Influence of bark pH on the occurrence and distribution of tree canopy myxomycete species Sydney E. Everhart Harold W. Keller 1 Joseph S. Ely Department of Biology, University of Central Missouri, Warrensburg, Missouri 64093 Abstract: This study compares the occurrence and distribution of myxomycete species in the canopy of living trees and neighboring grapevines. Corticolous myxomycetes of three temperate forests in southeast- ern USA were studied on six tree species (30 trees) and grapevines (30 vines) to determine distribution and occurrence of myxomycete species relating to geographic location, host species, and bark pH. The double-rope climbing technique was used to access the canopy and sample bark up to 16.5 m. Bark samples were examined in 580 moist chamber cultures and 44 myxomycete species were identified representing 21 genera, averaging 3.0 6 2.1 species per sample site. Jaccard’s coefficient determined community similarity between five individuals of six tree species, Acer saccharum, Fraxinus americana, Liquidambar styraciflua, Liriodendron tulipifera, Pla- tanus occidentalis and Tsuga canadensis, and neigh- boring grapevines, Vitis aestivalis and V. vulpina. Vertical variation in species richness was significantly different only for Platanus occidentalis and might be attributable to flaking of bark with increasing height in the canopy. Tsuga canadensis and neighboring grapevines had greatest community similarity. Cri- braria violacea was observed on all tree and grapevine species except T. canadensis and neighboring grape- vines. Occurrence and species assemblages of myxo- mycetes were associated with bark pH, not geographic location. Bark of V. aestivalis (pH 4.5) was more acidic than neighboring T. canadensis (pH 4.1), compared to grapevines of the same species neigh- boring other tree species. Results indicated that most species are not regionally restricted, and although some myxomycetes are associated with a certain pH range, others develop on any substratum. Future research protocols for corticolous myxomycetes should emphasize sampling adequate amounts of substrata in a local region from different host species that have a wide range of bark pH, ensuring a representative sample of species for an entire region. Key words: bark pH, Berea College Forest, corticolous myxomycetes, Daniel Boone National Forest, Great Smoky Mountains National Park, Kentucky, ordination, Tennessee, tree canopy, vertical distribution INTRODUCTION Myxomycetes are Protista characterized by a spore that gives rise to biflagellate swarm cells or myxamoebae which, after syngamy, form a multinucleate plasmodi- um that eventually gives rise to one or many fruiting bodies. Myxomycetes have been found throughout the world, with the highest diversity documented in temperate forests (Spiegel et al 2004). Within temper- ate forests some species inhabit only the forest floor on decayed wood, other species are found on decaying leaf litter and still others only occur on the bark of living trees and vines (Keller and Braun 1999). Myxomycetes that complete their life cycle from spore to fruiting body formation on the bark of living trees and vines are termed corticolous myxomycetes (Keller and Brooks 1973). Although certain species, such as Fuligo septica (L.) F.H. Wigg. and Lycogala flavofuscum (Ehrenb.) Rostaf., have been found fruiting up to 2 m on the bark of living trees, they typically complete their life cycle on decaying leaves and logs on the forest floor. For this reason collecting true corticolous myxomycetes involves sampling bark above 3 m, here defined as the tree canopy. Corticolous myxomycetes in the tree canopy were studied first by Snell and Keller (2003). They found a few common species among all tree species, and different species assemblages of myxomycetes were associated with different tree species and bark pH but showed no difference with respect to height in the canopy. On individual trees, distribution of myxomy- cetes on each tree was uneven and patchy, and the community of myxomycete species varied consider- ably from tree to tree. Using Sorensen’s community coefficient, Snell and Keller (2003) noted that myxomycete communities on trees of the same species had low similarities with one another, while trees of different species had some of the highest similarities. These results are counterintuitive and might indicate regional occurrence of myxomycete species. Unfortunately comparisons between trees from the same location were not made. Contrary to this hypothesis other studies suggest that species are Accepted for publication 23 January 2008. 1 Corresponding author. E-mail: keller@ucmo.edu Mycologia, 100(2), 2008, pp. 191–204. # 2008 by The Mycological Society of America, Lawrence, KS 66044-8897 191