During the few last decades dead, dying and living trees with hollows or dead limbs have been recognised as a key element of the forest ecosystem. The presence of dead wood strongly and positively affects their biodiversity, nutrient and water cycling, decomposition and soil development processes, and natural regeneration (e.g., Harmon et al. 1986; Muller & Schnell 2003; Bobiec et al. 2005; Stokland et al. 2012). Probably almost one quarter of all forest species are dependent on dead wood, and some of them, the socalled saproxylic, cannot live in woods devoid of old and dead trees (Siitonen 2001; Schuck et al. 2004). Speight (1989) defined a saproxylic organism as one “that is dependent, during some part of its life cycle, upon the dead or dying wood of moribund or dead trees, or upon wood-inhabiting fungi, or upon the presence of other saproxylics”. Such organisms use different kinds of dead wood not only as a source of food, but also as habitat, seasonal or daily shelter, construction material, a mating substrate or a location for other reproduction purposes. To date, the largest fraction of saproxylic arthropods are described and recognised within beetles (Coleoptera), flies, gnats (Diptera), wasps, ants and bees (Hymenoptera) (Harmon et al. 1986; Hammond 1997; Dajoz 2000; Buddle 2001; Siitonen 2001; Stokland et al. 2012). Nevertheless, knowledge of many other groups of dead wood-dependent invertebrates, especially microarthropods, is still incomplete and far from satisfactory.

Springtails (Collembola) belong to the smallest group of hexapodous arthropods (most are only a few millimeters in length) and are the most diverse and widespread among (Hopkin 1997). At present, more than 8000 species and 600 genera are described worldwide (Bellinger et al. 2014). Collembola are generally polyphagous, however, the majority of them consume fungal hyphae and fruiting bodies, decaying vegetation, organic detritus, algae and micro-organisms (Hasegawa & Takeda 1995; Chen et al. 1995; Ponge 2000). Due to their feeding habits and abundance in forest litter and soil (where their densities can reach more than 10,000 individuals/m²) springtails play an important role in decomposition and soil development processes (e.g., Heneghan et al. 1998; Berg et al. 2001; Bradford et al. 2002). Unfortunately, data with respect to Collembola and their dead wood dependence is very sparse, fragmented and notably incomplete.

To improve this unsatisfactory knowledge some research on saproxylic springtails was conducted in 2006, 2007 and 2009 at the H. J. Andrews Experimental Forest Long Term Ecological Research site, located in the western Cascade Range (Oregon, USA). A large part of this area is covered by old-growth forests characterized by a high amount (more than 1000m³/ha) of high quality and structurally complex dead wood. These forests, presently unique in the temperate zone of the Northern Hemisphere, seem to be one of the best places to study diversity of saproxylic assemblages. Analysis of materials collected from rotten logs allowed the identification of a hitherto unknown member of the genus Paleonura Cassagnau, 1982. This genus was established by the eminent French collembologist Paul Cassagnau in 1982 and later assigned by him to the tribe Paleonurini (Cassagnau 1989), one of the 6 tribes within the subfamily Neanuirnae. The type species of the genus, P. spectabilis Cassagnau, 1982, was described from Nepal (the Himalaya). Presently Paleonura is one of the most species-rich within the tribe and the whole subfamily, with most of 51 described taxa distributed in mountainous tropical regions. Previously, only 2 members of Paleonura have been identified in the Nearctic region, i.e., P. anops (Christiansen & Bellinger, 1980), previously identified in California (Petaluma, Sonoma County) and P. peterbellingeri found in Porter's Cave in Virginia (Bath Co.) (Palacios-Vargas & Simon 2007). Interestingly, the aforementioned cave is the northernmost locality of the whole genus. It should be also mentioned that P. anops was first described in the genus Paranura Axelson, 1902, and later transferred to Paleonura by Palacios-Vargas & Simon Benito (2007).

In addition, here we document the first occurrence of Galanura agnieskae Smolis, 2000 from Canada and North America. This member of the Paleonurini species was described recently from Poland (Europe) and has only been found in one locus typicus so far. The new finding notably extends the known distribution of the species. Specimens of the taxon were found in rotten logs of white birch in boreal forests of South Quebec (Canada) during an excursion of the International Symposium “On dynamics and ecological services of deadwood in forest ecosystems” (15-19-V-2011, Rouyn-Noranda).

Materials and Methods

Specimens were collected under bark and inside rotting wood using a mechanical aspirator. In addition, some deadwood samples were extracted in Tullgren traps. The specimens were cleared in Nesbitt's fluid, subsequently mounted on slides in Hoyer's solution and observed using a phase contrast microscope Nikon Eclipse 80iR. Figures were made using a camera lucida. Photographs were taken with a Canon 500DR (Taiwan) camera under a Nikon Eclipse 80iR (Tokyo, Japan). Image stacks were processed using Combine ZMR (Hadley 2010). Materials are deposited in the Department of Invertebrate Biology, Evolution and Conservation, Institute of Environmental Biology, Faculty of Biological Science, University of Wrocław, Poland (DIBEC) and in the Museum National d'Histoire naturelle in Paris, France (MNHN). Terminology for the description follows those of Deharveng (1983), Deharveng & Weiner (1984), Greenslade & Deharveng (1990) and Smolis (2008).

The following abbreviations were used in this study:

General morphology: abd.—abdomen, ant.—antenna, AOIII—sensory organ of antennal segment III, Cx—coxa, Fe—femur, Scx2—subcoxa 2, T—tibiotarsus, th. —thorax, Tr—trochanter, VT—ventral tube.

Groups of chaetae: Ag—antegenital, An— chaetae of anal lobes, ap—apical, ca—centroapical, cm—centromedial, cp—centroposterior, d—dorsal, Fu—furcal, vc—ventrocentral, Ve or ve—ventroexternal, Vea—ventroexternoanterior, Vem—ventroexternomedial, Vep—ventroexternoposterior, Vel—ventroexternolateral, Vec—ventroexternocentral, Vei—ventroexternointernal, Vi or vi—ventrointernal, Vl-ventrolateral.

Tubercles: Af—antenno-frontal, Cl—clypeal, De—dorsoexternal, Di—dorsointernal, Dl—dorsolateral, L—lateral, Oc—ocular, So—subocular.

Types of chaetae: Ml—long macrochaeta, Mc—short macrochaeta, Mcc—very short macrochaeta, me—mesochaeta, mi—microchaeta, ms—s-microchaeta, S or s—chaeta s, bs—s- chaeta on ant. IV, miA—microchaetae on ant. IV, iv—ordinary chaetae on ventral ant. IV, or— organite of ant. IV, brs—border s-chaeta on ant. IV, i—ordinary chaeta on ant. IV, mou—cylindrical s-chaetae on ant. IV, L'—ordinary lateral chaeta on abd. V, B4, B5—ordinary chaetae on tibiotarsi.

Results

Table 1.

Cephalic chaetotaxy of Paleonura saproxylica sp. nov.

Figs. 1–5.

Paleonura saproxylica sp. nov. 1, dorsal and lateral chaetotaxy of head and thorax 1 (holotype); 2, dorsal chaetotaxy of th. II–III; 3, dorsal chaetotaxy of abd. III; 4, dorsal chaetotaxy of abd. IV–VI; 5, ventral tube. Black arrows show axis of symmetry and orientation for position of the head. Black horizontal lines show borders of the segments.

Fig. 6.

Old-growth forest of Tsuga heterophylla zone, “Cougar 1” site, type locality of Paleonura saproxylica sp. nov.

Fig. 7.

Habitat of Galanura agnieskae in Quebec, Canada.

Table 2.

Postcephalic chaetotaxy of Paleonura saproxylica sp. nov.

Discussion

Knowledge of saproxylic Collembola is poor in quality and quantity. Fragmented and general data can be found in some taxonomic and faunistic papers (e.g., Fjellberg 1985; Christiansen & Bellinger 1998; Babenko & Fjellberg 2006). The latest studies suggest that Collembola occur in rotten wood, and many Collembola could and should be treated like typical saproxylic organisms (Stevens et al. 2007; Garrick et al. 2008; Smolis 2008; 2011) - including the species described in this paper. Despite sampling of soil and litter, both species have, so far, only been found in dead wood. Thus, we can clearly state that we are dealing with saproxylic organisms. Interestingly, the morphology of these 2 species is similar. It is characterized by lack of the pigment and total reduction of the eyes (in the subfamily Neanurinae usually 2 or 3 eyes are situated on each side of the head), and by the strong reduction of cuticular prominences. Two of the first mentioned characteristics are common for Collembola, but they are mostly limited to springtails living in deep layers of the soil (edaphic forms) and inside caves. Without a doubt these characteristics have developed to support a life inside dead wood, where the light is heavily limited. Many of the representatives of the subfamily Neanurinae have developed strong cuticular prominencies (named tubercles), which in extreme cases may take the form of fingerlike papillae (e.g. in Rambutanura genus). The reduction of these morphological structures in both described species can be also an adaptation to tight and narrow spaces in dead wood.

It is interesting that Paleonura saproxylica was found exclusively in wood of coniferous trees, despite of the presence of fallen trunks of deciduous trees at Mona.

Moreover, another species from a related tribe Neanurini - Xylanura oregonensis Smolis (Smolis 2011) was discovered in the same place (Mona) inside similar trunks of deciduous trees. Surprisingly, this species has well-developed eyes (3 eyes on each side of the head), yet, a hypodermal pigment is present (Smolis 2011). This raises the question of why this taxon shows none of the reduction observed in the saproxylic species, despite the fact that it has only been found in dead wood habitats and is deemed to be saproxylic. A comparison of the diam and the degree of decay of deciduous stumps harboring Xylanura versus conifers where P. saproxylica was found may provide a hint. The first of these, namely hardwood stumps have a much smaller diam. Moreover, the rate of decay of these hardwoods is several times faster than that of coniferous trees in the area. This means that populations of Xylanura are forced to move between the stumps far more often than those species living in coniferous logs. In addition, deciduous species in the area are much harder to find and can be found mostly in along stream channels and disturbed places affected by anthropogenic or natural factors such as landslides or avalanches (Franklin & Dyrness 1988).

Fig. 8.

Habitat of Galanura agnieskae in Poland (locus typicus).

Recent decades have brought about significant progress in our knowledge of the biodiversity of dead wood. We hope that this trend will expand to include small arthropods and will help with determining not only the species richness, but also the role that they play in this little-known habitat.