SYMBIOSIS (2009) 47, 61–76 ©2009 Balaban, Philadelphia/Rehovot ISSN 0334-5114
The ecology and evolution of fly dispersed dung mosses
(Family Splachnaceae): Manipulating insect behaviour
through odour and visual cues
Paul Marino1*, Robert Raguso2, and Bernard Goffinet3
1Department of Biology, Memorial University, St. John’s, NL A1B 3X9, Canada, Tel. +1-709-737-7497,
2Department of Neurobiology and Behavior, Seeley G. Mudd Hall, Cornell University, Ithaca, NY 14853-2702, USA;
3Department of Ecology and Evolutionary Biology, 75 N. Eagleville Road, University of Connecticut, Storrs,
CT 06269-3043, USA
(Received June 19, 2008; Accepted September 13, 2008)
The use of sensory attractants is central to most animal-mediated p ollination and seed dispersal interactions. Approximately
half the 73 species of mosses in the family Splachnaceae are entomophilous (have their spores dispersed by flies) and are
coprophilous (grow on feces and carrion). When mature, entomophilous species often produce brightly coloured, scented
sporophytes which, for several species, have been shown to attract flies. In a number of cases, sporophyte colours and
odours, as w ell as the flies that visit them, have been shown to be species-specific, suggesting that the mosses co-exist by
signal diversification, just as flowering plants are thought to reduce competition for pollinators. Analyses of scent chemistry
identified an odour contrast between generations; gametophytes were either unscented or weakly scented in most species,
whereas sporophyte odours were universally stronger per unit mass and much more chemically complex. Sporophyte odours
of North and South American species sampled were both complex and diverse, with an apparent inverse relationship
between the size and showiness of the apophysis and its odour complexity. Furthermore, phylogenetic evidence suggests
that fly dispersal of spores through visual and olfactory signals has evolved multiple times in the Splachnaceae and that
modifications of sporophyte morphology may have followed, rather than triggered, the transitions to coprophily and
entomophily. We review the ecological and evolutionary aspects of entomophily, with particular emphasis on the chemistry
of sporophyte odours and the means by which they mimic decaying organic matter.
Keywords: Directed dispersal, Diptera, key innovations, mosses, sensory signals, Splachnaceae, spore dispersal
One of the hallmarks of angiosperm flowers is their use
of sensory signals (colours, shapes, odours) to attract
animals as reproductive agents. Multicomponent floral
signals have long been appreciated for their diverse roles in
pollination, from the reduction of inbreeding (through
pollinator-mediated outcrossing) to the maintenance of
appropriate pollen placement, gene flow and species
boundaries (Proctor et al., 1996). High diversity of floral
colour, pattern and scent is thought to enhance the fidelity
of the animals recruited for dispersing the pollen through
*The author to whom correspondence should be sent.
increased floral constancy (Stebbins, 1970; Kevan and
Baker, 1983). In some cases, complex floral signals
promote a high degree of specificity between plants and
their animal pollinators, whether by specific attraction or
guidance of visitors to floral rewards (Johnson and Steiner,
2000) or to sensory exploitation of pollinators via floral
mimicry (Dafni, 1984; Schiestl, 2005; Raguso, 2008).
Aside from pollination, many seed plants also use
animals to promote the dispersal of their propagules. Like
flowers, seeds and fruits often use sensory signals to
advertise their presence to dispersal agents (Schaefer et al.,
2004). Seeds may be dispersed by frugivores (Janzen, 1983;
Levey et al., 2002), ants (Handel and Beattie, 1990; Hughes
and Westoby, 1992), by scatterhoarding birds and mammals
(Forget and Vander Wall, 2001; Vander Wall et al., 2006)
62 P. MARINO ET AL.
and by passive adhesion (Sorensen, 1986; Herrera and
Pellmyr, 2002). Propagule dispersal by adhesion differs
from other types of animal-mediated dispersal in that the
adhesive species does not actively recruit animals to the
parent plant through sensory cues (Sorensen, 1986).
An exception to the pattern that species having adhesive
propagules do not actively recruit their dispersers is found
in the moss family Splachnaceae, in which the spores of all
Splachnum, Tetraplodon and several Tayloria species are
dispersed to decaying substrates (feces, carrion) by diverse
families of flies (Koponen, 1990). The sporophytes of
these mosses are brightly coloured and distinctively
scented, but provide no nutritional reward. Thus, the
Splachnaceae, like some flowering plants, have combined
visual and olfactory signals to attract insects as dispersal
agents for their propagules through sensory exploitation and
deception. In this review, we explore ecological and
evolutionary aspects of this remarkable phenomenon, with a
special emphasis on the chemistry of sporophyte odours
and the means by which they mimic decaying organic
matter (Larson et al., 2001).
2. Splachnaceae Biology
The bryophyte family Splachnaceae is a globally
distributed (Koponen, 1990) monophyletic lineage
(Goffinet et al., 2004) of 73 species with genera distributed
among three subfamilies: Aplodon, Tetraplodon and
Splachnum (Splachnoideae), Moseniella and Tayloria
(Taylorioideae) and Voitia (Voitioideae) (Crosby et al.,
1999; Bai and Tan, 2000; Fife and Goffinet, 2003; Goffinet
et al., 2004). Most species are confined to north or south
temperate habitats, whereas others are endemics to isolated
mountains or montane forests at (sub)tropical latitudes.
Nearly half of these species (Table 1) possess two
noteworthy ecological features. First, their gametophytes
are “coprophilous”, growing on feces and occasionally
carrion and other animal matter (Koponen and Koponen,
1977; Marino, 1988a; 1991a;b; 1997). Second, their spores
are small and sticky making them suitable for insect
dispersal (Fig. 1a; Bryhn, 1897; Bequaert, 1921; Koponen,
1978; Vitt, 1981; Cameron and Troilo, 1982; Cameron and
Wyatt, 1986; Marino, 1988a; 1991a;b; 1997). Different
species of Splachnaceae vary in their life history strategies.
Entomophilous species grow quickly, occupying the entire
surface of a dropping in one or two summers and
reproducing within 2–3 years (Marino, 1988a). Populations
of entomophilous species of Tetraplodon in Alberta,
Canada that grow in dry habitats persist for four or more
years whereas populations of Splachnum that grow on
droppings in wet habitats, persist for only 2 years, as they
are rapidly overgrown by the surrounding bryophytic
vegetation (Marino, 1988a).
Like stinkhorn fungi (Sleeman et al., 1997; Tuno, 1998),
the Splachnaceae attract flies not to promote sexual
reproduction, but to disperse asexual spores to appropriate
substrates (Koponen, 1990). Although entomophily
previously has been demonstrated in only a small number
of taxa of the Northern hemisphere (Koponen and
Koponen, 1977; Pyysalo et al., 1978; Cameron and Troilo,
1982; Pyysalo et al., 1983; Marino, 1988a; Koponen, 1990;
Marino, 1997; Jofre, 2007), in all putatively entomophilous
Splachnaceae, sporophytes emit odours from the inflated
and/or elongated and often coloured (e.g., yellow, magenta,
red) apophysis, the sterile region of the capsule below the
sporangium (Fig. 1b; Pyysalo et al., 1978; 1983). Further-
more, fly-dispersed species produce small, thin-walled,
sticky spores that are extruded en masse by a false
columella (Demidova and Filin, 1994) as the capsule walls
shrink (Fig. 1b). This phenomenon is comparable to flowers
of Cypripedium orchids that present their pollen as sticky
yellowish amorphous masses (Banziger et al., 2005; Li et
al., 2006). Unlike stinkhorn fungi, entomophilous
Splachnaceae do not reward the flies they attract (Cameron
and Troilo, 1982; Marino, 1988a; 1991b; 1997). Pyrellia
(Eudasyphora) cyanicolor (Diptera: Family) are flies that
visit and leave mature sporophytes of Splachnum
ampullaceum with no observable time spent feeding,
grooming or thermal basking (Troilo and Cameron, 1981).
Rather, the sporophytes are like deceptive angiosperm
flowers that attract insects by using sensory cues that mimic
decaying organic matter used by such flies for oviposition
and for larval food (Dafni, 1984; Roy and Widmer, 1999;
Stensmyr et al., 2002; Jürgens et al., 2006; Salzmann et al.,
2006). However, unlike many carrion- or dung-mimicking
flowers, sporophytes of the Splachnaceae do not hold flies
captive; flies simply depart with spores adhering to their
bodies and fly to appropriate host sites, upon which spores
are dislodged and germinate into protonema (early
gametophytic developmental stage).
Among entomophilous Splachnaceae habitat and
substrate restrictions have been explored in detail for
several species. In central Alberta (Canada), sympatric
Splachnum ampullaceum, S. luteum, Tetraplodon
angustatus and T. mnioides segregate among droppings
along a moisture gradient; the two species of Splachnum
grow on droppings in wet habitats within peatlands whereas
the two species of Tetraplodon grow on droppings in dry
habitats either in elevated areas within peatlands or outside
peatlands (Marino, 1991a). Although congeners can share
droppings from the same kinds of animals, species of
different genera are never found growing together on the
same droppings (Marino, pers. obs.). In field growth
experiments Marino (1991a) demonstrated that habitat had
little influence on the competitive interactions between
congeners. However, habitat had a strong influence on the
interactions between species of different genera;
Tetraplodon spp. excluded Splachnum spp. in dry habitats
and Splachnum spp. excluded Tetraplodon spp. in wet
ECOLOGY AND EVOLUTION OF SPLACHNACEAE MOSSES 63
Figure 1. Sporophyte modifications in the Splachnaceae. A. Sporophytes [S] of Splachnum ampullaceum composed of elongated setae and
swollen apophyses above leafy gametophyte [G] tissue. B. S. ampullaceum spores are extruded as a sticky mass (arrow) that contacts
landing flies. The inflated apophysis (*) provides visual contrast and odour for fly attraction. Scale bar = 1 mm. Photos ©R.A. Raguso.
See cover illustrations.
Figure 2. Life cycles of Patagonian
Splachnaceae. Columns (left to right)
depict the life histories of Tayloria
mirabilis (left), T. dubyi (center) and
Tetraplodon fuegiensis (right). Rows
(top to bottom) show the habitat,
animal source and preferred fecal
substrate, protonema germinating on
the substrate and mature sporophytes.
All photos were taken in the XII
Region of Chile. ©R.A. Raguso.
64 P. MARINO ET AL.
Table 1. L ist of species comprising the Splachnaceae (Crosby et al., 1999; Bai and Tan, 2000; Goffinet and Shaw, 2002; Goffinet et al.,
2004); number in parentheses following the generic name indicates the species diversity.
Aplodon R. Brown (1): A. wormskioldii (Hornemann) R. Brown
Tetraplodon B.S.G. (10): Te. angustatus (Hedw.) B.S.G.; Te. blyttii Frisvoll; Te. caulescens (Brid.) Lind.; Te. fuegianus Besch.;
Te. itatiaiae C. Müll.; Te. mnioides (Hedw.) B.S.G.; Te. pallidus I. Hagen; Te. paradoxus (R. Brown) I. Hagen;
Te. tomentosus Sehnem; Te. urceolatus (Hedw.) B.S.G.
Splachnum L. ex Hedw. (11): S. adolphi-friederici Broth.; S. ampullaceum Hedw.; S. austriacum Brid.; S. luteum Hedw.;
S. melanocaulon (Wahl.) Schwäg.; S. pensylvanicum (Brid.) H. Crum; S. resectum (Haller ex Bridel) P. de Beauv.;
S. rubrum Hedw.; S. sphaericum Hedw.; S. vasculosum Hedw.; S. weberbaueri Reimers
Voitia Hornsch. (2): V. hyperborea Grev. & Arnott; V. nivalis Hornsch
Moseniella Broth. (2): M. brasiliensis Brotherus; M. ulei (C. Müller ex Brotherus) A. Kop.
Tayloria Hook. 1 (45):
Subg. Tayloria (10): Ta. acuminata Hornsch.; Ta. delavayi (Besch.) Besch.; Ta. froelichiana (Hedw.) Broth.;
Ta. grandis (Long) Goffinet & Shaw; Ta. kilimandscharica Broth.; Ta. rudimenta Bai & Tan;
Ta. rudolphiana (Garov.) B.S.G.; Ta. serrata (Hedw.) B.S.G.; Ta. splachnoides (Schwägr.) W.J. Hook.;
Ta. tenuis (With.) W.P. Schimp.
Subg. Pseudotetraplodon A.K. Kop. (7): Ta. altorum Herzog; Ta. dubyi Broth.; Ta. novo-guinensis E.B. Bart.;
Ta. octoblepharum (W.J. Hook.) Mitt.; Ta. scabriseta (W.J. Hook.) Mitt.; Ta. stenophysata (Herz.) A. Kop.;
Ta. tasmanica Broth.
Subg. Eremodon (Wils.) Broth. (5): Ta. callophylla (C. Müll.) Mitt.; Ta. gunnii (Wils.) J.H. Willis;
Ta. magellanica (Brid.) Mitt.; Ta. mirabilis (Card.) Broth.; Ta. purpurascens (W.J. Hook. & Wils.) Broth.
Subg. Cyrtodon (R. Br.) Lindb. (6): Ta. alpicola Broth.; Ta. hornschuchii (Grev.& Arn.) Broth.;
Ta. jacquemontii (B.S.G.) Mitt.; Ta. lingulata (Dick.) Lind.; Ta. nepalensis Iwat. & Steere;
Ta. pseudoalpicola Iwat. & Steere
Subg. Brachymitrion (Tayl.) Broth. (6): T. cochabambae Müll. Hall.; T. immersa (Goffinet) Goffinet, Shaw & Cox;
T. jamesonii (Tayl.) Müll. Hall.; T. laciniata Spruce; T. moritziana Müll. Hall.; T. pocsii A.K. Kop.
Subg. Orthodon (R. Br.) Broth. (11): Ta. arenaria (C. Müll.) Broth.; Ta. borneensis Dix.; Ta. chiapensis H. Crum;
Ta. indica Mitt.; Ta. isleana (Besch.) Broth.; Ta. longiseta E. B. Bart.; Ta. orthodonta (P. de Beauv.) Wijk & Marg.;
Ta. sandwicensis (C. Müll.) Broth.; Ta. solitaria (Card.) T. Kop. & W.A. Weber;
Ta. squarrosa (W.J. Hook.) T. Kop.; Ta. subglabra (Griffith) Mitt.
Incertae sedis *Ta. rubicaulis A. Kop. (1990). Ta. braithwaitiana Tixier
1Only subg. Brachymitrion and subg. Orthodon are considered monophyletic (Goffinet et al., 2004).
habitats. Nevertheless, all species grew well when grown
alone from spores on droppings in wet habitats. Thus,
although Splachnum spp. may be physiologically restricted
from colonizing dung in dry habitats, Tetraplodon spp.
were either excluded from dung in wet habitats by
Splachnum spp. or were restricted from dung in wet
habitats by dispersal constraints. As the chemistry of
droppings also differs between dry and wet habitats, these
habitat-sensitive growth differences between genera may
result from differences in moisture availability and the
chemistry of droppings (Marino, 1991a).
Are different species of Splachnaceae associated with
particular types of dung? There is observational evidence
suggesting non-random substrate utilization in the field,
although not in the laboratory (Marino, 1991a). For
example in the field, T. mnioides (Alaska, Alberta and
Labrador) has always been found growing on carnivore or
omnivore droppings whereas S. ampullaceum (Alberta; Isle
Royale, Michigan; Newfoundland, Labrador) has always
been found growing on herbivore droppings (Marino, pers.
obs.). In Chilean Patagonia, Tayloria mirabilis was found
growing on cattle dung (Goffinet et al., 2006), Tayloria
dubyi on goose droppings (Jofre, 2007) and Tetraplodon
fuegiensis on carnivore droppings (Marino et al., pers. obs.;
Fig. 2). Both spore germination and protonemal
development appear to be influenced by pH (Armentano
and Caponetti, 1972; Cameron and Wyatt, 1989) although
this early growth stage sensitivity to pH does not result in
substrate restriction (Marino, 1997). The fact that several
species appear to be restricted to specific types of droppings
in the wild, but are capable of initiating spore germination
and gametophyte growth on various types of droppings and
levels of pH suggests that directed spore dispersal by
different flies may play a critical role in the observed
patterns of substrate restriction.
Several further observations suggest that directed spore
dispersal may promote substrate restriction among
entomophilous Splachnaceae. First, sympatric species of
Splachnaceae vary greatly in the fly faunas that they attract
(Marino, 1991b) (Table 2; Fig. 3). Second, the fragrances
emitted by the mature sporophytes of Splachnaceae differ
greatly in chemical composition with different species
ECOLOGY AND EVOLUTION OF SPLACHNACEAE MOSSES 65
Table 2. Matrix representation of the qualitative pair-wise interactions derived from the trapping experiment at Ft. Assiniboine, Alberta
(see Fig. 3), with additional data showing the mean (± 1 S.D.) number of spores (× 103) carried by trapped flies by species. The number of
flies carrying spores is shown in parentheses. Fly species in which only a single individual carried spores are not shown (Marino, 1991b).
The matrix has been shuffled to indicate generalized fly species (top rows) and visually highlight “modules” of species (low er rows) that
specialized on a specific moss (see Olesen et al., 2007). Families: (A) = Anthomyiidae, (C) = Caliphoridae, (M) = Muscidae,
(S) = Sarcophagidae, (Sc) = Scatophagidae, (Se) = Sepsidae.
Fly species Tetraplodon angustatus Tetraplodon mnioides Splachnum ampullaceum Splachnum luteum
Eudasyphora cyanocolor Zett. (M) 29±17 (10) 24±30 (2)
Helina cothurnata Rondani (M) 52±39 (11)
Phormia terrae-novae R.D. (C) 16±5.3 (2) 20±20 (9)
Scatophaga furcata Say (Sc) 26±27 (6) 32±22 (6) 16±24 (9)
Calliphora vomitoria L. (C) 46±50 (11) 29±12 (3) 16±13 (4)
Pegoplata patellans Pand. (A) 23±19 (26) 14±14 (18)
Phormia regina Meigen (C) 42±50 (4)
Ravinia sp. 1 (S) 6.2±1.8 (6) 12±9.1 (16)
Sepsis sp. 1 (Se) 5.8±3.8 (3)
Cynomyopsis cadaverina L. (C) 30±27 (7)
Hydrotae meteorica L. (M) 17±7.7 (7)
Muscina assimilis Fallen (M) 20±8.2 (4)
Lucilia sp. 1 (C) 23±13 (4)
Fannia spathiophora Mall. (M) 24±35 (3)
Pegohylomyia sp. 1 (A) 14±12 (2)
Mydaea sp. 1 (M) 25±23 (5)
Scatophaga suilla Fab. (Sc) 29±22 (5)
Hebecnema nigricolor Fallen (M) 40±48 (5)
Hydrotae militarus L. (M) 45±65 (3)
Phaonia curvipes L. (M) 15±14 (2)
Polietes orichalceoides Huck. (M) 69±19 (2)
Myospila meditabunda Fab. (M) 3.5±2.2 (5)
Pegoplata nigriscutellata Stein (A) 6.2±1.8 (2)
Hydrotae scambus Zett. (M) 3.7±1.8 (2)
Hylemyza partita Meigen (A) 6.2±1.8 (2)
emitting odours characteristic of either carrion or dung (see
below). Taken together these observations suggest that
specific species of Splachnaceae and specific substrate
types are likely attracting relatively distinct guilds of flies,
which may in part explain the observed substrate restriction
among various species of Splachnaceae.
3. Diversification of Spore Dispersal Strategy
Koponen and Koponen (1977) trapped flies in Finland
using traps baited with the sporophytes of S. ampullaceum
and S. luteum, S. vasculosum and T. mnioides and using
non-baited control traps. Traps baited with Splachnaceae
captured 92 coprophilous flies (families Sepsidae,
Scatophagidae, Muscidae and Sarcophagidae) whereas
control traps captured only 8 such individuals from these
families. The majority (77%) of flies trapped were Pyrellia
and Orthellia spp. (Muscidae) and sepsid flies. The latter,
in particular, were associated with mixed populations of
S. ampullaceum and S. luteum. Marino (1991a) studied
sympatric moss assemblages of S. ampullaceum, S. luteum,
T. angustatus and T. mnioides in central Alberta, Canada
and quantified the number of spores carried by individual
flies. Consistent with the findings of Koponen and Koponen
in Finland (1977), Marino (1991b) also found that most
flies trapped on Splachnaceae with spores on their bodies
were coprophilous species, and that Pyrellia spp. and
sepsids were especially attracted to S. ampullaceum.
However, few of the relatively small, hairless sepsid flies
carried spores. Unlike the trapping results of Koponen and
Koponen (1977), Marino (1991b) also trapped large
numbers of spore carrying anthomyiid and calliphorid flies
(e.g., Calliphora vomitoria). The larvae of most
anthomyiids are phytophagous but some are saprophagous,
feeding on decaying material (McAlpine et al., 1981).
Adults, however, feed on nectar and are important plant
pollinators, particularly at high latitudes and altitudes
(Kearns and Inouye, 1993; Larson et al., 2001; Zoller et al.,
2002). Anthomyiids thus would appear to be ideal potential
66 P. MARINO ET AL.
vectors for spores of Splachnaceae, as the adults are
attracted by flower-like visual displays and odours but they
also visit fresh dung or carrion on which to lay their eggs.
Indeed, the showiest sporophytes of Splachnaceae (e.g.
those of Splachnum luteum, S. rubrum) approximate small
flowers in colour display and physical dimensions (Fig. 5).
However, an important departure from the pollination
metaphor is that insect-mediated spore dispersal in the
Splachnaceae requires that sporophytes attract flies only
once, after which they fly to appropriate host sites, whereas
successful cross-pollination requires flowers to attract
pollen vectors at least twice.
Marino (1991b) trapped flies visiting four sympatric
moss species (Tetraplodon angustatus, T. mnioides,
Splachnum ampullaceum, S. luteum) and identified
overlapping but distinctive patterns of fly visitation per
species. Fig. 3 shows his data in a bipartite interaction
network, revealing that only one third of the possible moss-
by-fly interactions actually were observed. When shown as
a matrix (Table 2), the data reveal “modules” or distinctive
Figure 3. Bipartite graph representing 33
observed pair-wise interactions (out of 100
possibilities) between Splachnaceae mosses and
spore-carrying fly species trapped while visiting
them (Marino, 1991b). Flies were collected in
1986 and 1987 near Fort Assiniboine, Alberta,
visitor faunas for each moss. The visitor spectra differed
among the four species by 77–92% between species
(Table 2). For example, the carrion-scented sporophytes of
Tetraplodon mnioides relied heavily upon calliphorid flies
as spore vectors, of which only Calliphora vomitoria also
was observed to visit Splachnum spp. (Fig. 3). In contrast,
the bright yellow, weakly scented sporophytes of
Splachnum luteum were visited by five families of flies,
including anthomyiids (Fig. 3). Moreover, fly faunas
captured on dung were more similar to each other than were
those captured on Splachnaceae; all flies trapped on
Splachnaceae were also trapped on dung (Marino, 1991b).
Overall, there appear to be at least three niche dimensions
that may be responsible for coexistence of regionally
sympatric species: 1) habitat and substrate type (e.g.,
between spp. of Tetraplodon and Splachnum), 2) fruiting
phenology (e.g., between spp. of Tetraplodon) and 3) the
‘potential’ for sensory filtering and targeted dispersal by
different vectors, using colour-odour combinations
(Marino, 1991a;b; 1997).
ECOLOGY AND EVOLUTION OF SPLACHNACEAE MOSSES 67
Evolutionary modifications of visual and olfactory
stimuli in different mosses may lead to diversification
within the larger syndrome of fly-dispersed moss spores.
Entomophilous Splachnaceae differ greatly in the size,
shape and colour of the sporophytes as well as the height of
the seta (Fig. 2). Cameron and Troilo (1982) assayed
landing by Pyrellia cyanicolor on coloured disks placed
among sporophytes of S. ampullaceum in Michigan; the
flies showed a clear 20-fold preference for yellow over blue
or red. More recently, Marino et al. (in prep.)
experimentally decoupled olfactory and visual cues (e.g.,
Roy and Raguso, 1997) to explore their relative importance
in fly attraction to S. ampullaceum in Newfoundland,
Canada. In that study, flies were trapped on pure
populations of: A) S. ampullaceum with mature
sporophytes (olfactory + visual cues), B) S. ampullaceum
with mature sporophytes covered with green dyed cheese
cloth (olfactory cues only), C) scentless models of
S. ampullaceum sporophytes placed adjacent to immature
(gametophyte only) populations of S. ampullaceum (visual
cue only), and D) scentless artificial models of
S. ampullaceum placed adjacent to cheese cloth covered
S. ampullaceum with mature sporophytes (model + odour,
the reconstituted combination of cues).
Of the approximately 1000 flies trapped, significantly
more flies were trapped on the S. ampullaceum positive
control (A) and the model + odour treatment (D) than either
the model alone (C) or the odour alone (B; Fig. 4a),
indicating that neither sporophyte odour nor colour are
sufficient to attract the full spectrum of flies. However, the
relative importance of olfactory and visual cues differed
among the eight families of flies trapped. Nearly all taxa
caught are commonly associated with dung or other
decaying organic matter (McAlpine et al., 1981). The
exceptions were the Dolichopodidae, whose adults are
predacious (Ulrich, 2005) and whose larvae are thought to
be predators or scavengers and the Sciomyzidae, whose
larvae are parasitic on molluscs (Rozkosny, 1984). The
most abundantly trapped flies were in the families
Anisipodidae, Muscidae and Sepsidae. Anisipodids and
sepsids were most attracted to the S. ampullaceum positive
controls and to a lesser extent to the odour only and model
+ odour treatments, whereas muscids and anthomyiids were
equally attracted to all three scented treatments (Fig. 4b).
As the relatively large and hairy muscids and anthomyiids
have been shown to carry many spores (Marino, 1991b),
these results suggest that, for S. ampullaceum, odour is a
key factor facilitating the dispersal of spores to fresh
patches of dung by flies.
4. Diversification of Odour Production
Sporophyte odour in the Splachnaceae is, to our
knowledge, novel among mosses and constitutes a key
adaptation associated with entomophily in this family
(Koponen, 1990). In angiosperm evolution, key adaptations
such as floral oil (Anderson, 1979; Buchmann, 1988),
nectar spurs (Hodges, 1997) and bilateral floral symmetry
(Donoghue et al., 1998) are thought to have triggered
evolutionary shifts to novel pollination strategies. When
such shifts occur, complementary phenotypic
modifications, such as species-specific odours in figs
(Grison-Pigé et al., 2002) and euglossine bee-pollinated
orchids (Dodson et al., 1969) are thought to promote
speciation. However, some angiosperm fragrances are
complex blends in which some chemical compounds are
pollinator attractants and others are more indicative of
phylogenetic relationships or plant defense (Levin et al.,
2003). Thus, identification of sporophyte odours from
entomophilous Splachnaceae and comparison with those
emitted by their anemophilous relatives might best be
considered within the context of an independently derived
phylogeny. Such an approach could provide inference as to
the number of origins of entomophily and indicate which
combinations of sporophyte odour and colour might have
been responsible for such transitions.
Early observations by Bryhn (1897) suggested that
sporophyte odour chemistry should differ between carrion
fly and dung fly-dispersed Splachnaceae species. For
example, Bequaert (1921), Wettstein (1921) and Marino
(1991b) found carrion-feeding flies (e.g., Lucilia, Phormia,
Cynomyopsis) only on Tetraplodon mnioides, which, unlike
Splachnum species, typically colonize carnivore dung in dry
habitats (Marino, 1988a; 1991a). Lucilia are attracted to
dimethyl disulfide and similar sulfurous compounds
emitted by rotting meat and by flowers that mimic carrion
(Borg-Karlson et al., 1994; Patiño et al., 2000; Stensmyr et
al., 2002). Thus, either the sporophytes of T. mnioides or
their substrates should emit oligosulfides. In contrast, the
brightly coloured, parasol-like sporophytes of
S. ampullaceum and S. luteum attract more diverse fly
spectra, including pollen and nectar-feeding anthomyiids
(Marino, 1991b). Several lines of evidence reviewed above
(see Fig. 4a) suggest that interspecific variation in
sporophyte odour chemistry should be found in the
Pyysalo et al. (1978) extracted sporophyte volatiles
from several species of entomophilous Splachnaceae in
Finland using pentane and diethyl ether solvents for
analysis with gas chromatography-mass spectrometry (GC-
MS). They identified a series of sour-smelling organic acids
(in Splachnum vasculosum, Tetraplodon mnioides and
Tayloria tenuis) and mushroom-scented octane-derived
alcohols and ketones (particularly in Splachnum luteum),
whereas no volatiles were detected from sporophytes of
Tayloria lingulata, an anemophilous species collected from
the same habitats as their target species. A follow up study
by the same authors (Pyysalo et al., 1983), collected
volatiles from intact and dissected sporophytes of three
68 P. MARINO ET AL.
Figure 4. a) The total mean number of flies trapped/day on S. ampullaceum, S. ampullaceum models, S. ampullaceum odour alone and
S. ampullaceum models + odour; b) The mean number of anisipodid, anthomyiid, muscid and sepsid flies trapped/day on S. ampullaceum,
S. ampullaceum models, S. ampullaceum odour alone and S. ampullaceum models + odour.
boreal species, Splachnum sphaericum, S. vasculosum and
Aplodon wormskioldii. They localized octanol and octanone
production to the setae of each species, with species-
specific organic acids (e.g. benzoic acid) and alcohols (e.g.
benzyl alcohol) present in the apophyses. Their studies
confirmed that volatiles could be identified from
sporophytes of entomophilous species, blend compositions
probably are distinct enough to promote vector
specialization, and scent emissions are localized to the part
of the sporophyte most likely to affect spore transfer to the
body of an insect visitor. However, solvent extraction of
cut plant material frequently produces artifacts associated
with plant wounding (Raguso and Pellmyr, 1998; Dobson et
al., 2005), Thus, odour chemistry in the Splachnaceae must
be studied using less invasive methods, in which odours
emitted by live, intact plants equilibrate in small
“headspace” chambers, adsorb onto the surface of solid
phase microextraction (SPME) fibres, and are desorbed
directly onto the GC column for GC-MS analysis (see
Dafni et al., 2005 for details).
To date, we have collected volatiles from several North
American (T. mnioides, S. ampullaceum, S. luteum,
S. pensylvanicum, S. rubrum and S. sphaericum), South
American (Tetraplodon fuegiensis, Tayloria dubyi and
T. mirabilis) and Australasian (Tayloria gunnii and
T. octablepharum) species of Splachnaceae using living
sporophytes and gametophytes growing on mammal dung
and bone collected in the field and transported to analytical
chemistry laboratories for analysis. First, we collected total
volatiles from small populations (50–100) of living sporo-
phytes placed within small headspace chambers as
described by Raguso et al. (2003) (see also Dafni et al.,
2005). Additional samples were collected simultaneously
from moss substrates and non-Splachnaceae mosses that
were present, to control for background odours. Second, we
determined the sources of different volatiles by separating
sporophytes from gametophytes (up to 50 per species
whenever possible) and further dissecting sporophytes into
setae and apophyses + capsules.
Our analyses of scent chemistry identified informative
patterns on several levels. Haploid and diploid generations
of the moss differ in their odours with gametophytes were
either unscented or weakly scented in most species. When
odours were present in gametophyte tissue, they were
chemically restricted to two classes: sesquiterpenoid
hydrocarbons (ubiquitous in terrestrial plants) and the
octane-derived odours identified by Pyysalo et al. (1978).
The octanols and octanones also were present in
gametophytes of Sphagnum and an outgroup, Pleurozium
schreberi, and thus constitute background odours in the
habitats we studied. In contrast, sporophyte odours were
universally stronger per unit mass and more chemically
complex. With the exception of T. mnioides, the apophyses
+ capsules were the primary sources of sporophytic odours,
again consistent with the findings of Pyysalo et al. (1983).
All substrates tested were scentless, as we would expect
given that Splachnaceae sporulate during the second year
after protonemal germination, by which time substrates are
no longer in a state of decay.
Sporophyte odours among North American species are
ECOLOGY AND EVOLUTION OF SPLACHNACEAE MOSSES 69
complex and diverse, with an apparent inverse relationship
between the size and showiness of the apophysis and its
odour complexity (Fig. 5). For example, the small,
brownish coloured apophyses of Splachnum sphaericum
constitute one of the least visually conspicuous sporophytes
in their genus, but emit over 50 volatiles from several
biosynthetic classes (Fig. 5), with specific compounds
indicative of fermenting sugar, floral odours, herbivore
feces and, remarkably, moose urine (see Whittle et al.,
2000). At the opposite extreme were the large, bright
yellow sporophytes of S. luteum, whose odour consisted of
little more than fungal octane-derivatives plus trace levels
of butanoic acid and indole, a nitrogenous compound
common to night-blooming flowers (Kaiser, 1993) and
bacterial metabolism of feces (Jürgens et al., 2006).
Sporophytes of the closely related S. rubrum are similar in
size to those of S. luteum, but are almost iridescent ruby-red
in colour (see Fig. 5) and emit a uniquely pungent blend of
odours. These compounds, which include indole and phenol
(herbivore feces), benzyl alcohol and 2-phenylethanol
(flowers) and the alcohols and esters of propanoic and
butanoic acids (fermenting sugar), seemingly represent a
generalized strategy of targeting diverse spore vectors
attracted to a broad spectrum of foods or hosts; however, all
of these compounds can be found in cow dung (Kite, 1995).
As predicted from fly visitation records, Tetraplodon
mnioides was the only species found to emit dimethyl
disulfide, a known indicator of carrion and carnivore dung,
and a universal attractant of calliphorid flies (Stensmyr et
al., 2002; Jürgens et al., 2006). Dissections consistently
identified setae, rather than apophyses, as the source of
sulphurous volatiles in Tetraplodon sporophytes.
Another pattern emerging from our studies is that
populations of related species sometimes grow intermixed
on the same droppings where their geographic ranges are
sympatric (Marino, 1988b). The frequent co-occurrence of
S. luteum and S. sphaericum on individual droppings
combined with the dramatic difference between the two
species in visual vs. olfactory display, respectively, is
suggestive of a mutually advantageous relationship with
respect to spore dispersal. In contrast, in eastern
Newfoundland, Canada S. ampullaceum with its large,
yellowish sporophytes grows sympatrically and often in
mixed populations on the same droppings with
S. pensylvanicum which has relatively tiny reddish
sporophytes yet both species produce strong, although very
different, scents. Manipulative field experiments would
elucidate the extent (if any) to which these related species-
pairs benefit from each other’s presence in mixed
Sporophyte odours from a smaller sample of Southern
Hemisphere species suggested similar themes of
generational contrast and chemical mimicry through the
emission of sporophyte-specific volatile compounds.
Tayloria mirabilis was found growing on cow dung in the
understory of Nothofagus forest on Isla Navarino, in
Patagonian Chile. Its pearl-like greenish white apophyses
emitted a simple, fetid blend of phenol, cresol and indole
indicative of cow feces (Kite, 1995). The related but much
less conspicuous T. dubyi was found growing on goose
dung on hummocks of Sphagnum moss in exposed peat
bogs. The spindle-shaped, burgundy coloured sporophytes
of T. dubyi emitted a sharply unpleasant, fecal blend of
phenol, cresols and methyl p-cresol with smaller amounts
of indole. Populations of Tetraplodon fuegiensis were
found in the same peat bogs as Tayloria dubyi but were
restricted to fox feces and bone substrates. As was found
for its relatives half a planet away, the sporophyte odour of
T. fuegiensis was dominated by sulphurous volatiles
(dimethyl disulfide and trisulfide) consistent with a strategy
of carrion mimicry. All three taxa emitted octane-derived
compounds from all parts and complex sesquiterpenoid
blends from gametophytic tissues. The closest extant
relatives of the Patagonian Taylorias are found in Tasmania
and New Zealand (see below). Remarkably, two of these
putatively entomophilous species, Tayloria gunnii and
T. octablepharum, were found to emit different ratios of
phenol, p-cresol and indole, as well as octane-derivatives
and organic acids from their apophyses.
Together, the findings described above, however
preliminary, suggest that chemical mimicry of herbivore
dung is a common strategy for spore dispersal within and
between lineages of entomophilous Splachnaceae, effective
at high latitudes of Northern and Southern Hemispheres,
wherever large populations of herbivores and their
attendant fly faunas flourish. Our data suggest that fly
trapping experiments comparing synthetic blends of
sporophyte odours with positive controls of closely related
species in different microhabitats can help to determine the
relative contribution of species-specific odour blends to
substrate limitation and reduced competition among
Splachnaceae. What we lack at present is a comparative
study in which sporophyte architecture and chemistry are
characterized for key sister and outgroup species as well as
members of the putatively anemophilous lineages of
5. Phylogeny of the Splachnaceae
The classification and hence the intuitive phylogenetic
relationships among mosses have historically been drawn
primarily from variation in the architecture of the
sporophyte (e.g., Brotherus, 1924; Vitt, 1984). In essence,
the moss capsule is to the bryologist what the flower is to
the angiosperm systematist. Splachnaceae differ most
conspicuously from one another in the aspects of their
capsule. Their leafy stems are orthotropic, scarcely if at all
branched and their leaves, which compose heteroblastic
series along the stem, vary in their shape and outline,
70 P. MARINO ET AL.
Figure 5. Sporophyte odours identified from Northern Hemisphere Splachnaceae. Numbers in parentheses are the number of distinct
volatile compounds identified per species. A. 1-octene-3-ol and 3-octanone common to peat moss and mushrooms. B. 6-methyl-5-hepten-2-
one and -ol, d erived from carotenoids. C. Butanoic acid and other short-chain fermentation products. D . cyclohexane carboxylic acid esters
and heptanal, both common to mammalian urine. E. Dimethyl disulfide and -trisulfide, common to rotting flesh. F. Indole, G. Phenol and
cresol, all common to herbivore dung. H. 2-phenyl ethanol and benzyl alcohol; floral odours. Photos of S. sphaericum and T. mnioides
©B. Goffinet; others ©R.A. Raguso.
ECOLOGY AND EVOLUTION OF SPLACHNACEAE MOSSES 71
Figure 6. Phylogeny of the Splachnaceae inferred from trnL-F and rps4 (cpDNA) sequence data (redrawn from Goffinet et al., 2004).
Solid black circles identify copro/entomophilous species; open grey diagnose coprophilous species with indehiscent capsules.
72 P. MARINO ET AL.
although not along generic boundaries. Patterns in the
variation of capsule attributes (e.g., differentiation, size and
colour of the apophysis, and architecture of the peristome,
the ring of teeth lining the capsule mouth) have hence been
interpreted as reflecting the affinities among species.
Application of such morphological and phenetic taxon-
concept led Koponen (1982) to recognize 7 genera,
distributed among three subfamilies: Aplodon, Tetraplodon
and Splachnum (Splachnoideae), Brachymitrion,
Moseniella and Tayloria (Taylorioideae) and Voitia
(Voitioideae). Indehiscent capsules lacking a peristome
diagnosed the latter, whereas expanded hypophyses
characterized the Splachnoideae. Brachymitrion is
distinguished by the 16 erect peristome teeth, and
Moseniella by the small and immersed capsules lacking a
peristome. The genus Tayloria exhibits the broadest
variation in capsule shape and peristome architecture,
which provided the foundation for recognizing five
subgenera with this genus. Koponen (1982) considered that
the conspicuous anatomical differentiation among the
subgenera may justify their recognition at the generic rank.
Such a concept may be further supported by the lack of a
single known diagnostic feature uniting all of Tayloria
(Goffinet et al., 2004).
The Splachnaceae exhibit a broad spectrum of
sporophytic complexity: from an inconspicuous sterile neck
to a prominent apophysis, from a well-develop double
peristome to a naked capsule mouth, or even from a giant
seta to an extremely short stalk at the base of the capsule.
Reduction in sporophyte architecture is considered rampant
during the diversification of mosses (Vitt, 1981), and hence,
features such as the lack of peristome offer little signal
beyond the immediate phylogenetic vicinity of species
complexes. Among the Splachnaceae, two lineages have
acquired indehiscent sporangia lacking a peristome: the
ancestor to Voitia and Tayloria grandis (Goffinet and
Shaw, 2002). Other morphological characters potentially
subject to reduction include traditional taxonomic features
such as the differentiation of the apophysis, the architecture
of the peristome, and the length of the seta. Whether or not
such characters compose robust phylogenetic indicators was
first tested by Goffinet et al. (2004) who reconstructed the
phylogenetic relationships among 49 species of
representing all but one genus of Splachnaceae (i.e.,
Moseniella is only known from the original type material
dating from Brotherus (1918) based on inferences from
variation in sequences of two chloroplast loci (Fig. 6). The
phylogenetic inferences provided strong support for: 1) the
monophyly of Brachymitrion, Splachnum, Tetraplodon, the
Taylorioideae and the Voitioideae, 2) no support for the
affinities of Aplodon to the Splachnoideae, and 3) strong
support against a shared unique ancestry of the
Splachnoideae, Tayloria, and any subgenus of Tayloria,
except subg. Orthodon (all sensu Koponen, 1982).
Although species of Tayloria fail to aggregate into
monophyletic groups congruent with either subg.
Cyrotodon, Eremodon, Pseudotetraplodon, or Tayloria, the
species compose at least two major robust lineages within
the Tayloria complex. The sole taxonomic consequences
implemented by Goffinet et al. (2004), are the reduction of
the Voitioideae into synonymy with the Splachnoideae and
the resurrection of subg. Brachymitrion to accommodate
species with no differentiated apophysis, and a peristome of
16 erect to incurved teeth, within Tayloria. The family
Splachnaceae is, thus, at present composed of 2 subfamilies
accommodating four and two genera, respectively
Inferences from variation in nucleotide sequences yield
hypotheses regarding the relationships among Splachnaceae
that are at least partially incongruent with inferences from
patterns in morphological variation, suggesting that
excessive homoplasy in critical taxonomic characters
obscures the relationships among species. Sequence data,
however, fail to offer a unique, fully resolved and robust
topology, because of the scarcity of fixed character
transformations marking the divergences of the major
lineages; scarcity possibly reflecting a rapid radiation. As a
consequence of the extensive convergence and parallelism
in morphological characters traditionally used to define
taxa, the newly resolved lineages lack diagnostic features.
Furthering our understanding of the diversification of the
Splachnaceae rests on inferring the phylogenetic
relationships from a much broader sampling of loci, and
exploring de novo the morphological and ontogenic
character space within the Splachnaceae and its sister-
group, the Meesiaceae, in order to identify unique
transformations associated with major cladogenic events.
6. Evolution of Entomophily
The recruiting of insects for the dispersal of spores
relies on critical anatomical and biochemical modifications
of the plant body to attract the vector and ensure that it
comes into contact with the spore mass (Table 3). Benefits
driving the evolution of entomophily may include the high
rates of colonization success through targeted dispersal to
uncolonized resource patches and lower investments in
spore production (fewer and smaller spores). Directed
dispersal of animal-dispersed seeds receives similar benefits
(Wenny and Levey, 1998; Nathan and Muller-Landau,
2000; Wenny, 2001). Until recently, empirical testing for
entomophily has been limited to taxa from the Northern
Hemisphere (Koponen and Koponen, 1977; Cameron and
Wyatt, 1986; Marino, 1988a; 1991a;b; 1997) but can be
inferred from the strictly coprophilous habitat for several
species from tropical and austral latitudes (Koponen and
Koponen, 1977). Koponen (1978) proposed that these
innovations were acquired after the divergence of
Brachymitrion, a lineage of anemophilous taxa that she
ECOLOGY AND EVOLUTION OF SPLACHNACEAE MOSSES 73
Table 3. List of anatomical and chemical characters of entomophilous Splachnaceae and their possible significance for the syndrome.
Morphology of sporophyte1
Continuous growth of seta after sporogenesis for intrapopulation competition for the vector
Hypertrophy of apophysis through elongation or widening for visual display
Pigmentation of apophysis for visual attraction
Recurved peristome teeth (by means of thickened outer surface for exposing the spore mass when dry
Apoptosis of guard cells for favouring the release of volatiles during dry periods
Differential thickening of the exothecial cell-walls for capsule contraction upon drying
Persistence of columa for exposing spores to vector
Well developed pseudocolumella below the sporangium for pushing the spore sac upward while capsule contracts
Size reduction of the sporogenous mass for allowing resource transfer to the expanded apophysis
Form of gametophyte
Reduction in spore size and number for maintain dispersal capability while reducing the sporangium
Compound holding the spores together favouring attachment of spore to vector
Volatile compounds emitted for attracting the vector
1Note: None of the morphological characters are true innovations per se, occurring in o thers mosses, too.
argued to be the most primitive exhibiting only generalized
traits (Koponen, 1983). Furthermore, she considered
entomophilous taxa to be closely related, and to have
originated from a single ancestor derived from one of the
lineages of anemophilous Splachnaceae (Koponen, 1979).
In other words, wind-dispersed taxa compose a paraphyletic
group subtending a monophyletic clade of entomophilous
species. As a corollary of these hypotheses, the
Splachnoideae, in particular the genus Splachnum may be
viewed as the most derived taxon characterized by a
combination of most if not all derived character-states, and
Tayloria, the sole genus of the Splachnaceae comprising
both anemophilous and entomophilous species, would be of
polyphyletic origin or be paraphyletic.
The phylogenetic inferences presented by Goffinet et al.
(2004) reveal that neither wind- or insect-mediated spore
dispersal defines a homogeneous monophyletic lineage,
suggesting that the transformations associated with biotic
spore dispersal occurred more than once. Multiple shifts
characterize particularly the diversification of the Tayloria-
complex, wherein entomophilous species are resolved
within two of the three main lineages. The polarity of the
transformations within the Taylorioideae remains
ambiguous, as scenarios involving the acquisition of
entomophily followed by reversals to wind dispersal
compete with hypotheses of multiple forward shifts toward
insect-mediated spore dispersal.
Elucidating the evolution of entomophily within the
Splachnaceae requires the reconstruction of a robust
phylogeny and critically assessing the diversity of
characters and hence the homology of character-states. The
phylogenetic hypothesis proposed by Goffinet et al. (2004)
suffers from lack of robustness at critical nodes: the
monophyly of the family, and in particular the affinities of
Aplodon to the Splachnoideae are uncertain, and the
relationships within the main lineages of the Tayloria-
complex currently are incongruent among optimal
topologies. These symptoms may be diagnostic of a rapid
radiation of the Splachnaceae, or at least of the
Taylorioideae. The rapid succession of cladogenic events
constrains the opportunities for multiple changes to
accumulate in discrete regions of the genome and hence its
resolution relies on surveying a broader sample of loci, each
harbouring only a few substitutions (Givnish and Sytsma,
1997). A robust phylogeny is also imperative for
reconstructing the transformations of morphological and
chemical traits, to test for character-state correlation and to
identify traits essential to insect-mediated spore dispersal.
Following Koponen (1978) only two morphological
character-states are intuitively linked to entomophily,
namely grooved and pitted spores and the development of a
“apophysis that is wide or long, and distinctly coloured”.
The adaptive value of spore-wall ornamentation is dubious
and the occurrence of the trait may merely reflect a genetic
linkage with a character-state essential to insect-mediated
spore dispersal. The apophysis serves as the site of
synthesis and emission of volatile compounds, and its
hypertrophy may reflect further improvement of the
syndrome rather than mark a modification that is essential
to recruiting flies. Differentiated sterile necks occur also in
other mosses and in some e.g., Pleurophascum; (Fife and
Dalton, 2005) the capsules of non-entomophilous mosses
also may be brightly pigmented. The fundamental
adaptations to entomophily in the Splachnaceae may thus
be purely biochemical, physiological and ontogenetic, to
ensure the attraction of the vector using olfactory cues, the
efficient uptake of nutrients from a eutrophic substrate and
rapid completion of the life cycle. The Meesiaceae, which
74 P. MARINO ET AL.
share a unique common ancestor with the Splachnaceae
occur in the same mesohabitat, but seem to be long-lived
perennials (at least in comparison to the Splachnaceae) and
are never found on fresh dung. Their capsule presents a
sterile neck, and, except for Leptobryum, their peristome is
reduced in that it lacks well developed cilia between the
endostome segments. Whether or not they produce any
immediate metabolic precursors to the volatile compounds
found in Splachnaceae (e.g., short chain acids), or could
grow on dung and complete their life cycle within a year of
vegetative growth is not known , and their significance in
the evolution of entomophily in immediate ancestor to the
Splachnaceae is ambiguous.
Innovations in the architecture of the sporophyte may
have followed the transition to coprophily and to
entomophily, and the diversity of sporophytic architecture
may be uninformative for inferring the origin of evolution
of insect-mediated spore dispersal in the Splachnaceae.
However, changes in the size and colour of the apophysis,
in the responses of the capsule to drying air, in the
complexity of the odours emitted, and in the substrate
preferences may be correlated to the partitioning of the
abiotic and biotic niche, and hence reflect specific
adaptations to the vector or the microhabitat. Our efforts
currently focus on biochemical diversification and its
correlation to the vector recruitment and its potential for
Of the entomophilous Splachnaceae that have been
explored, it appears that visual and chemical sensory
attractants play key roles in facilitating the dispersal of
spores to new patches of substrate. Variation in the quantity
and diversity of chemical and visual signals among species
is high, thus, the relative importance of each sensory
channel of the putative dispersal agents in promoting
effective dispersal of spores among different species
remains unclear. However, further study of chemical and
visual communication between entomophilous
Splachnaceae and their insect vectors would further our
understanding of the sequence of events by which plants
evolve the ability to communicate with their animal vectors
and, more specifically, the role of signal diversification in
promoting coexistence among sympatric species of
Splachnaceae. With the exception of biochemical
innovations, none of the putative characters thought to be
associated with entomophily (Koponen, 1990) constitute
clear examples of key innovations, as: a) their adaptive
value has not been demonstrated, and b) they all occur,
even if slightly modified, in other mosses. Rather, we
propose that Splachnaceae may possess a unique
combination of characters as pre-adaptations promoting
shifts to entomophily, with volatile biochemistry being the
key innovation. If this scenario proves to be true, it raises
several intriguing questions. Is there an obvious metabolic
step that unites the biosynthesis of the various compounds?
What is it about those pathways that constrain their
acquisition among moss lineages? How could volatiles that
are almost never encountered in plant odours (e.g.
cyclohexane carboxylic acids), which would in theory
depend upon very specific models for chemical mimicry
(mammalian urine) have evolved in this system? Mosses of
the Splachnaceae provide a remarkable and unique
opportunity to explore the patterns and processes
underlying how plants with adhesive propagules attract
animal vectors and the importance of this relationship in
promoting anatomical, chemical, niche and taxonomic
diversification within this remarkable family of plants.
For assistance in the field we thank Aadra Bhat, Kristie
Quarles, Louis Goffinet and Dheeraj Busawon for
assistance in North America and Anna-Maria Caicheo,
Jocelyn Jofre and Louis Goffinet for assistance in Chile.
Dr. Francisca Massardo (Universidad de Magallanes and
Fundación Omora) and Fundación Omora assisted greatly
in helping coordinate field research in Chile. For assistance
in analysis of odour chemistry we thank Aadra Bhat, Kristie
Quarles, Carolina Mendoza and Shaniece Charlemagne.
We also thank Dr. Thomas Clausen (University of Alaska,
Fairbanks) and Dr. Hermann Niemeyer (Universidad de
Santiago) for generously assisting in the use of their GC-
MS facilities and instrumentation. We gratefully
acknowledge financial support from NSERC grant 327414
(Marino), National Geographic Society grant 7942-05
(Raguso, Goffinet and Marino), two South Carolina
BRIN/EPSCoR CRP grants to Raguso and Marino (US
National Science Foundation (NSF) EPSCoR Grant No.
EPS-0132573 and US National Institute of Health (NIH)
Grant RR-P20 RR 016461, in association with the BRIN
Program of the National Center for Research Resources),
NSF grant DEB-0089633 (Goffinet), NSF grant DEB-03
Anderson, W.R. 1979. Floral conservatism in neotropical
Malpighiaceae. Biotropica 11: 219–223.
Armentano, T.V. and Caponetti, J.D. 1972. The effect of pH on
the growth of protonema of Tetraplodon mnioides and Funaria
hygrometrica. The Bryologist 75: 147–153.
Bai, X.L. and Tan, B.C. 2000. Tayloria rudimenta (Musci,
Splachnaceae), a new species from Nintgxia Huizu autonomous
region of China. Bryologie 21: 3–5.
Banziger, H., Sun, H., and Luo, Y.-B. 2005. Pollination of a
slippery lady slipper orchid in south-west China: Cypripedium
guttatum (Orchidaceae). Botanical Journal of the Linnean
Society 148: 251–264.
ECOLOGY AND EVOLUTION OF SPLACHNACEAE MOSSES 75
Bequaert, J. 1921. On the dispersal by flies of the spores of certain
mosses of the family Splachnaceae. The Bryologist 24: 1–4.
Bickel, D.J. 1985. A revision o f the Nearctic Medetera (Diptera:
Dolichopodidae). U .S. Department of Agriculture, Technical
Bulletin 1692: 1–109.
Borg-Karlson, A.K., Englund, F.O., and Unelius, C.R. 1994.
Dimethyl oligosulphides, major volatiles released from
Sauromatum guttatum and Phallus impudicus. Phytochemistry
Brotherus, V.F. 1918. Moseniella, un nouveau genre des mousses
du Brésil. Arkiv för Botanik utfivet av K. Svenska
Vetenskapsakademien 15: 1–3.
Brotherus, V.F. 1924. Splachnaceae. Die natürlichen Pflanzen-
familien 10: 333–343.
Bryhn, N. 1897. Beobachtungen über das Ausstreuen der Sporen
bei den Splachnaceen. Biologisches Centralblatt 17: 48–55.
Buchmann, S.L. 1988. The ecology of oil flowers and their bees.
Annual Review of Ecology and Systematics 18: 343–369.
Cameron, R.G. and Troilo, D. 1982. Fly-mediated spore dispersal
in Splachnum ampullaceum (Musci). Michigan Botanist 21: 59–
Cameron, R.G. and Wyatt, R. 1986. Substrate restriction in
entomophilous Splachnaceae: role of spore dispersal. The
Bryologist 89: 279–284.
Cameron, R.G. and Wyatt, R. 1989. Substrate restriction in
entomophilous Splachnaceae: II. Effects of hydrogen ion
concentration on establishment of gametophytes. The Bryologist
Crosby, M.R., Magill, R.E., Allen, B., and He, S. 1999. A
Checklist of the Mosses. Missouri Botanical Garden.
Dafni, A. 1984. Mimicry and deception in pollination. Annual
Review of Ecology and Systematics 15: 259–278.
Dafni, A., Kevan, P.G., and Husband, B.C., eds. 2005. Practical
Pollination Biology. Enviroquest, Cambridge, Canada.
Demidova, E.E. and Filin, V.R. 1994. False columella and spore
release in Tetraplodon angustatus (Hedw.) Bruch et Schimp. in
B.S.G. and T. mnioides (Hedw.) Bruch et Schimp. in B.S.G.
(Musci. Splachnaceae). Arctoa 3: 1–6.
Dobson, H.E.M., Raguso, R.A., Knudsen, J.T., and Ayasse, M.
2005. Scent as an attractant. In: Practical Pollination Biology.
Dafni, A., Kevan, P.G., and Husband, B.C., eds. Enviroquest,
Cambridge, Canada, pp. 197–230.
Dodson, C., Dressler, R., Hills, H., Adams, R., and Williams, N.
1969. Biologically active compounds in orchid fragrances.
Science 164: 1243–1249.
Donoghue, M.J., Ree, R.H., and Baum, D.A. 1998. Phylogeny and
the evolution of flower symmetry in the Asteridae. Trends in
Plant Science 3: 311–317.
Fife, A. and Goffinet, B. 2003. The identity of Tayloria maidenii.
The Bryologist 106: 309–310.
Fife, A. and Dalton, J. 2005. A reconsideration of Pleurophascum
(Musci: Pleurophascaceae) and specific status for a New Zealand
endemic, Pleurophascum ovalifolium stat. et nom. nov. New
Zealand Journal of Botany 43: 871–884.
Forget, P.M. and Vander Wall, S.B. 2001. Scatter-hoarding
rodents and marsupials: convergent evolution on diverging
continents. Trends in Ecology & Evolution 16: 65–67.
Givnish, T. and Sytsma, K. 1997. Molecular Evolution and
Adaptive Radiations. Cambridge University Press, Cambridge,
Goffinet, B., Buck, W.R., Massardo, P., and Rozzi, R. 2006. The
Miniature Forests of Cape Horn. Universidad de Magallanes,
Punta Arenas, Chile.
Goffinet, B. and Shaw, A.J. 2002. Independent origins of
cleistocarpy in the Splachnaceae: analyses of cpDNA sequences
reveals polyphyly of the Voitioideae. Systematic Botany 27:
Goffinet, B., Shaw, A.J., and Cox, C. 2004. Phylogenetic
inferences in the dung moss family Splachnaceae from analysis
of cpDNA sequence data and implications for the evolution of
entomophily. American Journal of Botany 91: 748–759.
Grison-Pigé, L., Hossaert-McKey, M., Greeff, J.M., and Bessiere,
J.M. 2002. Fig volatile compounds – a first comparative study.
Phytochemistry 61: 61–71.
Handel, S.N. and Beattie, A.J. 1990. Seed dispersal by ants.
Scientific American 262: 76–83.
Herrera, C.M. and Pellmyr, O., eds. 2002. Plant-Animal
Interactions: An Evolutionary Approach. Blackwell, Oxford
Hodges, S.A. 1997. Floral nectar spurs and diversification.
International Journal of Plant Science 158: S81–S88.
Hughes, L. and Westoby, M. 1992. Fate of seeds adapted for
dispersal by ants in Australian sclerophyll vegetation. Ecology
Janzen, D.H. 1983. Dispersal of seeds by vertebrate guts. In:
Coevolution. Futuyma, D.J. and M. Slatkin, M., eds. Sinauer
Associates Inc., Sunderland, pp. 232–262.
Jofre, J.P. 2007. Tayloria dubyi: fenología y dispersión de esporas
en una especie endémica de musgo de turbera en la Reserva de
Biosfera Cabo de Hornos. Abstract III Reunión Binacional de
Johnson, S.D. and Steiner, K.E. 2000. Generalization versus
specialization in plant pollination systems. Trends in Ecology &
Evolution 15: 140–143.
Jürgens, A., Dötterl, S., and Meve, U. 2006. The chemical nature
of fetid floral odors in stapeliads (Apocynaceae-
Asclepiadoideae-Ceropegieae). New Phytologist 172: 452–468.
Kaiser, W.M. 1993. The Scent of Orchids: Olfactory and
Chemical Investigations. Elsevier, Amsterdam.
Kearns, C.A. and Inouye, D.W. 1993. Techniques for Pollination
Biologists. University of Colorado Press, Boulder, CO.
Kevan, P .G. and Baker, H.G. 1983. Insects as flower v isitors and
pollinators. Annual Review of Ecology and Systematics 28: 407–
Kite, G.C. 1995. The floral odor of Arum maculatum. Biochemical
Systematics and Ecology 23: 452–468.
Koponen, A. 1978. The peristome and spores in Splachnaceae and
their evolutionary and systematic significance. Bryophytorum
Bibliotheca 13: 535–567.
Koponen, A. 1979. Entomophily and the classification of the
Splachnaceae (Musci). Abstracta Botanica 5 (Supplement): 61.
Koponen, A. 1982. The classification of the Splachnaceae. Nova
Hedwigia Beiheft 71: 237–245.
Koponen, A. 1983. Studies on the generic concept in the
classification of the moss family Splachnaceae. Publications
from the Department of Botany, University of Helsinki 11: 1–48.
Koponen, A. 1990. Entomophily in the Splachnaceae. Botanical
Journal of the Linnean Society 104: 115–127.
Koponen, A. and Koponen, T. 1977. Evidence of entomophily in
Splachnaceae (Bryophyta). Bryophytorum Bibilotheca 13: 569–
Larson, B.M.H., Kevan, P.G., and Inouye, D.W. 2001. Flies and
flowers: taxonomic diversity of anthophiles and pollinators. The
Canadian Entomologist 133: 439–465.
Levey, D.J., Silva, W.R., and Galetti, M. 2002. Seed Dispersal
and Frugivory: Ecology, Evolution, and Conservation. CABI
76 P. MARINO ET AL.
Levin, R.A., McDade, L.A., and Raguso, R.A. 2003. The
systematic utility of floral and vegetative fragrance in two
genera of Nyctaginaceae. Systematic Biology 52: 334–351.
Li, P., Luo, Y.B., Bernhardt, P., Yang, X.Q., and Kou, Y. 2006.
Deceptive pollination of the Lady's Slipper Cypripedium
tibeticum (Orchidaceae). Plant Systematics and Evolution 262:
Marino, P.C. 1988a. Coexistence on divided habitats: mosses in
the family Splachnaceae. Annales Zoologici Fennici 25: 89–98.
Marino, P.C. 1988b. The North American distribution of the
circumboreal species of Splachnum and Tetraplodon. The
Bryologist 91: 161–166.
Marino, P.C. 1991a. Competition between mosses (Splachnaceae)
in patchy habitats. Journal of Ecology 79: 1031–1046.
Marino, P.C. 1991b. Dispersal and coexistence of mosses
(Splachnaceae) in patchy h abitats. Journal of Ecology 79: 1047–
Marino, P.C. 1997. Competition, dispersal and coexistence of
Splachnaceae in patchy habitats. In: Advances in Bryology.
Longton, R.A., ed. J. Cramer, Berlin.
Marino, P.C., R aguso, R.A., and Goffinet, B. In prep. Chemical
communication, signal mimicry and exploitation of flies by
McAlpine, J.F., Peterson, B.V., Shewell, G.E., Teskey, H.J.,
Vockeroth, J.R., and Wood, D.M., eds. 1981. Manual of the
Nearctic Diptera. Volume 1, 2. Agriculture Canada Monograph.
Nathan, R. and Muller-Landau, H.C. 2000. Spatial patterns of seed
dispersal, their determinants and consequences for recruitment.
Trends in Ecology & Evolution 15: 278–285.
Olesen, J.M., Bascompte, J., Dupont, Y., and Jordano, P. 2007.
The modularity of pollination networks. Proceedings National
Academy of Science 104: 19891–19896.
Patiño, S., Grace, J., and Bänziger, H. 2000. Endothermy by
flowers of Rhizanthes lowii (Rafflesiaceae). Oecologia 124:
Proctor, M., Yeo, P., and Lack, A. 1996. The Natural History of
Pollination. Timber Press, Portland, Oregon.
Pyysalo, H., Koponen, A., and Koponen, T. 1978. Studies on
entomophily in Splachnaceae (Musci). I. Volatile compounds in
the sporophytes. Annales Botanici Fennici 15: 293–296.
Pyysalo, H., Koponen, A., and Koponen, T. 1983. Studies on
entomophily in Splachnaceae (Musci). II. Volatile compounds in
the hypophysis. Annales Botanici Fennici 20: 335–338.
Raguso, R.A. 2008. Ecology and evolution of floral fragrances.
Annual Review of Ecology, Evolution, and Systematics 39: in
Raguso, R.A., Levin, R.A., Foose, S.E., Holmberg, M.W., and
McDade, L.A. 2003. Fragrance chemistry, nocturnal rhythms
and pollination “syndromes” in Nicotiana. Phytochemistry 63:
Raguso, R.A. and Pellmyr, O. 1998. Dynamic headspace analysis
of floral volatiles: a comparison of methods. Oikos 81: 238–254.
Roy, B.A. and Raguso, R.A. 1997. Olfactory versus visual cues in
a floral mimicry system. Oecologia 109: 1432–1939.
Roy, B.A. and Widmer, A. 1999. Floral mimicry: a fascinating yet
poorly understood phenomenon. Trends in Plant Science 4: 325–
Rozkosny, R. 1984. The Sciomyzidae (Diptera) of Fennoscandia
and Denmark. E.J. Brill/Scandinavian Science Press Ltd.,
Salzmann, C.C., Brown, A., and Schiestl, F.P. 2006. Floral scent
emission and pollination syndromes: evolutionary changes from
food to sexual deception. International Journal of Plant Science
Schaefer, H.M., Schaefer, V., and Levey, D.J. 2004. How plant-
animal interactions signal new insights in communication.
Trends in Ecology & Evolution 19: 577–584.
Schiestl, F.P. 2005. On the success of a swindle: pollination by
deception in orchids. Naturwissenschaften 92: 1432–1904.
Sleeman, D.P., Jones, P., and Cronin, J.N. 1997. Investigations of
an association between the stinkhorn fungus and badger setts.
Journal of Natural History 31: 983–992.
Sorensen, A.E. 1986. Seed dispersal by adhesion. Annual Review
of Ecology and Systematics 17: 443–463.
Stebbins, G.L. 1970. Adaptive radiation of reproductive
characteristics in angiosperms. I: Pollination mechanisms.
Annual Review of Ecology and Systematics 1: 307–326.
Stensmyr, M., Urru, I., Collu, I., Celander, M., Hansson, B.S., and
Angioy, A.M. 2002. Rotting smell of dead horse arum florets.
Nature 429: 625–626.
Troilo, D. and Cameron, R.G. 1981. Comparative behavior of
Pyrellia cyanicolor (Diptera: Muscidae) on the moss Splachnum
ampullaceum and on substrates of nutritional value. Great Lakes
Entomologist 14: 191–195.
Tuno, N. 1998. Spore dispersal of Dictyophora fungi (Phallaceae).
Ecological Research 13: 7–15.
Ulrich, H. 2005. Predation by adult Dolichopodidae (Diptera): a
review of literature with an annotated prey-predator list. Studia
Dipterologica 11: 1–60.
VanderWall, S.B., Borchert, M.I., and Gworek, J.R. 2006.
Secondary dispersal of bigcone Douglas-fir (Pseudotsuga
macrocarpa) seeds. Acta Oecologica 30: 100–106.
Vitt, D.H. 1981. Adaptive modes of the moss sporophyte. The
Bryologist 84: 166–186.
Vitt, D.H. 1984. Classification of the Bryopsida. In: New Manual
of Bryology. Vol. 2. Schuster, R.M., ed. Hattori Botanical
Laboratory, Nichinan, Japan, pp. 696–759.
Wenny, D.G. 2001. Advantages of seed dispersal: a re-evalution
of directed dispersal. Evolutionary Ecology Research 3: 51–74.
Wenny, D.G. and Levey, D.J. 1998. Directed seed dispersal by
bellbirds in a tropical cloud forest. Proceedings National
Academy of Science 95: 6204–6207.
Wettstein, F.V. 1921. Splachnaceen Studien. I. Entomophilie und
Spaltöffnungsapparat. Oesterreich Botanik Zeitschrift 70: 65–77.
Whittle, C.L., Bowyer, R .T., Clausen, T.P., and Duffy, L.K. 2000.
Putative pheromones in urine of rutting male moose (Alces
alces): Evolution of honest advertisement? Journal of Chemical
Ecology 26: 2747–2762.
Zoller, H., Lenzin, H., and Erhardt, A. 2002. Pollination and
breeding system of Eritrichium nanum (Boraginaceae). Plant
Systematics and Evolution 233: 1–14.