J. Acarol. Soc. Jpn., 14 (2): 117-122. November 25, 2005  The Acarological Society of Japan http://acari.ac.affrc.go.jp/ 117

[SHORT COMMUNICATION] Wind-based Dispersal of Oribatid (: ) in A Subtropical Forest in Japan

1* 2 3 1 Shigenori KARASAWA , Kenshi GOTOH , Takeshi SASAKI and Naoki HIJII 1 Laboratory of Forest Protection, Graduate School of Bioagricultural Sciences, Nagoya University, Nagoya 464–8601, Japan 2 Environmental Information Sciences Laboratory, Graduate School of Agriculture, University of the Ryukyus, Nishihara 903–0213, Japan 3 Museum, University of the Ryukyus, Nishihara 903–0213, Japan (Received 23 May 2005; Accepted 27 September 2005)

Key words: canopy, dispersal mode, oribatid mites, Ryukyu archipelago, window trap

INTRODUCTION

Oribatid mites (Acari: Oribatida) are found not only on the forest floor and in arboreal habitats (e.g., Aoki, 1973; Behan-Pelletier and Walter, 2000; Travé, 1963), but also in littoral habitats (e.g., Karasawa and Hijii, 2004a, b; Luxton, 1992). Some species constantly live in bird’s feathers (Krivolutsky and Lebedeva, 2004a, b). Thus, clarifying the habitat use of oribatid mites can contribute to an understanding of the relationships between habitat diversity and biodiversity. The modes of dispersal used by individuals can determine whether or not a species can use a given habitat, and consequently contribute to defining the community structure of a given habitat (Prinzing and Woas, 2003). Thus, knowledge of the dispersal modes used by oribatid mites can help us to evaluate the importance of colonization processes in the development of their communities in various habitats. Oribatid mites can climb trees (Murphy and Balla, 1973), and some can even jump more than 10 cm (Krisper, 1990). However, they are known to use several other dispersal methods. Small mammals and insects pick up and disperse the mites passively (Miko and Stanko, 1991; Norton, 1980), and birds are also an important disperser for oribatid mites. Birds carry oribatid mites directly on their bodies, or transport them indirectly by including -infested litter in their nest materials (Aoki, 1966; Jacot, 1930). In addition, seawater and freshwater are likely to play critical roles in the dispersal of oribatid mites in both littoral and terrestrial environments (Coulson et al., 2002; Woodring

日本の亜熱帯林内におけるササラダニの風分散様式 1* 2 3 1 1 唐沢 重考 ・後藤 健志 ・佐々木 健志 ・肘井 直樹 （ 名古屋大学大学院生命農学研究科森林保護学 2 研究室，〒 464–8601 名古屋市千種区不老町； 琉球大学大学院農学研究科環境情報科学研究室，〒 903– 3 0213 沖縄県中頭郡西原町； 琉球大学資料館，〒 903–0213 沖縄県中頭郡西原町） * Corresponding author: e-mail: [email protected]; fax: +81–52–789–5518 DOI: 10. 2300/acari. 14. 117 118 Shigenori KARASAWA et al. and Cook, 1962). In arboreal environments, however, oribatid mites are most likely to be dispersed from one patch of habitat to another by wind currents (Behan-Pelletier and Winchester, 1998). For example, Behan-Pelletier and Winchester (1998) pointed out that active dispersal by random movement is an important mode of colonization of canopy habitats by oribatid mites. Ichisawa (2001) reported that oribatid mites associated with arboreal habitats sometimes arrived on the roofs of buildings in Japan, carried there by wind currents. However, there is currently no information available on wind-based dispersal of oribatid mites in forests in Japan. In the present study, we examined the dispersal mode of oribatid mites dispersed by wind currents using window traps established at two heights in a subtropical forest in Japan.

MATERIALS AND METHODS Study area We conducted our study in an old-growth evergreen broad-leaved forest in the Yona Experimental Forest at the University of the Ryukyus, in the northern part of Okinawa Island, southwestern Japan (26°49'N, 128°5'E; 250 to 330 m above sea level). The mean annual temperature is 23.0°C, and annual precipitation between 1992 and 2003 averaged 2330 mm (Yona Experimental Forest, University of the Ryukyus). The bedrock is composed of sandstone and slate, and the forest soil is classified as a yellow soil (Y) according to the classification methods of forest soils in Japan (Morisada, 1999). This forest is dominated by Castanopsis sieboldii with a maximum height of less than 20 m. There is no history of logging or other artificial disturbance in this area during the past 50 years (Enoki, 2003). Window traps The window traps used in our study comprise a roof, two transparent vanes (30×30 cm) that are joined at a 90° angle, and a yellow plastic bucket that forms the floor of the trap (29 cm diam. and 23.5 cm deep). The bucket contains a saturated solution of sodium sorbate (ca. 2 L) plus a small quantity of detergent. On a tower built around a single C. sieboldii tree at valley, a window trap was positioned in the canopy (middle of the foliage), 15 m above the ground, and a second trap near the forest floor (1 m above the ground). Because this was an exploratory study rather than an attempt to statistically compare mite dispersal mecha- nisms at various level within the forest, we used only a single sampling location. We collected samples from both traps almost every week from 26 February 2003 to 25 April 2004 (102 samples, on 51 sampling dates). We separated oribatid mites from other collected in the solution under a binocular dissecting microscope with a magnifi- cation of 30X, and then identified the individuals to the morphospecies level and counted the numbers in each morphospecies under a binocular microscope with a magnification of 400X. Oribatid faunas in the arboreal and forest-floor habitats The oribatid faunas in this forest have previously been investigated in seven habitats at a plot ca. 200 m northeast of the sampling location in the present study (Karasawa and Hijii, unpublished data). We obtained the following results: leaves (1,165 adults in 8 species), branches (816 adults in 13 species), and bark (3,918 adults in 85 species) of C. sieboldii; Wind dispersal of oribatid mites 119 forest-floor litter (4,608 adults in 113 species) and soil (4,164 adults in 89 species); and litter (4,023 adults in 90 species) in and roots of (26,308 adults in 105 species) bird’s nest ferns attached at heights of 0.1 to 10.5 m above ground level. That study collected oribatid mites from litter, soil, and ferns using Tullgren funnels and from leaf, branch, and bark samples by washing with dilute detergent and then filtering the solution through a 35-µm nylon mesh (Karasawa and Hijii, 2005).

RESULTS

We collected a total of 313 individuals in 49 morphospecies from the traps in the canopy (n=29) and near the forest floor (n=44) during the sampling period (Table 1). On several sampling dates (n=29), we collected no oribatid mites. The numbers of species and individuals of oribatid mites were both larger for the trap near the forest floor. Most of the oribatid species (87.8%) collected by both traps had already been identified in this forest (Table 1). Large proportions of the oribatid species previously collected from canopies (leaves and branches) and bark were also collected by the trap near the forest floor (Table 2). In contrast, the ratios of the number of the species common between those collected by the trap near the forest floor and those recorded from the forest-floor litter and soil to the number of species recorded from the forest-floor litter and soil were much lower, and the ratios with respect to the bird’s nest fern habitats were intermediate (Table 2). The ratios with respect to the number of oribatid species collected by the trap in the canopy followed a similar trend: the ratios for the three arboreal habitats were larger than those from the bird’s nest fern and the forest floor, and all values were considerable lower than for the forest-floor trap (Table 2).

DISCUSSION

Most of the oribatid species that inhabited the canopy (leaves and branches) and bark were caught by both the window traps (Table 2). Among the three habitats, the ratio of the number of common species between collected by these traps and recorded from the bark to the number of oribatid species recorded from the bark was lower than the corresponding ratios for the leaves and branches. However, more than 80% of all the oribatid species collected by both traps were also found on the bark. These results suggest that dispersal by means of wind currents is an important mode of colonization for oribatid mites living in these three arboreal habitats in this forest. On the other hand, oribatid mites inhabiting the forest-floor habitats and suspended soil (the litter in and roots of bird’s nest ferns) may disperse with more difficulty using wind currents than oribatid mites living in the three arboreal habitats, although some are nonetheless likely to be dispersed by wind currents (Tables 1, 2). The fact that larger numbers of species and individuals of oribatid mites were collected by the trap near the forest floor indicates that most of the wind-dispersed oribatid mites may not move upward, but instead fall to the ground. Ichisawa (2001) also suggested that most oribatid species that inhabit the forest floor are unlikely to move upward from the forest 120 Shigenori KARASAWA et al.

Table 1. Families and species of oribatid mites collected by window traps at two heights (canopy, 15 m above the ground; forest floor, 1 m above the ground) and their presence ( ○ ) and absence (—) in a previous study (Karasawa and Hijii, unpublished data) from seven habitats at a nearby site in the forest.

Window trap Canopy Bird’s nest fern Forest-floor Family Species Bark Floor Canopy Leaf Branch Litter Root Litter Soil Cosmochthoniidae Cosmochthonius reticulatus Grandjean 1 1 ——— — ○ —— Hoplophthiracarus sp. 1 ——○○ — ○○ ¶ Malaconothridae Trimalaconothrus spp. 1 ——○○ ○ —— Liodidae Liodes zimmermanni Sellnick 1 ——○○ — ○ — Cepheidae Ommatocepheus sp. 2 — ○○ ———— Carabodidae Austrocarabodes sp. 1 1 ——○ — ○○— Carabodes ikeharai Aoki 3 1 ——○ ———— Otocepheidae Fissicepheus coronarius Aoki 1 ——○○ ○ —— Eremellidae Eremella sp. 1 ——○ ———— § Machuellidae Machuella spp. 1 ——○ — ○ — ○ Oppiidae Oppiella nova (Oudemans) 4 ——○○ ○ ○○ Subiasella incurva (Aoki) 1 ——○○ ○ ○○ Suctobelbidae Suctobelbidae sp. 1 ——— — — —— ¶ Suctobelbidae spp. 1 ——○○ ○ ○○ Cymbaeremaeidae Scapheremaeus sp. 1 3 1 — ○○ ——○ — Scapheremaeus sp. 2 1 ——○ ———— Scapheremaeus sp. 3 4 — ○○ ———— Scapheremaeus sp. 4 2 ——○ ———— Scapheremaeus sp. 5 1 — ○○ ———— Micreremidae Micreremus sp. 6 ——○○ ——— Parakalummidae Neoribates roubali (Berlese) 3 ——○○ ○ ○○ Mochlozetidae Unguizetes sp. 1 ——○○ — ○○ Symbioribatidae Symbioribates sp. 5 5 ——○ ———— Oribatulidae Dometorina sp. 1 4 11 ○○○ ○ ——— Dometorina sp. 2 14 5 ○○○ ○ — ○○ Haplozetidae Incabates sp. 1 ○ — ○ ——○ — Peloribates ominei Nakatamari 60 5 ——○○ ○ ○○ Rostrozetes ovulum (Berlese) 5 1 ——○○ ○ ○○ Haplozetidae sp. 2 ——○○ — ○ — Scheloribatidae Scheloribates sp. 2 ○ — ○○ ○ —— Tuberemaeus singularis Sellnick 15 3 ○○○ ○ ○ ○— Oripodidae Cosmopirnodus pulcherrimus Balogh 3 4 ○○○ ○ ——— Oripoda sp. 1 22 20 ○○○ ○ ○ —— Oripoda sp. 2 1 ——— — — —— Oripoda sp. 3 1 ——○ ———— Pirnodus sp. 2 1 ——○ ———— Truncopes moderatus Aoki and Ohkubo 17 ○○○ ○ ○ —— Truncopes sp. 1 13 5 — ○○ ○ ○ ○— Truncopes sp. 2 1 ——— — — —— Birobatidae Brachyoripoda sp. 1 ——○ ———— Phenopelopidae Eupelops acromios (Hermann) 2 ——— — — ○ — Galumnidae Trichogalumna sp. 1 ——○○ ○ ○— Galumnidae sp. 1 ——○○ ○ —— Ceratokalummidae Cultrobates nipponicus Aoki 1 ——○○ ○ ○— Unidentified Unidentified sp. 1 1 ——— — — —— Unidentified sp. 2 4 ——○ — ○ —— Unidentified sp. 3 1 ——— — — —— Unidentified sp. 4 4 ——○ ———— Unidentified sp. 5 1 ——— — — —— Juvenile 16 7 ○○○ ○ ○ ○○ Total number of oribatid species 42 21 8 11 41 24 20 19 10 ——0* 2* 44* 66* 85* 94* 79* ¶ : multiple species belonging to the two taxa were regarded as one species because it was difficult to identify them to species level. § : included not less than the two species, M. ventrisetosa and M. lineata. *: the numbers of oribatid species that were not collected by the two window traps but were only recorded from each habitat. Wind dispersal of oribatid mites 121

Table 2. Ratios of the number of the species common between those collected by the two window traps and those recorded from each habitat in a previous study (Karasawa and Hijii, unpublished data) to the number of species recorded from each habitat. Larger values represent smaller differences in the number of oribatid species between categories.

Window trap Habitat Forest floor Canopy

Canopy Leaf 100.0 62.5 Branch 76.9 61.5 Bark 42.4 21.2 Bird’s nest fern Litter 23.3 12.2 Root 18.1 7.6 Forest floor Litter 15.0 8.0 Soil 9.0 5.6 floor into the arboreal habitats using wind currents. It is likely that most of the oribatid mites that inhabit the arboreal habitats can move horizontally between patches in the arboreal habitats or can move down toward the forest floor on wind currents, whereas few oribatid species that inhabit the forest floor can move upward from the forest floor into the arboreal habitats using these wind currents.

ACKNOWLEDGEMENTS

We thank Dr. T. Enoki of the University of the Ryukyus for permission to work at the Yona Experimental Forest at the University of the Ryukyus. We thank Mr. I. Kawashima in Okinawa and Mr. N. Shimada of the University of the Ryukyus for supporting the field work; and Mr. N. Ohkubo of the Mie Plant Protection Office for valuable advice on oribatid mites; and Drs. E. Shibata and H. Kajimura, and all members of the Laboratory of Forest Protection, Nagoya University, for helpful suggestions.

REFERENCES Aoki, J. (1966) Oribatid mites from bird’s nests on midway island (Acari: Cryptostigmata). Pacific Insects, 8: 770– 776. Aoki, J. (1973) Soil mites (Oribatids) climbing trees. In: Proceedings of 3rd international congress of acarology, Prague, 31 August to 6 September 1971. (eds., Daniel, M. and B. Rosicky), pp. 59–64. Junk, Hague. Behan-Pelletier, V. and N. Winchester (1998) Arboreal oribatid mite diversity: colonizing the canopy. Applied Soil Ecology, 9: 45–51. Behan-Pelletier, V. and D. E. Walter (2000) Biodiversity of oribatid mites (Acari: Oribatida) in tree canopies and litter. In: Invertebrates as webmasters in ecosystems. (eds., Coleman, D. C. and P. F. Hendrix), pp. 187–202. CAB International, Wallingford. Coulson, S. J., I. D. Hodkinson, N. R. Webb and J. A. Harrison (2002) Survival of terrestrial soil-dwelling arthro- pods on and in seawater: implications for trans-oceanic dispersal. Functional Ecology, 16: 353–356. Enoki, T. (2003) Microtopography and distribution of canopy trees in a subtropical evergreen broad-leaved forest in the northern part of Okinawa Island, Japan. Ecological Research, 18: 103–113. Ichisawa, K. (2001) A taxonomical and ecological study on arboreal collembolans (Insecta: Collembola) and oribatid mites (Acari: Oribatida). Ph. D. thesis. 121 p., Yokohama National University, Kanagawa, Japan. (in 122 Shigenori KARASAWA et al.

Japanese) Jacot, A. P. (1930) Biological notes on the moss-mites. The American Naturalist, 64: 285–288. Karasawa, S. and N. Hijii (2004a) Effects of microhabitat diversity and geographical isolation on oribatid mite (Acari: Oribatida) communities in mangrove forests. Pedobiologia, 48: 245–255. Karasawa, S. and N. Hijii (2004b) Morphological modifications among oribatid mites (Acari: Oribatida) in relation to habitat differentiation in mangrove forest. Pedobiologia, 48: 383–394. Karasawa, S. and N. Hijii (2005) Efficiency of sampling oribatid mites (Acari: Oribatida) from the canopies of broad-leaved trees in Japan: a comparison between the washing and direct methods. Journal of the Acarological Society of Japan, 14: 19–24. Krisper, G. (1990) Das sprungvermögen der milbengattung Zetorchestes (Acarida, Oribatida). Zoologische Jahr- bucher. Abteilung fur Anatomie und Ontogenie der Tiere, 120: 289–312. Krivolutsky, D. A. and N. V. Lebedeva (2004a) Oribatid mites (Oribatei, Acariformes) in bird’s feathers: non- Passerines. Acta Zoologica Lituanica, 14(1): 26–40. Krivolutsky, D. A. and N. V. Lebedeva (2004b) Oribatid mites (Oribatei) in bird’s feathers: Passeriformes. Acta Zoologica Lituanica, 14(2): 19–38. Luxton, M. (1992) Oribatid mites from the marine littoral of Hong Kong (Acari: Cryptostigmata). In: The Marine Flora and Fauna of Hong Kong and Southern China III. (ed., Morton, B.), pp. 211–227. Hong Kong University Press, Hong Kong. Miko, L. and M. Stanko (1991) Small mammals as carriers of non-parasitic mites (Oribatida: Uropodina). In: Modern Acarology. Vol. I. (eds. Dusbábek, F. and V. Bukva), pp. 395–402. Academia, Prague. Morisada, K. (1999) Classification methods of forest soils. In: Forest Environment Examinations. (ed., Japanese Society of Forest Environment), pp. 28–32. Hakuyu-sha, Tokyo. (in Japanese) Murphy, P. W. and A. N. Balla (1973) The bionomics of Humerobates rostrolamellatus Grandjean (Cryptostigmata- Ceratozetidae) on fruit trees. In: Proceedings of 3rd international congress of acarology, Prague, 31 August to 6 September, 1971. (eds., Daniel, M. and B. Rosicky), pp. 59–64. Junk, Hague. Norton, R. A. (1980) Observations on phoresy by oribatid mites (Acari: Oribatei). International Journal of Acarology, 6: 121–130. Prinzing, A. and S. Woas (2003) Habitat use and stratification of collembola and oribatid mites. In: Arthropods of tropical forests. (eds., Basset, Y., V. Novotny, S. E. Miller, and R. L. Kitching), pp. 271–281. Cambridge University Press, Cambridge. Travé, J. (1963) Écologie et biologie des Oribates (Acariens) saxicoles et arbicoles. Vie et Milieu, Supplement, 14: 1–267. Woodring, J. P. and E. F. Cook (1962) The biology of Ceratozetes cisalpinus Berlese, Scheloribates laevigatus Koch, and Oppia neerlandica Oudemans (Oribatei), with a description of all stages. Acarologia, 4: 101–137.