Composition of the invertebrate fauna
The invertebrate fauna is in many ways similar in composition to that of Ontario and northern Europe but there are also some important differences.
Taxonomically it is dominated by insects and Collembola, followed by arachnids. Mollusca form a small but significant component that is often overlooked. These findings are consistent with those from the two previous eastern Canadian regional surveys (Table 3). Pooled results from all three studies indicate that in eastern Canadian caves, insects and Collembola typically constitute ~60% of all invertebrate species, arachnids ~20%, molluscs ~5% and all other invertebrates ~15%.
In the taxonomic groups Diptera, Aranea, Opiliones and Mollusca, as well as in the overall total of all taxa, the number of taxa recorded in the present survey is significantly lower than that collected in southern Ontario caves (Peck, 1988) (Table 3). This almost certainly reflects lower species diversity in the present survey area, particularly Nova Scotia which has a somewhat impoverished provincial fauna due at least in part to the Tantramar Marshes zoogeographical barrier. More intensive collecting will certainly significantly extend the New Brunswick list. The four taxonomic groups referred to are dominated by species from the parietal assemblage (see below) which constitutes a major faunal association in caves and mines in both Ontario and the Maritimes. In the Acari and the Collembola the opposite is the case: the local cave fauna is more species rich than that of Ontario. Both these taxonomic groups are wellrepresented in communities living in porcupine dung, a habitat not reported in southern Ontario caves by Peck (1988).
In general, there is little specific overlap between the invertebrates found associated with different organic substrates. Vegetable litter is generally richer in terrestrial troglophiles than is porcupine dung. Rotting timbers in abandoned mines and elsewhere provide a habitat for a few earthworms, Acari and Collembola, but the fauna found in them does not appear to be very rich in this geographical area. Some invertebrate species appear to be almost exclusively associated with oligotrophic sites e.g. Heteromurus nitidus and Arrhopalites hirtus.
As in virtually all cave ecosystems porcupine dung communities are simplified with two fundamental trophic levels, primary producers and herbivores, absent. They differ from most other cave guano ecosystems in that the porcupine is herbivorous feeding preferentially on the cambium of trees, especially conifers. The remaining two classic trophic levels, decomposers and predators, are easily recognized. This is usually true of cave guano communities but may not be the case in oligotrophic cave communities where decreasing resources are associated with blurring of the distinction and even disappearance of obligate predators in favour of omnivores (DeHarveng & Bedos, 2000). Quedius s. spelaeus adults and larvae and various predatory mites (e.g Alliphis, Geolapsis) are the common predators in porcupine dung communities.
There is an unusual suite of parietal species that is ecologically fundamentally different in its nature from the original sense of Jeannel (1926). It constitutes a distinct faunal component derived directly from the porcupine dung fauna. The larvae of several Diptera (Limonia cinctipes, Trichocera maculipennis, Chaoborus, various sciarids, Psychoda, and Leptocera) live in porcupine dung, and adult Scatopsciara, Chaoborus and Leptocera have been observed apparently ovipositing in very fresh pellets, thus, unlike almost all the “traditional” parietal insects, these Diptera are probably able to complete their life cycle underground. This is reminiscent of the situation in the humid tropics where there is frequently an assemblage of arthropods derived from guano communities on cave walls deep inside caves (Deharveng & Bedos, 2000).
A ‘parietal association’ in Jeannel’s (1926) sense can also be readily recognized in Maritime caves. It is very similar to that found in Ontario (Peck, 1988), northern Europe and other cold temperate regions of the world and comprises the associations of arthropods living or resting on cave walls and other rock surfaces together with the spiders that prey upon them. Parietal predators tend to be specialized forms found only in this and similar dark humid habitats such as cellars, and are usually considered to be troglophiles. Most of the parietal association however consists of habitual trogloxenes that are using the habitat temporarily for shelter, summer aestivation, hibernation, overwintering, or other purposes that are not yet fully understood. Adult Diptera predominate both in number of species and number of individuals. Overwintering fauna includes Nelima elegans (Moseley & Hebda, 2001), female Culex spp. and some species of Bolitophilia, Rymosia and Tarnania (Mycetophilidae). Some of the Exechiini (Mycetophilidae) along with the adult Limniphilidae occasionally seen in New Brunswick caves are probably there for summer aestivation. Two spiders (Meta ovalis and Nesticus cellulanus) specialise in predating other parietal fauna. Ceuthophilus brevipes uses caves and mines for shelter and forages outside. There are also two moths Scoliopteryx libatrix and Triphosa haesitata. Adult helomyzids (Scoliocentra, Helomyza, Amoebaleria, and Tephrachlamys) are common and are present year round but their purpose in entering caves is unknown: their larvae are not found in this habitat.
Several other arthropods (Oniscus asellus [Isopoda], Polydesmus angustus [Diplopoda]) and gastropods (Arion, Deroceras laeve, Trichia hispida) are commonly found on rock surfaces in the threshold but rarely much further in. Another gastropod Zonitoides arboreus appears to be somewhat more cavernicolous in its behaviour than the other recorded mollusca, and is sometimes found well inside the dark zone. These species are also common in dark moist habitats on the surface and unlike the animals traditionally included in the parietal association are apparently not in caves for a specific purpose. However, because they are characteristically present in this habitat and the arthropods are preyed upon by parietal spiders, it seems logical to include them as parietal fauna. The subterranean aquatic fauna is undercollected and the statistical dominance of terrestrial (80%) versus aquatic (20%) taxa in the lists certainly partly reflects this collecting bias.
Mammals, amphibians and fishes
Most of the recorded vertebrates enter caves for definite purposes and thus may be considered to be habitual trogloxenes.
Two fishes, Salvelinus fontinalis and Pungitius pungitius, are seen frequently enough to suggest that in behaviour they are the most cavernicolous vertebrates in the region. Beaver as well as Porcupine use underground sites as dens. McAlpine (1977) reports records of beaver in caves from Alabama, Missouri and New Brunswick. Two small mammals, the Smokey Shrew (Sorex fumeus) and the Deer Mouse (Peromyscus) are known from research elsewhere to predate hibernating bats in caves (Banfield, 1974; Trevor-Deutsch, 1973) and are thought to enter local caves primarily for this purpose, whilst Raccoon and Mink enter them to forage.
The remaining vertebrates are probably accidentals. This is certainly the case for the adult frogs seen occasionally, perhaps entering during flooding. They have however been observed to survive for some time where a stream brings in food and offers suitable habitat (e.g. CB). Phoxinus eos was observed in HC for several years, but this fish has not been seen in any other cave and collected specimens were pallid and undernourished. A single extra-limital winter record of the migratory bat Lasionycteris noctivagans in PT probably represented a stranded storm-driven animal (Hebda, pers. comm.)
Energy inputs: porcupine dung
Porcupine and extensive deposits of their dung were first reported in caves in California (Graham, 1962). Calder & Bleakney (1965, 1967) investigated the ecology of a porcupine-inhabited cave in Nova Scotia. Despite the wide distribution of the porcupine in North America the ecological significance of the porcupine dung habitat in caves has been overlooked or neglected. Graham (1962) stated that it was poor in invertebrate life. In a recent review of cave guano communities Gnaspini & Trajano (2000) do not mention porcupine dung.
Substantial and sustained inputs of porcupine dung are the principle energy source supporting subterranean communities in New Brunswick and mainland Nova Scotia. These communities often display considerable species diversity. This is well illustrated by the St. Croix group of caves (FC + F2 + WB) in Nova Scotia where some 89 taxa in 64 families have been identified, and is similar to many caves in the humid tropics where inputs of externally derived organic energy sources such as bat and cricket faeces support some of the most diverse cave ecosystems known (Gunn et al., 2000).
There are further ecological and faunal similarities with tropical guano caves. Although species-rich in comparison to other cave habitats, guano communities are in general simpler than those above ground (Gnaspini & Trajano, 2000). Calder & Bleakney (1965) demonstrated that the microarthopod fauna of FC is less diverse than that of nearby epigean habitats. Acari, as noted by Gnaspini & Trajano (2000) in Brazil and elsewhere, are the most diversified group followed by Collembola. Mites are almost never dominant organisms in caves except in guano. The presence of an unusual parietal association derived directly from dung communities has already been referred to.
Accordingly this fauna can be seen as a rare coldtemperate North American analogue of tropical guano cave faunas. It must be emphasized that porcupine dung does not only directly support a guanobious fauna, but indirectly contributes substantially to other aquatic and terrestrial communities through the export of this fauna to other areas of the cave. The habitat thus has a more important status regionally, biospeleologically and as a reservoir of significant species diversity than hitherto recognized.
Historically, the guano habitat has not been seen as a “true” cave habitat and guanophiles have been categorized as “false cave-dwellers” (see e.g. Gnaspini & Trajano, 2000 and references therein). This view can be challenged as an extreme extension of the a priori assumption that the cave environment is per se oligotrophic (itself an extension of the “troglocentric” focus briefly discussed in the Introduction). In fact, dismissal of the guano fauna in tropical caves seems to be at least in part an unconscious derivation from a belief, now known to be false, that there were no or very few troglobites in the tropics.
Troglobites are in fact not restricted to oligotrophic habitats. In Hawaii a rich fauna of specialized troglobites is found in the food-rich habitat provided by tree roots penetrating cavities in lava flows (Howarth, 1972). Some animals which are dependent on guano are traditionally accepted as troglophiles. This is almost certainly the case for instance with Trichocera maculipennis, a circumpolar fly which is distinctly uncommon in surface habitats (Jefferson, 1981). It also seems somewhat illogical to exclude organisms that complete their life cycle underground as not cavernicolous whilst accepting those which for example merely use caves as temporary seasonal shelter. Finally, in order to reach and colonise guano piles organisms must be able to orient and survive in the wider cave environment (Gnaspini, 1992).
Thus, as proposed by Gnaspini (1992), guano should be treated simply as a substrate within a cave. All animals regularly found in and utilizing caves and related subterranean habitats ought to be classified as cavernicolous. Those guanophiles regularly occurring in subterranean habitats should be treated as one ecological category of cave fauna.
Other energy inputs
Vegetable debris, detritus and flood debris are significant food sources. They are presumed to be the main energy source in Cape Breton sites, where the porcupine is absent. Spring snowmelt flood debris constitutes a special habitat because it is seasonally pulsed. Rotting timbers are available at some sites, especially abandoned mines.
Insectivorous bat droppings are only of incidental importance as a food source, being localized and never forming substantial guano deposits. Where bat colonies occur they are small and droppings only accumulate in hibernacula for a few weeks during the autumn.
Seasonality
Circannual rhythms in cavernicoles have not been observed or investigated in the Canadian Maritimes. However the cave communities are subject to several seasonal environmental changes and cues. These include annual temperature cycles, pulsed inputs of organic matter into stream caves during the spring thaw and bat droppings in late summer. The composition of parietal assemblages is seasonal. Origin of the subterranean fauna The present survey indicates that the subterranean fauna of the region consists of communities of nonobligate species which have all arrived in the area and colonised hypogean habitats at various different times and via several routes during the past ~11,000 years, or perhaps in some cases earlier during deglaciation (~21,000-11,000 BP).
No convincing example of a preglacial survivor has been found. The apparently disjunct distribution of Q. s. spelaeus suggests that this troglophilic beetle may have arrived from the emergent land areas that existed on the present-day Atlantic continental shelf during deglaciation (Moseley et al., 2006) but evidence based on faunal distribution is always difficult to interpret. The subterranean aquatic fauna needs more sampling as it is possible that one or more aquatic troglobites (stygobites) could have survived and reinvaded Nova Scotia groundwaters from Atlantic refugia. An unidentified species of Cavernocypris (Ostracoda) may hint at this possibility. Stygobitic amphipods of the genus Stygobromus are found on Vancouver Island and on offshore islands as far north as southeastern Alaska (Shaw & Davies, 2000) and in glaciated regions of Alberta (Holsinger, 1980; Bousfield & Holsinger, 1981).
One of the most interesting findings is the number of recent invasive non-indigenous taxa present in terrestrial cave communities. Of the cavernicolous terrestrial invertebrates 17% are probably European in origin (Table 1). Notable examples are the European threshold troglophile Nesticus cellulanus which is now established here, being found in cellars and similar dark damp places in the region, and the springtail Folsomia fimitaria. Records from FC are the first cave records of N. cellulanus in North America (Ewing, pers. comm.) and the same cave is the only confirmed North American locality for F. fimitaria (Christiansen & Bellinger, 1980). It is also worth noting that specimens of the circumpolar Protaphorura armata (Collembola) from local caves are morphologically very similar to European examples from anthropogenic (agricultural) habitats (Pomorski pers. comm. 2006) and thus may represent another non-indigenous population. Some care must be taken in accepting a taxon as nonindigenous. Some North American species previously thought to be co-specific with European animals have later been found to be distinct, and there are examples in the Acadian cave fauna. The spider Meta ovalis was distinguished from the European M. menardi only recently (Marusik & Koponen 1992) and the widespread North American collembolan Folsomia stella was long confused with the morphologically very similar Old World F. fimitaria (Christiansen & Bellinger, 1980). In other cases species previously thought to be introduced have later been shown to be native e.g. the earthworm Dendrodrilus rubidus (Schwert, 1979). Nevertheless, taken as a whole, the list of probable introductions is convincing.
The eastern Canadian seaboard has been subject to European exploration and colonization since Viking times, and has long been known as the probable point of introduction of many exotic invertebrates as a consequence of human migration and trade. Many of these are now widespread in surface habitats. Apparently a subset of “pre-adapted” species that have arrived and successfully established in the Maritimes has subsequently been able to enter and survive in subterranean habitats here.
Many of the invasive species were collected in FC and/or F2 (Table 1), but it is uncertain whether this is related to the fact that these caves are located in one of the earliest areas of Acadian French settlement in Nova Scotia or is a result of collecting bias.
Although recent work with epigean invasive species elsewhere suggests that many have colonized environments that are radically different from their sources (Lee, 2002) no evidence of this has been found in our non-indigenous cave-inhabiting fauna: all the introduced species listed in Table 1 are also reported either as common in caves (as troglophiles or habitual trogloxenes) or as guanophiles in their region of origin. There can be little doubt that rather than competing with established fauna, some or most of these exotic species are exploiting empty biotopes in the subterranean environment. In the case of the Isopoda and Diplopoda there are no native species in the lists, all are introductions. Most of the cave-collected earthworms and mollusca are non-indigenous taxa, and almost half the terrestrial beetles are European in origin. The widespread Nearctic threshold spider Nesticus pallidus was not found: it is replaced in this habitat by the closely related European N. cellulanus. The existence of previously unfilled ecological niches must be at least in part due to the taxonomic impoverishment of the Nova Scotian fauna resulting from zoogeographical isolation of the province. Several authors (e.g. Chapman, 1993) have pointed out that cave faunas in formerly glaciated regions are in an early dynamic phase of recolonisation and adaptation. It appears that recolonisation of hypogean habitats in Nova Scotia has been relatively even more delayed and thus that subterranean communities are in general at an earlier stage of this process than those in other otherwise similar geographical areas.
In Maritime Canada we also have the unusual situation where there are eutrophic dung communities in this early phase: guano caves in the humid tropics are ancient systems never subjected to glaciation. It is believed that the porcupine is a relatively late postglacial arrival in Nova Scotia (Calder & Bleakney, 1965).
Colonisation, adaptation and speciation in cave faunas can be rapid processes that may be taking place on a human timescale. Lava tubes on the geologically very young island of Hawaii, which is less than 700,000 years old, already harbour a rich fauna of highly-adapted troglobites consisting of representatives of native groups in the process of adaptive radiation (Howarth, 1972).
Significance of the subterranean fauna
Before 1970 only a few local caves were well-known to naturalists and cave ecosystems in the region were assumed to be isolated, localised and restricted. Maritime Canadian cave fauna was not recognised as ecologically significant. However, subsequent documentation of many more caves together with the important insight from elsewhere that most so-called “cave” fauna is not restricted to caves (an anthropocentric concept) as such but is widely distributed throughout mesocavernous voids and the MSS (“milieu souterrain superficial”) (Juberthie, 1984 and references therein) means that we must now recognize subterranean ecosystems as provincially and regionally notable. The porcupine dung habitat is exceptional and thus particularly important.
Underground habitats are in themselves unusual and fascinating, and cave ecosystems are natural systems that in most cases have not been directly modified by man. Also, due to the protection that they offer from large oscillations in climate, caves and subterranean waters are habitats where species normally living further to the north or to the south may sometimes be found (Gunn et al., 2000).
The cave fauna of Nova Scotia is now probably the most extensively and comprehensively sampled and documented of any geographic region in Canada. Records go back to the 1960s and large documented collections of both terrestrial and aquatic invertebrates have been made since then in caves and abandoned underground mines around the province.
Invasive and other species in Maritime caves may have high potential for investigating and testing biospeleological evolutionary theories with populations at a very early stage of active colonisation and adaptation to the subterranean environment.
ACKNOWLEDGEMENTS
Any prolonged wide-ranging study is impossible without the help and participation of many individuals, and it is unfortunately impossible to name them all here. I would especially like to express my gratitude to the many taxonomists who identified specimens: there is a list in Moseley (1998). Mention must also be made of those individuals who accompanied me on field trips: especially M. Hamilton, C. Murphy, the Proctor brothers, Dr. D. Sawatzky, and the late Dr. P. Schwinghamer. My colleagues amongst the permanent staff and Research Associates of the Nova Scotia Museum of Natural History have also helped in many ways. Dr. Dale Calder (Royal Ontario Museum) commented on the manuscript and also provided useful unpublished data from Frenchman’s Cave. Additional unpublished records were generously provided by Dr. Don McAlpine (NB Museum) as well as by Calum Ewing and Andrew Hebda (NS Museum) who also prepared Fig. 1. A mention needs to be made of Harry Bassett and Fred St. Peters who skillfully recovered decades of notes and raw data from a set of corrupted computer diskettes. Finally, thank you to the two anonymous reviewers who critiqued the draft manuscript and made a number of very useful and constructive suggestions. Fieldwork in 1996/97 was supported in part by a grant from the endowment fund of the Nova Scotia Museum Board of Governors. Fig. 1 was generated from the NS Museum MIMS database.
Answers to all your Biology Questions
