubulose and have 

percurrent costae, or they can be narrowly acute with a 

strong mucro, but always at least some of the apices are 

naviculate and frequently most are cucullate. The 

species shows more variability in Europe, with cucullate 

apices in some populations to flat apices in others, but 

all with erect to incurved margins, lack of tomentum, 

larger leaf cells, and central strand present. 

The proximal cell region in larger leaves is 

sharply distinguished from the distal laminal region by 

abruptly elongated smooth cells in contrast with the 

distal quadrate papillose cells. However, in many 

smaller leaves the transition is completely gradual in 

cell size and cell wall thickness such that the V-shaped 

region is defined more by the papillae than by the cell 

dimensions, and the proximal margin has a border of 

more abruptly elongated proximal cells. This may 

account to some extent for European published reports 

of the taxonomic proximity of this species with Tortella 

nitida, which has gradually transitional cells in all of its 

leaves with a sometimes poorly defined proximal 

marginal border of elongate cells. This situation is also 

rather characteristic of Tortella humilis and may align 

these three species, together with the leaf shape, stem 

central strand and undifferentiated perichaetial leaves. 

The peristome of Tortella flavovirens is 

somewhat short, but not rudimentary: it is about half the 

size of that of T. humilis, for example, and is apparently 

an example of reduction for the genus in North 

America. It is erect or somewhat inclined and 

reminiscent of Trichostomum peristomes, which are

smooth, papillose or striolate and generally erect (some 

species in the world may be spiculose, Zander 1993). 

Tortella humilis overlaps in range that of T. 

flavovirens on the beaches of the southeastern United 

States, the only other species of the genus to do so, and 

characters that separate them are critical. Most 

confusion in herbarium specimens are between these 

two species. The surest way to differentiate them is the 

nearly always present autoicous perigonial bud of T. 

humilis which is absent in the dioicous T. flavovirens. 

Otherwise, the stem of T. flavovirens is elongate, 

usually two distinct annual whorls of leaves, with 

rhizoids few, coarse and confined to the base. Tortella 

humilis typically has a rosulate habit, with short, 

densely foliose, darker stems densely and finely 

tomentose up the stem. The former species rarely fruits 

and fruiting material is thus a strong candidate for 

Tortella humilis. When Tortella flavovirens does fruit, 

the peristome teeth are erect or somewhat oblique, an 

unusual feature for the genus, whereas T. humilis is 

nearly always in a fruiting condition, is autoicous, male 

buds are always conspicuous on dissection, and the 

peristome teeth are long and twisted 2–3 times. Tortella 

flavovirens has larger leaf cells, 10–12(–14) µm wide, 

and narrowly incurved apical margins grading to 

cucullate apices that are keeled. The cells of T. humilis 

are 6–7(–9) µm wide, and it has broadly acute apices 

with plane or erect margins. If a doubtful T. flavovirens 

grows on anything but sand, it is probably T. humilis 

instead. 

In North America Tortella flavovirens may be 

separated from T. inclinata s.l. by distribution: T. 

flavovirens is a southern maritime species whereas T. 

inclinata s.l. is northern (the Great Lakes region to 

Alaska). Otherwise Tortella inclinata s.l. has no central 

strand, no quadrate, papillose cells on the adaxial 

surface of the costa, and has more tomentum on the 

stem. Its proximal V-shape of hyaline cells is distinct in 

all leaves, it has strongly differentiated perichaetial 

leaves when archegonia are fertilized, and a long 

peristome twisted perhaps twice in recently 

deoperculated capsules. It is due to these characters that 

T. inclinata is probably more aligned with T. tortuosa, 

and T. flavovirens with T. humilis. 

Tortella nitida, when it is confused with T. 

flavovirens in Europe, usually may be first distinguished 

by substrate. The latter species requires coastal sand, if 

it is not an obligate halophyte. The former occurs on 

cliff faces, rocks and walls, apparently not on sand or 

sandy soil. The leaves of typical T. nitida will not have 

sharply differentiated proximal cells in most of its 

leaves, and the proximal border of elongate hyaline cells 

will be confined to the margin of the leaf. Sections of 

the leaf of T. nitida will display occasional bistratose 

pairs in the guide cell region; T. flavovirens will never 

have this. Note that the apices of T. nitida may have 

inflexed margins, but resemble more T. humilis than T. 

flavovirens. The leaf cell size of T. nitida is the same as 

T. humilis, around 7–10 µm, rather than the 10–12(–14) 

µm range of T. flavovirens. 

11 

If one accepts that the peristome illustrated 

here for Tortella nitida is the true one, in contradiction 

to published reports (see discussion of T. nitida below), 

then it and that of T. flavovirens are nearly the same in 

character: shorter by about half that of Tortella tortuosa 

and other Tortella species with long-twisted peristomes, 

erect and densely spiculose, more like that of the genus 

Trichostomum as currently understood. 

Sterile Weissia species may be confused with 

T. flavovirens, being pale yellow with the same cell size, 

and appearing to have a cucullate apex with the 

marginal proximal cells somewhat extending up the 

leaf: these will, however, have no marginal extension on 

some leaves and the distal leaf margins will be tightly 

inrolled or incurved to a stronger degree than in any 

Tortella species; the substrate is frequently 

inappropriate in these specimens, such as bark at the 

base of trees or some other woody or humic substrate, 

likewise inland stations. 

Weissia jamaicensis (Mitt.) Grout is 

occasionally confused with Tortella flavovirens. Both 

have strong central strands and a lemon-green color, 

with hyaline cells running somewhat up the proximal 

margins. Weissia jamaicensis is distinguished by its 

long, linear-lanceolate leaves and much more sharply 

incurving margins running from proximal mid-leaf to 

the apex in every leaf. Tortella flavovirens is restricted 

to coastal stations whereas Weissia jamaicensis may be 

found inland. 

Trichostomum crispulum, with which Tortella 

flavovirens has occasionally been confused, has smaller 

leaf cells, a broader, chestnut-colored costa and has a Ushaped 

proximal differentiated area of cells in all leaves. 

As discussed above, Tortella flavovirens var. 

glariecola (Christens.) Crundw. & Nyholm is 

distinguished from the typical variety by having cells in 

the distal part of the leaf regularly 10–14 µm wide. In a 

European specimen of this variety that was seen, the 

areolation is indeed distinctive because most of the cells 

are of these dimensions, and the cell walls appear to be 

thicker. Specimens labeled as this from North America, 

however, have smaller cells overall, characteristic of the 

typical variety, with only occasional cells reaching 14 

µm. Crundwell and Nyholm (1962) suggested that 

larger cell size in this species is correlated with northern 

location. Nothing identifiable as the var. glariecola was 

found in American populations of T. flavovirens var. 

flavovirens. 

Smith (1978) gave a record for Tortella 

flavovirens var. glariecola occurring in Canada, most 

likely based on the citation by Crundwell and Nyholm 

(1962) of a depauperate specimen of var. glariecola 

from Alberta collected by C. D. Bird 3624, identified by 

him as T. tortuosa. They state that "the occurrence of 

any variety of T. flavovirens in Alberta is exceedingly 

strange." These plants "with longly tapering acutely 

pointed leaves" may be one of the peculiar instances of 

depauperate, large-celled T. tortuosa specimens (see 

discussion under that species). T. rigens, with leaf cell 

sizes to 14 µm, also may have long-pointed leaves, but

the absence of quadrate cells on the adaxial surface of 

the costa is characteristic of that species. The substrate, 

at the base of a cottonwood tree, is not typical of that 

species. 

One alternative is that the collection seen was 

Tortella alpicola. Crundwell had seen and annotated 

several specimens of that species from North America 

in North American herbaria, speculating that they had 

an affinity with Tortella rigens due to their leaf cells 

attaining 14 µm. Tortella alpicola has been recently 

reported from Alberta (Eckel 1997). Ireland et al. (1987) 

in their checklist of the mosses of Canada considered 

the variety Tortella flavovirens var. glareicola to be 

excluded from Canada. 

3. TORTELLA ALPICOLA Plate 3 

Tortella alpicola Dix., Ann. Bryol. 3: 54. 1930. 

Sarconeurum tortelloides S. W. Greene, Sci. 

Rep. Brit. Antarct. Surv. 64: 38. 1970. 

Tortella tortelloides (S. W. Greene) Robins. in 

Llano, Antarct. Terr. Biol., Antarct. 

Res. Ser. 20: 170. 1972. 

Tortella fragilis var. tortelloides (S. W. 

Greene) Zand. & Hoe, Bryologist 82: 

84. 1979. 

Plants scattered or in soft, loose or close tufts, 

light or dark and vivid or clear green above, pale buffbrown 

below with glistening white leaf bases, elongate, 

not rosulate. Stems thin, slender, 0.5–1(–1.5) cm, 

branches few to several, central strand present and 

conspicuous, not tomentose except in perichaetiate 

plants. Stem leaves loosely foliose, densely so in 

perichaetiate plants, appearing soft but actually rather 

rigid, fragile, closely to loosely aggregated, uniform in 

size, apices in sterile plants (except the youngest) 

usually fallen, incurved-circinate and weakly contorted 

when dry, erect-spreading, occasionally patent when 

moist, gradually long-lanceolate, 1.5–2 mm; base 

undifferentiated or somewhat broader than the limb, 

elliptical; margins of some leaves weakly undulate, 

constricted, lobed in scallop-shapes in several places 

distally, erect to incurved; apex narrowly acuminate, 

occasionally sharply contracted into a subula in the 

apical 1/3, this a papillose cylinder about the size of the 

costa, composed of a series of barrel-shaped 

constrictions disarticulating in segments, intact subula 

and tip of propaguloid apex with an apiculus of several 

cells, this usually dentate and tipped by one or two, 

elongate, sharply pointed clear cells, in sterile plants 

leaf apices caducous along zones of weakness, youngest 

leaves at the stem tips composed entirely of a serially 

constricted, multistratose, awl-shaped propagulum, 

these usually absent in mature leaves; costa shortexcurrent, 

in leaves of fertile plants adaxial surface 

covered distally by an epidermis of quadrate, papillose 

laminal cells, back of the costa smooth throughout—in 

leaves of fragile sterile plants abaxial costa surface 

smooth only below apical subula, densely papillose on 

12 

both adaxial and abaxial surfaces in the distal 

subulate region, in cross section cells weakly 

differentiated and frequently rather chaotic, adaxial 

epidermis always present, rather thin-walled adaxial and 

abaxial substereid layers present, the adaxial stereid 

cells often disappearing in the distal region of the leaf, 

guide cells in one row, often chlorophyllose; proximal 

laminal cells abruptly differentiated from distal cells, 

distinct in color, cell size, cell wall thickness and 

papillosity, pale white-hyaline and transparent, shaped 

marginal angles of differentiated proximal cells steep; 

distal laminal cells relatively large, 10–14 µm wide, 

lamina unistratose but apparently bistratose at juncture 

of lamina and costa, especially along the distal costa in 

propaguloid leaves of sterile plants; marginal cells 

undifferentiated, papillose-crenulate throughout. 

Asexual reproduction: modification for asexual 

reproduction in two modes, a general fragility of the leaf 

lamina as well as smaller, soft, multistratose, 

propaguloid deciduous leaf apices articulated by regular 

constrictions, falling early in units of approximately 

uniform length. Leaf apices generally absent due to the 

modified apices falling off when the young leaves were 

formed, rarely retained on mature leaves. Sexual 

condition: apparently dioicous. Perigonia not seen. 

Perichaetiate stems tomentose, perichaetia terminal on 

successive perichaetial innovations; outer perichaetial 

leaves differentiated, especially evident when dry, 

longer than the cauline leaves, long-lanceolate to linearlanceolate, 

fragile, tipped with long, rigid, subulate, 

smooth awns; these leaves rise above the contorted stem 

leaves when dry, proximal region broad, generally 

concolorous with the lamina, lamina quickly 

disappearing into the costa, or appearing to clothe the 

costa in a margin one or two cells wide for a distance 

before disappearing altogether, not propaguloid, 

unbordered; inner perichaetial leaves long-triangular. 

Sporophytes unknown. 

Associated with shaded or exposed, wet or dry 

rocks, crevices and ledges of granite, quartzite, schist, 

sandstone and calcareous outcrops on cliffs and in 

canyons, in cracks in a limestone gully in a cavern on 

Ellesmere Island, but also in wet, mesic tundra (in the 

Yukon), and on a wet log (in Montana), in Quebec on 

dry limestone cliff face facing north; elevation 25–3300 

m; Alta., N.W.T., Que., Yukon; Alaska, Ariz., Colo., 

Idaho, Mont., Nebr., Utah, Wyo.; South America in 

Colombia, Asia in India, Pacific Islands in Hawaii, 

Antarctica (Alexandra I.). I have verified all extra- 

American records here cited. 


DUKE, MICH, MNA, MO, NY, UBC. 

The species in North America is associated 

with a western corridor of both montane and valley 

habitats, including the western Great Plains, extending 

in an arc from Arizona to Ellesmere Island. The location 

of far eastern stations in the Gasp‚ Peninsula, Quebec 

(Mont Commis, St.-Donat de Rimouski, Lepage 3412 

MT; Compt‚ Gasp‚-Est, Cap-des Rosiers-Est, Forillon 

National Park, Brodo 18646 CANM) seems strikingly

anomalous, yet the disjunction is nearly identical with 

that of Molendoa sendtneriana (Bruch & Schimp. in 

B.S.G.) Limpr. This latter species is decidedly western 

cordilleran (Zander 1977) with an outlier in the province 

of Newfoundland (Belland and Brassard 1981 and Fife 

specimen 2380 BUF). Tortella alpicola is the only 

species of the genus known for Colombia (Churchill & 

Linares 1995). 

Tortella alpicola is the smallest of the North 

American Tortellae. The plants are noted by their size, 

absence of tomentum on typical (sterile) stems, broken 

tips on most of the leaves, vividly shining white leaf 

bases on the stem in contrast with the deep or bright 

green of the limb, and large leaf cells (to 14 µm wide). 

In the two populations seen composed of 

perichaetiate plants, the stems were considerably more 

densely foliose and the color somewhat more sordid, the 

ordinarily glistening leafbases yellowish and the 

tomentum very evident. These plants strongly resemble 

small stems of Tortella fragilis, due to the leaves with 

uniformly broken leaf tips surrounding a cluster of stiff 

and erect subulate leaves, which are, in fact, the 

perichaetial leaves. There never is a typical T. fragilis 

propagulum in the leaf apices of stem or perichaetial 

leaves. 

Small fertile plants of T. tortuosa and T. 

inclinata have a similar aspect; these also have stiff, 

awned perichaetial leaves distinct from the cauline 

leaves. 

The large leaf cells and presence of a distinct 

stem central strand in Tortella alpicola will differentiate 

ambiguous specimens from similarly small stems of T 

tortuosa and T. fragilis. The presence of such a feature 

aligns T. alpicola with species such as T. humilis, T. 

flavovirens in the United States, and T. nitida in Europe 

as well as certain species in the genus Trichostomum. 

This character may indicate that Tortella alpicola is 

ancestral to species in the genus with no stem central 

strand and similar leaf shape and perichaetial leaves. Its 

very broad distribution in the world may indicate that 

the species is older than at first suggested (Eckel 1991) 

and that it does not derive from T. fragilis or T. tortuosa 

(as per Zander & Hoe 1979; Eckel 1991). 

The consistently large leaf cell size (on average 

14 µm) is also distinctive, although note must be made 

of the smaller cells in one Arizona collection, which 

retained all other characteristics, especially the peculiar 

propagula displayed at the leaf apex. Tortella rigens has 

leaf cells this large, but lacks quadrate, papillose adaxial 

epidermal cells on the costa, presenting a generally 

smooth surface in an adaxial groove throughout the leaf 

length and has no stem central strand. 

These young leaves are actually the emerging 

apices of the developing leaf. In one specimen from the 

Northwest Territories (Ellesmere Island, Brassard 4202 

CANM) the apical propagula had not fallen away and 

were retained until the leaf became fully mature, 

producing a plant seemingly covered with worm-like 

apical appendages, the constrictions of these somewhat 

irregularly offset from one another. The plants were 

13 

sterile and quite as foliose and several-branched as 

perichaetiate plants. The rare occurrence of leaves with 

intact propaguloid apices is likely due to their protected 

situation: "in cracks of limestone gully, deep inside 

cavern." 

The sense that the leaf is structurally more 

prone to deterioration in Tortella alpicola parallels 

lamina characteristics of such species conspicuous in 

their erose leaves as Trichostomum tenuirostre and 

Tortella nitida, with irregularly scalloped, subdentate 

margins, the apex of the marginal sinus marking a point 

of weakness. Species without fragile leaves have 

generally straight and regular margins. 

Tortella alpicola apparently has dimorphic 

leaves: one kind in sterile plants, emphasizing fragility 

and asexual reproduction, and another when fertile and 

perichaetiate, in which the leaves are more intact. When 

sterile, the distal and abaxial costal surface in fragile 

subulate leaf apices is densely papillose and hence has a 

dull appearance; thus T. alpicola is the only species in 

the genus in North America to regularly have anything 

but a smooth, shining abaxial costal surface in the distal 

portion of the leaf. On the other hand some taxa in the 

genus here under examination have a rarely expressed 

tendency to form elongate leaf apices that are tubulose 

or even cylindrical, with the suspicion of caducous 

tendencies in the leaf apex. These do develop abaxial 

papillae in this area and the condition is rarely seen in T. 

tortuosa var. fragilifolia. In Europe, T. nitidum var. 

irrigatum Winter possesses this characteristic 

(according to May 1986). 

The costal anatomy of leaves of Tortella 

alpicola shows two sets of characteristics, one for sterile 

plants and one for fertile (perichaetiate). In sterile 

plants, the costal layers (epidermis, stereids and guide 

cells) are weakly differentiated and in many cases, not 

differentiated at all. It is perhaps this characteristic that 

led S. W. Greene to originally place T. alpicola in the 

genus Sarconeurum Bryhn, in which, as Robinson 

(1972) noted, the costa "has less differentiation 

throughout and no differentiation in the cells adaxially." 

Sarconeurum also has a deciduous apical propagulum 

and a stem central strand. However, the proximal cells 

are "completely different and are more like those of 

Barbula" (Robinson 1972) and as Robinson suggested, 

other genera in the Pottiaceae have fleshy, deciduous 

leaf apices, Didymodon Hedw. being a good example. 

In sterile T. alpicola, the costa is undifferentiated in the 

proximal part of the leaf where the laminae are fully 

extended and unistratose. The laminae are never fully 

bistratose but occasionally exhibit bistratose patches. In 

sterile plants, the epidermal and guide cell layers may 

be chlorophyllose and there are papillae on the abaxial 

surface of the costa. In Tortella fragilis, the costal cells 

become undifferentiated in the non-laminate subula 

while the lamina just below the propagulum is 

bistratose. 

In perichaetiate plants the costa is not so 

amorphous. In sterile plants the abaxial stereid band 

disappears along with the other layers. In perichaetiate

plants the abaxial and adaxial stereid bands are definite, 

with thick-walled cells. The adaxial stereid layer, 

however, appears to quickly disappear in the distal 

region of the leaf, rather than the epidermal layer, the 

reverse of every Tortella species in North America 

(excepting many specimens of T. tortuosa var. 

fragilifolia), leaving a costal anatomy that resembles 

that of the two Mexican endemic species of 

Streptocalypta, with either a rather undifferentiated 

costa excepting the persistent abaxial stereid band (S. 

santosii (Bartr.) Zand.) or one with in which the adaxial 

stereid is absent and the adaxial epidermal cells are not 

differentiated from the guide cells (S. tortelloides 

(Card.) Zand.). Other than this peculiarity of the costa, 

however, Tortella alpicola cannot be shown to have any 

other similarity with Streptocalypta other than in 

characters shared by both genera, such as the high angle 

of the proximal cell region. Zander (1993) interpreted 

these adaxial costal areas as multistratose guide cells 

with an adaxial epidermis. 

The two other Tortella species in North 

America with stem central strands, Tortella humilis and 

T. flavovirens, have two well-differentiated, strong 

stereid bands abaxial and adaxial to the guide cell layer. 

Both Tortella fragilis and T. tortuosa, when 

small, typically have rufous tomentum extending up the 

stem. Their leaf bases are more yellow. In tiny plants of 

T. fragilis, the margins are regular and sharply defined, 

almost stiff, rather than the crenulate-papillose and 

scalloped or indented margins of T. alpicola. Both 

species have leaf cells smaller than 14 µm and neither 

has a stem central strand. 

The propaguloid modifications of the leaves 

are different in kind between Tortella fragilis and T. 

alpicola. The propagules of the former species fall as a 

single unit from leaves disposed all along the stem, 

whereas those of the latter fall in numerous, fragile, 

barrel-shaped caducous units of about equal length from 

the leaf apices. 

Tortella nitida (Lindb.) Broth. has been 

ascribed to the North American flora by Haring (1938) 

and Flowers (1973) among others (see also discussion 

below under T. tortuosa var. fragilifolia). These 

specimens have been suggested to be a variant of T. 

tortuosa by Crum and Anderson (1981), who also state 

that material cited as T. nitida by Haring was in fact 

either T. fragilis or T. tortuosa, and material from Utah, 

described by Flowers, was not seen by them. A few 

specimens labeled T. nitida from various herbaria in 

North America were in fact one or the other of those 

two species. However, the specimens cited by Haring 

(1938) were variously either T. tortuosa var. fragilifolia 

or T. alpicola, and of the two cited for Utah by Flowers 

(1973), the one available for study was T. alpicola. 

None of the three specimens cited by Haring was T. 

tortuosa or T. fragilis. 

Tortella nitida, a European species, has 

proximal cells gradually, not abruptly, differentiated 

from the laminal cells, a shining costa on the abaxial 

leaf surface with no distinctive subulate propaguloid 

14 

leaf apex, and leaf cells to 10 µm wide. Its leaves are 

usually broadly lanceolate to almost oblong-ligulate, 

whereas those of all of the taxa just cited are lanceolate 

to linear-lanceolate. Only T. alpicola and Trichostomum 

tenuirostre are like it in its laminal fragility and stem 

central strand. 

North American material examined for the 

present study, named Tortella nitida, is variously T. 

alpicola, T. fragilis, T. tortuosa and Trichostomum 

tenuirostre, with Tortella alpicola specimens 

predominating in the western United States of Utah and 

Colorado. Tortella alpicola differs from authentic 

European T. nitida immediately in the brightly distinct 

clear proximal cells set off from the green laminal cells 

as well as the other characters mentioned. There is a 

tiny specimen collected by R. S. Williams, "Columbia 

Falls, Montana, on wet log, Williams 329, Nov. 4 1895 

[NY]." It was identified by him as Trichostomum 

cylindricum (Brid.) C. Müll. (= Trichostomum 

tenuirostre), a species with forms that closely resemble 

Tortella alpicola. His analytical description of the plant 

fits the description given above for Tortella alpicola, 

but more particularly, his careful and minute drawings 

of stem and leaf cross sections in the specimen packet 

show perfectly the undifferentiated costal cells and the 

stem section with central strand of that species. This is 

the Montana specimen cited by Haring (1938) as T. 

nitida. 

Specimens of Tortella alpicola from North 

America have also been determined to be T. rigens N. 

Alberts. (Weber 1973) due to the wide leaf cells (ca. 14 

µm), but plants of that species have elongate, smooth 

cells on the adaxial surface of the costa throughout the 

leaf length, have no stem central strand, are 1.5–3 cm 

high, tomentose, and lack differentiated apical 

propagula, although the leaf apices are fragile (see also 

Nyholm 1989). 

The perichaetiate plants of Tortella alpicola 

were found in a Nebraska collection from "rocky cliffs 

at edge of prairie, moss common, exposed on dry sand 

covered rock" on May 19 (Churchill 7302, COLO, 

MICH); also from Alberta on "soil, steep east-facing 

limestone outcrop, white spruce-aspen-birch woods, 

4500' [1364 m]" (Bird & Glenn 11046, CANM). 

Although the epithet "alpicola" indicates alpine stations, 

it is possible that the perichaetiate populations are 

associated with short-grass plains, and the sterile forms 

are associated with montane stations such as in the 

Rocky Mountains, but not enough material is available 

to make generalities (see analogy with T. inclinata var. 

densa below). 

4. TORTELLA TORTUOSA Plates 4–5 

Tortella tortuosa (Hedw.) Limpr., Laubm. Deutschl. 1: 

604. 1888. 

Tortula tortuosa Hedw., Sp. Musc. 124. 1801. 

Barbula tortuosa (Hedw.) Web. & Mohr, Ind. 

Musci Pl. Crypt. 2. 1803.

Mollia tortuosa (Hedw.) Lindb., Musci Scand. 

21. 1879. 

Trichostomum tortuosum (Hedw.) Dix., Handb. 

Brit. Mosses ed. 1. 221. 1896. 

Plants in dense tufts, often forming deep and extensive 

sods, dull, green, yellow-green or yellow-brown above, 

brown below, becoming reddish at higher latitudes and 

altitudes, elongate. Stems 1–6 cm high, leaves distantly 

disposed along the stem with the shining leaf bases 

usually apparent, branching tending to concentrate in 

the distal part, central strand nearly always absent, 

rarely present, sclerodermis moderately developed, 2– 

3(–4) cells deep, cells of the central cylinder rather 

thick-walled, stems visibly tomentose with dense redbrown 

radicles, rarely nearly eradiculose in very small 

stems. Stem leaves rather soft, uniform in size, strongly 

crisped or contorted with spirally curled tips when dry, 

flexuose- to wide-spreading when moist, longlanceolate 

to linear-lanceolate, broadly to narrowly 

concave or nearly plane below to more or less keeled in 

the apical region, (2–)3–6.5(–7) mm long; base 

somewhat broader than limb, oblong; margins usually 

shortly and strongly undulate, evenly crenulate by 

papillae, gradually subulate-acuminate, apex 

acumination confluent with the mucro, leaves at the 

extreme stem apex surmounted by a stout, multicellular 

mucro, sometimes the mucro longer than the incipient 

lamina of young leaves at the stem apex; costa 

excurrent as a long, smooth or denticulate mucro or 

short awn, usually composed of 5–10 rhomboidal cells, 

yellow or reddish and shining, adaxial cells of the costa 

variable, costa above the leaf base to the distal median 

region covered by an epidermis of quadrate to shortrectangular 

(2:1) papillose cells, in the distal adaxial 

region variously with a narrow or broader central 

groove of exposed, smooth, elongate (8:1) stereid cells, 

occasionally the groove conspicuous and extensive; 

15 

cross section lunate and broad, nearly flat on the 

adaxial surface in the median leaf region, rounder 

distally on the leaf, adaxial and abaxial stereid bands 

present, guide cells present in one row; proximal 

laminal cells abruptly differentiated from distal cells in 

color, cell size, cell wall thickness and papillosity, with 

proximal cells hyaline, laxly thin-walled, marginal angle 

steep; distal laminal cells unistratose, quadrate, 7–10 

(12–13) µm wide, marginal cells undifferentiated. 

Asexual reproduction: Modifications for asexual 

reproduction none except possibly through fragility of 

the lamina in some populations, or weakness toward the 

apex. Sexual condition: dioicous, seldom fruiting. 

Perigoniate plants rare; perigonia apparently few per 

stem, inner perigonial bracts ovate and abruptly 

apiculate, scarcely longer than the antheridia, 0.5 mm 

long. Perichaetiate plants common; perichaetia 

numerous on the stem; perichaetial leaves differentiated 

even in unfertilized perichaetia, slender and erect at the 

base, long, 5–5.5 mm, somewhat sheathing, distal part, 

consisting mostly of costa, setaceous-subulate, erect, in 

fertile plants, stiff and slightly flexuose, distinct and 

conspicuous above the tightly crisped cauline leaves 

when dry, occasionally with a border of clear, elongate, 

thick-walled cells extending down from the apex in all 

or some leaves, occasionally the leaves without 

quadrate, papillose epidermal cells anywhere on the 

adaxial surface of the costa in some or all leaves. Seta 

red below, paler above, 0.9–2.7 (rarely 3.5) cm long. 

Capsule 1.5–3.3 mm long; annulus not vesiculose; 

operculum 1.5–2 mm long, nearly as long as the 

capsule; peristome teeth long and spirally wound 2 or 3 

times. Calyptra cucullate. Spores 10–12(–13) µm, 

papillae small and somewhat irregular in size and 

distribution, not punctate or fine. 

Spores mature in late spring and summer (late 

June through August). 

KEY TO VARIETIES OF TORTELLA TORTUOSA 

1. Plants red-green, appearing black below, densely foliose with a thick apical coma, leaf bases hidden; stems 

appearing atomentose but tomentum hidden in the bases of branch innovations; leaves not fragile, intact; leaves 

often strongly squarrose-recurved when wet, plane, not undulate; proximal laminal cells thick-walled and brownish, 

intergrading in shape and size with the distal cells, which are often non-papillose in the area of merger; leaves 

broadly concave in section; costa at midleaf exposed adaxially by as much as four stereid cells . . . . 4b. Tortella 

tortuosa var. arctica 

1. Plants green or yellow-green, appearing brown below, loosely foliose, some leaf bases exposed, only slightly 

comose at stem apex; stems conspicuously tomentose; leaves fragile or not, erect- to erect-spreading when wet; 

undulate or plane; proximal laminal cells thin-walled and hyaline, sharply differentiated in shape and size from the 

papillose distal cells, which are papillose in the area of contact; leaves keeled in section, costa in apical region 

exposed adaxially by up to two stereid cells in width or completely covered with an adaxial epidermis of quadrate 

papillose cells . . . . 2 

2. Leaves in tight, complex spirals when dry, appearing soft or lax throughout the stem length, not fragile 

or erose, leaf tips nearly all present; conspicuously undulate; leaf cross section without bistratose areas 

beside the costa, the lamina uniformly unistratose, lamina intact, the costa always differentiated into guide 

cells, stereids and epidermal cells, adaxial stereid layer never disappearing toward the apex, adaxial 

epidermal layer typically absent apically in a medial groove to two stereid cells in width 4a. . . . . T. 

tortuosa var. tortuosa 

2. Leaves in loose, simple spirals or once circinate when dry, appearing firm or rigid, often fragile and 

erose, leaf tips often absent; inconspicuously undulate, especially when dry; leaf cross section with

16 

bistratose areas beside the costa, the lamina irregularly bistratose in patches, lamina tattered, costa 

occasionally appearing undifferentiated in apical region of the leaf, adaxial stereid layer occasionally 

disappearing toward the apex, epidermal layer may be continuous throughout the leaf length . . . . 4c. T. 

tortuosa var. fragilifolia 

4a. TORTELLA TORTUOSA VAR. TORTUOSA 

Plate 4 

Leaves in complex spirals when dry, appearing soft or 

lax, not fragile or erose, leaf apices nearly all present; 

conspicuously undulate; distal laminal cells rarely 

attaining 12 µm, usually less, quadrate, papillose 

adaxial cells on the surface of the costa present in the 

median leaf region or higher; leaf cross section without 

bistratose areas beside the costa, the lamina uniformly 

unistratose, costa always differentiated into guide cells, 

stereids and epidermal cells, adaxial stereid layer never 

disappearing toward the apex, adaxial epidermal layer 

typically absent to two stereid cells wide near the stem 

apex. 

Common in calcareous regions in a diversity of 

habitats such as exposed or forest-shaded rock crevices, 

boulders, ledges of mountains or low, peaty soil and 

rotten wood, dry wooded hillsides or wet areas such as 

Thuja swamps, banks of streams over humus, river 

margins, in northern regions in wet tundra and 

solifluction lobes; elevation 100–3800 m; Greenland; 

Alta., B.C., N.B., Nfld., N.W.T., N.S., Ont., Que., Sask., 

Yukon; Alaska, Calif., Colo., Idaho, Ill., Ind., Iowa, 

Maine, Mass., Mich., Minn., Mont., Nebr., Nev., N.H., 

N.Mex., N.Y., N.C., Ohio, Oreg., Penn., S.Dak., Tenn., 

Tex., Utah, Vt., Va., Wash., Wisc., Wyo.; Mexico, 

Guatemala, Europe, Asia, n Africa. "Iceland; Svalbard, 

widely distributed in Europe; eastward across northern 

and Arctic Asiatic USSR; Central Asia; Japan (Steere 

1978). 



Tortella tortuosa in North America is a species 

of Temperate-North Temperate, Boreal and Subarctic 

distribution and of higher elevations in lower latitudes. 

In northern regions where their ranges overlap, T. 

tortuosa is more frequently found mixed with T. 

fragilis. This is especially true in Canada. 

Numerous specimens growing on islands and 

the coastal mainland of British Columbia and southern 

Alaska in presumably hyperoceanic climatic situations 

show anomalies of substrate and certain morphological 

characteristics, such as deep sods of long-stemmed, 

densely comose stems, thick-walled proximal cells and 

an epiphytic habitat. These collections seem worthy of 

further study, but their relationship to Tortella tortuosa 

seems clear. 

In Europe the species is best developed and 

frequently fruiting in montane forests in the northern 

Alps (Braunmiller et al. 1971). These authors further 

report that the plant is sterile near the borders of its 

range and it is nearly absent in the Lowlands. The 

species is discussed as a possible adventive in the 

Netherlands by Ruber (1973). 

Hyvönen (1991) has reported a bipolar 

distribution of this species (s.l.) based on reports for 

southern South America, such as by Seki (1974) who 

indicated its distribution there as "West & South 

Patagonia and Tierra del Fuego." It was noted also by 

Matteri (1985, 1986). In the latter publication Matteri 

reported that Tortella tortuosa possessed a bipolar 

distributional pattern that included "Boreal-Patagonian 

taxa absent from Australasia.". One specimen of T. 

tortuosa seen from Patagonia was autoicous with 

stalked, flattened perigonial buds, and the suggestion 

was made that this specimen was Tortella knightii 

(Mitt.) Broth. instead, an Australasian species (Eckel 

1997). Re-examination of other specimens from 

southern South America in the light of this possibility 

may indicate that T. tortuosa is confined to the Northern 

Hemisphere. 

The report of Tortella tortuosa from Colombia 

by Churchill (1989) has been emended to 

Pseudosymblepharis schimperiana (Paris) Crum 

(Churchill & Linares 1995). These two species are 

notoriously difficult to distinguish where their ranges 

overlap. Delgadillo et al. (1995) in their bibliographic 

catalogue of neotropical mosses report T. tortuosa from 

Colombia and Peru. 

The epithet refers to the most striking 

characteristic of the species, its crisped leaves in spirals 

when dry which, together with the strongly undulate 

margins, is unlike other species in the genus or related 

genera. The long, multicellular, vitreous awn confluent 

with the lamina is also distinctive. 

Only in very depauperate forms is Tortella 

tortuosa confused with T. fragilis, which has leaves 

more rigid, more or less erect, not or only slightly 

contorted when dry and regularly fragile with 

propaguloid modifications in the leaf apex. 

Occasional specimens of Tortella tortuosa 

have large laminal cells, on average 12 µm wide and 

attaining 14 µm in individual cells. They seem to be 

distinguished by no other character (British Columbia, 

damp outcrop, Range Lake, 59°52N', 137°50'W, 1100 

m, W. B. Schofield 98194, UBC; New York State, 

Essex Co., 2500–2800 ft. (760–850 m), vertical rock, 

Redfearn 13444, UBC). 

Tortella tortuosa leaves are never with 

incurved margins nor are apically cucullate as in T. 

inclinata or in the younger leaves of T. rigens. 

Throughout the range of T. tortuosa occasional puzzling 

specimens may be found with cell sizes to 14 µm in the 

distal region of the leaves (Nebraska, Alberta, Arctic 

specimens). One specimen from Newfoundland 

(Waghorne, 25, NY) had plants with leaf cell sizes to 14 

µm in a collection that also contained rather typical T. 

tortuosa plants, as did specimens from Alberta (Crum & 

Schofield 5260, MICH; Bird & Lakusta 16219, MICH)

and Quebec (Ireland 11186, MICH). These specimens 

may have the mucro reduced to a thin, weak, sharp 

apiculus with broader leaves, or the lamina reduced and 

the long mucro retained; the cell walls may appear 

thicker than usual and the marginal papillae seem more 

sharply salient, the crenulations more prominent. 

Absence of adaxial epidermal cells on the costa 

throughout much of the distal portion of the leaf may 

suggest T. inclinata s.l., but epidermal cells may be 

found at least in the mid-proximal regions of most 

leaves. R. Zander (pers. comm.) suggested these forms 

may be associated with the perichaetium and be 

modifications due to sex hormones. Examination of 

these specimens showed them all to be non-fruiting 

perichaetiate plants. In some of the thinnest of the 

leaves there may be a marginal border of epapillose 

cells somewhat indicative of T. fragilis leaves, and in 

cross section an occasional bistratose patch on the 

lamina—: these in T. tortuosa are only in the 

perichaetial leaves or leaves associated with the 

perichaetium. An occasional perichaetiate population of 

T. tortuosa looked strikingly like T. fragilis when the 

setaceous perichaetial leaves of T. tortuosa dominate the 

leaves at the stem apex—especially when they had 

differentiated border cells. Frequently several of these 

perichaetial leaves have clear (non-papillose) elongate, 

thick-walled cells along the margins toward the leaf 

apex, and sometimes these may be sharply serrulate by 

the projecting distal ends of the marginal cells. Cross 

sections of these leaves, however, readily demonstrate 

that they are not bistratose and that these are not 

propaguloid leaves, they occur only at the stem apex 

where they envelop archegonia, while the stem leaves 

below them are typical of the species. Occasional 

specimens determined as T. fragilis are richly 

perichaetiate specimens of T. tortuosa with extremely 

long perichaetial leaves throughout the stem resembling 

the subulate propagula at the leaf apices of the former 

species. The tips of these leaves, however, are not 

swollen at the distal ends, but are either uniform in 

length or diminish in circumference toward the leaf 

apex. Occasional perichaetiate populations of this type 

may have no quadrate, papillose cells anywhere along 

the length of the costa on any leaves. 

Cell modification in perichaetiate leaves 

includes a tendency to thicker-walled cells, especially 

striking in the transitional area between the proximal 

and distal laminal cells where very thick walled cells 

extend far down toward the base adjacent to the costa. 

The transitional area between distal and proximal cells 

may be extensive. In some leaves on the same stem with 

more typical leaves, laminal cells in the proximal region 

complete with papillae may extend nearly to the leaf 

insertion, with the elongate, lax, smooth, hyaline cells 

restricted to the leaf margins. 

Leaves of Tortella tortuosa with no quadrate 

papillose cells on the adaxial surface of the costa 

throughout their length are always perichaetial leaves: 

sterile leaves just below them on the stem have the 

typical quadrate cells on the adaxial surface, or the stem 

17 

may bear only perichaetia in which case many or all 

of the leaves may have elongate smooth cells on the 

adaxial costa surface for the entire leaf length. 

Fertility of Tortella tortuosa does not appear to 

be enhanced by wet conditions; fruiting and fertile 

plants are found on dry habitats, such as on rock and 

cliff surfaces, as well as wet substrates, such as soil in 

bogs. 

Most published illustrations of the costa of 

Tortella tortuosa are taken from sections in the middle 

region of the leaf. The costa in fact is very reduced in 

size toward the leaf tip and the adaxial epidermis is 

usually absent there by a width of two cells. Short, 

depauperate plants of this species with correspondingly 

reduced costae in their leaves may then show anomalous 

cross-sections much resembling that of T. inclinata, i.e. 

keeled, with small and nearly circular costae without an 

adaxial epidermal layer. 

The distinctively keeled leaf cross-section of T. 

tortuosa is especially evident in the distal portion of the 

leaf. Occasionally in the median leaf region it is so 

broadly keeled that the leaf appears canaliculate. 

Occasionally, the leaf apices may be longsubulate 

and suggest T. fragilis, but the deciduous 

apices of that species are usually obtusely thickened at 

the narrowed end. The narrow tips of T. tortuosa leaves 

narrow gradually to their glistening points, whereas the 

leaves of T. fragilis are more abruptly narrowed into the 

base of a subulate leaf tip. The mucro of T. tortuosa is 

longer, usually to 10 cells, whereas that of T. fragilis is 

shorter, with to 5 or 6 cells or less. 

Small plants to 1 cm may be confused with T. 

alpicola, but the characters of stout mucro in subulate 

leaves, corkscrew twisting apices when dry, and 

undulate leaf margins in addition to the intact, nonpropaguloid 

leaf apices are diagnostic of T. tortuosa as 

is the absence of a stem central strand and smaller 

laminal cells. 

Occasionally Tortella tortuosa has shorter, 

broad leaves with a reduced mucro, resembling T. 

humilis. If there are no autoicous buds, then the 

presence of a stem central strand will confirm these as 

T. humilis. If there are no quadrate papillose cells on the 

adaxial surface of the costa on non perichaetiate (sterile) 

leaves, reference might be made to T. rigens and both 

varieties of T. inclinata. 

Tortella tortuosa exceptionally has a distinct 

central strand but this trait is generally absent. Stem 

sections often are large, to 13 cells across the stem 

diameter, or more. The cell wall thickness gradually 

diminishes toward the center and what can be 

interpreted as a central strand seems to be only the very 

thinnest walled cells at the center (usually two). When 

T. tortuosa has a definite central strand, the strand is 

like that of T. humilis or T. flavovirens where it is more 

clearly distinguished by numerous cells smaller and 

with much thinner walls in contrast to the surrounding 

cells. 

Long-leaved Bryoerythrophyllum 

recurvirostrum (Hedw.) Chen can resemble T. tortuosa

macroscopically in its spiral, twisting leaf apices and 

glistening leaf bases due to inflated or lax, clear cells, 

but it never has the long, needle-like mucro and the 

margins are strongly recurved, with the costa dull on the 

back. There are usually one or more coarse, irregular 

teeth associated with the short mucro or leaf apex. 

Trichostomum tenuirostre, with leaves also 

spirally twisted when dry, has hyaline proximal cells 

that extend indistinctly only a short way up the margin 

by a few cells in width, and the stem is not radiculose. 

In fruiting specimens, Tortella tortuosa will have long, 

spiraled peristome teeth. The teeth of Trichostomum are 

erect or slightly inclined and often rudimentary. 

Trichostomum tenuirostre usually has a large and 

distinct stem central strand, and the apex of the leaf 

often has low and distant teeth. 

The traditional treatment of Tortella tortuosa 

in North America is remarkable for its lack of problems 

in taxonomy. It leaves one unprepared for the 

presentation of the species in the European literature. In 

Europe, a continent of physiographically complex 

microregions, the variation is either divided into several 

species or is considered to be a single species of great 

morphological variability united by a welter of 

intergrading forms. Podpera's list (1954) of subspecific 

taxa shows five forms for the var. tortuosa and 22 

additional infraspecific taxa for a total of 27, 

demonstrating that the species is "oecomorphis dives" 

indeed. 

But how is it that North American Tortella 

tortuosa has been reported as so uniform in its 

characteristics throughout the continent, and can this 

fact contribute to resolving some European problems by 

assuring systematists that some of the European 

infraspecific taxa are, in fact, species? For example, 

there is the issue of the central strand and its utility in 

giving specific rank to such taxa as Tortella 

bambergeri, T. brotheri (Broth.) Broth., and T. 

fleischeri (Bauer) Amann, all considered by various 

authors as varieties, forms, subspecies or synonyms of 

T. tortuosa, yet all three have a distinct stem central 

strand. 

Meylan (1921) wrote that Tortella fleischeri 

and T. bambergeri in the French Jura Mountains "are 

probably only forms or races of the polymorphic T. 

tortuosa. While studying the constancy of their principal 

differentiating character, which is the presence of the 

central strand, I can say in my opinion, this character 

has little value for this group. Typical T. tortuosa also 

sometimes presents a stem in which the central strand is 

completely absent, sometimes, on the contrary, a central 

strand with various degrees of development. I have 

seen, for example, specimens well fruiting and 

representing the typical species to include a large, 

distinct central strand. Limpricht and other authors, 

mainly Germans, have accorded a very great importance 

to the presence or absence of a central strand, mainly in 

various groups." 

In my experience with North American species 

of Tortella, the central strands, when they occur, have 

18 

little "various degrees of development." Cells of the 

central cylinder in species characterized as having no 

central strand may get thinner-walled in the center of 

the stem, but the definition of a central strand used here 

is a group of cells (more than two) in the stem center 

that are abruptly smaller than the surrounding cells and 

have thinner walls. 

Throughout the range of Tortella tortuosa and 

its varieties arctica and fragilifolia, the occurrence of a 

definite stem central strand was only found in a 

specimen from South Dakota in an otherwise typical 

plant, and another in Montana, one from Michigan 

(Emmet Co. on shady soil, Cecil Bay, 4 miles W of 

Mackinaw City, R.R. Ireland 4379, UBC); also in var. 

fragilifolia: Vermont, Haring 1939, NY, CANM (see 

discussion below). 

Sections were also made at the bases of the 

longest stems examined (to 6 cm) to see whether a 

central strand may be expressed in the initial stages of 

growth, and lost later, or lost in branches, but all such 

specimens proved to lack a central strand anywhere 

throughout the stem length. 

The treatment of Tortella tortuosa in Mexico 

(Zander 1994d), where a central strand is reported for 

the species, is somewhat problematical due to the 

confusion of several specimens of Pseudosymblepharis 

schimperiana, which does have a central strand, with 

specimens used in preparation for that description of T. 

tortuosa. In fact, the illustration for the latter species is 

probably that of Pseudosymblepharis, and not T. 

tortuosa. 

Ascribing too much "polymorphism" to a 

single species, Tortella tortuosa s.l., in Europe may 

disguise the floristic conclusion that Europe is an 

important center of diversity in the genus Tortella in the 

North Temperate Zone. 

4b. TORTELLA TORTUOSA VAR. ARCTICA Plate 5 

Tortella tortuosa var. arctica (Arn.) Broth. in Fedch., 

Fl. As. Ros. 13: 160. 1918. 

Mollia tortuosa var. arctica Arn., Ark. f. Bot. 

13(2): 51. 1913. 

Tortella arctica (Arn.) Crundw. & Nyh., Trans. 

Brit. Bryol. Soc. 4: 187. 1963. 

Plants red-green or yellow-green with a coppery sheen 

above to black below, in coarse, dense, stiff tufts or 

deep sods, elongate. Stems 0.8–7 cm high, leaves 

densely disposed on stem or in dense annual whorls 

distinctly separated by less foliose regions, terminating 

in a thick multi-branched comal tuft, leaf bases hidden 

in foliose regions, exposed in elongated ones, branch 

innovations disposed throughout the stem, central strand 

absent or occasionally present, sclerodermis robust, 4– 

5(–6) cells thick, cells of central cylinder relatively 

thick-walled, tomentum absent or occasional, especially 

at the bases of innovations, or inconspicuous and hidden 

in the leaf axils. Stem leaves coarse, uniform in size 

along the stem, somewhat larger at the stem apex, 2.3–6

mm long, rigid, not fragile, when dry erect, becoming 

crisped at the leaf tips, when wet erect-spreading, 

occasionally squarrose-recurved, narrowly longlanceolate, 

quickly narrowing above an enlarged 

proximal region, forming a variably distinct and 

stiffened limb, gradually tapering toward and confluent 

with the apex, canaliculate to tubulose, never keeled, 4– 

5(–6) mm long; base oval to rectangular-oblong, 

generally broader than the limb; margins erect and 

somewhat incurved, not to somewhat undulate, evenly 

crenulate by papillae or smooth, gradually subulateacuminate, 

apex: acumination narrowly confluent with 

the mucro; costa excurrent as a long, smooth or 

denticulate mucro or short awn, usually composed of 

more than five rhomboidal cells, concolorous with the 

lamina, adaxial cells of the costa variable, costa in the 

median region usually covered medially by quadrate to 

short-rectangular (2:1) papillose cells (epapillose if the 

lamina has no papillae), and in the distal adaxial region 

with a narrow central groove of exposed, smooth, 

elongate (8:1) stereid cells; cross section lunate and 

broad, nearly flat on the adaxial surface in the median 

leaf region, adaxial epidermis present and usually 

continuous in median leaf region, absent by about 4 

cells breadth in the distal region, adaxial and abaxial 

stereids present, guide cells present in one row; 

proximal laminal cells often gradually differentiated 

from distal cells, displaying a zone of cells intermediate 

in color, cell size and papillosity, with an intermediate 

zone of thick-walled porose proximal cells, yellow, 

elongate, firm, seldom laxly thin-walled except at the 

insertion, smooth; marginal angle steep and defined by 

cell length and smoothness, with a line of longer cells 

extending a short way up the lamina; distal laminal 

cells unistratose, rounded-quadrate or hexagonal to 

rounded (with increase in wall thickness), (7–)10–12(– 

13) µm wide, incrassate, strongly papillose or the 

papillae apparently undeveloped at the expense of 

strongly thickened cell walls, the papillae when present 

small to massive and obscuring the cell lumen; marginal 

cells undifferentiated. Asexual reproduction: 

modifications for asexual reproduction none. Sexual 

condition: dioicous. Perigonia not seen. Perichaetia 

terminal, inner perichaetial leaves long-linear, laxly 

subulate, the costa sometimes extending beyond the 

reduced lamina into a naked subula, perichaetial leaves 

longer than the stem leaves, to 10 mm in length, distal 

lamina confluent with a long or short subula. 

Sporophyte: fruiting plants not seen. 

Usually associated with more moisture than the 

typical variety: depressions in mesic to wet tundra, soil 

over limestone and sandstone above timberline in Arctic 

Canada and mountain peaks in the Rocky Mountains, 

below snowfields, in hillside and mountain seepage and 

seepage channels, bordering streams and ponds, wet 

rock surfaces, fens, wet frost boils, frequently on ridges 

in the wetter areas; in Carex-Eriophorum meadows, 

Dryas communities, Cassiope heaths; wet sandy 

Eriophorum triste meadow (N Greenland), in the Arctic 

19 

from sea level to 800 m, in the Rocky Mountains to 

3600 m. 

Greenland, Lab., N.W.T., B.C., Nfld., Yukon; 

Alaska, Colo., Maine, Siberia; in North America in the 

Arctic "with a disjunct distribution southward in the 

western mountains to northern British Columbia and 

Colorado. Distributed across Arctic Asiatic USSR and 

in northwest Yunnan" (Steere 1978). 


DUKE, MICH, MO, NY, UBC. 

Other than high altitude reports from 

mountains in the north temperate zone, this variety is 

best and most abundantly present in the wet lowlands of 

the coasts of seas, gulfs and bays of the Arctic Ocean. 

The variety is easily distinguished when the 

leaf cells are extremely thick-walled and obscured by 

massive papillae, and the proximal cells grade so 

gradually into the distal laminal cells as to be 

indistinguishable except for their length. However, 

intermediates in anatomical characters are abundant in 

more temperate situations in a zone between north 

boreal regions and the Arctic (extreme north). 

Intermediate specimens are also found at higher 

elevations in the south Boreal zone and in wet 

continental (coastal) margins in the extreme northern 

Arctic and Greenland. 

The most distinctive representative specimens 

are a blackened brick-red, with very long stems, and so 

densely foliose with rigid leaves as to appear to felt the 

substrate on which they grow with woolly mats. 

However, in sites that have more temperate conditions, 

especially at the southern limit of its range, the species 

varies such that the stems may be less foliose, hence 

less dense except in innovating whorls and at the apex, 

the dry leaves become more contorted at the stem apex, 

the red color subsides toward the yellow-green of T. 

tortuosa, the sclerodermis recedes to resemble that of 

the typical variety and the distal laminal cells have 

thinner walls, tiny papillae and are strikingly set off 

from the proximal cells which are thin and lax, clear and 

free of papillae. Such plants then have more tortuous 

leaves when dry. 

In such specimens, one must rely for 

identification on the relatively denser habit, at least at 

the stem apex, the leaf shape, with its broader leaf base 

and more abruptly contracted, more rigid distal lamina, 

the tubular or canaliculate leaf cross section (not keeled) 

and the tomentum hidden in the bases of the leaves 

(rather than exposed). Note that the proximal cells of the 

dry leaves are generally not distinct macroscopically on 

the stem by their coloring (being concolorous with the 

distal laminal cells), whereas in Tortella tortuosa var. 

tortuosa the glistening, relatively colorless (pale yellow) 

leaf bases are usually prominent. Proximal cells of var. 

arctica are usually the same color as the lamina only 

paler due to their transparency. There is a zone of 

smooth proximal cells that are as thick-walled as the 

distal laminal ones, whereas in var. tortuosa there is 

usually an abrupt difference between the chlorophyllose 

distal laminal cells and the hyaline proximal cells, and

an abrupt change between the thicker walled papillose 

distal laminal cells and the lax, thin-walled smooth 

proximal ones. 

Crundwell and Nyholm (1963) give an 

excellent account of this taxon, treating it at the species 

level. Although Tortella tortuosa var. arctica may 

replace var. tortuosa in the high Arctic, the variety does 

range to the south into Maine and Colorado. In latitudes 

where the two varieties overlap, their differentiation 

becomes highly problematical. Arctic and high alpine 

material of var. arctica easily show their special 

characteristics. Unfortunately, as Crundwell and 

Nyholm suggest, there are many specimens in more 

southern latitudes that exhibit clearly and sharply 

differentiated proximal cells, more lax and strongly 

undulate leaves, more greenish coloration to make it 

very difficult to differentiate the two taxa with 

confidence. 

The further away from tundra conditions, the 

less highly colored do specimens of var. arctica become. 

This red coloration was discussed by Steere (1976: 54– 

55) for individual variation, but some points might 

apply to latitudinal and altitudinal adaptations of taxa at 

the varietal level: "To sum up, tundra bryophytes, with 

very few exceptions, are wholly tolerant of insolation, 

and many of them develop a 'sun-red' pigmentation as 

the result of full insolation, as Mesoptychia sahlbergii 

and Climacium arcticum, which are often associated and 

yet which may lack the red pigment when protected 

from the direct sun, as on steep north-facing 

mountainsides, where they grow in the shade of slump 

terraces. In fact, most tundra bryophytes tend to be 

considerably more pigmented in full sun than when 

growing in shade." 

The more highly colored reddish characteristics 

of most populations of Tortella inclinata var. densa may 

also be a response to insolation. 

Crundwell and Nyholm (1963) also discuss in 

detail the ecology of var. arctica, emphasizing the 

association of this species with wet soil in seepage or 

poor drainage, areas that may be dry late in the season. 

The difficulty of verifying these ecological generalities 

is the inadequacy of the data on labels. Many mention 

no substrate at all and an equal number mention basalt, 

granite, limestone outcrops, ridges and cliffs, areas one 

would assume were dry, except if the collector 

happened to mention seepage. 

The sclerodermis of var. arctica can be a thick 

rind around the central cylinder. For the Russian moss 

flora, Savicz-Ljubitzkaja and Smirnova (1970), who 

likewise considered var. arctica a variety of Tortella 

tortuosa, illustrated the small stem central strand that is 

infrequently present in the American Arctic (it is as rare 

as in var. tortuosa.) They indicated in their drawings 

that both varieties can have the thick sclerodermis. 

Fruiting plants were not seen, and fertile 

specimens of var. arctica were not often encountered: 

three perichaetiate plants were seen, one from Maine, 

another from Melville Island and one from the Yukon. 

20 

Intermediate populations occur in Oregon 

(Saddle Mountain, Clatsop Co., Schofield & Christy 

81863, UBC). These are dark red plants with thickened 

cell walls with the proximal region concolorous with the 

distal lamina, but their distal leaf cross section is 

intermediate as is their leaf shape, degree of leaf torsion 

when dry, etc. Intermediate plants can be readily found 

in Alaskan collections south of the Brooks Range and 

Greenland. 

Of the dried leaves of Tortella tortuosa var. 

arctica, some may appear stiff and straight, resembling 

those of T. fragilis. The branch innovation-tufts on the 

sides of the stem appear like small dark claws or talons 

with the stiff limbs of the young leaves. Others may be 

perichaetial leaves protruding from among the 

vegetative leaves, appearing as stout glassy mucros. In 

T. fragilis, the leaves will be slightly thickened at the 

tips and most of the distal leaves will have lost their 

tips. 

The plants in similarly dense tufts of Tortella 

tortuosa var. tortuosa are thinner (less foliose) when 

dry, the leaves more crisped-chaotic to the base of the 

stems, which are also conspicuously felted with red 

tomentum. Stems in tufts of var. arctica are usually 

more regimented and regular in their leaf stance with 

only the young tips crisped, the tufts are less open and 

more closely packed. Although they appear not 

tomentose, the tomentum is actually inconspicuous and 

hidden in the leaf bases. T. inclinata var. densa is 

similar to this, with tomentum hidden or reduced to 

nearly absent in well-developed forms. 

Stems of Tortella tortuosa var. arctica in tufts 

in the Arctic are frequently rigidly upright due to a 

varnish of what is perhaps cyanobacteria. The stems are 

so tightly packed that there is no room for flexion 

except at the stem tips. The regimented, tightly packed 

stems in orderly linear packets are also found in Arctic 

specimens of Tortella fragilis—so dense as to be devoid 

of admixtures of other species. 

No species of Tortella so far studied from the 

North American Arctic have stem central strands except 

very rarely. Trichostomum arcticum Kaal. in the Arctic 

has a strong central strand, a useful character to 

distinguish it from specimens of Tortella tortuosa var. 

arctica when there appear to be no differentiated 

marginal proximal cells. Trichostomum arcticum also 

has no quadrate papillose cells on the adaxial surface of 

the costa, whereas Tortella arctica does, for the most 

part. Trichostomum tenuirostre, on the other hand, has a 

variable central strand, sometimes even among sections 

of the same stem: the horizontal transition of distal 

laminal to proximal cells is approximately flat across 

the leaf, not at an oblique angle, and the leaf apex is 

frequently not a long acumination, but one much 

shorter, arising above an abrupt contraction in the leaf 

blade forming small shoulders in the lamina just before 

the apex. 

The thick habit of the var. arctica is 

occasionally seen in more southern latitudes: a 

specimen from Tennessee had leaves thickly covering

the stem with an apical coma, as did another from Iowa 

(Jackson County, along Brush Creek east of Andrew, on 

high limestone ledge, H. S. Conard, Oct. 13, 1945, NY). 

The Labrador specimen from Carroll Cove, J. 

A. Allen, 1882 (NY) cited by Crundwell and Nyholm as 

Tortella arctica and annotated by Crundwell 

demonstrated that the distinctions are not as obvious as 

might be on first sight. The specimens had a reddish 

cast, were in dense, uniform sods, had thick-walled cells 

including proximal cells excepting a median area near 

the leaf insertion and a relatively indeterminate 

transition area between the papillose laminal cells and 

the clear proximal ones, the impression that the costa 

was broader than typical T. tortuosa var. tortuosa: yet 

the leaf shape was not that as illustrated for the variety 

(an inflated proximal area and a differentiated distal 

lamina), the margins were undulate, the plants soft and 

more contorted in the lower stem regions, some of the 

proximal cells were distinct and thin-walled, the stems 

were not densely foliose, tomentum was clearly evident 

from the leaf bases. Perhaps the most decisive 

characteristic was the circular costa section with only 

two exposed stereid cells. The specimen was var. 

tortuosa. 

A specimen from the Cape Parry was 

distinctive because the leaves were long-subulate and 

whip-like, stiff and not erect but flexuose-circinate in 

broad arcs and hardly undulate, the tomentum hidden. 

The color was a dark, dull green, without any evidence 

of reddness except in the lower part of the stem. The 

specimen was so peculiar that at first glance it appeared 

to be a specimen of Tortella fragilis. The laminal cells 

were thin-walled, the proximal ones distinctly set off 

from them in color and smoothness, but were as thick 

walled. The leaf shape distal to the proximal cell region 

was abruptly set off but extremely long with very long 

mucros. It appeared to be Tortella tortuosa var. tortuosa 

except for the peculiar elongate leaf cross-section and 

was ultimately determined to be var. arctica. The leaf 

section appeared reniform, rather than nearly circular as 

in var. tortuosa, apparently due to the number of 

buttressing or supporting cells in the costa adjoining the 

lamina. 

Greenland: west, on dry basaltic rock, Nugssuaq Pen: 

Marrait, 70°30'N, 54°12'W, Sept. 4, 1956, K. 

Holmen 14.806 (NY); head of a fjord S of 

Marmorilik 71°06'N, 51°12'W; Nuggsuaq 

Peninsula: Ikorfat, 70°45'N, 53°07'W; N. 

Greenland, Heilprin Land, Brönlund Fjord 

82°10'N, 31'W. 

B.C.: ca. 3 mi. beyond N. end of Summit Lake, cliff 

shelf, W.B. Schofield 66186, July 23, 1977 

(DUKE). 

Nfld.: South Branch Condroy River, St. George's, Port 

au Port; about 47°54'N, 58°58'W, July 2–4, 

1949, R. Tuomikoski 2432 (MICH). 

N.W.T.: Axel Heiberg Island, E of alluvials of Crusoe 

River 79°23–24'N, 91°01'07"W, in peat, ca, 50 

m., June 20, 1967, M. Kuc M204 (NY); Baffin 

21 

Island: Head of Clyde Inlet, 1950, Baird 

Expedition, P. Dansereau 500820–0254 (NY). 

Yukon: alpine communities N of Carpenter Lake, 4000', 

Wernecke Mtns, 64°33'N, 135°07'W, July 24, 

1972, Scotter 17929 (NY). 

Colo.: Clear Creek Co., dry rocky tundra below 

Argentine Pass, 13,000 ft. (3940 m), 7 miles 

SSW of Georgetown, Aug. 28, 1981, F. J. 

Hermann 29063 (NY); Boulder; Larimer; 

Summit. 

Maine: Somerset Co., ledges by Falls of the Carrabasset 

River, near Anson, July 21, 1896, J. Franklin 

Collins 1311 (NY) (perichaetiate plant). 

4c. TORTELLA TORTUOSA VAR. FRAGILIFOLIA 

Plate 6 

Tortella tortuosa var. fragilifolia (Jur.) Lindb., Laubm. 

Deutschl. 1: 605. 1888. 

Barbula tortuosa var. fragilifolia Jur., 

Laubmfl. Öst. Ungarn 123. 1882. 

Tortella fragilifolia (Jur.) Roth, Hedwigia 49: 

218. 7 f. 2. 1910. 

Tortella tortuosa fo. fragilifolia (Jur.) Monk., 

Laubm. Eur. 270. 1927. 

Trichostomum tortuosum var. fragilifolium 

(Jur.) Dix., Stud. Handb. Brit. Moss. 

221. 1896. 

Barbula tortuosa var. pseudofragilis Thér., 

Rev. Bryol. 25: 20. 1898. 

Leaves in simple spirals or once circinate when dry, 

appearing firm or rigid, laminae often fragile, erose, leaf 

tips often absent; distal laminal cells variable: (6–)7– 

10(–17) µm, inconspicuously undulate, especially when 

dry; leaf cross section with bistratose areas beside the 

costa, the lamina irregularly bistratose in patches, costa 

occasionally appearing undifferentiated in apical region 

of the leaf, adaxial stereid layer occasionally 

disappearing toward the apex, epidermal layer 

continuous. 

Herbaria examined: BUF, CANM, MICH, NY, 

NYS, UBC. 

The year after Haring's treatment of the North 

American species of Tortella (1938) appeared, she 

collected a curious specimen in a soapstone quarry on 

rock from the town of Newfane, Vermont, where A. J. 

Grout lived. This was a rather peculiar example of 

Tortella and she sent it to Grout and to H. N. Dixon in 

England for verification. Both men confirmed the plant 

to be Tortella nitida and, as such, it entered the ongoing 

tradition in the literature that this species occurs in 

North America (see discussion below under Tortella 

nitida). 

USA, Vermont: town of Newfane, soapstone 

quarry, on rock in more or less open woods, Aug., 1939, 

Inez M. Haring, s.n. "checked by Dr. A. J. Grout & H. 

N. Dixon" (NY; duplicate CANM). The CANM 

specimen was annotated as Tortella tortuosa by K.

Holmen, Sept. 59 (there are no mixtures with other 

species). 

A second collection of nearly identical material 

was discovered in Quebec, determined by R. T. 

Wareham as Tortella nitida. Wareham did the 

treatments of Pterygoneurum Jur. and Pottia (Reichenb.) 

F rnr. for Grout's North American moss flora (1939): 

Quebec, Mont Commis, St-Donat de Rimouski, 

rocher calcaire, 2-7-42, Ernest Lepage 3412 (MT), 

being mixed with Myurella julacea, Encalypta 

streptocarpa and Hypnum cupressiforme. This was also 

annotated by Kuc in 1968 as Tortella tortuosa: "T. 

nitida is a critical species, this material doubtfully 

belongs to this taxon.") This collection also contained 

plants of T. alpicola (discussed below). This specimen 

may be one basis for the Quebec citation of T. nitida in 

the Canadian checklist of Ireland et al. (1987). 

This Quebec specimen had setae and capsules, 

most of which, due to their age, no longer retained their 

peristomes or spores. The single capsule with a battered 

peristome showed that to be eroded and quite short, 

short enough to resemble that of Tortella flavovirens or 

T. nitida. The setae were 1 cm long, the capsules 1.2–2 

mm; spores nearly smooth (10–)12(–14) µm. 

Neither of these interesting collections is in 

fact referable to Tortella nitida, yet, except for the stem 

central strand in the Vermont specimen, they were 

nearly identically distinctive. Both had extremely fragile 

leaf apices such that nearly all were absent, there were 

leaf cross-section irregularities (bistratose areas 

juxtacostally), both had relatively broad leaves that were 

rather shortly mucronate and had quadrate cells on the 

adaxial surface of the costa completely through to the 

leaf apex. The adaxial stereid band disappeared distally, 

but the adaxial epidermal layer remained intact, an 

unusual characteristic for species of the genus. 

Additional specimens with similar 

characteristics were discovered during the course of the 

present study in Michigan material at higher elevations 

in the northern part of the state. These include: 

Michigan: Baraga Co. soil in crevices of basalt outcrop 

on bank of Silver River Falls, ca. 6 miles ENE of 

L'Anse, July 9, 1970, F. J. Hermann 23173 (MICH); 

Keweenaw Co., open ground on summit of mountain, 

Mt. Lookout, May 14, 1949, Charles D. Richards 38 

(MICH); also Mt. Bohemia, July 18, 1950, Charles D. 

Richards 212 (MICH); also gravelly field, site of old 

schoolhouse, Copper Harbor, Aug. 30, 1960, F. J. 

Hermann 16352 (MICH); Ontonagon Co., talus blocks, 

Porcupine Mountains, G. E. Nichols & W. C. Steere, 

Aug. 20–27, 1935 (MICH). 

The stem central strand so evident in Haring's 

Vermont specimen was absent in additional collections. 

More specimens then came to light that were not fragile, 

that had large laminal cells (to 17 mm) or else tiny, 

crazed ones. The quadrate, papillose adaxial costal cells 

may disappear in the apex and retain the adaxial stereid 

band. But the many independently variable characters 

had one element in common: they appeared to be 

22 

variations on the basic set of characters for Tortella 

tortuosa. 

One similar specimen from central Alaska had 

leaves sufficiently erect and narrowed as to resemble 

Tortella fragilis. On dissection, it appeared that the 

narrowed leaves were actually typical leaves of T. 

tortuosa, but so inrolled as to be tubulose in crosssection, 

hence its narrow appearance when dry. In 

addition, the adaxial epidermal layer remained intact 

throughout the leaf-length, but the adaxial stereid layer 

was absent. Otherwise, in all other macro- and 

microscopic characters, the plants were T. tortuosa 

("moist, rather sunny seepage on S face outcrop in 

tundra," (65°23'N, 145°57'W, Norris 54115, NY). 

A hitherto European taxon, Tortella tortuosa 

var. fragilifolia (Jur.) Limpr. was examined in the light 

of this variability and all of the above specimens 

conform to descriptions and specimens of that variety. 

The type of var. fragilifolia has no stem central strand 

and also possesses a central groove on the adaxial 

surface of the costa in the distal portion of the leaf, 

exposing the adaxial stereids. This particular specimen 

is so similar to the typical variety as to initially cast 

doubt on its distinctiveness. 

Juratzka (1882) originally described the variety 

fragilifolia as sparsely tomentose, the leaf apices fragile, 

the costa whitish on the back and strongly shining. 

"Diese Form verh„lt sich zu der gewöhnlichen wie 

[Tortella] nitida zu [T.] inclinata und kann leicht mit T. 

fragilis verwechselt werden," by which I understand 

him to mean that the variety stands in relation to the 

typical variety as T. nitida does to T. inclinata. The 

variety is easily confused with T. fragilis, having a 

setaceous, deciduous (but non-propaguloid) apex. 

Among the substrates he noted for the variety were 

sunny calcareous rocks, on trees (an Baumen bei 

Gostling) and sandstone walls. 

Demaret and Castagne (1964), for Belgium, 

separate the variety fragilifolia from the typical one by 

the shorter leaves, 1.4–5 mm, more narrow and fragile, 

the leaves less crisped in the dry state, the leaf point 

more shortly acuminate, the stem without tomentum. 

This is an excellent description for North American 

material in general. However, leaves in both American 

and European material of this variety reach the longest 

length the typical variety can attain. There occur, as 

well all degrees of tomentum on the stem, and lengths 

of mucro. The European authors suggest that the 

distribution of the variety is the same as the species 

whereas it seems to be associated with upland 

(subalpine) stations in North America. 

Albertson (1946), in his key to the Swedish 

species, indicated that the var. fragilifolia was the same 

as the var. tortuosa except that the leaves were less 

strongly spiralled, less undulate and strongly fragile. 

Additional named material from Europe was 

examined. The most stable character that unites North 

American variability with the European is the 

anomalous character of the leaf cross section tending 

toward bistratose juxtacostal regions that may penetrate

into the costa section. Sometimes the adaxial stereid 

layer is missing toward the apex and sometimes the 

quadrate, papillose cells on the adaxial surface of the 

costa extend completely to the leaf apex. The lamina 

may have bistratose patches; some lumina can have 

diagonal cross-walls; in one section there was a 

bistratose area at the leaf margin and in many these 

variations were not symmetrical across the leaf blade, 

but occurred on only one lamina. Occasionally the costa 

itself becomes undifferentiated. 

Sections in the distal leaf more than any other 

single character are critical or fundamental to the 

determination of the variety. The var. fragilifolia, then, 

is essentially a cryptic variety in the sense that the 

definitive characters derive mostly from the anatomy of 

the leaf. It is this variation in Tortella tortuosa as well 

as the characteristics of T. alpicola that together have 

contributed to the concept of T. nitida in North America 

(see discussion below). Also numerous specimens in 

American herbaria identified as T. tortuosa with a query 

can be referred to this variety. Recognition of this 

variety in North America should contribute to stability 

in the determination of specimens. 

Plants are as in the typical variety: in dense 

tufts, interwoven at the stem apices when dry by the 

contortion of the leaves, with the same coloration; 

although the stems in section appear rather larger than 

typically, a distinctive central strand is rarely present. 

The stems usually have a dense rufous tomentum. The 

leaves differ by being little or not undulate or crisped 

when dry, but may be undulate when wet; they are 

typically only once circinate and without the delicate 

appearance of the typical variety. The leaves, within the 

range of variation of the species, tend to be somewhat 

shorter and broadly lanceolate in the proximal twothirds 

(not setaceous), and the laminae can seem to 

deteriorate and appear fragile; yet they may also be 

extremely long and narrow (setaceous), especially in 

western North American populations. 

While some leaves have the typical long, 

glossy mucro of var. tortuosa, in others the mucro is just 

a small tip surmounting what appears to be a 

prolongation of the leaf apex into a canaliculate 

extension, this frequently broken. Infrequently some 

apices have a solid and circular form which, like a 

finger, together with a vague border of cells free of 

papillae, show a striking affinity with T. fragilis. That 

the abaxial surface of the costa in the apical portion of 

the leaf is often roughened with low papillae is also 

indicative of the propaguloid modifications of T. 

fragilis, as is the tendency toward undifferentiated cells 

in the costa in the apical region. 

The leaf midsections occasionally seem fragile 

or brittle, but also it is frequently only the leaf tips that 

are prone to breakage. These appear to taper into thinner 

apical limbs, especially in some leaves, facilitating this 

condition. In some collections, many of the leaf apices 

are broken off (absent). The leaves may or may not have 

a vague display of more or less elongate smooth cells on 

the distal margins, this contributing to a more rigid and 

23 

erect leaf apex, and it hints at the deciduous nature of 

the apex, on analogy with Tortella rigens and more 

conspicuously in T. fragilis. 

The character of the costa is frequently 

different from that of the typical variety: the adaxial 

epidermis can be retained continuously through to the 

apex without a central groove in the distal region: the 

adaxial stereid band can disappear distally. Occasionally 

the quadrate cells are lost in the far distal portion of the 

leaf, but the quadrate cell layer usually extends further 

towards the apex than in the typical variety. In some 

cases the abaxial stereids become substereid and the 

entire costa appears undifferentiated. Note that both the 

epidermal cells and the guide cells should be present in 

the leaf cross section to assess whether the adaxial 

stereid layer has disappeared. If the guide cells are the 

only cells at the adaxial surface of the leaf, the section is 

made too far into the extreme apex to be diagnostic. 

The laminal cells in the distal half of the leaf 

can become very small and obscure: 5–7 µm on average 

with transversely flattened marginal cells. The laminal 

cells may also be very large, reaching 17 µm (the case 

in Macoun 605; see below). The leaf cross section also 

frequently resembles a smaller celled version of Tortella 

rigens: small marginal cells grading into larger ones 

beside the costa, there being a bi-to multi-stratose area 

juxtacostally. The proximal cells may be thicker walled 

than the typical lax cells, the distal laminal cells may 

also be somewhat thicker-walled. The Quebec specimen 

cited above is the only example seen so far of the var. 

fragilifolia in fruit. 

Dixon (1924) wrote that European Tortella 

tortuosa var. fragilifolia is very close in resemblance to 

T. nitida. This is further evidence at least as to fragility 

that the var. fragilifolia described here is indeed within 

the variation of the var. fragilifolia in Europe. Dixon 

described the var. fragilifolia in England as growing in 

"short small tufts; leaves smaller [than in the typical 

variety], shorter, less finely tapering, very fragile..., 

leaves when dry less strongly contorted, but more 

closely incumbent." In specimens seen from both 

Europe and North America, the var. fragilifolia may 

have leaves fully as long or even longer than Tortella 

fragilis or T. tortuosa and the tufts in which they grow 

are neither relatively short nor small. 

Dixon's statement that the var. fragilifolia 

grows on "exposed alpine rocks" seems consistent with 

the tendency of North American var. fragilifolia to 

favor higher elevations and rocky substrates. 

Throughout Dixon's treatment of Tortella (as 

Trichostomum) there is a degree of uncertainty about 

species delimitations concerning T. inclinata, T. fragilis, 

T. nitida and the varieties of T. tortuosa. Dixon, upon 

studying numerous puzzling variants of T. tortuosa, 

postulated that the var. fragilifolia and Boulay's var. 

rigidum (= T. tortuosa var. rigida (Boul.) Limpr.) "are 

not more than dwarf and somewhat starved forms 

induced by their habitat, which is usually on exposed 

mountain rocks." He noted that while T. nitida is fragile

in the lamina, the var. fragilifolia is usually broken at 

the leaf tip. 

Dixon (1924) was also not sure of the 

distinction of Tortella fragilis from T. tortuosa. He 

would have liked to have been definite but, "I have, 

however, received specimens from two or three 

localities in Labrador showing a distinct approach to [T. 

tortuosa]; the leaves on some of the plants, and even the 

lower leaves on some stems in which the distal ones are 

quite typical, being strongly contorted when dry, crisped 

and undulate at the margin, and when moist slightly 

flexuose and undulate; hardly, indeed, to be recognised 

from the var. fragilifolia...except by the leaves 

somewhat firmer, the very shining nerve and the 

presence (usually) of some rigid and less curved 

leaves." He speculated that T. fragilis would be found to 

be close to T. tortuosa "by intermediate forms." 

Polunin (1947), when reporting Tortella 

fragilis for the Canadian Eastern Arctic, quoted the 

same lines from Dixon (1924), adding "This puzzling 

situation that Mr. Dixon has so ably elucidated is to be 

found also in the specimens from Baffin Island" 

although some of these specimens, upon examination, 

may turn out to be T. tortuosa var. arctica. 

Tortella tortuosa var. fragilifolia in North 

America appears to be a montane and boreal taxon. The 

var. fragilifolia in North America is apparently not in 

montane elevations much south of the latitude of the 

Canadian border with the conterminous United States. 

TORTELLA NITIDA IN NORTH AMERICA 

Sources of North American reports of Tortella nitida 

perhaps began with Renauld and Cardot (1889: 99) (as 

Trichostomum nitidum) "United States, without locality, 

collected by James, communicated by Mr. Bescherelle." 

Barnes (1897: 275–276) next reported the 

species, quoting Renauld and Cardot and basing his 

description on Limpricht (1890, 1: 581) where the 

description of the proximal cells indicates their being 

"less sharply marked off from chlorophyllose cells." 

Assuming James, of the Renauld and Cardot quote, to 

be Thomas Potts James (1803–1882), a request was 

made to the Farlow Herbarium for a specimen that 

might correspond, but without success. A James 

specimen at NY collected by him in Canada, July 1850, 

was Tortella tortuosa var. tortuosa. 

Next, there is the exsiccat specimen in the 

Canadian Musci by John Macoun, number 605, labeled 

as Barbula nitida. In 1892, Macoun and N. C. Kindberg 

published the label data (number 196, p. 52). Kindberg 

(1897) affirmed its occurrence ("Can.: Macoun.), 

presumably from Macoun's exsiccat number 605. 

Macoun 605 is labeled as Barbula nitidum [sic] (Lindb.) 

Jur., of the Canadian Musci series, "on wet earth at 

Hector, Rocky Mountains, B.C. [British Columbia, 

Canada], Aug. 14th, 1890" and is Tortella tortuosa var. 

fragilifolia (one specimen at FH, two others at NY). It is 

as tall as the typical variety gets, has a broad stem 

section but no central strand. Some apices are deciduous 

and there is conspicuous fragility of the lamina. The 

24 

crucial juxtacostal bistratose character of the leaf 

cross section is present. The cells of the lamina are 

unusual for the species, on some leaves regularly ca. 14 

µm, some even to 17 µm in diameter, with distal 

proximal cells thick-walled and porose. Some leaves on 

the same stem have adaxial quadrate cells covering the 

costa nearly to the apex, in others the adaxial surface of 

the costa is exposed in a narrow groove from base to 

apex. 

Subsequent reports of Tortella nitida include 

Haring's (1938) treatment in Grout's moss flora of North 

America, Flower's treatment for the state of Utah 

(1973), Ketchledge's publication on the moss flora of 

New York State (1957), Steere and Scotter's (1978) 

report for the Nahanni Preserve, in the Mackenzie 

District of the Northwest Territories, and the recent 

checklist of Canadian mosses (Ireland et al. 1987). 

Haring (1938) cited Macoun's number 605 in 

the Canadian Musci as Tortella nitida; also a specimen 

from Montana in NY, and one by Homer House, a 

specimen from Harris Lake, New York. Haring and 

perhaps Grout, were strongly influenced in their 

expectations by the Macoun specimen. This was at a 

time when it was not yet accepted that some of Macoun 

and Kindberg's determinations were perhaps doubtful. 

Kindberg's determinations of Macoun's specimens 

"were often hasty and consequently wrong and 

Macoun's own determinations were often based on 

superficial study and thus are open to doubt" (Schofield 

1965), and yet Bescherelle and Dixon also made similar 

independent determinations. 

Examination of the specimen of Tortella nitida 

that Haring cited for Lake Harris (New York, Essex Co., 

Newcomb, surface of boulder on shore (perichaetiate: 

without sporophytes), July 10, 1927, H.D.House (NYS), 

shows this specimen also to be Tortella tortuosa var. 

fragilifolia. The Harris Lake specimen was fragile in 

some stems, but on the whole seems more typical of T. 

tortuosa than most examples of the variety fragilifolia 

(it had no stem central strand). The leaves on some 

stems, however, were more broadly lanceolate with 

larger distal laminal cells (to 12 µm). Some of the leaf 

cross sections showed some bistratose areas near the 

costa in the distal median portion of the leaf. 

The Harris Lake specimen showed no 

annotation by Haring or Grout. There was a query on 

the label, and the note, "Doubtful !E.H.K" [Edwin H. 

Ketchledge] written on the packet. Ketchledge's 

subsequent publication on the moss flora of New York 

State (1957) listed Tortella nitida for the New York 

flora, but in the revised edition (Ketchledge 1980), there 

was no citation of that species. 

Another specimen of the var. fragilifolia, from 

New York on limestone pavement in Jefferson Co. in 

association with Tortella rigens, is cited below in the 

treatment of that species. 

Haring's third representative specimen was 

from Montana at NY. A specimen from Montana 

collected by R. S. Williams, 329 at NY was determined 

here to be Tortella alpicola (see discussion of this

specimen under that species). There is no indication on 

the sheet that Inez Haring saw this, but there is an 

annotation on the sheet "= Tortella!! A. J. G. Probably 

T. nitida (Lindb.) Broth." The annotation appears to be 

by A. J. Grout. If this is one of the specimens cited by 

Haring in her treatment of the genus, then her (and 

possibly Grout's) concept of North American T. nitida is 

that of T. alpicola, in part (the specimen is sterile). 

A specimen (MNA, FH) determined by Haring 

as "possibly Tortella nitida" collected in Arizona, was 

also T. alpicola. She noted on the packet that the plants 

had central strands, strongly crisped leaves, which were 

much broken at tips and were sometimes undulate. 

The large size of the two specimens of var. 

fragilifolia is in marked contrast to the tiny dimensions 

of T. alpicola in the four specimens on which she based 

her concept of T. nitida, but it is clear that the 

description of T. nitida by Haring (1938) included a 

conflation of T. alpicola and T. tortuosa var. fragilifolia. 

The var. fragilifolia occurs throughout the middle range 

of the var. tortuosa, and the range of T. alpicola is 

decidedly western (see discussion under that species). 

It is truly extraordinary that in this context the 

specimen from Quebec mentioned above, identified by 

Warham as Tortella nitida, actually contained two 

separate Tortella entities: growing beside the var. 

fragilifolia in this specimen was T. alpicola: a 

significant range extension, indeed, from its western and 

Arctic populations and presumably associated with a 

southern extension of a boreal-Arctic floristic element. 

The tiny T. alpicola had its typical stem central strand, 

the var. fragilifolia plants did not. The single specimen 

determined by S. Flowers as Tortella nitida and 

collected in Utah from Duchesne county turned out to 

be T. alpicola (see Flowers 1973). 

As discussed under Tortella alpicola above, 

citations in the checklist of the mosses of Canada for the 

presence of T. nitida were derived from the literature. 

Those from Quebec are most certainly Tortella tortuosa 

var. tortuosa or var. fragilifolia. For example, another 

specimen at CANM (Quebec, Beauce Co., Beauceville, 

Apr. 20, 1939, Br. Marie-Anselm 2875 (CANM) 

determined by the collector as T. nitida is T. tortuosa 

var. tortuosa. The Macoun specimens (Canadian Musci 

605) on which the British Columbia citation may be 

based are Tortella tortuosa var. fragilifolia. The Steere 

and Scotter (1978) specimen of Tortella nitida cannot 

be located, but there was doubt cast on this 

determination when it was published (see discussion 

under T. nitida below). 

25 

Tortella alpicola differs from T. tortuosa var. 

fragilifolia by its small size, large leaf cells (averaging 

14 µm) and development in sterile material of a 

caducous propagulum at the leaf apex: primarily in the 

same characters that distinguish it from typical T. 

tortuosa. What brings Tortella nitida strongly to mind 

when viewing Haring's Vermont specimens is the nearly 

complete absence of leaf tips in the collections. The 

distinct central strand happens here to be merely one of 

the rare instances of its occurrence in the species, s.l.. 

The eastern North American plants differ from 

European T. nitida by the long-lanceolate to nearly 

subulate leaf, not ligulate (not resembling T. humilis), 

the rather consistently long mucro, the fragility confined 

to the leaf apex rather than throughout the leaf, and the 

distinct and abruptly differentiated proximal cells rather 

than the gradual transition in the European plant. The 

cross section of the costa of true T. nitida has a convex 

adaxial silhouette, massive abaxial and adaxial stereid 

layers and a striking adaxial palisade of papillose 

epidermal cells. This resembles no Tortella species in 

North America; it is more like Pseudosymblepharis 

schimperiana of Latin America in this respect. 

Another feature distinguishing the Vermont 

specimens from Tortella nitida is that they are stiffly 

erect when dry and are only twisted in the apices of the 

leaves at the apices of the stems, such as actually retain 

their apices. Tortella nitida has stiffly incurved leaves 

when dry throughout the stem. 

The V-shaped region of proximal leaf cells of 

this variety place the Vermont specimens clearly in 

Tortella, as no other genus in North America has this 

characteristic. This feature distinguished the material 

from any species of the genus Trichostomum, yet the 

specimens resemble that genus in two characters: there 

is an elaborate, unambiguous stem central strand, and 

the surface of the costa is completely covered with 

quadrate, papillose cells, where there is no exposure 

whatsoever. Tortella tortuosa has an adaxial groove 

where the stereid cells are exposed to no more than two 

stereid cells and these only at the distal end of the leaf. 

Most other taxa in the United States lose the adaxial 

epidermal layer in the apex of their leaves except for 

Tortella alpicola. In that species the quadrate cells of 

the epidermal layer also persist while the adaxial stereid 

layer disappears and in fertile stems there is no stereid 

layer at all in the distal portion of the leaf—reminiscent 

of the leaf section of species of the South American 

genus Streptocalypta or the Antarctic Sarconeurum. 

KEY TO NORTH AMERICAN TORTELLAE WITH DECIDUOUS APICES 

1. Stem with central strand, leaf cells generally 14 µm, quadrate papillose cells on the adaxial surface of the costa; 

very small plants to 1.5 cm; with distinctive caducous propagula at modified leaf apex when sterile, plants of the 

West, disjunct to Quebec . . . . Tortella alpicola 

1. Stem without central strand, leaf cells to 14 µm or less, quadrate papillose or elongate smooth cells on the adaxial 

surface of the costa; plants typically larger than 1.5 cm, plants on limestone pavements or of general distribution 

throughout northern North America . . .. 2

26 

2. Leaf cells generally 14 µm; elongate, smooth stereid cells exposed on the adaxial surface of the costa 

throughout the leaf length in all cauline leaves; plants of limestone pavements in the Great Lakes region . . . 

. Tortella rigens 

2. Leaf cells to 12 µm or less, adaxial surface of the costa partially covered by quadrate, papillose 

epidermal cells in stem leaves, plants of North Temperate, Boreal and Arctic distribution . . . . 3 

3. Leaves erect and stiff, lamina intact, leaf developing a morphologically distinct deciduous 

apical propagulum, leaves in section regularly bistratose in the apical portion . . . . Tortella 

fragilis 

3. Leaves generally crisped, lamina tattered, leaf not developing an apical propagulum, merely 

disintegrating in pieces, leaves in section irregularly bistratose in patches . . . . Tortella tortuosa 

var. fragilifolia 

Specimens examined, with annotations: 

Alta: 6000 ft. (1820 m) in tundra, Crum & Schofield 

5742 (MICH, UBC) ( with extremely long, 

slender, tattered leaves to 7 mm). 

Alta: rather dry boulders near Athabasca Falls, 20 mi. S. 

of Jasper, 4100 ft. (1241 m), June 25, 1955, H. 

Crum 3700 (MICH); tundra, slopes of Forum 

Peak, above Cameron Lake, ca. 6000 ft. (2000 

m), Waterton Lakes National Park, July 30, 

1955, H. Crum & W. B. Schofield 5742 

(MICH). 

B.C.: outcrop crevices Mt. Liumchen-Church Mt area, 

49°02'N, 121°50'W, Schofield 63200 (UBC). 

(The only thing to associate this specimen with 

the variety, other than general aspect, was the 

distal leaf cross section which displayed a 

brightly chlorophyllose adaxial epidermis 

covering the adaxial stereid layer, this latter 

vanishing toward the apex, leaving the 

quadrate cells intact. There was no central 

strand and the diameter of the stem in section 

was small). 

B.C.: Sn. Cariboo Mtns 52°N, 120°W., moist hummock 

on river shore, distal Tsuga zone, Leena & 

Teuvo Ahti 13730 (UBC). (Identified by the 

leaf cross section: all leaves intact.) 

Alaska: southeast; above tree line (2200–3160') on 

Harbor Mtn. above Sitka, Baranof I., mostly 

slopes of exposed rocks in heathers, shaded 

rock angle, Aug. 4, 1968, I. A. Worley & F. M. 

Boas 10066 (UBC). 

Calif.: Siskiyou Co., moist, rather shaded rock outcrop 

in seepage, in open alder, incense cedar and 

white fir forest on south slopes of Red Butte 

near Cook & Green Pass, 6500 ft. (2170 m), 28 

Sept., 1977, D. H. Norris 50265 (UBC). (The 

specimen in leaf cross section had the quadrate 

adaxial epidermis intact, the adaxial stereid 

disappeared and some of the cells were with 

bistratose patches, no stem central strand.) 

Wash.: North Cascades, Rainy Pass area, Lewis Glacier 

and Lake, subalpine meadows, seepy slope 

above lake, under water in temp. snowmelt 

stream in subalpine, 48°N 121'W, June 29, 

1982, J. Spence 1077 (UBC). 

A form of variation seen in only one specimen 

was the presence of quadrate, chlorophyllose papillose 

cells on the back of the costa in the leaf apex (Alberta, 

Crum 3512 UBC) which was acicular and nearly 

cylindric. There is a variety of Tortella nitida in Europe 

(var. irrigatum (Winter) Zand., see below) that has this 

feature, but the specimen cited here is the only occasion 

in T. tortuosa s.l. noticed in the present study of this 

feature. 

5. TORTELLA FRAGILIS Plate 6 

Tortella fragilis (Drumm.) Limpr., Laubm. Deutschl. 1: 

606. 1888. 

Didymodon fragile [sic] Drumm., Musci Amer. 

(Rocky Mts.) 127. 1828. 

Trichostomum fragile (Drumm.) C. Müll., Syn. 

Musc. 1: 586. 1849. 

Barbula fragilis (Drumm.) BSG, Bryol. Eur. 6: 

157. 1855 (fasc. 62–65 Mon. Suppl. 4: 

1.), hom. illeg. 

Tortula drummondii Mitt., Jour. Linn. Soc. 

London Bot. 1(Suppl.): 27. 1859. 

Mollia fragilis (Hook. & Wils.) Lindb., Musci 

Scand. 21. 1879. 

Plants coarse, rigid, dark green to yellowish or 

blackish-brown in dense tufts or deep sods, elongate. 

Stems (0.5–)1–5 cm high, leaves distantly disposed 

along the stem with shining leaf bases usually apparent, 

central strand absent, densely and visibly radiculose the 

entire stem length, rarely somewhat tomentose when 

very small. Stem leaves firm, coarse, rigid, uniform in 

size and shape along the stem, not appreciably larger at 

the stem apex, lowermost leaves mostly without leaf 

tips, leaves typically fragile only at base of deciduous 

apical subula, stiffly erect-appressed, incurved (or stemapical 

leaves somewhat twisted around the stem when 

dry); erect and weakly spreading when moist, narrowly 

lanceolate to linear, narrowing gradually above the 

proximal region until gradually or quickly contracting 

into a narrower propaguloid region up to 1/2 the distal 

leaf length; broadly to narrowly concave, keeled toward 

the apex in non-propaguloid leaves, 4–6 mm long; base 

broadly oblong; margins plane to erect, not to 

somewhat undulate; apex narrowly acute, those of nonmodified 

leaves shallowly channeled to subnaviculate, 

not cucullate, those of modified leaves long-subulate, 

fragile, the subula obtusely thickened and multi-stratose 

distally and often broken off, youngest leaves arising

from the stem apex fully propagulose and subulate; 

costa short-excurrent as a smooth or weakly denticulate 

mucro of 1–5 rhomboidal cells, in propaguloid leaves, 

internal cells in section appearing undifferentiated in the 

subula; adaxial surface covered by quadrate to shortrectangular, 

papillose laminal cells throughout the leaf 

except in the propaguloid leaf portion where a narrow 

medial groove is covered by elongate (8:1) smooth 

cells; cross section lunate and broad, adaxial epidermal 

cells may be interrupted in the center, exposing the 

adaxial stereid layer by about two cells in width, adaxial 

and abaxial stereid bands present, guide cells present in 

one layer, in propaguloid leaves appearing 

undifferentiated in the subula; proximal laminal cells 

rather abruptly differentiated from distal cells, usually 

lacking a zone of cells intermediate in color, cell size, 

and papillosity, occasionally with an intermediate zone 

of thick-walled somewhat porose cells, yellow-hyaline, 

elongate, firm to laxly thin-walled, smooth; marginal 

angle of differentiated proximal cells steep, with a 

narrow line of firm to thin-walled cells extending a 

short way up the lamina or to mid-leaf; distal laminal 

cells frequently bistratose on one or both laminae beside 

the costa, or equally extending toward the margins in 

various degrees, and becoming completely bistratose 

and ultimately undifferentiated in the, deciduous leaf 

tips, rounded-quadrate or hexagonal, (9–)10–12(–13) 

µm wide, strongly papillose, areolation more or less 

distinct; marginal cells in median leaf region 

undifferentiated, papillose-crenulate, along the distal 

median margin and into the subula differentiated as one 

or two rows of smooth, thick-walled, short-rectangular, 

(4–3:1), epapillose, yellowish cells leading into the 

subulate region and extending to the apex distinctly 

visible in at least some leaves, sometimes present only 

in young leaves on mature stems. Asexual 

reproduction: modification for asexual reproduction by 

deciduous, subulate, rigid, smooth-sided, distally 

thickened apices of all leaves. Sexual condition 

dioicous, seldom fruiting. Perigonia not seen. 

Perichaetial leaves setaceous, fragile, with propaguloid 

modifications as in stem leaves. Seta red-orange below, 

paler above, 1.5–2 cm long, occasionally two per 

perichaetium. Capsule 1.8–3 mm long; annulus none or 

of 1–2 rows of weakly vesiculose cells, operculum ca. 

1.5 mm long; peristome teeth long, spirally wound 

together, twisted 1, 2 or 3 times. Calyptra cucullate. 

Spores 8–10 µm, nearly smooth. 

Spores mature in summer (June). 

A common moss of northern regions on 

various substrates probably desiccated at least during 

some of the year, on acid or calcareous rock, in seepage, 

and in crevices and ledges of cliffs in maritime 

situations, dry prairies, on logs or peaty humus in 

springs, fens, Thuja swamps, on the open sands of lake 

shores, in the north or in mountains, in tundra, alpine 

and willow bogs, frost boils, margins of solifluction 

lobes, snow patch margins; elevation sea level to 3600 

m; Greenland; Alta, B.C., Lab., Man., N.B., Nfld., 

N.W.T., N.S., Ont., Que., Yukon; Alaska, Colo., Iowa, 

27 

Mich., Minn., Mont., [N. Car.], N.J., N.Y., N.Dak., 

Oreg., S.Dak., Tenn., Vt., Wash., Wyo.; Europe, Asia, 

Africa, Australasia in New Zealand; Antarctica; 

"reaching Nevada in the western mountains. Iceland; 

Svalbard, widely distributed in the mountains of 

Europe; through the Himalayan chain to China; across 

northern and Arctic Asiatic USSR; Japan" (Steere 

1978). 


DUKE, FLAS, MICH, MO, UBC. 

A North Temperate, boreal and Arctic species 

whose southern range hovers around the Canadian- 

United States border, but extending south in the 

mountains. 

Tortella fragilis is characteristic of the high 

Arctic. It is the only representative of the genus in 

Arctic Europe (Arctic Scandinavia and Spitsbergen) 

according to Crundwell and Nyholm (1964). In Arctic 

Asia and North America it is joined by Tortella tortuosa 

var. arctica, and by var. tortuosa throughout its range. 

Numerous collections from the Northwest Territories 

were found mixed with T. inclinata var. densa. 

In Europe, the occurrence of Tortella fragilis, 

rather like T. tortuosa, is concentrated in the Alps and 

their foothills (Braunmiller et al. 1971). Stations in the 

north, such as southern Sweden, may be the result of 

dispersal by apical fragmentation. Both Braunmiller et 

al. (1971) and Düll and Meinunger (1989) map the 

concentration of this species in Germany in the extreme 

south in the highlands near the Swiss-Austrian border 

and with few stations elswhere in that country. 

Tortella fragilis has a bipolar distribution. 

Specimens examined from the Southern Hemisphere 

(New Zealand, Tierra del Fuego, South Africa) show the 

same range of stature and variation as North American 

plants of this species. Hyvönen (1991) recently 

described the species from southern South America in 

Tierra del Fuego from sea level to 610 m on "bare soil 

and schistose gravel." The species has additionally been 

reported in the Southern Hemisphere: New Zealand 

(Fife 1984; Vitt 1974) and South Africa (Magill 1981), 

(South African specimen J van Rooy 3127BB, BUF). I 

have also seen specimens of Tortella fragilis at BUF 

collected from Antarctica (R. I. Lewis-Smith 7674B; 

9277D). 

Hyvönen (1991) discussed specimens from the 

Antarctic, originally described as Sarconeurum 

tortelloides S.W.Greene, as T. fragilis. However, my 

examination of this material proved it to be Tortella 

alpicola and not T. fragilis. Tortella alpicola always has 

a stem central strand. This character is absent in T. 

fragilis. Tortella fragilis has been reported for Hawaii 

(as var. tortelloides (S. W. Greene) Zand. & Hoe 

(Zander & Hoe 1979). The Hawaiian material is also T. 

alpicola, and so T. fragilis is excluded from that Pacific 

island. The only species of Tortella occurring there, 

other than T. alpicola, is T. humilis (Hoe 1974). 

In northern regions, Tortella fragilis is 

frequently found with Distichium capillaceum (Hedw.) 

B.S.G, which also has long, subulate leaves, but which

has no proximal cells differentiated higher on the 

margins, with leaves distichous on the stem, and short 

peristome teeth that are not elongated and twisted. 

The long, subulate, rigid, usually fragile leaves 

with propaguloid tips that are not or only slightly 

contorted when dry are distinctive of Tortella fragilis. 

In ambiguous cases, the bistratose cross section of the 

distal leaves is diagnostic, but usually all that is 

necessary is to demonstrate the smooth, elongate, 

thicker-walled cells along the leaf margin near the apex, 

a characteristic trait illustrated by Abramova et al. 

(1961: 296). That such a border is somewhat discernible 

in the deciduous apices of some leaves of Tortella 

rigens is one indication of that species possibly sharing 

genes with T. fragilis (see discussion under T. rigens). 

It is possible that the thick-walled marginal 

cells which occur in the distal leaf regions and in the 

propagules, rather than continuously along the margins 

such as might be expected from simply an extension of 

the proximal marginal cells higher up the leaf, 

contributes to the efficiency by which the plant 

maintains the structural integrity of the propagule when 

it detaches. Regions of the leaf that have no elongate 

marginal cells appear to have no particular diaspore 

function, and rigidity may not be necessary. In fact, the 

absence of strong marginal cells contributes to the 

detachability of the rigid diaspore, hence the middle 

third of the leaves has neither proximal nor distal-leaf 

marginal cells. No specimens of Tortella fragilis were 

found without at least some leaves displaying 

differentiated distal marginal cells. 

Note that some robust leaves of Tortella 

inclinata var. densa have elongate, smooth-walled 

laminal cells, but more frequently in the proximal part 

of the leaf and without evident correlation with a 

diaspore. 

In Tortella fragilis the leaf cell size is not 

necessarily uniform throughout the leaf: while the 

quadrate cell size may be 7–10 µm in the medial region 

of the lamina, it may reach 14 µm in the deciduous 

apical leaf region. 

A greater range of variability in the leaves of 

Tortella fragilis was seen than is usually described: the 

leaves may be stiff and merely arcuate in some plants, 

in others variously to strongly contorted-circinate, but 

not crisped. Such plants may be difficult to distinguish 

from Tortella tortuosa var. fragilifolia, which, however, 

has a fragile leaf lamina in addition to fragile leaf 

apices. 

In other collections there is a distinctive 

dimorphism in the leaves, especially in innovating 

branches, with strongly crisped-contorted leaves without 

propaguloid subulae but instead short, denticulate sharp 

mucros subtending typical stiff, subulate-propaguloid 

leaves toward the shoot apex. 

Occasionally a specimen will occur that does 

not develop the deciduous leaf apices; the leaves are 

merely partially bistratose near the costa and hence 

identification is problematic (usually identified in 

herbarium specimens as Tortella tortuosa). The overall 

28 

rigidity of the leaves on the stem, wet or dry and the 

elongate cells on the leaf margins are, however, 

definitive. Such plants, especially in mountain areas in 

the United States and northern North America, rarely 

exhibit some leaves with rather large cells (to 14 µm) in 

the apical portion of the leaves and have been identified 

as T. rigens (see discussion below). 

In Quebec, the species is often represented by 

tiny, depauperate plants with short, ovate-lanceolate 

leaves, yet with the distinctive distal border. In the most 

robust stems of these plants the cylindrical leaf-apical 

propagulum emerging from the stem tip in the youngest 

leaves helps identification (see illustration). As one 

might expect, leaves from these specimens are only 

partially bistratose beside the costa. Specimens such as 

these are usually identified in herbarium specimens as 

Tortella tortuosa. 

In specimens of Tortella fragilis where the 

propaguloid apices are not very well developed, the 

leaves tend to be twisted instead of stiff. One specimen 

(Quebec, Macoun, Aug. 26, 1883, CANM) had long, 

subulate leaves with distal marginal borders but with the 

propagula restricted to only some leaves and then only 

near the extreme leaf tips, expressed as a short, round 

section below the mucro, hence generally fewer leaf 

apices were missing. Occasionally (Alberta, Crum & 

Schofield 4609, UBC) there is no propagulum 

development at all and one must be guided by the clear, 

smooth-celled (rather than papillose-crenulated) distal 

margins, the bistratose area near the costa in section, 

and the erect, non-undulate leaves. 

Younger leaves and the leaves of young plants 

of Tortella fragilis may be tipped with a multicellular (5 

or 6 cell) mucro much like that of T. tortuosa, but the 

bistratose margins and the elongate clear cells on the 

leaf margin (these cells often absent in mature leaves) in 

addition to fragile, propaguloid leaves are characteristic. 

The leaf apices of T. tortuosa have very long mucros, 

mostly more than 5 cells in length. 

Quadrate, papillose epidermal cells across the 

adaxial portion of the costa occur throughout the leaf of 

Tortella fragilis except in the proximal cell region and 

in the propaguloid-setaceous part of the leaf where the 

costa is often exposed in a continuous groove of 

elongate cells. This character is useful in distinguishing 

the species from T. rigens, which has a continuous band 

of smooth, elongate cells on the adaxial surface of the 

costa throughout the length of the leaf. 

The abaxial dorsal stereid band is without 

quadrate, papillose cells throughout the length of the 

leaf. Transverse sections throughout the subulate 

propagulum in the distal leaf portion reveals a reduction 

in the the different layers of the leaf as well as their 

surface characteristics, and yet, excepting for the guide 

cell layer, they can still be somewhat differentiated in 

various degrees by the smoothness of the surficial cell 

walls: the marginal cells can be identified by their 

smooth marginal surface, the stereid cells, whether of 

the abaxial or adaxial band, can also be identified by 

their smoothness in at least one or two cells. The

laminal cells can be identified by the papillosity of the 

cell surfaces. 

In Tortella alpicola, the lamina is never 

bistratose except juxtacostally. The lamina is abruptly 

rather than gradually contracted into a more readilly 

deciduous propagulum. While the lamina of T. alpicola 

is unistratose, only the costa becomes undifferentiated 

in section, quadrate papillose cells clothe the dorsal 

surface of the costa in the apex of leaves at the distal 

end of the stem. Quadrate papillose cells at the back of 

the costa at the leaf tip of plants of T. nitida have been 

distinguished by a varietal name (var. irrigata (Winter) 

Zand., see discussion below). 

Tortella alpicola is further distinguished by the 

stem central strand, large leaf cells, absence of 

tomentum and sterile stems (see discussion above). 

North American specimens identified as 

Tortella rigens by the late F. J. Hermann were 

somewhat odd expressions of T. fragilis in wet tundra 

(see discussion under T. rigens). Specimens of other 

Tortella species with stiff, unbroken, subulate leaves at 

the stem tip often resemble T. fragilis. Usually these are 

perichaetial leaves at the stem tips of T. tortuosa, T. 

inclinata var. inclinata or T. alpicola. Perichaetial 

leaves of these species often show differentiated 

elongate smooth cells on the margin as in T. fragilis, but 

they are not modified into propagula at the leaf tips, nor 

are they bistratose. 

A reduced form of Tortella fragilis seen from 

New Zealand approximated T. alpicola. However, 

elongate, smooth marginal cells on propaguloid leaves 

could be demonstrated on a few of the leaves, at least 

some of the stems were rufous-tomentose, and the 

proximal cells were yellow (not hyaline). When stems 

are 1 cm or less in length, the surest way to differentiate 

T. alpicola is to examine the primordial leaves at the top 

of the stem: in T. alpicola they will be rough with 

papillae, with regular 'barrel-shaped' caducous 

segments; in T. fragilis they will have straight, clean 

edges bordered by elongate, smooth cells, and will 

comprise a single unit. 

Specimens of Tortella rigens from the Baltic 

region, where it was hitherto been considered endemic, 

were examined and were found to share many of the 

characteristics of T. fragilis, except that the leaves did 

not fully develop the distinctive subulate propaguloid 

leaf apices. Although much more contorted than typical 

T. fragilis, the leaves of T. rigens are, as the epithet 

suggests, more rigid than other species in the genus. 

Specimens of T. rigens examined with large leaves also 

had a border of elongated, thicker-walled, and smoother 

marginal cells in the distal third of the leaf. Dissection 

of the larger, more developed leaves in the collections 

revealed them to have a cross section similar to that of 

T. fragilis: bistratose juxtacostally, but the distal laminal 

cells were much larger and young leaves at the stem tips 

showed immature leaves without propaguloid 

modifications. Although T. fragilis may develop stems 

that are to one centimeter tall without losing its 

principle characteristics, small T. rigens plants appear to 

29 

lose most of their T. fragilis-like features and 

resemble more and more a different kind of plant 

altogether. However, even in populations of very small 

plants of T. rigens, larger leaves may be extracted that 

show the somewhat elongate distal marginal cells that 

indicate its affinity to T. fragilis. These leaves, on 

sectioning, will be bistratose next to the costa, 

sometimes imperfectly so (e.g. on only one lamina). 

Variation in the leaf-shape is extensive among different 

stems in the same collection, such that leaves that 

appear consistent with the description are found, but 

there are always those stems that bear T. fragilis-like 

modifications, such as the abrupt acumination or 

constriction at the base of a subuloid distal leaf region 

(to 1/3 the leaf length), that is similar to T. fragilis. 

These plants are strikingly different in low 

magnification: their true affinities are most apparent on 

examination of the cross-section of the leaf: T. fragilis 

will be strongly bistratose with quadrate, papillose cells 

on the adaxial surface of the costa in the median-leaf 

region, T. rigens only juxtacostally bistratose with a 

groove of elongate, smooth cells on the adaxial surface 

of the costa throughout. 

The occasional specimen of Tortella fragilis in 

North America has been ascribed to T. nitida perhaps 

when all or most of the leaf apices of a specimen are 

absent (e.g. Alberta, C. D. Bird 5514, CANM, with a 

query). Tortella nitida has a strong stem central strand, 

poorly differentiated proximal cells except along the 

proximal margins of the leaf, its leaf is fragile 

throughout the lamina, rather than with a deciduous leaf 

apex, nor in section is the lamina anywhere bistratose. It 

is excluded from North America (see discussions above 

under T. tortuosa var. fragilifolia, and below under T. 

nitida.) 

6. TORTELLA INCLINATA Plates 8–9 

Tortella inclinata (R. Hedw.) Limpr., Laubm. Deutschl. 

1: 602. 1888. 

Barbula inclinata (R. Hedw.) Schwaegr., Spec. 

Musc. Suppl. 1(1): 131. 1811. 

Tortula inclinata R. Hedw., Beitr. Naturk. 1: 

123. 1805. 

Barbula fragilis ssp. inclinata (R. Hedw.) 

Boul., Muscin. France 422. 1884, 

nom. illeg. prior. ut spec. 

Mollia inclinata (R. Hedw.) Lindb. in 

Braithw., Brit. Moss Fl. 1: 251. 1887. 

Barbula inclinatula C. M. & Kindb. in Macoun 

& Kindb., Cat. Canad. Pl. 6: 50. 1892. 

Trichostomum inclinatum (R. Hedw.) Dix., 

Stud. Handb. Brit. Moss. 219. 1896. 

Tortella inclinatula (C. Müll. & Kindb. in 

Mac. & Kindb.) Broth., Engler & 

Prantl, Musci ed.1. 397. 1909. 

Plants in dense, low tufts or turves, dull yellow-green to 

yellowish brown above, brown below, elongate, not 

rosulate. Stems 0.5–1(–1.5) cm high, central strand

absent; few-branched, stem sparsely rufous radiculosetomentose 

at the base, essentially without tomentum 

above. Stem leaves somewhat soft or lax, loosely to 

more closely foliose, variably gradually somewhat 

larger and more crowded toward the stem tips or the 

reverse with the leaves somewhat larger below than 

above, loosely and irregularly curled to crisped when 

dry, erect-spreading to spreading when moist, oblonglanceolate, 

to lanceolate, sides generally parallel and 

scarcely tapered from the base, typically strongly 

keeled-concave or broadly channeled across the leaf 

adaxially, (1–)1.5–2 mm long, not undulate; base not 

differentiated or somewhat broader, elliptical; margins 

plane to erect, broadly incurved above, not undulate; 

apex typically cucullate, occasionally fistulose, obtuse 

to acute, occasionally abruptly narrowed and concaveacuminate; 

costa short-excurrent by 1–3 cells, cells of 

the adaxial surface cells entirely smooth and narrowly 

elongated, 8:1, throughout, adaxial epidermis absent, 

adaxial and abaxial stereid bands present, guide cells 

present in one row, in cross section nearly circular, 

without bistratose areas at juncture with lamina; 

occasionally abaxial surface roughened by projecting 

distal cell ends, these more dense at the leaf apex; 

proximal laminal cells abruptly differentiated from 

30 

distal cells in color, cell size, cell wall thickness and 

papillosity, proximal pale yellow, transparent, elongate, 

laxly very thin-walled and fragile, smooth; V-shaped 

marginal angle steep; marginal cells undifferentiated, 

sharply crenulate-papillose; distal laminal cells densely 

and sharply papillose, quadrate, 10–12 µm wide. 

Asexual reproduction: modifications for asexual 

reproduction none. Sexual condition: dioicous, seldom 

fruiting. Perigonia not seen. Perichaetial leaves with 

unfertilized archegonia hidden in the circumperichaetial 

stem leaves; perichaetial leaves of fertilized archegonia 

conspicuously differentiated, elongate, 2–5.5 mm long, 

erect from a broad, long and hyaline sheathing 

sometimes inflated base, not or somewhat curled at the 

tips when dry, rising above the crispate stem leaves 

below, narrowly lance-subulate with a long excurrent 

costa, in some leaves with an interrupted border of 

elongate, thick-walled smooth cells in the distal portion 

of the leaf. Seta orange, paler above, 1.5–2.7 cm long. 

Capsule 1.5–2 mm long, yellow-orange, red at the 

mouth; annular cells not vesiculose; operculum 1–2 mm 

long; peristome teeth long (to 0.7 mm), spirally wound 

more than once. Calyptra cucullate. Spores 9–12 µm, 

appearing smooth or with minute papillae. 

KEY TO THE VARIETIES OF TORTELLA INCLINATA 

1. Stems short, to 1 cm, not to few-branched, tomentum usually conspicuous; leaf cross section with adaxial and 

abaxial superficial walls the same width as the cross-walls; papillae not elevated, leaf cells 10–12 µm, stems yellow 

or green, rather pale, brown below, yellow in KOH, leaves irregularly twisted and incurved when dry; majority of 

leaf apices obtuse, strongly cucullate, mucros of 1–3 cells; in low sandy alluvium of beaches and streams . . . . 6a. 

Tortella inclinata var. inclinata 

1. Stems tall, to 2 cm, many-branched, tomentum hidden in leaf axils; leaf cross section with thick abaxial and 

adaxial superficial walls, but thin cross-walls; papillae appearing pedestaled on thick walls, leaf cells 7–10 µm, 

stems dull and often strongly orange-green, orange, or dull green above and brownish-black below, orange in KOH; 

leaves when dry somewhat uniformly twisted on the stem (funaliform), erect below; leaf apices cucullate, acute, to 

narrowly acuminate, mucros of 5–7 cells; in rock crevices, hillsides and upland alluvium . . . . 6b. Tortella inclinata 

var. densa 

6a. TORTELLA INCLINATA VAR. INCLINATA 

Plate 8 

Leaves yellow-green, incurved and irregularly curled 

when dry, appearing soft or lax; apices generally 

cucullate, distal laminal cells to 12 µm, not obscure; cell 

walls uniformly medium, thin-walled, with dense, 

unpedestaled, rather high, solid papillae in sharp relief. 

The variety is known to fruit in Alberta and 

British Columbia. 

This plant is associated with exposed 

calcareous silt, sand or other loosely consolidated 

substrates where it functions as a pioneer species. It 

frequently grows near bodies of inland fresh water, 

gravel bars, among pebbles on river banks, mud flats, 

sand dunes on the Great Lakes. It also is found in gravel 

pits near bogs and sandy clearings in mixed deciduous 

woods or spruce-pine groves, and highway ditches; 

elevation 200–1830 m; Alta., B.C., N.W.T., Ont., Que., 

Yukon; Alaska, Iowa, Mich., Minn., Mont., Ohio; 

reported from South America, Europe, Asia, Africa, 

Australia. "A widespread species in Eurasia" (Steere 

1978). 

Spores mature in spring (late May). 



In the course of the present study, many 

specimens in North American herbaria of what has 

hitherto been determined to be Tortella inclinata have 

been redetermined as two other taxa: T rigens and T. 

inclinata var. densa. The three taxa are very similar in 

their tubulose and frequently cucullate leaves, and the 

groove of elongate, smooth cells on the adaxial surface 

of the costa extending throughout the leaf. These three 

taxa are rather substrate specific and this was a crucial 

element in their differentiation during this study. In 

North America, Tortella inclinata var. inclinata appears 

to be restricted to "calcareous silt, usually in the flood 

zone of streams" (Steere (1978). The var. inclinata also 

enjoys larger sediments, such as the coarse sands along 

the Great Lakes beaches and rivers. It characterizes 

lower areas in flood zones than T. inclinata var. densa,

which also colonizes sandy soil, but in this case in the 

crevices of rocks in hilly stations—perhaps in areas 

flooded only when the waters are very high. 

Steere (1978) mentioned the occurrence of 

Tortella inclinata in Ontario based on a citation of 

Williams (1968). Her habitat designation, "bare open 

limestone flats in drier locations than Tortella tortuosa" 

suggests strongly that her specimen was T. rigens (see 

discussion below). Although in the alvar regions of 

southern Sweden, on the islands of Öland and Gotland, 

T. inclinata var. inclinata may grow on limestone, so far 

this habitat has not produced specimens in North 

America, hence it has not been found growing mixed 

with T. rigens. That T. inclinata is found growing in 

mixed populations with T. fragilis in Alberta (T. 

inclinata being fertile), and frequently so in the Nahanni 

River area (based on numerous specimens at NY), is 

suggestive when one considers the possible hybrid 

nature of T. rigens. 

Steere (1978) discussed three disjunct areas of 

distribution in North America for Tortella inclinata, and 

these were mapped and discussed by Miller (1976). The 

specimens mapped follow roughly the southern limits of 

maximum Wisconsinan glaciation. Specimens seen by 

Miller were examined in this study from CANM, MICH 

and NY, (but not FH or US), and some of them were 

redetermined to be either T. rigens or T. inclinata var. 

densa. Fossil specimens of T. inclinata, noted by Miller 

(1976) from the Two Creeks Forest Bed peats of 

Wisconsin, dated 12,000 to 11,850 years old, and from 

"sediments deposited during final ice withdrawal...and 

commencement of postglacial (Holocene) time about 

10,000 years ago" (Miller 1976) in northwestern New 

York State (Miller 1973) should be re-examined to see 

if they are not T. inclinata var. densa. The northernmost 

station noted on Miller's (1976) map from Alaska in the 

Brooks Range, for example, is T. inclinata var. densa 

(see discussion under that taxon). 

Braunmiller et al. (1971) report that Tortella 

inclinata var. inclinata is "ecologically rather sharply 

delimited on open, unprotected [unshade], strongly 

insolated and quickly desiccated calcareous gravelly 

soil." They indicate it is a colonizer and dominant 

species in a pioneer-association: "zu der auch andere 

Tortella-Arten sowie die Flechten des Bunten 

Ardflechtenvereins...gehören" (see Reimers 1951). It is 

not found on humus. It is found on calcareous gravels of 

rivers in the Alps and other calcareous uplands. They 

state that a large part of the stations are anthropogenic, 

colonizing gravel pits, excavations, ballust on railway 

embankments and similar substrates. They emphasize 

that like T. fragilis the center of distribution is in the 

Alps and its foothills. Many stations are along the banks 

of the Danube and Rhein rivers. Stations in the German 

north are suggested to be adventive. Interestingly, these 

authors are uncertain about the altitudinal boundary of 

the species in the Alps and they suggest that in higher 

altitudes, stations reported as T. inclinata may be T. 

densa instead. The present treatment resolves these 

31 

issues by presenting the taxa as a single variable 

species. 

Tortella inclinata var. inclinata has an 

enormous area of distribution throughout the temperate 

latitudes according to information assembled by Düll 

(1984). He restricted the variety densa (as T. densa) to 

Europe, in boreal-montane situations. 

Braunmiller et al. (1971) mapped for central 

Europe the distribution of var. inclinata as nearly 

uniform from the Alps in the south to the margins of the 

Lowland countries to the north. The var. densa was 

closer to the distribution of Tortella fragilis: restricted 

to the mountanous highlands in the south. Düll and 

Meinunger (1989) give nearly identical distributions 

(but var. inclinata is not mapped). 

In Europe the ranges of Tortella flavovirens 

and T. inclinata s.l. are so close as to appear to overlap, 

whereas in North America the ranges are disjunct from 

the coast of North Carolina to the Great Lakes region. If 

T. inclinata cannot tolerate the salty conditions of the 

substrates that support populations of T. flavovirens, 

then it is possible the ranges are contiguous and do not 

really overlap. However, Rubers (1973) reported two 

collections in the Netherlands of T. inclinata (possibly 

from a single locality) in the dunes by Overveen. 

Extra-North American reports of this species in 

South America, Asia, Africa and Australia seem 

problematic to some extent when compared to the 

restriction of Tortella inclinata in North America to 

terrain at the Wisconsinan glacial boundary. One would 

anticipate similar restrictions and distributional histories 

in other areas of the world. 

The epithet "inclinata" refers to a curving 

capsule, however at least some capsules of every 

species in the genus in the flora area also have inclined 

capsules. Some of the infrequent North American 

sporophyte material shows curving capsules atop a seta 

that curves or is geniculate at or near the juncture with 

the capsule: the two characters in combination show a 

striking departure from an erect posture. Haring (1938) 

emphasized the "capsule narrowly cylindrical, nearly 

straight, suberect" for American populations, possibly to 

distinguish them from European specimens. The present 

study, however, found variation similar to European 

collections: examples of this species among W. P. 

Schimper's duplicates at BUF have capsules suberect, 

somewhat curved to strongly curved in single 

populations. 

As Crum and Anderson (1981) indicated, 

Tortella inclinata var. inclinata in the Great Lakes 

region is rather readily separated from T. tortuosa and 

T. fragilis, which occupy similar habitats on partially 

stabilized, calcareous dune sands in the Great Lakes 

region, by its shorter leaves that are incurved and 

moderately contorted when dry with, broad, cucullate 

tips. Tortella humilis, which might possibly be confused 

with these species, has a distinct stem central strand. 

The absence of any quadrate, papillose adaxial 

epidermal cells in cross sections of the distal or middle 

part of the leaf will differentiate specimens of T.

inclinata var. inclinata from the three species just 

named. Specimens with an adaxial groove of elongate, 

smooth cells on the surface of the costa throughout the 

leaf and with leaves that are not cucullate but only acute 

with a short mucro, may be T. inclinata var. densa 

instead. Thickened abaxial and adaxial superficial cell 

walls seen in lamina cross-section will also indicate that 

variety. 

No small, terminal propaguliferous shoots, as 

mentioned for English plants of Tortella inclinata var. 

inclinata by Smith (1978), were observed or recognized 

in North American material. Any terminal filaments 

proved to be associated with archegonia, and comprised 

the distinctive setaceous apices of perichaetial leaves. 

Lack of such shoots in American material may indicate 

that European plants could be a somewhat different 

taxon. Nyholm (1989) for example, stated that in 

Scandinavia: "Short fragile yellow-green apical shoots 

are characteristic for this species." Nyholm indicated 

that such shoots are important in differentiating the 

species from T. densa and T. rigens. It is possible that 

these shoots are simply elongations of the main stem 

axis and are no more deciduous than the whorls of the 

var. densa which disarticulate readily upon dissection. 

Tortella inclinata s.l. has up to the present been 

the only taxon of the genus in North America to have no 

epidermal layer on the costa anywhere on the leaf. Only 

the narrow, elongated, smooth stereid cells are exposed 

on both surfaces of the costa. In Europe, this species 

shares this characteristic with T. rigens, T. densa and T. 

limosella (Stirt.) Rich. & Wall. (Smith 1978). Tortella 

limosella is separated from T. inclinata in the British 

flora in part by having distal leaf cells 8–12 µm wide, 

whereas cells of the latter species are 8–10 µm. North 

American T. inclinata var. inclinata regularly attains the 

larger cell dimensions, however. In addition to larger 

cells, T. limosella has thicker cell walls and these are 

unipapillose. It was collected once (1906) from dense 

patches from a single population on the seacoast in 

western Scotland (Smith 1978) and maybe synonymous 

with a more inclusive concept of the variability of T. 

inclinata s.l. 

Persson and Weber (1958), when they first 

reported Tortella inclinata from North America, 

expressed doubts that the previously described North 

American Barbula inclinatula C. M. & Kindb. was 

distinct from it. Dixon (1924) reported a different 

ecological and distributional regime for European T. 

inclinata: "calcareous banks and rocks, usually in 

mountainous districts." North American material, for 

example, never grows on rocks, but T. inclinata var. 

densa does. 

Crundwell and Nyholm (1962) stated that the 

Müller and Kindberg type of Tortella inclinatula was 

identical with European T. inclinata as were two 

additonal specimens, one from Ontario and from British 

Columbia. Steere and Scotter (1978) still, however, 

expressed some hesitation based on a sense that the 

perichaetial leaves of the type of Barbula inclinatula 

were different from European material. "Macoun's 190 

32 

in the National Museum of Canada (CANM) has 

perichaetial leaves that are conspicuously larger and 

more hyaline than those on European material 

examined" (Steere & Scotter 1978). 

Although Macoun 190 was not available in the 

loan received from CANM, three other North American 

fruiting specimens were examined: Alberta, John 

Macoun, June 27, 1904 (CANM); B.C., Lake of 

Hanging Glaciers, Wilmer, July 21, 1928, MacFadden 

4252 (NY); and also what is apparently an isotype of 

Tortella inclinatula: B.C., Illicillewaet near Revelstoke, 

May 27, 1890, Macoun 188 (NY). These were 

compared with fruiting specimens of T. inclinata from 

Europe (Strasbourg in western France, Switzerland and 

Saxony in eastern Germany) among the W. P. Schimper 

duplicates at BUF. No significant transoceanic 

differences between the perichaetial leaves of these 

plants were found. None of the American perichaetial 

leaves exceeded those of the European ones excepting 

one specimen (Macoun) in which the leaves were 

perhaps 0.4 mm longer, an unimpressive magnitude of 

difference. The laminae of the perichaetial leaves appear 

white-hyaline in all specimens even at low 

magnification. If the American material is different 

from European specimens, the perichaetial leaves are 

not decisive. 

The strikingly differentiated perichaetial leaves 

of Tortella inclinata var. inclinata only occur in 

association with fertilized archegonia. When adjacent 

perichaetiate unfertilized stems in specimens with 

fertilized stems were examined, the perichaetial leaves 

were less distinct from the stem leaves—for example, 

they were not longer. Numerous specimens of 

unfertilized perichaetiate plants of the var. densa from 

North America showed much variability of the 

perichaetial leaves but nothing to distinguish them from 

those of unfertilized, perichaetiate var. inclinata. 

In Tortella inclinata var. inclinata the 

setaceous perichaetial leaves may be conspicuous on 

dry stems where they rise more stiffly above the curled 

stem leaves—as is true of those of T. alpicola, T. 

fragilis and T. tortuosa. Such perichaetiate plants, 

though uncommon, are easy to confuse with sterile or 

fertile T. fragilis, but the setaceous leaves of that species 

have propaguloid modifications in the apex whether 

barren or fertile. Fertile T. inclinata var. inclinata has 

also been confused with T. tortuosa but in the latter 

species the lower leaves are also long-mucronate. The 

lower leaves in T. inclinata var. inclinata are all 

relatively blunt or cucullate. 

The presence in some perichaetial leaves (with 

fertilized archegonia) of an interrupted border of 

elongate, thick-walled smooth cells in the distal portion 

of the leaf is reminiscent of perichaetial leaves of 

Tortella tortuosa where this character is more 

pronounced. This characteristic is very important in the 

subulate-propaguloid leaves of both sterile and fertile 

(perichaetial) leaves of T. fragilis and seems to 

contribute to the rigidity of the leaves that possess these 

borders in all three species.

In her key to the species of Tortella in North 

America north of Mexico, Haring separated T. inclinata 

from T. humilis and T. flavovirens by cucullate leaf 

apices and the leaves "very concave." Neither Macoun 

and Kindberg (1892), or Haring (1938) compared T. 

inclinatula with T. inclinata of Europe, as noted by 

Persson and Weber (1958). The cucullate leaves of T. 

inclinata in Europe (Sweden) are reported as critical to 

distinguish that species from T. densa and T. rigens, 

which are "narrowed into a sharp point" (Nyholm 1989). 

Cucullate leaf apices characterize the var. inclinata 

presented here. 

An extreme of this kind of apex is shown in a 

specimen from Minnesota that has leaves broader than 

those of typical T. inclinata and apices that are 

completely cucullate. Latzel (1934, p. 174) described a 

variety of T. inclinata from Hungary as var. mutica: 

"Folia etiam humecta partim fere fistulosa, obtusa vel 

obtusiuscula, costa ante apicem vel in apice 

evanescens." The Minnesota plants possess fistulose 

leaves in the sense that fistulose implies a tube with one 

end closed. The description suits this specimen, except 

the costae are both percurrent and excurrent as delicate 

mucros, many of which are eroded. Just as the 

acuminate-leaved plants have long-stems, this strongly 

obtuse-cucullate plant has short ones. An even better 

representation of the var. mutica is represented in a 

specimen from British Columbia that is better identified 

as T. inclinata var. densa (see below) at least by the 

thick cell walls and orange coloration. The var. mutica 

seems not worthy of recognition, merely representing a 

sporadic unimportant extreme of variation. 

The differentiated proximal cell region in 

Tortella inclinata is short among its congeners. Haring 

(1938) for example indicated it as 1/5–1/4 the leaf 

length, whereas it is longer (1/4–1/3) in both T. humilis 

and T. flavovirens. This short proximal cell region is 

displayed in at least some leaves of T. inclinata var. 

inclinata and var. densa. Other leaves on longer stems 

in T. inclinata occupy more of the leaf length, to 1/3, 

and in some leaves the region appears to extend quite 

high on the margin in all three closely related species. 

6b. TORTELLA INCLINATA VAR. DENSA Plate 9 

Tortella inclinata var. densa (Lor. & Mol.) Eckel, 

comb. nov. [note post publ, hom. superf.] 

Barbula inclinata var. densa Lor. & Mol. in 

Lorentz, Moosstud. 90. 1864, 

basionym. Type: Germany, Bayern, 

Oberbayern, Landkreis Garmisch- 

Partenkirchen: "Platt an der Zugspitz, 

6000–6500'; 10. Juli 1862, L. 

Molendo" (M—lectotype). 

Tortella inclinata fo. compacta ROll, 

Hedwigia 56: 142. 1915. 

Tortella densa (Lor. & Mol.) Crundw. & Nyh., 

Trans. Brit. Bryol. Soc. 4(2): 188. 1, 

a–c. 1962. 

33 

Tortella tortuosa fo. curta Alberts. 

Österplana hed ett alvarområde på 

kinnekulle. Upsala. 207. 1946. [name 

invalid: without Latin description. 

Albertson meant to use the form-name 

provisionally ("provisoriskt": 193.)] 

Plants in dense sods or cushions of high, erect stems, 

dull pale green with reddish highlights (coppery-sheen) 

in youngest leaves at stem apex, dull brownish-orange 

below or throughout, elongate, not rosulate. Stems (0.5– 

)1.5–2 cm high, central strand absent, longer stems with 

multiple branches, elongate lateral shoots with 

primordial or small reduced leaves occasionally present; 

distal stem with inconspicuous rufous radiculosetomentum 

hidden at the base of the leaf axils, this 

sometimes appearing absent in well-developed forms, 

large coarse rhizoids scattered along the stem, 

sometimes concentrated at the bases of whorls. Stem 

leaves somewhat stiff, closely and densely foliose, 

appearing imbricated, larger, broader and nearly 

uniformly erect at the base and smaller and narrower at 

the stem apex and more irregularly contorted there when 

dry, most middle leaves loosely curled in a counterclockwise 

direction when dry, or counter-clockwise 

below and clockwise above, giving a rope-like 

(funaliform) appearance to the habit, erect and 

spreading when moist, individual leaves as seen from 

above twisted clockwise, shortly triangular-lanceolate 

and tapered from the base, tubulose (margins broadly 

incurved), variously strongly keeled-concave or broadly 

channeled across the leaf above, (1–)1.5–2 mm long, 

not undulate; base shortly ovate or sharply ovate below 

a cincture at the juncture of distal laminal cells; 

margins crenulate-papillose, apex cucullate to concaveacute 

to acuminate; costa excurrent into a slender, sharp 

mucro of 5–7(–9) cells in acuminate leaves confluent 

with the lamina, covered on the adaxial surface by 

elongate, smooth-walled cells throughout the leaf 

length, shape in section circular distally to lunate in 

middle and proximal regions, the adaxial surface convex 

to concave in the median leaf region, especially when 

there are bistratose cells at junction of lamina with 

costa, adaxial epidermis absent, abaxial and adaxial 

stereids present, somewhat more massive than in T. 

inclinata, costa and adjacent lamina in cross section 

often bistratose in the middle region of the leaf; 

proximal laminal cells sharply differentiated, proximal 

angle steep, although filling only the proximal fifth or 

less of the leaf, the cells in some leaves tend to rise 

somewhat high on the margins, smooth, hyaline, 

somewhat thick-walled (not lax) near junction with 

distal laminal cells, marginal cells somewhat reduced in 

size, sharply crenulate with high papillae to smooth and 

crenulate by projecting cell walls, usually 

undifferentiated, occasionally on longer leaves with a 

marginal border of elongate (2–3:1), thinner-walled, 

smooth cells extending distally from the hyaline 

proximal cell region; distal laminal cells somewhat 

larger towards the apex, obscured by thick superficial

cell walls and massive pedestaled coroniform papillae, 

the papillae sometimes reduced and the lumen capped 

by a smooth, thick salient, 8–10 µm, in one collection 

with some cells reaching 14 µm. In cross section the 

abaxial and adaxial cell walls thicker than the crosswalls 

together with abaxial and adaxial papillae, both 

thicker than the height of the lumen itself. Asexual 

reproduction: sterile lateral shoots with minute or 

primordial leaves distant on the axis developing into 

foliose stems, possibly for vegetative dispersal. Sexual 

condition: dioicous. Perigonia in sessile buds at stem 

apices, inner perigonial bracts ovate and abruptly 

apiculate, scarcely longer than the antheridia, 0.5 mm 

long. Perichaetia as many as six on a stem, the stem 

elongating by multiple, successive, short subperichaetial 

innovations; outer perichaetial leaves undifferentiated 

and appearing more acute than stem leaves by closely 

infolded distal margins; inner perichaetial leaves 

variable, equal to or shorter than surrounding leaves, 

differentiated as narrow, abbreviated laminae ending in 

a long, subulate awn above an inflated, sheathing 

hyaline base. Sporophytes: no fruiting material seen. 

A species of temperate and boreal associations 

on sandy soil of rocky uplands or hills, rock slides, soil 

at the base of sandstone outcrops, cliff crevices, ledges 

and bases in mesic to moist stations, probably 

associated closely with old post-glacial alluvium in 

drainage basins of lakes and rivers; in Europe on Baltic 

alvars and in montane or alpine stations; in North 

America 934–1560 m., B.C., N.W.T., Alaska, Mich, 

Wisc.; Europe. 

Herbaria examined: BUF, CANM, NY, UBC. 

Of the five specimens of var. densa in 

Molendo's herbarium (M—comm. H. Hertel), one (see 

above) was selected as lectotype, being collected before 

the date of publication and fitting well the original 

authors' concept. It is beautifully represented by copious 

material giving both short stems (0.5 cm) and long (4 

cm), described on the label as var. "densa, foliis brevius 

acuminatis, statura compacta robustiore insignis." The 

Latin implies there are other variations that are more 

long-acuminate. The leaves on the specimen are 

intermediate-acuminate, without the long, narrow apices 

with long mucros of many specimens seen. It conforms 

perfectly with material from stations in northwestern 

North American, especially in the strongly concave 

adaxial surface of the costa in section with multistratose 

areas beside the costa (cf. illustration, Plate 9). The 

European specimen exhibits the extremely deep sods 

that the species often attains in the Alps, compared to 

shallower sods in the comparatively drier mountains of 

North America (Rocky Mountains). 

Specimens of Tortella inclinata having 

substrates not directly associated with alluvium may 

reveal var. densa which seems to be associated with 

older, stable habitats, rather than colonizing relatively 

more recent ones, such as does the var. inclinata. While 

European descriptions of T. inclinata include sandy soil 

associated with river banks, they also include walls and 

rocks (Demaret & Castagne 1964), stations not known 

34 

for T. inclinata var. inclinata in North America. Most 

North American specimens are identified as T. inclinata 

if bearing a cucullate apex, and T. tortuosa if an 

acuminate one. 

Braunmiller et al. (1971) discussed Tortella 

densa as a relatively new species after the publication 

by Crundwell and Nyholm (1962). After examination of 

a series of specimens, these authors had apparently 

decided it to be a good species, independently noting the 

homogeneous torsion of the leaves in well developed 

plants, reminiscent of that of Grimmia funalis Schimp. 

The authors associated the species with (calcareous) 

rock crevices, unlike the stations noted above in var. 

inclinata and they felt hindered in their generalities due 

to the lack of adequate herbarium specimens or their 

mistaken inclusion in the circumscription of other 

species. A station found in Bavaria by Paul (1943) is 

noteworthy for its occurrance on glacial moraine, "one 

of the richest stations for circumalpine plants in the 

foothills of the Alps...the appearance of the species 

there suggests a glacial foothills dispersal." 

The occurrence of this variety in the Alps of 

Europe leads one to expect it to occur in the Rocky 

Mountains in North America south of the Canadian 

border. North American stations, however, are as 

hitherto reported for Tortella inclinata var. inclinata, 

that is, within the southern and western limit of 

Wisconsinan glaciation, but not south of it. The 

specimens on which the North American occurrence of 

var. densa is here based perhaps fail to show the full 

range of characters possible, based on extremes 

examined in European collections mainly in the greater 

degree of acumination, but are very like the type 

specimen. Re-examination of specimens of T. tortuosa 

in the Rocky Mountains may ultimately reveal a more 

widespread distribution of T. inclinata var. densa on this 

continent. 

The single report for Tortella inclinata in the 

Arctic was a citation by Steere (1978) of a specimen 

reported from the Brooks Range in Alaska, collected by 

Louis Jordal (2050, CANM; see specimen citation 

above) and published by Miller (1976). The specimen 

was originally identified as Tortella tortuosa but was 

redetermined by Miller with the following note: 

"probably T. inclinata...but material not typical. Distal 

leaf cells small, ca. 10 µm; apices of even young leaves 

scarcely cucullate" (det. April 1975). Crum, in 1964, 

and Zander (1997) in a recent treatment of Tortella for 

Arctic North America, also examined the specimen and 

determined it to be that species. However, the stems are 

reddish-orange, the leaves are densely foliose, in cross 

section they have thickened superficial cell walls on 

both sides of the leaf section, forming pedestals on 

which the papillae are developed, and are redetermined 

here to be T. inclinata var. densa. Most of the leaves are 

acuminate, rather than cucullate, perhaps because the 

stems are richly perichaetiate. 

Even in some specimens with short stems (less 

than 0.5 mm), the very thick superficial walls in 

transverse section and acute apices and orange color

were diagnostic (Vitt 18784, citation below). The Jordal 

specimen is most like the discussion of Tortella densa 

(as T. tortuosa forma curta) by Albertson (1946). The 

leaves are narrowly acuminate, most only subcucullate, 

with marginal borders extending from the proximal 

region nearly to the apex. 

Another specimen from Arctic Alaska, cited by 

Steere (1978) as Tortella tortuosa and by Zander (1997) 

as T. inclinata from the Ambler River area (Brooks 

Range, Lewis 1768; see citation below), is also the var. 

densa by the same criteria. These plants have perigonia. 

The specimen includes stems with short, acute leaves 

with short mucros, mostly eroded, in dense erect tufts, 

with papillae with high salients. The leaf cross-section 

shows a strongly keeled leaf with the adaxial stereid 

layer exposed throughout the leaf. The label data 

indicate the collection came from a "dry calcareous 

ridge," and not from sandy riverine deposits typical of 

T. inclinata. The stems are coated with sandy particles 

and there are horizontal layers of what appear to be 

deposits made when the collections were in wet 

conditions, perhaps of high water from the Ambler 

River. 

As these two specimens comprise the only 

reports of Tortella inclinata from Arctic North America, 

the var. inclinata is considered here to be excluded from 

that region. 

Crundwell and Nyholm (1962) described 

Tortella densa as tall, to 4 cm. Limpricht (1890) 

described stems to 8 cm in deep sods. Perhaps the most 

telling description, with respect to North American 

populations, is of the leaves on the stem: "in well-grown 

shoots twisted slightly round the stem in a manner 

rather like that in Grimmia funalis." The leaves are 

hardly crisped excepting the youngest at the stem apex. 

The species is "widely distributed in western, northern 

and central Europe. Although only the Norwegian 

locality is north of the Arctic Circle all the central 

European ones are from mountains, many from high 

altitudes." Limpricht (1890) described it as a very 

distinctive high-alpine form growing above 2,000 m in 

Europe. 

Crundwell and Nyholm (1962) reported that 

"We have seen rather much North American material of 

Tortella, and if T. densa occurs there it must certainly 

be rare." Note that all of the specimens cited above were 

collected after their 1962 article. 

In North America, Tortella inclinata var. 

inclinata is usually said to be confined to a rather 

narrowly defined substrate: essentially rather recently, 

hence partially, stabilized alluvium associated with 

beaches in the Great Lakes and (perhaps recently) 

abandoned river channels (Haring 1938; Crum & 

Anderson 1981). "Recent" here refers to Holocene, or 

post-glacial sediments. Specimens hitherto labeled as T. 

inclinata in North American herbaria, however, also 

indicate their occurrence on limestone pavements 

(alvar) not associated with alluvial sediments. These 

particular specimens proved to be T. rigens (see 

discussion below). 

35 

Another suite of specimens, mostly recently 

collected in the less accessible northern areas of Alaska, 

British Columbia and the Northwest Territories, 

revealed a third set of substrates for var. densa as rock 

crevices in more elevated stations that those of 

previously reported stations of Tortella inclinata. Even 

though among rock, these stems were embedded in 

sand-sized particles, indicating some association with 

localized watersheds: riverine environments with a 

historical character, perhaps with once higher water 

levels or with more recent stream channel abandonment, 

in post-glacial fens. Tortella inclinata was first 

discovered in North America in abandoned river beds 

"along old channels of the Illicillewaet" in British 

Columbia (Macoun and Kindberg 1892)—a perhaps 

more recently abandoned situation, as might be dunes 

correlated with relatively recent historic reductions in 

lake water levels. Such upland stations tend to suggest a 

refugial character. 

Steere (1976: 54) in his discussion on the 

ecological relationship of light to bryophytes in Arctic 

Alaska reported that: "The distribution of tufts of mats 

of bryophytes on a vertical rock face seems more to 

reflect the presence of water drainage channels than a 

response to the intensity of light. In Arctic Alaska, a few 

species of bryophytes are generally restricted in their 

habitat to deep rock crevices. However, this preference 

appears to indicate a requirement for the higher 

humidity that occurs out of the wind and direct sun, and 

not a reaction to light intensity, per se....My evidence 

for this conclusion is that in more temperate climates 

with normally higher atmospheric humidity, these very 

same species may grow in the open on the trunks of 

trees and on the sides of rocks, not in full sunlight, but 

certainly not restricted to crevices for protection against 

drying winds." 

Tortella densa "In the British Isles and in 

Sweden...is a plant of rock crevices and of rocks with a 

thin covering of soil. We have never seen it on sand 

dunes. It frequently grows in the same habitat as T. 

tortuosa" (Crundwell & Nyholm 1962). In Scandinavia 

it grows "up to 4 cm high on calcareous rocks and soil" 

(Nyholm 1989). "...In the alvar vegetation of Öland and 

Gotland, where the underlying limestone is covered by 

only a very thin layer of soil [T. densa and T. inclinata] 

often occur together, associated with T. tortuosa, T. 

fragilis and T. rigens. These five species frequently 

grow in mixed tufts, without intermediates" (Crundwell 

& Nyholm 1962). Such a community has so far not been 

reported for the alvars of the Great Lakes region (see 

Catling & Brownell 1995) but may be anticipated. One 

may also wonder whether these taxa might not be 

present in the extensive limestone plateaus or 

geologically classic karst areas of the Adriatic. 

These species, but especially Tortella inclinata, 

T. densa and T. rigens apparently have close 

correlations with different substrates. When the latter 

three are said to grow in alvar, it must be borne in mind 

that there are numerous kinds of alvar vegetation, 

depending on the depth of the soil and most particularly

whether it is a wet or dry alvar (classification by Catling 

& Brownell 1995). Label data is usually insufficient 

from alvar regions to specify species preferences for 

specific ecological regimes. 

The most striking characteristic of the western 

North American specimens recently collected from 

rocky substrates was their tall, densely foliose, 

numerously branched stems in regimented tufts, unlike 

T. inclinata whose stems are usually smaller and 

indistinct, little branched, the leaves more chaotically 

twisted around the stem and half buried in the debris. 

The new specimens are noted by their darker (fuscous to 

black below) or more highly colored (orange) aspect, 

stiffer leaves, many or most narrowly acute yet still 

inrolled so that a subcucullate aspect to the leaf apex 

could be identified, a relatively orderly rope-like 

disposition of the leaves on dry stems and peculiar 

distinctly thickened superficial cell walls. Yet the 

specimens were clearly related to T. inclinata by reason 

of the absence of quadrate, papillose cells on the adaxial 

costa surface, tubulose leaves yielding cucullate apices 

on many leaves, high, sharp papillae, especially if that 

species were considered to be a depauperate form of the 

specimens from rocks. 

Schimper (1855) described Tortella inclinata 

stems as "fasciculato-ramosae," or branched in fascicles, 

and "densissime foliosae," very densely foliose— 

characters not presently associated with T. inclinata, 

which is scarcely branched and only loosely foliose, but 

which are characteristic of T. densa. Nyholm (1989) 

described the leaves of T. densa as "densely and 

homogeneously arranged along the stem." The substrate 

is described as calcareous soil and limestone pavement, 

but the latter substrate in North America is so far 

reserved for Tortella rigens. The substrate of 

European material (England) of Tortella densa is 

"shallow soil on limestone rocks and in crevices" (Smith 

1978). Smith indicated that the plants are not tomentose. 

He wrote that the lower leaves are straight or slightly 

curved and narrowly lanceolate. 

Tortella densa was originally described as a 

variety of T. inclinata (i.e. Barbula inclinata var. densa 

Lor. et Mol.). Tortella densa was provisionally 

described by Albertson (1946) as a form of Tortella 

tortuosa ("forma curta"), and not thought to resemble 

Tortella inclinata or T. rigens, which it does, however, 

by reason of the smooth, elongate cells on the adaxial 

surface of the costa throughout the leaf and by other 

characters so ably described by Crundwell and Nyholm 

(1962). 

Albertson (1946) described his form as "leaf 

short (3–4 mm), soft [i.e. not rigid], brownish green, 

dull, not gradually tapering to a cucullate apex, as a 

rule, with a hyaline border and an often toothed, manycelled 

mucro composed of the excurrent costa, many 

however, often are broken off [som dock ofta är 

avbruten]. Leaf cells small (7–10 µm broad), very 

obscure and without distinct cell walls due to the 

densely situated papillae." 

36 

A specimen seen from Sweden determined 

by Albertson showed leaf margins crenulated by 

extension of the leaf cells. These appeared to be smooth 

more by reason of erosion of the papillae than that they 

do not occur. Inspection of several additional specimens 

shows that often the papillae are not developed on 

various lengths of the margin, the marginal walls are 

smooth. 

The border is a distal extension of the top of 

the proximal cell region, but not composed of the 

elongate, lax, hyaline cells of the base. Rather, these are 

of short-quadrate smooth cells best exposed on the 

longest leaves on the stem and can extend most of the 

length of the leaf. 

In northern Europe the specimens of T. densa 

are described as having leaves "lanceolate, distal part 

gradually narrowed into [a] sharp point" (Nyholm 1989) 

and "narrowly lanceolate to linear-lanceolate, gradually 

tapering to acuminate apex" (Smith 1978). Limpricht 

(1890) described the plants as characteristic of the high 

Alps (1900–2470 m). They are different from Tortella 

inclinata by their very high and dense turves (to 8 cm) 

which are not tomentose. The leaves are shorter, more 

firm (or rigid). When wet, they differ by being erectspreading, 

rather than inflexed (eingebogen) above. In 

North American specimens the tomentum is reduced 

and hidden in the leaf axils, much like T. tortuosa var. 

arctica. 

The variety inclinata in North America appears 

to be a reduced form of the more wide-spread variety 

densa: the leaves are shorter (2–3 mm), so they are less 

likely to show the marginal border in the longer leaves. 

Shorter leaves are cucullate (see youngest leaves of var. 

densa which are often cucullate or naviculate). Var. 

inclinata has fewer branches, the stem is shorter, it has 

no lateral proliferations although in Europe there may 

be apical ones, it has thin walls. When the walls are 

thin, the cells appear bigger (to 12 µm). When they are 

thick, as in var. densa, they appear to be smaller and 

obscure. Small plants are less highly colored, just like 

the younger leaves and plants of var. densa. All of this 

is paralleled in Tortella tortuosa and its variety arctica. 

The var. arctica is more densely foliose with more 

branches at the apex, thickened cell walls and more 

color (orange). Frequently the papillae are reduced 

when the cell walls are thickened. For the association of 

redness with Arctic conditions see Steere's (1976) 

discussion under T. tortuosa var. arctica above. 

The var. inclinata of Europe is described as 

tomentose, and the var. densa as without tomentum. In 

North American populations the smaller stems of either 

variety tend to show rhizoidal tufts at the stem base, but 

not on the elongating distal stem region. When the stem 

is as well-developed as it is in the variety densa, the 

proximal rhizoids have been lost in the soil or by 

crowded leaves and most of the stem appears non 

tomentose. 

North American reports of plants to now 

designated as Tortella inclinata have not derived from 

the kinds of habitats where T. densa in Europe is known

to grow. With the addition to American herbaria of 

recent collections in more northern or upland stations in 

Canada and Alaska, a more complete range of variation 

was made available for analysis. 

The decision was made to associate Tortella 

densa with T. inclinata as a variety after the discovery 

that their ranges in North America overlapped 

intimately as did their morphological characteristics, 

leaving the var. inclinata as an ecological variant that is 

somewhat depauperate compared to the morphological 

complexity of var. densa, the former variety with the 

nature of a recent colonizer of relatively new substrates. 

While examining inland stations of 

Netherlands reports of Tortella flavovirens (all proving 

to be T. inclinata), Rubers (1973) made the interesting 

observation that he found plants with acuminate apices, 

so that they strongly resembled those of T. densa. He 

dismissed their identity because the habit of the 

specimen did not resemble that of the latter species and 

suggested that the form of the leaf tip of T. inclinata 

varied with variation in ecological circumstances. He 

suggested that it was due to acuminate leaf tips that the 

two species have been very difficult to identify. It is 

suggested in the present paper that both taxa are variants 

of T. inclinata. 

One important specimen that helps to reveal an 

amplified eastern distribution of var. densa is the 

following, cited by Crum and Anderson (1981: 308) 

under Tortella inclinata: USA, Vermont, Equinox Mtn. 

(Bennington Co.), rocks, William Randolph Taylor 

155a, 9.18.22 (MICH). The authors write "habitat 

unknown" although the substrate is given on the label at 

MICH, and it is "rock." The specimen is frustrating in 

two aspects: it does not show leaves that are cucullate, 

as the authors emphasize for the species, but are 

narrowly acute to acuminate. It did not even grow in a 

valley bottom where one might hope to find some kind 

of alluvium (Equinox Mt. attains 3816 ft., or around 

1000 m). Yet the specimens are undoubtedly related to 

T. inclinata s.l. but they were also undoubtedly on a 

mountain and on rock. They are, in fact, T. inclinata 

var. densa. This station suggests that the variety may 

occur elsewhere in the Appalachian Mountains in 

eastern North America. 

C. Müller and Kindberg's description of 

Tortella inclinatula (Macoun & Kindberg 1892) 

represents one extreme of the variability of T. inclinata 

s.l. in North America. The stems were very small, only 

0.5 cm, and had no branches. The leaves were cucullate, 

the costa was yellow and pellucid, the capsule "narrowcylindric, 

nearly straight, suberect," unlike the strongly 

inclined capsules of European T. inclinata, the 

description taken from fruiting material from the 

alluvium of abandoned river channels. 

Inez Haring's (1938) description of Torella 

inclinata included more specimens and more characters, 

and one can begin to see elements diverging from the 

var. inclinata: the stems attained 1 cm, the leaves were 

long, to 5 mm, "sublinear to ovate-lanceolate, usually 

narrowing abruptly to a point" but also "sometimes to an 

37 

acuminate apex" which is "usually cucullate." The 

costa is yellow, but "may be red at the base in mature 

leaves." The walls were thick, "giving a honey-comb 

appearance to the cells...." Seville Flower's 

accompanying illustration shows attenuate leaf apices, 

although if they represent leaves from a perichaetiate 

plant, the cucullate apices might be difficult to draw 

distinctively, as they are in sterile stems of var. 

inclinata. 

Appearing in her description are elements of 

the variety densa postulated here: the long stems, the 

longer leaves (to 5 mm) that are acuminate and not 

cucullate, although they are concave (tubulose), the 

tendency toward color, the thick walls obscuring the 

areolation. The low and fewer papillae might refer to the 

reduction to the papillae at the expense of thicker 

superficial walls. 

American material shows the thick superficial 

walls clearly as well as cucullation at the leaf apex. This 

thickened wall tendency of the species is seen in the 

relatively thick-walled proximal cells, at least in the 

proximal area distal to the leaf insertion. The high 

papillae and dense walls seem to have a similar 

function: to buffer the interior cell. The walls can be 

thick and nearly smooth, thick and highly papillose or 

only dense with tall, solid papillae. The papillae are 

solid (not hollow), as though they were a spongy cell 

wall, or to make the area porous to water and gas 

exchange and still protect the cell from extreme 

environmental conditions. This is reminiscent also of T. 

tortuosa var. arctica in which the laminal papillae seem 

to become reduced with increase in cell wall thickness. 

Both taxa may have smooth crenulate rather than 

papillose crenulate marginal cells. That var. densa 

occurs most frequently in rocky cliffs, one might 

speculate that the crevices of the rocks themselves 

provide similar protection necessary to the species, as 

per the discussion of Steere (1976) above. 

Tortella inclinata var. inclinata has a cross 

section showing an unistratose lamina adjacent to the 

nearly circular costa with walls that are not thick, and 

papillae that are sharply delineated. The var. densa 

differs by often being bistratose juxtacostally and with a 

lunate costa, frequently adaxially concave, the papillae 

on both surfaces collectively as thick as the lumen in 

cross section and tapering to the margin which often has 

a smaller, thin-walled smooth cell consistent with the 

tendency of the taxon to occasionally display a marginal 

border of smooth, short-rectangular cells. Except for the 

thicker superficial walls and prominent papillae, the 

cross section of the leaf strongly resembles that of T. 

rigens. 

If one postulates that Tortella inclinata var. 

inclinata is a small, lowland form, perhaps numerous 

other closely related taxa may share a similar 

morphological variation related to elevation: short, 

blunt-leaved forms re-colonizing lowland stations once 

blocked by glacial ice or melt water and where fruiting 

occurs, whereas long-leaved variants occur in the 

uplands. Plerurochaete luteola (Besch.) Thér., for

example in the United States is a highland plant, P. 

squarrosa (Brid.) Lindb. more of a lowland plant (both 

are presently considered synonymous). Anoectangium 

peckii (Sull. ex Peck) Sull. ex Aust. a highland, A. 

aestivum (Hedw.) Mitt. a lowland plant (both also 

considered synonymous). 

The stems of North American plants produce 

lateral axes or shoots on which primordial leaves are 

distributed at distant intervals, becoming more crowded 

at the apex. These shoots develop longer leaves, and on 

some stems may be detached at which time they 

resemble fully grown leafy stems with rhizoids. Note 

that Limpricht's (1890, p.603) illustration of Tortella 

inclinata shows a small branch arising laterally from the 

main stem in a manner rather similar to what is 

described here. Possibly the plants reproduce asexually 

in this way. Coarse rhizoids are present along mature 

stems. Also the stems are fragile and disarticulate in 

whorls. These whorls may or may not represent annual 

growth, but, as in the whorled branches of Tortella 

humilis, simply that the leaves are more obviously 

arranged in verticels densely distributed on the stem. 

One could imagine the turf expanding by a 

kind of branch-budding process in relatively loosely 

consolidated sediments of a large clast size, such as 

sand, relative to silt or clay. The fully grown shoot 

would be less likely to be buried in loose sand than 

other vegetative structures, such as bulbils, gemmae and 

leaf fragments. Other species of sandy habitats may 

reproduce this way, but without obvious shoots as in 

Tortella densa. Possibilities may include branched, 

highly tomentose or radiculose weedy species of 

Bryum, such as B. pallescens Schleich. ex Schwaegr. or 

Bryum caespiticium Hedw. It is often difficult in dense 

mats of these species to tell whether one has a branch or 

a fully formed plant. Note again that, at least in Europe, 

T. inclinata is said to generate deciduous shoots at the 

stem apex (Crundwell & Nyholm 1962; Smith 1978; 

Nyholm 1989). 

The populations in the American and Canadian 

West also develop stems of a significantly more gracile 

appearance alongside robust stems in or beside mature 

clumps, and perhaps these more delicate plants 

represent matured lateral stem shoots. Apical segments 

of the stem composed of similarly gracile leaves that are 

distinct from the broader and longer leaves lower on the 

stem are also visible in some stems, as are what appear 

to be densely foliose side branches composed of axes on 

which narrower leaves occur and which may be fully 

mature lateral shoots. Most Tortella species develop 

larger leaves near the stem apex, giving the habit a club 

shape. In Tortella densa the leaves are generally not 

enlarged at the stem apex in this way, perhaps to 

accommodate dispersal mechanisms involving the lower 

stem. 

In other species of Tortella, the unfertilized 

perichaetia are usually conspicuous at the stem apex, 

especially in species with differentiated perichaetial 

leaves such as T. tortuosa, T. fragilis and T. inclinata. In 

the few perichaetiate stems seen of T. densa in North 

38 

America, the perichaetia were not at the turf surface 

but along the lower stem, persisting from their initial 

apical development. 

As discussed above under Tortella inclinata 

var. inclinata, the striking perichaetial leaves of fruiting 

specimens cannot be compared because var. densa is 

not known to fruit. Little difference could be found 

between perichaetia of unfertilized archegonia of both 

varieties. 

While many species of the genus Tortella may 

disperse through leaf fragmentation, T. densa, upon 

further study of additional specimens and populations in 

the field, may be found to disperse instead through stem 

modifications. Why var. densa does not fruit in North 

America is a mystery because both perigoniate and 

perichaetiate populations exist, occasionally in 

contiguity. The var. inclinata of more moderate 

ecological regimes, however, does fruit. 

Tortella inclinata s.l. might once have had a 

range that extended into areas where fertilization was 

successful (dependent on an environmental condition 

such as temperature). When this condition was lost, 

vegetative propagation such as fragmentation became 

important for dispersal. This perhaps enabled the plants 

to (re)colonize areas once under water and fruiting was 

again possible: the alvars, limestone areas with thin to 

absent soil and historically newly exposed substrates in 

temperate regions. 

This possibility may also be suggested for 

Tortella alpicola which fragments in the highlands and 

is fertile (perichaetiate) in lower elevations. 

Tortella inclinata var. densa, according to 

European specimens from S and NY determined by 

Albertson, Crundwell and Nyholm, when it is found on 

alvar, is reduced in the expression of its characteristics: 

while many leaves on a stem are narrowly acute, many 

others are decidedly cucullate. The hyaline proximal 

cells of plants from this habitat also become peculiar in 

that they intergrade on many of the leaves, or the 

proximal hyaline region is rectangular (not V-shaped), 

but are in some leaves strongly differentiated in a Vshape 

in leaves from the same stem. The stems may also 

be strongly reflexed or squarrose-recurved. The cells are 

larger and the walls thicker than in robust specimens 

from upland stations. Plants from alvar vegetation in the 

Great Lakes region may display similar characteristics, 

but so far, similar plants from these areas are T. rigens 

only. 

At the other extreme, in Europe, a specimen 

(England: Westmoreland Co., limestone rocks on 

northwest side of Beacon Hill, Orton, 19/3/1961, 1200 

ft. (400 m), A. C. Crundwell, NY) was of a fuscous to 

black coloration, long leaves (to 3.5) on relatively short 

stems (1 cm) with thick superficial laminal walls, 

bistratose juxtacostally in the midleaf region. The leaf 

shape strongly resembles those of Tortella tortuosa var. 

arctica, being rather abruptly contracted just above the 

hyaline proximal cells into a straight, somewhat rigid 

limb, the laminae evenly tapered to and confluent with a 

quite long, multicellular mucro of 2–6 elongate, smooth

cells. Another robust specimen from Ireland (East 

Mayo, E. Hegewald 2632, NY) with similar 

characteristics also shows little tomentum, or this very 

sparse along the stem. In both specimens the leaves did 

not detach as easily into whorls or whorled fascicles as 

in the American material examined. 

In Japan, Saito (1975) described montane 

specimens of Tortella tortuosa that strongly resembled 

T. inclinata. The specimens had "costae whose adaxial 

surface cells are much elongated and smooth, forming a 

sharp break in the areolation of lamina." He indicated 

that these plant had "apical portions" being "not involute 

and gradually narrowed into an attenuate apex. In 

addition, the stems of these plants are rather long (10– 

20 mm long) and the rhizoids are developed from the 

upper portion of the stem." Some aspects of this 

description resemble characteristics of T. inclinata var. 

densa (the upland (non-alluvial) habitat and the long 

stem in addition to the smooth, elongate cells of the 

adaxial costal surface). 

Short-mucronate, sterile forms of Encalypta 

Hedw. may be confused with T. densa: they are reddishgreen, 

tubulose with a proximal cell region similar to 

those forms of T. densa where the proximal V-shape is 

not strong. Encalypta species are immediately 

differentiated by the quadrate, papillose cells on the 

adaxial surface of the costa and the distinctive thickened 

cross-walls in contrast to thin vertical walls of the 

hyaline proximal cells. Encalypta procera Bruch 

usually is accompanied by brown filiform brood bodies. 

A collection from Emmet Co., Michigan 

(Ireland 4362, CANM) was first thought to contain both 

T. tortuosa and T. inclinata in mixture, but actually the 

plants represented one variable collection with leaf cells 

up to and exceeding 14 µm. Smaller plants in the 

collection resembled T. humilis in their very broad 

leaves, flat cross-section and the reflexed mucro, rather 

than inflexed mucro typical of other collections and of 

most leaves in this collection. This collection was also, 

according to the label, anomalously growing "On sandy 

soil near lake." The leaves were not rigidly erect. 

British Columbia: foot of Mabel Mountain, 

1300 ft. (400 m), on sand in river (with T. tortuosa) 

Aug. 24, 1950 V. J. Krajina (UBC)—this specimen 

shows the clearest relationship to T. inclinata var. 

mutica Latzel (see discussion above under var. 

inclinata). The leaves of the proximal whorl are very 

short (2:1) and fistulose, resembling the leaf of Aloina. 

The costa in the lowermost leaves also disappears 

before the apex. In succeeding whorls, however, the leaf 

elongates, is successively less fistulose, finally deeply 

cucullate with an excurrent costa. The Wisconsin 

material was var. inclinata, the BC material here var. 

densa: massive superficial cell walls (pedestaled 

papillae). The leaves were brick red, as was the T. 

tortuosa growing intermixed (distinguished by the 

quadrate papillose cells on the ventral surface of the 

costae and the long mucro, the narrowly acute apex and 

dense tomentum throughout the stem). 

39 

In Europe, the opposite extreme, and one to 

be anticipated in North America, may be represented 

by: Scotland, on limestone rock, alt. 2000 ft. (606 m), 

Wallace s.n. June 30, 1963 (NY), with cell sizes very 

small, costa highly colored brown-orange. Yet in the 

youngest leaves (these are large) they are pale green, as 

one might expect—with very large stems. These are 

very acuminate but one can easily see they can become 

cucullate when the leaves are reduced. No tomentum 

was found on European stems examined. 

A specimen originally identified as Tortella 

tortuosa from southern Wisconsin was at first thought 

here to be T. inclinata, yet it occurred in disturbed soil 

on an exposed hillside, associated with a fen and oak 

opening (Christy 4272 UBC)—an atypical substrate and 

habitat. The subcucullate, obtuse leaf apices, and naked 

adaxial surface of the costa precluded it from being T. 

tortuosa, yet many of the leaves were very narrow 

throughout the leaf and into the apex, the leaf cells only 

came to 10 µm (not 12 µm as in T. inclinata. The 

pedestaled papillae were diagnostic. Although the 

leaves seemed yellowish, their orange coloration was 

brought out after soaking in dilute KOH. 

Specimens examined: 

B.C.: above Muncho Lake, Alaska Highway mile 465, 

rock slide. Aug. 26, 1967, F. M. Boas 67032 

(UBC). 

B.C.: Range Lake, N.E. of Tweedsmuir Glacier, damp 

crevices of cliff, 1100 m., 59–52°N, 137– 

50°W, July 26, 1992, Schofield, Godfrey & 

Goward 98203 (UBC). 

B.C.: Mt. Stalin Area, Tuchodi Lakes, W Lake; near SE 

end of Lake on S shore. 58°12'N, 124°31'W, 

934 m, on steep, seepy calcareous cliffs on 

rock faces, on ledges and in crevices, and in 

adjacent Picea glauca-Alnus mesic forest with 

wefts of Thuidium, Hylocomium and Entodon 

concinnus July 25, 1977, D. H. Vitt 19794 

(BUF). 

N.W.T.: Mackenzie District: alpine Dryas integrifolia, 

Cetraria tilesii, Potentilla community, Nahanni 

National Park, 61°16'N, 124°12'W; 1200 m, 16 

July 1976, S. Talbot T 6077–18 (NY) (as T. 

inclinata); Mackenzie District: Logan Mtns, 

lake close to S. Nahanni River, 1560 m., 

62°34'N, 128°31'W; on earth slopes at cliff 

base, Aug. 15, 1978, Schofield, Vitt & Horton 

70886 (UBC). 

Alaska: among pebbles on high gravel banks of river 

north of Wiseman, 28 June 1949, Louis H. 

Jordal 2050 (as Tortella tortuosa) (CANM). 

Alaska: Ambler River Region, upper Ambler River, 

unnamed peak N. of "Carnes Creek" near its 

junction with Ambler River; ca. 4 miles S. of 

mouth of Ulaneak Creek; ca. 67°20'N, 

156°51'W, dry calcareous ridge, 1600 ft. (490 

m), 6 July 1976, M. Lewis 1768 (NY, F) 

(perigoniate plants).

Mich.: Emmet Co., on sandy soil near lake, Sturgeon 

Bay, Wilderness State Park, July 26, 1961, R. 

R. Ireland 4362 (CANM) (resembles T. 

tortuosa in habit). 

Wisc.: Waukesha Co.: Genesee Oak Opening and Fen 

Scientific Area, 1.6 km W of Genesee T6N, 

R18E, NE 1/4 SW 1/4 S28 (fen), SW 1/4 NE 

1/4 S28 (oak opening), 280–290 m, on 

disturbed soil of exposed hillside, oak opening, 

full exposure, with Euphorbia corollata, Stipa, 

Cladonia, July 9, 1982, J. A. Christy 4272 

(UBC). 

7. TORTELLA RIGENS Plate 8 (fig. 11–13) & 9 

Tortella rigens N. Alberts., Act. Phytogeogr. Suec. 20: 

197. 15, 16. 1946. 

Plants firm, somewhat rigid, occasionally brittle when 

dry, in low, dense, dark brown to black to emeraldgreen 

tufts, compact, elongate, not rosulate. Stem 1–3 

cm, rarely to 4 cm high, usually branched, densely 

foliose, leaf bases not evident between the leaves, 

central strand absent, tomentum rufous, dense, visible 

between the leaf bases especially along the lower stem. 

Stem leaves firm, somewhat rigid, stiffly erect, 

especially proximal leaves, to loosely twisted and 

moderately contorted when dry, erect and somewhat 

spreading when moist, gradually larger and more 

crowded toward the stem tips, distal-most leaves usually 

more spreading than the erect leaves or leaf bases 

below; leaf shape variable within clumps and on the 

same stem: short stems with leaves short, ovatelanceolate, 

margins inwardly tapering to acute apices, 

longer stems with longer, more narrowly lanceolate 

leaves with more acuminate apices, the proximal leaves 

frequently narrowly lanceolate, distal leaves ovate or 

broadly ovate-lanceolate, generally narrowing 

gradually, but often some distal-most leaves with an 

abrupt narrowing with parallel sides in the distal 

quarter, leaf tips fragile, the older ones usually eroded; 

deeply concave to canaliculate-tubulose in longer 

leaves, 1.5–3(–4) mm long; base undifferentiated in 

shape to broadly ovate below a gradually to abruptly 

lanceolate distal lamina; margins variously plane, erect, 

to incurved, infrequently undulate in some leaves, apex 

fragile, acute to acuminate, abruptly ending before a 

sharp apiculus or narrowly acute and confluent with a 

conical, nearly mucronate apiculus; in shorter leaves 

naviculate to nearly cucullate, distinctly cucullate in the 

awl-shaped youngest leaves arising from the stem apex; 

longer, narrow leaves with erect to broadly incurved 

margins extending to the leaf apex; costa shortexcurrent 

into a sharp, narrow, denticulate or smoothly 

conical mucro of 1–5(–10) cells, adaxial surface 

covered by quadrate, papillose laminal cells except for a 

narrow groove of elongate (8:1) smooth cells 

continuous throughout length of the leaf, in cross 

section, adaxial epidermal cells interrupted in the center, 

exposing the adaxial stereid layer by about two cells, 

40 

adaxial and abaxial stereid bands present throughout, 

guide cells present in one row; proximal laminal cells 

abruptly differentiated from distal cells, lacking a zone 

of cells intermediate in color, cell size, and papillosity, 

yellow-hyaline, elongate, 6–8:1, firm to laxly thinwalled, 

smooth; marginal angle of differentiated cells 

steep due to the quadrate laminal cells extending 

juxtacostally far down into the proximal cell region, 

with a marginal row of firm to thin-walled, quadrate to 

short-rectangular smooth cells extending a short way up 

the lamina or to mid-leaf; distal laminal cells 

frequently bistratose on one or both laminae 

juxtacostally or extending marginward as one to two 

pairs of cells, but never extending to the margins even 

in longer leaves with narrowed apices, cells gradually 

smaller toward the margins; rounded-quadrate or 

hexagonal, with slightly thickened walls, (12–)14(–17) 

µm wide, strongly papillose, areolation more or less 

distinct, differentiated distal marginal cells absent 

except in longer leaves possessing a somewhat 

differentiated and deciduous apex, these having obscure 

(due to erosion) to distinct thicker-walled, less papillose 

to smooth, elongate marginal cells. Asexual 

reproduction: vegetative propagation apparently by 

means of fragile leaf tips in the longer leaves. Sexual 

condition: apparently dioicous: only sterile and 

perichaetiate plants seen. Perichaetia terminal, leaves 

long-lanceolate, costa excurrent into a long awn or 

subula. Sporophytes: unknown. 

Forming shag-like patches on limestone alvar 

pavement (but not limestone bluffs, boulders and other 

calcareous landforms), mostly near sea level (to 100 m). 

In North America: Ont.; Mich., N.Y., Ohio. Distributed 

in the Baltic region of Sweden where it is considered 

rare (Nyholm 1989), the Baltic republics of the former 

U.S.S.R. (Ignatov & Afonina 1992), with one report of a 

specimen seen from Czechoslovakia (Düll 1984). Pilous 

(1965) has reported numerous collections in 

Czechoslovakia. 

Herbaria examined: BUF, CANM, DUKE, 

MICH, NY, S, UBC. 

The species is confined to temperate climate 

regimes on limestone, particularly in alvar pavements, 

rather than in colder austral or boreal areas and their 

altitudinal equivalents as has been indicated on some 

misdetermined specimens from the Rocky Mountains. 

So far, it is only known in Europe from the alvars of the 

circum-Baltic states and Czechoslovakia and may be 

expected in other areas with appropriate landforms. 

Braunmiller et al. (1971) doubted that the species was 

endemic to Scandinavia and Czeckoslovakia and 

anticipated its discovery in central Europe. By 1989, 

however, it has not been discovered in Germany (Düll 

& Meinunger 1989), or other areas in Europe. 

In North America, alvars appear confined to 

substrates of Ordovician and Silurian limestones and 

dolomites, being areas of exposed rock adjacent to 

bodies of water. These areas were once under water 

when sea and inland water levels were higher. The 

exposures occur around the Great Lakes and St.

Lawrence River and associated lakes, such as Lake 

Simcoe in Ontario, Canada and Lake Champlain in the 

states of New York and Vermont. These areas may be 

expected to harbor more populations of this species (see 

Catling & Brownell 1995 for the distribution of the 

alvars in the Great Lakes Region). 

Since these two continental regions are most 

likely to have a similar post-glacial history and 

geomorphology, details on the distribution and substrate 

of the triad of closely related taxa, Tortella inclinata 

var. densa, T. inclinata var. inclinata and T. rigens, may 

contribute to an understanding of the relationship of 

their biogeography and evolution. Tortella inclinata var. 

densa, in montane and boreal situations in Europe and 

in the North American west, occupies the most 

potentially stable of the three habitats associated with 

these taxa, and it may have been the most likely to find 

a refugium during glacial periods. The distribution of T. 

inclinata may reflect the availability of recently 

stabilized habitats in temperate situations. Alvars have a 

more restricted distribution as does T. rigens, which is 

apparently confined to it. 

Alvar, it should be stated, is not a geological 

term but an ecological one: "the plant community 

consisting typically of mosses and calciphilous 

herbaceous plants that grows on steppelike shallow 

alkaline soils overlying Scandinavian limestones" (it is a 

Swedish word: Ålvar; Gove et al. 1976). Alvar is 

distinct from the geological term karst which relates to 

subsurface conditions and is "a limestone region marked 

by sinks, abrupt ridges, irregular protuberant rocks, 

caverns, and underground streams" (Gove et al. 1976). 

In North America there are karst regions in New 

Mexico and Kentucky and in Europe there are the karst 

regions of the Dalmatian coast in the Mediterranean, but 

the surficial deposition of sediment is apparently deep 

enough to preclude them from supporting an alvar 

vegetation. Areas stripped of topsoil are associated with 

areas once glaciated or recently underwater, especially 

in northern latitudes. Plants of alvars do not necessarily 

match those of karst regions, which are most 

represented in warm subtemperate to subtropical 

latitudes. 

Of the numerous alvar vegetation types 

delineated by Catling and Brownell (1995), only in the 

"Pavements and Pavement Edges" type are bryophytes 

mentioned: Riccia sorocarpa Bisch. for areas with some 

moisture, and Tortella tortuosa, Tortula ruralis (Hedw.) 

Gaertn., Meyer & Scherb., and Ceratodon purpureus 

(Hedw.) Brid. on the "drier pavements with a...soil 

depth of less than 2 cm." See discussion under T. densa 

for associated Tortella species in European stations. 

The epithet refers to the somewhat rigid 

character of the leaves (that is, not as rigid as in Tortella 

fragilis). The leaves lower on the stem when dry, 

however, may be as rigidly erect and straight as those of 

T. fragilis. 

The identification of this species for the North 

American flora, just as with Tortella inclinata var. 

densa discussed above, came about by attempting to 

41 

reconcile morphologically anomalous specimens and 

substrates of traditionally understood Tortella inclinata 

with what might be expected from the literature. The 

few specimens of T. rigens are scattered in five North 

American herbaria, making it difficult to characterize 

American populations, as opposed to the relatively 

abundant European specimens represented in Swedish 

collections (at S). 

Two specimens originally determined as 

Tortella tortuosa from Ohio and New York state had 

undeniable characteristics of T. tortuosa: lanceolate 

leaves, margins rather undulate, with rather long 

acuminate apices tipped with mucros. However the 

leaves had no papillose or quadrate epidermal cells on 

the adaxial surface of the costa throughout the leaf, and 

some of the leaf tips had inrolled leaf margins and were 

cucullate. They were at first renamed T. inclinata. Inez 

Haring's (1938) description of that species included 

leaves "sublinear to ovate-lanceolate, usually narrowing 

abruptly to a point." Seville Flower's accompanying 

illustration showed attenuate leaf apices, much like the 

Ohio and New York specimens (see treatment above 

under T. inclinata var. inclinata). 

However, both specimens were not exactly 

characteristic of that species either. They seemed to 

have broader leaves, be greener, with leaf cell sizes to 

14 µm. Perhaps most tellingly, they both occurred on 

limestone pavements, the specimen from Ohio growing 

with Opuntia and Liatris species—hardly the alluvium 

expected with T. inclinata var. inclinata. Williams 

(1968) collected a specimen of Tortella inclinata from 

an apparently anomalous substrate in Ontario: "bare 

open limestone flats in drier locations than Tortella 

tortuosa, Carden flats 1185." The anomalous substrate 

in the Williams citation was striking and T. inclinata 

collected by Williams was searched for in our CANM 

loan without result. However, several other specimens 

of T. inclinata in that loan were noted as growing on 

limestone pavement, rather than alluvium. 

Earlier in 1997, A. C. Crundwell (pers. comm.) 

suggested that Tortella rigens might be found in the 

grasslands of central Canada. The above specimens 

were examined to see if they fit the concept of that 

species sensu Nyholm (1989). This, in fact, is what they 

turned out to be. 

Specimens of Tortella inclinata have no 

epidermal cells on the adaxial costal surface, the stereid 

cells usually exposed by about three cells. Tortella 

rigens is very similar. It may have an adaxial epidermis 

in the proximal part of the leaf and the adaxial stereid 

layer is less exposed (usually by two stereid cells), 

hence there is a somewhat more narrow groove or 

channel running the length of the leaf. Tortella inclinata 

var. inclinata is regularly cucullate whereas T. rigens is 

distinctly cucullate usually only in the smallest leaves or 

on the smallest stems, or in the first whorl of leaves at 

the stem base. The leaf apices of T. inclinata var. 

inclinata may be acute to narrowly so, the distal laminal 

cells, however, are only 10–12 µm. In section, the 

leaves are not bistratose juxtacostally, the leaves keeled

eside the costa and broadly incurved above. Tortella 

rigens has bistratose cells juxtacostally in the median 

portion of the leaf and therefore the laminae spread out 

in a broadly concave leaf cross section, the distal cells 

14 µm or more in width. 

Although Tortella rigens was not recorded as 

part of the Canadian moss flora by Ireland et al. (1987), 

it does occur in the most recent checklist of the mosses 

of North America north of Mexico (Anderson et al. 

1990). The species has previously been reported from 

the United States for Colorado by Weber (1973), and 

several specimens have been collected by Hermann and 

determined as that species. Since all of these collections 

occurred in the Rocky Mountains far from alvar 

substrates, it was felt necessary to examine European 

material to clarify the concept of the species. 

Specimens from Scandinavia of Tortella 

rigens, including authentic material identified by 

Albertson, were loaned to the author from the Museum 

of Natural History, Stockholm (S). The specimens were 

compared with the species description as given by 

Nyholm (1989). 

R. Zander (pers. comm.) pointed out the 

elongate, smooth cells on the apical margins of the 

leaves of some plants of Tortella rigens in Swedish 

material—but these occurred only on specimens with 

long-lanceolate leaves that showed a narrowed region in 

the apex of the leaf. Many reduced stems with 

correspondingly reduced leaves occur in these Öland 

and Gotland collections. Larger stems and leaves start to 

display an abrupt apical limb, which may be interpreted 

as an apical propagulum since these are deciduous. It is 

only these apical elements that are fragile. The apical 

leaf regions also have elongate smooth clear cells on the 

margins and, in fact, together with the stiffly erect 

leaves and tomentose stems, strongly resemble T. 

fragilis. (Note that these marginal cells are frequently 

eroded and not evident, but many samples exist where 

they are as distinctive as in T. fragilis.) These 

collections include populations with short stems and no 

apical differentiation, grading to long stems that are 

hardly distinguished from the latter species. Smaller 

leaves without apical modifications do not appear to be 

fragile at the apex. 

Tortella inclinata has frequently been 

associated with T. fragilis through "intergrading forms" 

(Limpricht 1890; Dixon 1924). Tortella rigens was 

probably one of those forms before being removed from 

the concept of T. inclinata. Persson (1947) indicated 

that Tortella rigens may have a "hybrid origin in the 

Postglacial" since (in Europe) it is morphologically 

intermediate between T. fragilis and T. inclinata. The 

deciduous apices and marginal cells noted above seem 

to substantiate this suggestion. 

In the few American specimens available for 

study, the apex may become narrowed, but so far no 

bordering tendency has been observed in distal portions 

of the leaf. 

Tortella rigens appears to be much more 

polymorphic than one might expect from the key and 

42 

description of it given by Nyholm (1989). The 

species can resemble T. inclinata s.l. in having short 

plants with only cucullate leaves. It can resemble T. 

tortuosa in having lanceolate leaves that are undulate 

and tipped with a short mucro. These can have 

anomalous leaf cross sections that are not bistratose 

juxtacostally but have the large leaf cells. Taller plants 

can resemble T. fragilis by reason of the fragile leaf tips 

of specimens with elongated leaf apices and overall by a 

more rigid leaf-stance in dry material. The stablest 

characters are those that show the affinity with leaves of 

T. inclinata s.l.: particularly the relatively naked adaxial 

surface of the costa, the inrolled to tubulose to cucullate 

leaf margins as well as the impression that the proximal 

cell region occupies less of the overall area of the leaf. 

Definitive characteristics are the larger size of the leaves 

(than T. inclinata), the bistratose area beside the costa in 

leaf cross-sections resulting in a broadly tubulose 

outline, and the alvar substrate. The plants are generally 

taller, and of a darker green (less yellow) color. 

Specimens noted to date from North America 

more resemble the Tortella tortuosa facies in larger size 

and leaf characteristics, especially in undulate leaves, 

which are not fragile. The leaves are all tubulose, 

however, with occasional cucullate or subcucullate leaf 

apices, with distinctive cross sections alread noted 

(bistratose juxtacostally with large lumens beside the 

costa becoming sharply reduced in size toward the 

margins). Specimens in herbaria identified from the 

American and Canadian west, possibly because of an 

expectation that the species would resemble T. fragilis 

based on European material and literature, were actually 

depauperate examples of that latter species (discussed 

above). 

Tortella inclinata is a rather small plant, 

occasionally attaining 15 mm in height. Plants that 

resemble that species, but which are much larger 

overall, such as to 25 mm, and which display variable 

leaf apices, rather than the uniformly stubby, nearly 

fistulose leaves of T. inclinata, together with their 

association with a limestone pavement substrate, rather 

than gravel, sand and silt, are most likely to be T. 

rigens. 

The observations listed in the above 

description of the species parallel in many respects the 

Latin description of the type by Albertson (1946), which 

is roughly translated as follows: 

"Tufts robust more or less rigid, 1.5–3, more 

rarely to 5 cm tall composed of stems not or little 

branched with red-brown radicles loosely scattered. 

Moistened leaves erect-patent, dry ones strongly 

contorted or crisped; in young plants short, lanceolatetriangular 

prominently canaliculate, mitriform in the 

apex, mature leaves 4–5 mm long (at the most to 6 mm, 

in T. inclinata normally 2–3 mm, at most 4 mm long), at 

least in the proximal part canaliculate oblonglanceolate, 

more or less elongate in the distal part, yet 

not subulate, very often broken, suddenly contracted. 

The nerve strong, in quite mature plants usually redbrown, 

in all parts distinctly separated from the cells of

the lamina, when dry more or less shining. Cells of the 

leaves formed as in T. inclinata, indeed normally 

somewhat larger (around 11–14 µm broad) and at least 

in the distal part of the leaf less papillose.... Margin 

lightly (more lightly than in T. inclinata) crenulated, 

often smooth or provided with a row of hyaline 

narrowed cells. The rest unknown." 

Much of Albertson's discussion leading up to 

the establishment of specimens from southern Sweden 

as a distinct and new species involved removing 

specimens with certain characters from the 

circumscription of Tortella inclinata in previous reports 

from his study area. As long as specimens like T. rigens 

are excluded from collections identified as T. inclinata, 

that latter species is readily identifiable, and so it is in 

North American populations that half of the specimens 

of T. rigens were found among herbarium collections of 

T. inclinata. 

Neither Albertson (1946) nor Crundwell and 

Nyholm (1962) stated that populations with morphology 

intermediate between any of the Tortella species in 

Sweden occur in southern Scandinavia, although "in the 

alvar vegetation of Öland and Gotland, where the 

underlying limestone is covered by only a very thin 

layer of soil, they [T. inclinata and T. densa] do often 

occur together, associated with T. tortuosa, T. fragilis, 

and T. rigens. These five species frequently grow in 

mixed tufts, without intermediates" (Crundwell & 

Nyholm 1962). Whether intermediate plants will be 

found in North America requires additional field work. 

After examining numerous Scandinavian 

specimens from S, several identified by either 

Crundwell or Nyholm or both, the following may be 

said: 

1. The stature of Swedish plants is variable, 

from short to rather long. When the stems (so far only 

seen in Europe) are as long as those of Tortella fragilis, 

that is usually what the plant is. Characteristic of T. 

rigens populations is intergradation between forms short 

enough to pass as T. inclinata to material that may pass 

as T. fragilis or T. tortuosa. Neither T. inclinata, T. 

fragilis, or T. tortuosa populations show this 

intergradation of stature. 

2. The leaf variation in herbarium specimens 

ranges from short incurving to cucullate leaves to longer 

leaves that begin to show a differentiation in the distal 

region into a not very well-developed deciduous 

propagulum, usually one or two leaves of which are 

marginally bordered by thick-walled, smooth, elongate 

cells, similar to that of Tortella fragilis. 

3. The leaf cross-section is intermediate 

between Tortella fragilis, with its bistratose distal leaf 

section in the region where the propagulum is forming, 

and T. inclinata, with a relatively broad area (to two 

stereid cells wide) on the adaxial surface of the costa 

that has no quadrate papillose cells. This is not as broad 

as the exposed area of T. inclinata, which often three 

stereid cells wide. 

43 

4. Neither Tortella fragilis nor T. inclinata 

has a stem central strand: no such feature has been 

found in specimens of Tortella rigens either. 

5. Leaf cell sizes are consistently 14 µm or 

larger. Those of Tortella inclinata are rather large (to 12 

µm) and those of T. fragilis occasionally can attain that 

dimension. 

The specimens on which Weber (1973) based 

his description and discussion of Tortella rigens from 

the American West (Colorado) were redetermined to be 

T. alpicola—a species with leaf cell sizes identical to 

those described for the former species, and which also 

have deciduous leaf apices. The structure of Tortella 

alpicola is quite different from T. rigens (see discussion 

above). 

Depauperate plants of Tortella fragilis account 

for the other specimens and reports of T. rigens, hence 

the present study presents the first genuine reports for 

North America. A specimen from Larimer County, 

Colorado (Hermann 27641, MICH, NY) from the 

Rocky Mountain National Park in moist tundra at 

11,600 feet had plants of a vivid green. It resembled 

Tortella rigens superficially in the small plants, in the 

foliose habit with less stiff leaves and larger leaf cells 

(from 11 to 14 µm). However, none of the leaves were 

tubulose or subcucullate, all had quadrate, papillose 

cells on the adaxial costa surface in the median leafregion. 

As the collector himself noted, the apex was 

fully bistratose and lanceolate-subulate leaves with 

complete propaguloid modifications together with the 

border of elongate, thicker walled, smooth cells typical 

of T. fragilis were found in at least some leaves. This 

latter characteristic precluded the specimen being 

identified as T. rigens, although the plant was striking 

for the unusually few leaves with such modifications 

and the relatively short subulae, when they did occur. 

Habitat differences may also be brought to bear: 

Tortella rigens occurs on flat calcareous alvar 

pavements near sea level. 

Hermann 27264 (DUKE, MICH) from 

Gunison County, Colorado at 12,162 ft. (3685 m), in 

"wet, peaty vertical bank of pond on tundra" was also 

similarly modified Tortella fragilis with the same 

characters as noted above. 

All specimens of Tortella fragilis in North 

America and identified as T. rigens always had long- to 

linear-lanceolate leaves (typical of T. fragilis) and 

always had a fully completely developed apical 

propagulum in at least a few leaves and quadrate, 

papillose cells across the adaxial surface of the costa. 

Even minute specimens of this species from Quebec 

showed the typical cylindrical apical propagula at the 

tips of the youngest leaves emerging from the stem 

apex. 

The cells sizes are undoubtedly large in 

Tortella rigens (14 and even to 17 µm), but such cells 

sizes (to 14 µm) correlated with reduced apical 

propagula can be demonstrated in the two American 

specimens of T. fragilis collected by F. Hermann. Large 

cell sizes can be demonstrated in some northern

collections of T. tortuosa as well and, as noted by 

Crundwell and Nyholm (1962) above, regarding T. 

flavovirens var. glariecola, large cell size may be a 

northern condition of species with otherwise smaller 

cell sizes. Depauperate specimens of Tortella fragilis 

with large leaf cell sizes may be demonstrated in the 

Southern Hemisphere as well, such as in specimens 

from Campbell Island (Eckel 1997). These specimens 

have quadrate, papillose cells covering the adaxial 

surface of the costa. 

Specimens examined: 

N.Y.: Jefferson Co., limestone pavement, east of 

Rosiere, ca. 340 ft. (110 m), July 9, 1968, S. J. 

Smith & W. V. Glider 42789 (MICH). 

This specimen is intriguing in that it is a mixture of 

Tortella rigens and a perichaetiate plant that at first 

appeared to be an arrested form of T. fragilis: the latter 

plant had everything but the propaguloid leaf apices and 

the completely bi-multi-stratose distal leaf cross section 

of that species. The plant with stiff leaves, deciduous 

apices, the marginal elongate, non-papillose cells was 

identified as T. tortuosa var. fragilifolia, complete with 

bistratose cells juxtacostally and leaf laminal cells 

gradually smaller toward the margins. The var. 

fragilifolia is rigid, lacks leaf apices on most mature 

leaves and is a bleached yellow-brown with small cells 

and quadrate papillose cells on the adaxial surface of the 

costa; the T. rigens is broadly lanceolate and a dull 

greenish black with all its apices intact and with large 

leaf cells and quadrate, papillose cells absent in a 

narrow groove on the adaxial surface of the costa 

throughout the leaf. 

Additional specimens examined: 

Ont.: Renfrew Co.: Eganville, Grattan Township, 

45°31'N, 77°00'W, limestone bluffs along river 

and adjacent conifer-mixed wood habitat, 

limestone pavement in clearing beside river, R. 

R. Ireland 22735, July 20, 1987 (NY) (also 

22737 UBC, same area and time: dark greenblack). 

Ont.: Cochrane District, along Stone Rapids, Scovil Tp., 

600 ft. (180 m), growing along rocky shore, 

Aug. 1976, Garry A. Shea (CANM). 

Ohio, Ottawa Co. Marblehead quarry, limestone rock, 

Opuntia barren, C. M. Boardman, Aug. 7, 

1953, C. M. Boardman, Aug. 8, 1955 (DUKE) 

(plants perichaetiate). 

Mich.: Chippewa Co., on box limestone area, very 

abundant, just N of Fork Kampe Y in road, 

Sec. 34 T43N R6E, Drummond Island, June 

28/9, 1948, L. F. Koch 2630 (CANM, very 

robust specimen). 

N.Y.: Jefferson Co., limestone pavement, east of 

Rosiere, ca. 340 ft. (130 m), July 9, 1968, S. J. 

Smith & W. V. Glider 42789 (MICH) (plants 

perichaetiate). 

44 

A protonematous specimen from Michigan 

(Emmet Co. "on quartz pebbles on sandy beach" Ireland 

4362a (CANM), identified as T. inclinata is clearly 

anomalous: it has very large leaf-cells, to 14 µm and 

larger, has undulate leaves and acute to acuminate 

apices. It is unlikely to be T. inclinata and might be 

referable to T. rigens, but the specimen is too 

depauperate and too peculiar to assess with confidence. 

8. TORTELLA NITIDA Plate 9 

Tortella nitida (Lindb.) Broth., Nat. Pfl. 1(3): 397. 

1902. 

Tortula nitida Lindb., Öfv. K. Vet. Ak. Ferh. 

21: 252. 1864. Type: Gibraltar, A. F. 

Regnell. 1839 (S—isotype). 

Barbula alexandrina Lor., Abh. Ak. Wiss. 

Berlin 1867: 31. 6, 7. 1868. [The date 

of publication of this species was 

1867, fide Stafleu & Cowan 1981: 

159.] 

Trichostomum diffractum Mitt., J. Bot. 6: 98. 

77 f. 5, 6. 1868. 

Trichostomum nitidum (Lindb.) Schimp., Syn. 

ed. 2: 179. 1876. 

Mollia nitida (Lindb.) Lindb. in Braithw., Brit. 

Moss Fl. 1: 250. 37 A. 1887. 

Plants in dull olive-green to yellowish-green, dense, 

rounded cushions, compact, the habit elongate, not 

rosulate. Stems 0.5–2 cm high, forked, densely foliose, 

leaf bases hidden, tomentum confined to the base of the 

stem, hyalodermis well developed, central strand strong, 

in older stems often pigmented around a central aperture 

lacking tissue. Stem leaves once-circinately incurved 

when dry, erect to erect-spreading when moist, typically 

very fragile to moderately so, distal part of leaf usually 

missing, to 4 mm long, ligulate, oblong, oblongspathulate 

to broadly lanceolate, appearing more 

narrowly lanceolate in strongly tubulose leaves, mostly 

broadly acute to acute, broadly channeled, nearly flat to 

somewhat to strongly tubulose in the distal portion of 

the leaf, leaves frequently naviculate; base ovate, 

somewhat broader than above, often differentiated from 

the distal part of the leaf by a constriction (wasp-waist) 

or, if the distal lamina is strongly incurved, appearing 

with shoulders, if proximal cells are particularly lax, 

appearing narrower than the distal portion of the leaf 

from the sides of the leaf folded inwards at the base; 

margins broadly to strongly incurved-involute, then 

erect just before the apex, often appearing cucullate at 

least in some leaves; apex obtuse to broadly to narrowly 

acute, sometimes appearing emarginate due to the 

erosion of the mucro, the mucro often preceded by brief 

channel or excavation on the adaxial surface of the 

costa, this condition contributing to the tip of the apex 

often bent backwards, occasionally the distal third 

region of the leaf becoming long-acuminate and narrow 

above a broader proximal two-thirds, the costa excurrent 

into a longer, more sharply tipped mucro; costa thick,

elatively broad, a brown, rather than stramineous color, 

short-excurrent into a broad, bluntly conic mucro of 3–5 

cells, in leaves of young plants the mucro may be 

extended into a sharp, denticulate subhyaline point, or in 

specimens from regions at higher elevations, the mucro 

may be longe and composed of many cells; adaxial 

surface of the costa covered throughout by quadrate 

papillose cells, absent in a narrow groove at the extreme 

leaf apex; in section prominently rounded due to well 

developed abaxial and adaxial stereid layer and a 

distinctive and well-developed adaxial epidermis 

forming a palisade of often vertically elongated, narrow, 

thin-walled papillose cells throughout the leaf, the 

juncture of lamina and costa occasionally marked by 

incomplete double layers of either laminal, guide or 

adaxial epidermal cells, in cross section lunate and flat 

to convex adaxially, in the proximal median region 

occasionally with a partially bistratose adaxial 

epidermis, guide cells in a partially double row, nearly 

always with at least one extra guide cell abaxial or 

adaxial to the guide cell layer, i.e. in a bistratose guide 

cell pair; proximal laminal cells smooth, mostly thickwalled 

except for the proximal leaf margin, gradually 

differentiated in size and cell-wall thickness across the 

proximal portion, elongate, 6:1, somewhat inflated, 

hyaline-yellowish, rather firm-walled at the insertion, 

appearing bordered at the base by a marginal band of 

thinner-walled, firm to lax elongate cells, often to 7 

cells wide, reduced to one cell wide distally and 

continuing a short distance up the margins, or the 

margin not differentiated except for a proximal marginal 

border of somewhat more delicate and relatively 

elongate cells, distal laminal cells 7–10 µm wide, 

unistratose, marginal cells undifferentiated, evenly 

papillose-crenulate, some leaves variously undulate. 

Asexual reproduction by a fragile lamina. Sexual 

condition dioicous. Perigonia not seen. Perichaetia 

terminal, leaves little differentiated from cauline leaves, 

lamina of the inner perichaetial leaves somewhat 

narrowed apically, the mucro more pronounced. Seta 

dark orange below, becoming paler above, 1.7–1.8 cm 

long. Capsule 3–3.5 mm long, operculum 1–1.2 mm, 

conic, appearing more subulate when immature; 

peristome teeth elongate, ca. 0.5 mm, erect to omewhat 

inclined, orange-red (same color as the capsule), 

densely spiculose. Calyptra: not seen. Spores light 

brown, nearly smooth or distinctly papillose, 9–11 µm. 

Herbaria examined: BUF, CANM, MT, NY, 

NYS, S. 

Smith (1978) cited the substrate (in Britain) 

and the species distribution as: "On basic rocks and 

walls in W. and SW Britain, W Ireland and Jersey. C. 

Europe, S.W. Asia, N. Africa, N. America." It "is the 

most common Tortella species of the Macaronesian 

Islands (May 1986), where it occurs on lava cliffs, on 

sloping rocks, frequently on "calcareous walls" in towns 

(see May 1986 for details). Limpricht (1890) reported it 

as a species of the Mediterranean region as did 

Braunmiller et al. (1971) "with outliers in the warmest 

45 

regions of central Europe...," although these authors 

did not discuss or map its distribution there. 

In Britain, "Recognisable by the leaves tightly 

incurved when dry, the neat, rounded tufts and the 

fragile leaves. Has been confused with Tortella 

flavovirens but differs in habit, habitat and in the 

transition from proximal hyaline to chlorophyllose cells 

being not very abrupt" (Smith 1978). One might expect 

T. nitida to be found in our southern states, associated 

with and in the range of T. flavovirens as it is in the 

warmer parts of Greater Britain, but to date only North 

Temperate stations have been reported for North 

America, while the present study excludes it entirely. 

Braithwaite (Brit. Moss Fl. 1: 250. 37 A. 1887) 

appears to have been the first to include in his 

synonymy of Tortella nitida: "Trichostomum barbula 

(non Schwaeg.) Lange, in Bot. Tids. ii, 235 (1868)." 

Podpera (1954) also cited the same publication in his 

synonymy of Tortella nitida. The Index Muscorum 

citation (Van der Wijk et al. 1958–69) is: 

"Trichostomum barbula Lang., Bot. Tidskr. 2: 235. 

1868 hom. illeg. = Tortella nitida (Lindb.) Broth. fid. 

Grout, Moss Fl. N. Am. 1: 169. 1938." Lange's 

publication is a checklist of the mosses of Tuscany 

"Toscanske Mosser, et bryologisk Bidrag" on pages 

226–254 and on page 235 is a simple statement that 

Trichostomum barbula Schwaegr. occurs in Tuscany. It 

is not clear how authors have presumed Lange to have 

been proposing a nomenclatural change on this page 

having to do with Tortella nitida, a name published only 

a few years earlier. There seems to be no reason to 

assume that Trichostomum barbula Schwaegr. as used 

by Lange was not meant to represent Timmiella 

barbuloides (Brid.) Mönk., a species that also occurs in 

Italy. 

Tortella nitida and Timmiella barbuloides do 

resemble one another. May (1986) for Macaronesia, 

warned that "if one finds "Tortella nitida with 

sporophytes in abundance, one should check for 

Timmiella barbuloides, which looks similar in habit, but 

is easily recognised...by the bistratose, mamillose 

lamina of the leaf." Timmiella barbuloides also has a 

serrate leaf apex. 

The delimitation of Tortella nitida in Europe 

has been problematical for nearly a century. No 

extensive systematic study of Tortella species in Europe 

was undertaken in this study, hence no attempt was 

made to understand or conclusively resolve the 

somewhat uncertain status of this species. It was 

decided that a description based on the type specimen 

and one fruiting specimen that resembled the type in all 

of its characteristics was adequate as a basis for 

discussion, and the least potentially misleading, 

especially as the descriptive tradition in both the New 

and Old World literature is inadequate and contradictory 

in various degrees. Other material from Europe was also 

examined. 

Limpricht (1890) suggested that "the 

resemblance of the habit [of Tortella nitida] to Tortella 

inclinata etc. was the reason that one could long not

agree about this species." He put T. nitida in the genus 

Trichostomum in the basis of the presence of a stem 

central strand. 

The type specimen of Tortella nitida strongly 

resembles the autoicous Tortella humilis except for the 

massive costa, in section with two well-developed 

stereid bands. It is perhaps because of this that 

Limpricht (1890) wondered whether T. nitida might 

possess axillary male buds. Tortella nitida differs from 

T. humilis immediately by its long stem and elongate 

habit, rather than the typically much shorter stem and 

rosulate habit of T. humilis. The cross section of T. 

humilis in the middle of the leaf is adaxially flat. 

Although the type of T. nitida has leaves that appear 

adaxially flat, most specimens are more tubulose, with 

margins erect and broadly incurved. Although both 

species have gradually intergrading proximal cells, 

those of T. nitida are more strongly intergrading: the 

marginal cells in the proximal area are more 

conspicuously set off as a border of lax-walled, 

elongate, hyaline epapillose cells, in some specimens so 

reduced as to cast doubt that the specimen is a Tortella 

at all. Tortella nitida rarely fruits and has short, erect 

peristome teeth, whereas T. humilis, at least in North 

America, is rarely found without fruit, and its peristome 

teeth are long and in one or two spirals. 

Tortella nitida has a distinctive growth-form: 

in hemispherical cushions, rather like some Grimmia 

species, and is frequently represented in herbarium 

specimens as fan-shaped slices from the original polster. 

None of the species of Tortella that occur in North 

America assume this shape except T. inclinata var. 

densa. 

Tortella nitida has frequently been described as 

having a "white" or "pale" costa. Its costa seems, 

however, to be in color no different than others in the 

genus, except that the costa itself is more prominent 

than in other species, especially when the laminae are 

folded together when dry and the whole clump of leaves 

present their broad backs to view. This whiteness or 

paleness apparently is in reference to the "gloss" due to 

the smoothness of the costal abaxial surface. The costae 

of all species in the genus are yellow or yellow-brown 

to orange. 

The leaf cross section in highly colored 

specimens of Tortella nitida shows a palisade of 

quadrate papillose cells on the adaxial surface of the 

costa forming a green layer over the orange costa, 

confluent with the green color of both laminae. The 1-3 

bistratose pairs of guide cells, or incompletely bistratose 

guide cell layer, is characteristic of Pseudosymblepharis 

and some species of Trichostomum (Zander 1993). 

Specimens from Europe show Tortella nitida 

to give little indication of its placement in Tortella 

except for a narrow border of about two to several cells 

width of elongate, thin-walled, smooth hyaline cells, 

otherwise the proximal cells are all thick-walled, 

colored and gradually become differentiated from the 

distal laminal cells. Sometimes the hyaline border is 

46 

wider and distinct, resembling that of Plerurochaete 

squarrosa (Brid.) Lindb. 

The fragility of the leaf of Tortella nitida often 

expresses itself in moderately scalloped or lobed 

margins, with indentations at points of laminal fragility. 

The leaves fragment in rectangular units and along the 

length of the costa such that in many older leaves the 

stem is invested with naked costae divested of the 

laminae. 

None of the common ovate-lanceolate-leaved 

Trichostomum species have fragile leaves. 

Trichostomum tenuirostre, in North Temperate stations, 

may have fragile, tattered generally long-linear leaves 

with a sprawling, loosely tufted habit, occurring with or 

without a stem central strand. Tortella alpicola is 

similarly fragile and is separated by its diminutive size, 

large leaf cells and propagulum at the leaf tip. It is this 

fragility that helps to distinguish ambiguous specimens 

of Tortella nitida from Trichostomum brachydontium or 

Tr. crispulum, since some specimens of the latter 

species may show a tendency toward more delicate cells 

on the margins of an otherwise intergrading proximal 

cell region. 

Although most regional floras do not mention 

it, neither Trichostomum crispulum nor Tr. 

brachydontium are tomentose except at the base, and in 

this characteristic Tortella nitida is similar. 

William Mitten's isotype (NY) of 

Trichostomum diffractum Mitt., now included in the 

synonymy of Tortella nitida, was also examined and, 

although representing rather smaller plants with 

narrower leaves, perhaps characteristic of immaturity, it 

matches Lindberg's type. Other collections made by 

Mitten in Britain also closely resemble the Lindberg 

type. 

The description and illustration of fruiting 

material here is taken from a single specimen: 

Dalmatian plants collected by Jul. Baumgartner 16/III 

1904 (S). Clearly the description and illustration of the 

peristome of this specimen in the present manuscript 

does not match what is described by the following 

sources: "peristome teeth short, rather imperfect, very 

slightly oblique" (Dixon 1924); "peristome short, of 

rather imperfect, very slightly oblique, yellowish, 

papillose teeth" (Haring 1938); "Teeth of the peristome 

very short and irregular, truncate, yellow, papillose" 

(Braithwaite 1887). Mönkemeyer (1927) also indicated 

that the peristome of Tortella nitida was rudimentary. 

Limpricht (1890), in addition, reported that the cells of 

the operculum were straight, the peristome was 

rudimentary, yellow-red and smooth, although Philibert 

described the teeth as papillose, the basal tube scarcely 

or not at all exserted, the teeth linear, scarcely 

developed, the longest to 0.07 mm. He appears to have 

seen a fruiting specimen, although the species 

apparently did not fruit in Germany, Austria or 

Switzerland. 

The peristome teeth in Braithwaite's illustration 

(1887: pl. 37 and reproduced by Haring in Grout 1938: 

169, pl. 81) are very short with respect to the capsule

length, not at all like the illustration provided for the 

present paper, drawn from a collection that displays all 

the gametophyte characteristics of the type of Tortella 

nitida. The Dalmatian peristome in the present paper's 

illustration is no more rudimentary than that of T. 

flavovirens, which is never described as rudimentary; it 

is about half the length of that of T. tortuosa or T. 

humilis. It may be that the peristome of T. nitida is 

variable or that past authors have misidentified the 

specimens they described—more study is warranted. 

As stated above in the discussions under 

Tortella alpicola and T. tortuosa var. fragilifolia, North 

American material seen labeled as T. nitida proved to be 

T. alpicola, T. fragilis, T. tortuosa, Trichostomum 

tenuirostre and Tortella tortuosa var. fragilifolia. 

Some note must be made of a particular kind of 

variation in Tortella nitida in Europe as it may or may 

not contribute information of relevance to patterns of 

variation in T. tortuosa and T. inclinata in North 

America. Although the problem could not be addressed 

in the present study and so concluded to satisfaction, 

there is variation in T. nitida that seems to parallel that 

of T. tortuosa var. fragilifolia and T. rigens, the latter 

possibly a hybrid of T. inclinata with T. fragilis. 

Tortella tortuosa var. fragilifolia and T. rigens, 

primarily in Europe, are characterized by having 

deciduous modifications in the leaf tip. My own 

experience with this variation is that the leaf apices 

become more rigid, elongate (leaf apices becoming 

acuminate and the leaf appearing more linear-lanceolate 

than typical), parallel-sided and fragile. Also they may 

exhibit a single-layered border of elongate, smooth to 

smoother cells, a section showing bistratose juxtacostal 

cells, and the leaf tip may tend to become cylindrical, 

hence the back of the costa may exhibit quadrate 

papillose cells, all of these characteristics most 

strikingly displayed in Tortella fragilis. May (1986) in 

his discussion of T. nitida in Macaronesia described the 

var. irrigata as differing in having a more acuminate 

non-cucullate leaf than the typical variety and that at the 

apex "on the dorsal side in the upper part covered with 

quadrate, papillose lamina cells," and which is 

otherwise as in the typical variety. May also showed 

with excellent illustrations the characters of T. nitida. 

Another variety, T. nitida var. subtortuosa 

(Boul.) Jelenc, may indicate a similar variation, but 

further study is required. Any special deciduous nature 

of the apices of either variety would be obscured by the 

tendency of the species in general toward fragility. Both 

varieties are distinguished from Tortella tortuosa var. 

fragilifolia or T. rigens by the well-defined stem central 

strand. 

Both types of variation in Tortella nitida are 

indicated in a series of specimens examined in loans 

from NY and S and the kind of variation described by 

May above is evident in many of them: these have a 

longer (more multi-cellular), sharply pointed denticulate 

mucro, rather than the short, blunt, smooth, conical 

mucro of the typical plant and the two types examined, 

and the leaves have long-acuminate apical leaf-quarters 

47 

with approximately parallel sides. Again, it appears 

as though tendencies toward apical modifications that 

resemble those of Tortella fragilis are present in T. 

nitida. 

Boulay (1884: 445) was reported by Limpricht 

(1890) to distinguish three varieties of what is now 

called Tortella nitida: Tortella nitida var. obtusa (Boul.) 

Jelenc; T. nitida var. media (Boul.) Corb.; and (T. nitida 

var. subtortuosa (Boul.) Jelenc. What Limpricht did not 

say was that Boulay described these varieties as part of 

the variation of what is now considered to be Tortella 

flavovirens (i.e. Trichostomum flavovirens), not Tortella 

nitida. I doubt one should assume that these varieties 

actually represent significant variation in the 

morphology of T. nitida. They were originally intended 

to describe variation in the (now known to be) coastally 

restricted T. flavovirens. 

Upon examination of a series of European 

specimens of Tortella nitida from NY and S, it is 

possible to speculate that this species may be sensitive 

to altitudinal variation, perhaps similar to the variation 

here attributed to T. inclinata s.l. The further away from 

the Alps and the nearer to sea level, especially in the 

peripheral part of the range, the more the species 

resembles the types both of Lindberg and of Mitten with 

ligulate-lanceolate leaves with generally obtuse to 

broadly acute apices. The Lindberg type derived from 

the Gibraltar area and the Mitten type from England. 

With higher altitude the leaves undergo an elongation 

and elaboration of the leaf apex: the proximal region 

stays the same, but the distal third of the leaf becomes 

acuminate, with quadrate cells on the back of the costa, 

with a long denticulate mucro and so forth. The types 

thus may represent only variation at low altitudes, 

including islands in the Atlantic and the Mediterranean, 

coastal plains and interior plains regions rather than 

upland regions on the continent of Europe. Not enough 

material has been examined here to substantiate this. 

The smallest plants (in stem length) in 

specimens were seen from British stations. They were 

hard to separate from Trichostomum, especially Tr. 

crispulum and Tr. brachydontium, but at least some 

leaves had a proximal margin of cells that were longer 

and more lax than the inner proximal cells, and this 

margin extended above the proximal region. They had 

indented or scalloped laminal margins in the distal 

region of the leaf, indicating that the leaves were fragile 

and not simply eroded. The excurrent costa in these 

small forms is of a few cells and resembles more of an 

apiculus than a mucro. 

Belgian specimens were of typical and 

intermediate forms. In Switzerland, one sees plants with 

typical leaves (resembling those of Tortella humilis in 

shape: ligulate-lanceolate or broadly lanceolate) in the 

lower part of the stem. However, in the distal part of the 

stem, the younger leaves start to become more abruptly 

acuminate from a broader proximal 2/3 region. The 

distal acumination is further characterized by greater 

modifications for fragility (marginal indentations). The 

costa on the abaxial side has chlorophyllose, quadrate,

papillose cells (the var. irrigata, according to May 1986) 

and a longer mucro. This type of apical region may 

detach as a chlorophyllose propagule. 

Occasional specimens have been confused with 

Weissia species, whose proximal cells can resemble 

those of Tortella nitida, as a vaguely defined extension 

of the proximal marginal cells a short way up the 

lamina. While T. nitida can have broadly incurved distal 

margins in part, and sometimes this seems strong in the 

extreme apex which then appears cucullate, species of 

Weissia will have tightly and narrowly inflexed margins 

throughout the entire distal 3/4 of the leaf. 

Specimens of Trichostomum brachydontium 

from Europe are often confused with Tortella nitida. In 

the latter species, however, the leaves are generally 

tubulose and the apex of at least some leaves appears or 

is subcucullate. In the former species, the leaf is flat 

throughout or is bounded by relatively sharply or 

abruptly upturned margins. The costa of Tr. 

brachydontium is more excurrent than that of Tortella 

nitida, although appearing thick as is typical in that 

species, and regularly comes to a smooth sharp point 

like a hard spike. While examining specimens from S, 

there was a plant (Italy, A. Bottini May 1882) with the 

herbarium name "Tortula nitida var. acuminata 

Renauld," which had leaves with narrow apices and 

robust excurrent denticulate costae (see illustration), but 

whose other characteristics are typical. Trichostomum 

brachydontium is not fragile, its laminal margins are 

straight without indentations or scalloping, and it is 

difficult to demonstrate any marginal border of 

proximal cells in any leaf. 

In North America, doubt must be cast on the 

likelihood that Haring (1938) saw North American 

fruiting material or even that of Europe, and her 

descriptions of the peristome seem to derive from the 

literature, particularly of Braithwaite (1887). This 

conjecture is supported by the rarity of fruiting material 

of the species in Europe at the time of her publication: 

"...The fruit has only been found at 

Angouleme, in France, by M. Philibert in 1867, for 

some of whose specimens I am indebted to the kindness 

of M. Husnot" (Braithwaite 1887: 250–251, pl. 37A). 

Dixon acknowledged he had a "slight acquaintance with 

the fruiting plant" to hinder his final judgment on the 

placement of Tortella nitida: "indeed, it is probable that 

it will remain more or less doubtful until an opportunity 

is afforded of studying the fruit in good condition and in 

greater quantity. Hitherto it has only been found in two 

localities, the two plants presenting certain differences 

of some importance in the structure of their fruit and 

peristome..." (Dixon 1924). Since Dixon indicated that 

the species is sterile in Britain, one must assume that the 

two localities he mentioned are of the fruiting material, 

and perhaps they are from the same locality as 

Braithwaite's, if not the same specimens sent by 

Philibert. 

One might wonder if Philibert's specimens 

were in fact Trichostomum brachydontium Bruch., a 

species with a rudimentary peristome, but having no 

48 

fragility in the leaves, having lax, thin-walled 

proximal cells. It has a relatively abrupt transition 

between distal laminal cells and proximal cells, 

something Tortella nitida would not. Mitten (1868) 

indicated both Tortella flavovirens and T. nitida to 

closely resemble Trichostomum brachydontium. The 

most obvious differentiating character of Tr. 

brachydontium is the horizontal or weakly U-shaped 

line that defines the distal limit of the hyaline proximal 

cells. 

In this context it seems interesting that Mitten 

also noted Tortella flavovirens to be more close in 

resemblance to Tortella humilis "than any other moss." 

He did not indicate that Tortella nitida is also close. 

Mitten also indicated the close resemblance of Tortella 

nitida (= Trichostomum diffractum) with Trichostomum 

brachydontium. Also, the closeness of Tr. 

brachydontium with Tortella flavovirens indicates that 

the three form a natural kind of group: all three have 

stem central strands, all three have more or less vaguely 

differentiated proximal cells, all three have different 

peristomes (shorter and erect to rudimentary) than 

typical in the genus Tortella, which has long, once to 

twice spiralled peristome teeth. 

The name Tortella nitida as used by American 

authors (Haring 1938; Flowers 1973; Crum & Anderson 

1981) refered to North American plants all with 

abruptly differentiated proximal cells, and therefore 

which belong to other species. Crum and Anderson 

(1981) discussed North American reports of Tortella 

nitida, including those made by Haring in Grout's Moss 

Flora, as Tortella tortuosa var. nitida (Lindb.) Pilous. 

However, the structure of the stem in cross section with 

its distinctive central strand, the bistratose pairs in the 

guide cell layer of the leaf section, the gradually 

differentiated proximal cells, not to mention the 

anomalous peristome is a good basis for separating true 

T. nitida from T. tortuosa. Tortella tortuosa never 

develops the ligulate-lanceolate leaf at least of the type 

specimen of T. nitida. 

A specimen from Alaska collected by Hermann 

(21727, NY) identified as Tortella nitida is 

Trichostomum tenuirostre. Most of this collector's other 

specimens of T. nitida from North America are Tortella 

alpicola. 

The specimen of Tortella nitida from the 

Northwest Territories cited by Steere and Scotter (1978) 

76–620a could not be located at NY where the specimen 

was said to be deposited (Steere 76–620 is Lophozia 

badensis, W. Buck, in litt.). The authors acknowledge in 

their paper that, according to Crum et al. (1973) "some 

doubt is cast...on whether it is present in North America 

at all." The species was excluded from the North 

American flora by Crum et al. (1973). Although Crum 

and Anderson later (1981) mention a specimen of 

Tortella nitida from the Nahanni Preserve, they do not 

comment on it further, perhaps because the specimen 

was not available for verification at that time either. If 

that particular collection followed the concept of Haring

(1938) or Flowers (1973), it was most likely Tortella 

alpicola or T. tortuosa var. fragilifolia. 

A citation of Tortella nitida as a literature 

report by Ireland et al. (1987) for the Northwest 

Territories is probably a reference to the Steere and 

Scotter paper (1978). A specimen determined to be 

Tortella nitida from Saskatchewan (Willow Bunch, 

Armand Filion, 13 Aug. 1927 MT) is Tortula 

mucronifolia Schwaegr. One may with confidence 

exclude this species from the moss flora of Canada as 

well as the United States. 

EXCLUDED TAXA 

Tortella mollissima Broth. ex Bartr. 

The specimen, kindly loaned by P. L. 

Redfearn, (Missouri: Douglas Co., along Greasy Creek, 

2.5 miles SE of Ann, P. Redfearn 21330 (SMS), 

reported as Tortella mollissima Broth. ex Bartr. by 

Redfearn (1969) is Trichostomum tenuirostre (Hook. & 

Tayl.) Lindb. Zander (1994) has made Tortella 

mollissima (with a Mexican type) a synonym of 

Pseudosymblepharis schimperiana (Par.) Crum, a 

species of Mexico, Central America and Venezuela. 

Tortella nitida—see discussion above. 

ACKNOWLEDGEMENTS 

This paper is respectfully presented in honor of 

George Forman Goodyear in tribute of a lifetime of 

distinguished cultural service to the people of Buffalo, 

New York. Michael Smith, Director of the Buffalo 

Museum of Science, and Richard Zander, Curator of the 

Clinton Herbarium allowed me research facilities 

throughout the course of this study. 

I am grateful to Richard H. Zander for critical 

assistance throughout the preparation of this manuscript 

and for permitting access to an unpublished manuscript 

treatment of the genus Tortella in Arctic North 

America, and for help in translations from the French. 

Gert Steen Mogensen gave information useful in the 

evaluation of Tortella arctica, critical bibliographical 

assistance and help with translations from the Swedish. 

I also thank R. Zander for kindly sharing with me 

detailed information provided by Hannes Hertel of the 

Botanische Staatssammlung (M), Munich, for data 

critical to designating a lectotype and for the loan of 

specimens. A. C. Crundwell is thanked for useful 

comments on various species in the genus. 

Loans made by the curators of herbaria at the 

University of Alaska, the New York Botanical Garden 

(NY); the New York State Herbarium, Albany (NYS); 

the University of Michigan, Ann Arbor (MICH); the 

Missouri Botanical Garden (MO); Duke University 

(DUKE); the National Museums of Canada, Ottawa 

(CANM); the Herbier Marie-Victorin, Universit‚ de 

Montr‚al (MT); the Mus‚um National d'Histoire 

Naturelle, Paris; herbarium of the University of Alberta 

at Edmonton (ALTA); that of the University of 

Colorado (COLO), the National Science Museum, 

49 

Tokyo (TNS); the University of Florida, Gainesville 

(FLA); the Swedish Museum of Natural History, 

Stockholm (S); the Farlow Cryptogamic Herbarium 

(Harvard University), the herbarium of the University of 

British Columbia (UBC), the Herbarium of the 

University of Alaska (ALA), and the Botanical Institute, 

Christchurch, New Zealand (CHR) are gratefully 

acknowledged. Special acknowledgement is made of the 

outstanding service of the curators at NY, cryptogamic 

collections, in the loan of critical specimens. 

BIBLIOGRAPHY 

AND SELECTED REFERENCES 

Abramova, A. L., L. I. Savicz-Ljubitzkaia and Z. N. 

Smirnova. 1961. Manual of the Leafy Mosses 

of the Arctic U.S.S.R. V.L. Komarov Botanical 

Institute, Acad. Sci. USSR, Moscow, 

Leningrad. 

Albertson, N. 1946. Österplana Hed ett Alvaromr†de p† 

Kinnekulle. Uppsala. 

Anderson, L. E., H. A. Crum & W. R. Buck. 1990. List 

of the mosses of North America north of 

Mexico. Bryologist 93(4): 448–499. 

Barnes, C. R. 1897. Analytic Keys to the Genera and 

Species of North American Mosses., revised by 

F. D. Heald, University of Wisconsin Press, 

Madison Wisconsin. 

Boulay, J. N., Abbe. (1884). Muscinees de la France. 

Paris 1884–1904. 2 parts: Premiere partie: 

Mousses. Nat. Nov. Sep 1884. 

Braithwaite, R. 1887–1905. The British Moss Flora. 

London. 3 vols. 

Belland, R. J. & G. R. Brassard. 1981. New or 

additional moss records from Newfoundland 

VII. Bryologist 84: 560–563. 

Braunmiller, H., J. Poelt, W. Schultze-Motel. 1971. šber 

die verbreitung einiger arten der 

laubmoosgattung Tortella in Mitteleuropa. 

Arch. Naturschutz und Landschaftsforsch, 

11(3): 169–178. 

Brotherus, V. F. 1924–1925. Musci. In A. Engler & K. 

Prantl, Die nat rlichen Pflanzenfamilien, Ed. 

2. 10: 1–478; 11: 1–542. Leipzig. 

Catling, P. M. & V. R. Brownell. 1995. A review of the 

alvars of the Great Lakes region: distribution, 

floristic composition, biogeography and 

protection. Can. Field-Nat. 109(2): 143–171. 

Chen, P.-c. 1941. Studien ber die ostasiatischen Arten 

der Pottiaceae, I–II. Hedwigia 80: 1–76; 141– 

322. 

Churchill, S. P. 1989. Bryologia Novo Granatensis. 

Estudios de los musgos de Colombia. Tropical 

Bryology 1: 95–132. 

Churchill, S. P. & E. L. Linares C. 1995. Prodromus 

Bryologiae Novo-Granatensis. Instit. Cienc. 

Nat., Mus. Hist. Nat. Bibliot. "Jose Jeronimo 

Triana" 12. 2 Vols.

Crum, H. A. & L. E. Anderson. 1958. Taxonomic 

studies in North American mosses. VI-XIII. J. 

of the Elisha Mitchell Sci. Soc. 74: 32. 

Crum, H. A. & L. E. Anderson. 1981. 2 Vols. Mosses of 

Eastern North America. Columbia University 

Press, New York. 

Crum, H. A. & E. B. Bartram. 1958. A Survey of the 

Moss Flora of Jamaica. Institute of Jamaica, 

Science Series Bulletin 8. 

Crum, H. A. & W. C. Steere. 1957. Scientific Survey of 

Porto Rico and the Virgin Islands Vol. VII. 

Park 4. New York Academy of Sciences. 

Crum, H. A. & W. C. Steere & L. E. Anderson. 1973. A 

new list of mosses of North America, north of 

Mexico. Bryologist 76: 85–130. 

Crundwell, A. C. & E. Nyholm. 1962. Notes on the 

genus Tortella. I. T. inclinata, T. densa, T. 

flavovirens and T. glareicola. Trans. Brit. 

Bryol. Soc. 4: 187–193. 

Crundwell, A. C. & E. Nyholm. 1963. Notes on the 

genus Tortella. II. Tortella arctica. Bryologist 

66: 184–191. 

Delgadillo M., C., B. Bello & A. C rdenas. 1995. 

LATMOSS: A Catalogue of Neotropical 

Mosses. Monographs in Systematic Botany 

from the Missouri Botanical Garden 56: 1–191. 

Demaret, F. & E. Castagne. 1964. Flore Generale de 

Belgique. Bryophytes. Vol. 2, Fascicle 3. 

Ministere de L'Agriculture, Jardin Botanique 

de L'Etat. Bruxelles. 

Dixon, H. N. 1924. The Student's Handbook of British 

Mosses. London. 

Dixon, H. N. 1930. Additions to the Moss Flora of the 

North-Western Himalayas. Annal. Bryol. 3: 

51–70. 

Düll, R. 1984. Distribution of the European and 

Macaronesian Mosses (Bryophytina). Part I. 

Bryologische Beitr„ge Vol. 4. I. Düll- 

Hermanns Verlag. Rheurdt, Germany. 

Düll, R. 1992. Distribution of the European and 

Macaronesian Mosses (Bryophytina) 

Annotations and Progress. Bryologische 

Beitr„ge Vol. 8/9. I. Düll-Hermanns Verlag. 

Düll, R. & L. Meinunger. 1989. Deutschlands Moose. 

IDH Verlag, Bad M nstereifel Ohlerath. 

Eckel, P. M. 1991. Tortella tortelloides (Musci: 

Pottiaceae) new to North America. Bryologist 

94: 84–87. 

Eckel, P. M. 1996. Trichostomum tenuirostre var. 

gemmiparum (Schimp.) Zand. new to Japan. 

Bryologist 99(4): 460. 

Eckel, P. M. 1997. The moss Tortella alpicola new to 

Alberta and the Yukon Territory with a 

discussion of its range and comments on 

related species. Can. Field-Nat. 3: 320–322. 

Eckel, P. M. 1997. Tortella knightii, an Australasian 

species new to South America in Tierra del 

Fuego. Bryologist 100(2): 210–211. 

Fife, A. J. 1984. Records of new or otherwise 

interesting mosses in New Zealand, including a 

50 

new species of Racomitrium. New Zealand 

Journal of Botany 22: 1–6. 

Fife, A. J. 1995. Checklist of the mosses of New 

Zealand. Bryologist 98(3): 313–337. 

Flowers, S. 1973. Mosses: Utah & the West. Brigham 

Young Univ. Press, Provo, Utah. 

Gove, P. B., ed. & editorial staff. 1976. Webster's Third 

New International Dictionary of the English 

Language Unabridged. G. & C. Merriam Co, 

Springfield, Mass. 

Haring, I. M. 1938. Tortella. In A. J. Grout, Moss Flora 

of North America. I(3): 165–170. Newfane, 

Vermont. 

Hilpert, F. 1933. Studien zur Systematik der 

Trichostomaceen. Dresden. 

Hoe, W. J. 1974. Annotated checklist of Hawaiian 

mosses. Lyonia 1: 1–45. 

Holmgren, P. K., N. H. Holmgren & L. C. Barnett. 

1990. Index Herbariorum. Part 1. Edition 8. 

New York Botanical Garden, Bronx, New 

York. 

Hyvönen, J. 1991. Tortella fragilis (Pottiaceae) reported 

for southern South America. Bryologist 94: 

416–418. 

Ignatov, M. S. & O. M. Afonina. 1992. Check-list of the 

Mosses of the Former USSR. Arctoa 1: 1–85. 

Ireland, R. R. 1982. Moss Flora of the Maritime 

Provinces. National Museums of Canada 

Publications in Botany 13. Ottawa. 

Ireland, R. R., G. R. Brassard, W. B. Schofield & D. H. 

Vitt. 1987. Checklist of the mosses of Canada 

II. Lindbergia 13: 1–62. 

Juratzka, J. 1882. Die Laubmossflora von Österreich- 

Ungarn, J. Breidler & J. B. Förster, eds. 

Vienna. 

Ketchledge, E. H. 1957. Checklist of the Mosses of 

New York State. New York State Museum 

Bull. 363. State Education Department, 

Albany, New York. 

Ketchledge, E. H. 1980. Revised Checklist of the 

Mosses of New York State. New York State 

Museum Bull. 440, R. S. Mitchell, ed. State 

Education Department, Albany, New York. 

Kindberg, N. C. 1897. Species of European and N. 

American Bryineae (Mosses. Vol.2, Part 2. 

Linköping, Sweden. 

Kramer, W. 1980. Tortula Hedw. sect. Rurales De. Not. 

(Pottiaceae, Musci) in der Ostlicher Holarktis. 

Cramer, Vaduz. 

Lange, M. T., Pr„st: Toscanske Mosser, et bryologisk 

Bidrag. Botanisk Tidsskrift, Copenhagen, 2: 

226–254. 

Latzel, A. 1934. Beitrag zur Kenntnis der Moose des 

Komitats Baranya. Magyar botanikai lapok. 

Ungarische botanische Blätter. 33: 160–191. 

Limpricht, K. G. 1890. Die Laubmoose Deutschlands, 

Österreichs und der Schweiz. 1: 1–836. 

Leipzig. 

Macoun, J. & N. C. Kindberg. 1892. Catalogue of 

Canadian Plants. Part. VI. Musci. Geological

and Natural History Survey of Canada. 

Montreal. 

Magill, R. E. 1981. Bryophyta. Part 1 Mosses. Fascicle 

1. Sphagnaceae-Grimmiaceae. In O. A. 

Leistner (ed.), Flora of Southern Africa. 

Pretoria. 

Matteri, C. M. 1985. Catálogos de las briófitas. 

Catálogo de los musci: 265–298. In O. 

Boelcke, D. M. Moore & F. A. Roig (eds.), La 

Transecta Botánica de Patagonia austral. 

Consejo nacional de Investigaciones 

Cientificas y Ténicas (Argentina), Instituto de 

la Patagonia (Chile), Royal Society (Gran 

Bretaña). Buenos Aires. 

Matteri, C. M. 1986. Overview of the phytogeography 

of the moss flora from southern Patagonia, at 

51°–52° South Latitude. J. Hattori Bot. Lab. 

60: 171–174. 

May, R. 1986. Notes on some Macaronesian Tortella 

species. Bryologische Beiträge, R. Dull, ed., 

Duisberg. 6: 58–66. 

McIntosh, T. T. 1986. The Bryophytes of the Semi-Arid 

Steppe of South-Central British Columbia. 

Dissert., University of British Columbia, 

Vancouver, B.C., Canada. 

Meylan, Ch. 1921. Nouvelles contributions a la flore 

bryologique du Jura. Rev. Bryol. 48: 1–5. 

Miller, N. G. 1973. Lateglacial plants and plant 

communities in northwestern New York State. 

J. Arnold Arboretum. 54: 123–159. 

Miller, N. G. 1976. Quaternary fossil bryophytes in 

North America: a synopsis of the record and 

some phytogeographic implications. J. Hattori 

Bot. Lab. 41: 73–85. 

Mitten, W. 1868. New or rare British mosses. J. Bot. 

London. 6: 97–99, plate 77. 

Mönkemeyer, W. 1927. Die Laubmoose Europas. Vol. 

4. In L. Rabenhorst, Kryptogamenflora von 

Deutschland, Österreich und der Schweiz. 

Leipzig. 

Muller, Karl. 1878–79. Prodromus Bryologiae 

Argentinicae. I. Linnaea 42: 217–460. 

Noguchi, A. 1988. Illustrated Moss Flora of Japan. Part 

2. Hattori Botanical Laboratory, Obi, 

Nichinan-shi. 

Nyholm, E. 1956. Illustrated Moss Flora of 

Fennoscandia. II. Musci. Fasc. II. Gleerups, 

Lund, Sweden. 

Nyholm, E. 1989. Illustrated Flora of Nordic Mosses. 

Fasc. 2. Pottiaceae—Splachnaceae— 

Schistostegaceae. Nordic Bryological Society. 

Copenhagen. Paul, H. 1943. Nachträge und 

Bermerkungen zur Moosflora Beyerns. Ber. 

Bayer. Bot. Ges. 26: 118–133. 

Persson, H. 1947. Bryum arcticum och några andra 

mossfynd från Stora Karlsö. Svensk Bot. 

Tidskr. 41, I: 141–150. 

Persson, H. & W. A. Weber. 1958. The bryophyte flora 

of Mt. McKinley National Park, Alaska. 

Bryologist, 61: 214–42. 

51 

Pilous, Z. 1965. Fragmenta bryologica 56. Tortella 

rigens Albertson und Tortella densa Crund. et 

Nyh., zwei neue tschechoslowakische Moose. 

Preslia 37: 20–22. 

Podpera, J. 1954. Conspectus Muscorum Europaeorum. 

Nakladatelstvi Ceskoslovenske Akademie Ved 

Praha. 

Polunin, N. 1947. Botany of the Canadian Eastern 

Arctic. Part II Thallophyta and Bryophyta. 

Canada Department of Mines and Resources, 

National Museum of Canada Bull. 97, 

Biological Series 26. 

Redfearn, P. L., Jr. 1969. Bryophytes of the Interior 

Highlands XIV. Tortella mollissima new to 

North America. Bryologist 72: 60. 

Redfearn, P. L., Jr. 1993. List of Mosses of China. Flora 

Online (Buffalo Museum of Science, Buffalo, 

NY.) 29 (ver.1.0). Reimers, H. 1951. Beiträge 

zur Kenntnis der Bunten Erdflechten- 

Gesellschaft. I.–II. Ber. deutsch. bot. Ges. 63: 

147–156 

Renauld, F. & J. Cardot. New mosses of North America. 

II. Bot. Gaz. 14(4): 91–100. 

Robinson, H. E. 1972. Observations on the origin and 

taxonomy of the Antarctic moss flora. In G. A. 

Llano (ed.), Antarctic Terrestrial Biology, 

Antarctic Research Ser. 20: 163–177. 

American Geophysical Union, Washington, 

D.C. 

Rubers, W. 1973. [In Dutch]. Enkele opmerkingen over 

Nederlandse Tortella's. Lindbergia 2: 128–129. 

Sainsbury, G. O. K. 1955. A handbook of the New 

Zealand mosses. Royal Society of New 

Zealand Bull.5: 1–490. 

Saito, K. 1975. A Monograph of Japanese Pottiaceae 

(Musci). J. Hattori Bot. Lab. 39: 373–537. 

Savicz-Ljubitzkaja, L. I. & Z. N. Smirnova. 1970. 

[Handbook of the Mosses of the U.S.S.R. The 

acrocarpous mosses.] Acad. Sci. USSR, 

Komarov Botanical Institute, Leningrad. [text 

in Russian]. 

Schimper, W. P. 1855. Corollarium Bryologiae 

Europaeae, Conspectum Diagnosticum 

Familiarum, Generum et Specierum. Stuttgart. 

Schofield, W. B. 1965. Correlations between the moss 

floras of Japan and British Columbia, Canada. 

J. Hattori Bot. Lab. 28: 17–42 

Seki, T. 1974. A moss flora of Provincia de Ais‚n, 

Chile. Results of the second scientific 

expedition to Patagonia by Hokkaido and 

Hiroshima Universities, 1967. J. Sci. 

Hiroshima Univ. Ser. B, Div. 2 (Bot.) 15: 9– 

101. 

Sharp, A. J., H. Crum & P. M. Eckel, eds. 1994. Moss 

Flora of Mexico. Mem. New York Bot. Gard. 

69. 2 vols. 

Smith, A. J. E. 1978. The Moss Flora of Britain and 

Ireland. Cambridge University Press, 

Cambridge.

Stafleu, A. & R. S. Cowan. 1981. Taxonomic literature. 

Vol. III: Lh–O. ed. 2. The Hague, Utrecht. 

Steere, W. C. 1939. Tortula, pp. 228–246 In A. J. Grout 

(ed.), Vol.I, Moss Flora of North America 

north of Mexico. Published by the author, 

Newface, Vermont. 

Steere, W. C. 1940. Tortula in North America north of 

Mexico. Bryologist 43: 12–23, 45–56, 76–86, 

98–109. 

Steere, W. C. 1976. Ecology, phytogeography and 

floristics of arctic Alaskan bryophytes. J. 

Hattori Bot. Lab. 41:47–72. 

Steere, W. C. 1978. The Mosses of Arctic Alaska. J. 

Cramer. Vaduz. 

Steere, W. C. & G. W. Scotter. 1978. Additional 

bryophytes from Nahanni National Park and 

vicinity, Northwest Territories, Canada. Canad. 

J. Bot. 56: 234–244. 

Townsend, C. C. 1965. Bryophytes from Cyprus. Rev. 

Bryol. et Lich‚nol. NS 33: 484–493. 

Van der Wijk, R., W. D. Margadant & P. A. Florsch tz. 

1959–1969. Index Muscorum. Reg. Veg. Vols. 

17, 26, 33, 48, 65. 

Vitt, D. H. 1974. A key and synopsis of the mosses of 

Campbell Island, New Zealand. New Zealand 

J. Bot. 12: 185–210. 

Wareham, R. T. 1939a. Pottia. In A. J. Grout, Moss 

Flora of North America North of Mexico. 1(4): 

197–208. 

Weber, W. A. 1973. Guide to the Mosses of Colorado. 

Institute of Arctic and Alpine Research, 

University of Colorado, Occ. Paper 6. 

Williams, C. 1968. New and additional moss records for 

Ontario. Bryologist Vol.71: 282–284. 

52 

Zander, R. H. 1977. The tribe Pleuroweisieae 

(Pottiaceae, Musci) in Middle America. 

Bryologist 80: 233–269. 

Zander, R. H. 1978. A propaguliferous variant of 

Oxystegus tenuirostris in Europe, India, 

Mexico, Cuba and Brazil. Lindbergia 4(3–4): 

285–288. 

Zander, R. H. 1993. Genera of the Pottiaceae: Mosses of 

Harsh Environments. Bull. Buffalo Soc. Nat. 

Sci. 32. 

Zander, R. H. 1994a. Bryoerythrophyllum, pp. 273–283. 

In A. J. Sharp, H. A. Crum and P. M. Eckel 

(eds.), Moss Flora of Mexico. Mem. New York 

Bot. Gard. 69. 2 vols. 

Zander, R. H. 1994b. Oxystegus, pp. 233–235. In A. J. 

Sharp, H. A. Crum and P. M. Eckel (eds.), 

Moss Flora of Mexico. Mem. New York Bot. 

Gard. 2 vols. 

Zander, R. H. 1994c. Streptocalypta, pp. 238–242. In A. 

J. Sharp, H. A. Crum and P. M. Eckel (eds.), 


Gard. 69. 2 vols. 

Zander, R. H. 1994d. Tortella, pp. 235–238. In A. J. 

Sharp, H. A. Crum and P. M. Eckel (eds.), 


Gard. 69. 2 vols. 

Zander, R. H. 1997. Merceyoideae and 

Trichostomoideae. In G. S. Mogensen (ed.), 

Illustrated Moss Flora of Arctic North America 

and Greenland. Meddel om Groenland, 

Bioscience (in press). 

Zander, R. H. & W. J. Hoe. 1979. Geographic 

disjunction and heterophylly in Tortella 

fragilis var. tortelloides (= Sarconeurum 

tortelloides). Bryologist 82: 84–87. 

!" #$$%& ' ( )" ! * +,- 

. / / 

, 0 , 12##34#$#& ) * 

55 5 5& 5#& 

PLATES 1–11 FOLLOW

Plate 1. Tortella humilis. 1. Habit, wet and dry, short (typical) form. 2. Habit, wet and dry, elongate 

(japonica) form from British Columbia. 3. Stem cross section. 4. Five leaves from typical populations, central U.S. 

5. Four leaves from elongate populations, Texas. 6. Four leaves from elongate populations (as Tortella japonica), 

Japan. 7. Four leaves from elongate populations, Mexico. 8. Two leaves of elongate, robust specimen from Ontario. 

9. Five elongate, robust leaves from single stem, the type (at PC) of Tortella japonica, Japan. 10. Leaf apex. 11. 

Leaf cross sections, lower median to leaf apex. 12, 12. Stalked perigonial buds, two single, one double. 13. 

Sporophyte and calyptra. 14. Sporophyte and operculum. 15. Capsule and peristome. 

53

Plate 2. Tortella flavovirens. 1. Habit, wet and dry. 2. Stem cross section. 3. Four stem leaves. 4. Leaf cross 

section series from lower middle to apical region. 5. Leaf apex. 6. Leaf base. 7. Perichaetial leaves and base of seta. 

8. Two outer perichaetial leaves. 9. Peristome. 

54

Plate 3. Tortella alpicola. 1. Sterile habit, wet and dry. 2. Fertile habit, wet. 3. Stem cross section. 4. Six 

leaves. 5. Distal leaf margin showing scalloped indentations. 6. Leaf base. 7. Three cross sections of sterile stem 

leaves. 8. Three cross sections of fertile stem leaves. 9. Three views of propagula formation at stem apex. 10. Three 

examples of propagula, including apex. 11. Five perichaetial leaves. 

55

Plate 4. Tortella tortuosa var. tortuosa. 1. Habit, wet and dry of perichaetiate plant, showing differentiated 

perichaetial leaves. 2. Stem cross section 3. Eight leaves showing variation in size and recurvature. 4. Youngest 

leaves at stem apex showing long mucro emergent. 5. Leaf apex. 6. Leaf base. 7. Leaf cross sections from medial to 

distal region. 8. Perigonium and leaves. 9. Perichaetial leaves. 10. Areolation of some perichaetial leaves showing 

extensive border of smooth, elongate cells. 

56

Plate 5. Tortella tortuosa var. arctica. 1. Habit, wet and dry. 2. Stem cross section. 3. Five stem leaves. 4. 

Leaf base. 5,5. Leaf cross sections. 6. Marginal cells at junction of proximal cells and distal laminal cells. 7. 

Perichaetium. 8. Perichaetial leaf. 

57

Plate 6. Tortella tortuosa var. fragilifolia. 1. Habit, dry. 2. Stem cross section with central strand. 3. Seven 

stem leaves. 4. Marginal cells. 5. Leaf cross sections from various specimens showing various aberrations: bistratose 

juxtacostal areas (frequently asymmetrical), bistratose patches, loss of ventral stereid toward the leaf apex and 

undifferentiated costa. 6. Apex of fertile stem, Quebec specimen. 7. Perichaetial leaves. 

58

Plate 7. Tortella fragilis. 1. Habit, wet and dry. 2. Habit of reduced stem, Quebec. 3. Stem cross section. 4. 

Two subapical stem leaves. 5. Leaves at stem apex showing emergent propaguloid apex. 6. Three leaves not forming 

propagula, from branch innovations. 7. Three leaves of reduced stem, Quebec. 8. Emergent propaguloid leaf at stem 

apex in reduced plant, Quebec. 9. Apex of propaguloid leaf. 10. Two apices of non-propaguloid branch innovation 

leaves. 11. Distal margin showing smooth marginal border of elongate cells. 12. Seven leaf cross-sections from 

median to apical (propaguloid) regions. 

59

Plate 8. Tortella inclinata var. inclinata (1–10). 1. Habit, wet and dry. 2. Habit showing apical whorl. 3. 

Stem cross section. 4. Five leaves. 5. Leaf apex. 6. Elongate cells on the ventral surface of the costa in surface view. 

7. Four leaf cross sections. 8. Apical portion of fertile stem. 9. Capsule. 10. Inner perichaetial leaves. Tortella 

rigens (11–13): Ohio and New York plants. 11. Three leaves from New York specimen. 12. Three leaves from Ohio 

specimen. 13. Leaf apex, Ohio specimen. 

60

Plate 9. Tortella inclinata var. densa. 1. Habit, wet, showing maturing leafy lateral shoot. 2. Stem apex, 

dry. 3. Stem cross section. 4. Eight stem leaves. 5. Leaf cross sections. 6. Leaf apex. 7. Lateral shoot with primordial 

leaves. 8. Perichaetium. 9. Dissected perichaetium showing perichaetial leaves and archegonia. 10. Three leaves 

with narrowly acute apices. 

61

Plate 10. Tortella rigens: European specimens. 1. Habit, wet and dry. 2. Stem cross section. 3. Stem leaves, 

showing variation from short, cucullate-leaved forms to variously acuminate forms with differentiated apical 

regions. 4. Two leaf apices. 5. Cells at midleaf. 6. Leaf cross sections showing bistratose juxtacostal areas and 

broadly channeled aspect. 

62

Plate 11. Tortella nitida: all figures are drawn from Lindberg's type (sterile) at S except where noted, see 

text. 1. Habit, wet and dry. 2. Stem cross section. 3. Stem leaves. 4. Leaf of var. irrigata, stippled area showing 

region of quadrate, papillose cells on the back of the costa and two presumably deciduous apices from leaves on the 

same stem (non-type). 5. Two leaf apices, side view. 6. Leaf apex, front view. 7. Leaf apex of "var. acuminata" 

(non-type, see text). 8. Leaf base. 9. Stem leaf cross sections from median region to near apex. 10. Perichaetial 

leaves and base of seta (non-type). 11. Capsules operculate and deoperculate (non-type). 12. Peristome (non-type). 

63

re-evaluation of tortella - Missouri Botanical Garden

Create successful ePaper yourself

Delete template?

Save as template?