Eagle Hill Masthead

Southeastern Naturalist
    SENA Home
    Range and Scope
    Board of Editors
    Editorial Workflow
    Publication Charges

Other EH Journals
    Northeastern Naturalist
    Caribbean Naturalist
    Urban Naturalist
    Eastern Paleontologist
    Eastern Biologist
    Journal of the North Atlantic

EH Natural History Home


About Southeastern Naturalist


Spatial Distribution of Epiphytic Diatoms on Lotic Bryophytes
Jessica M. Knapp and Rex L. Lowe

Southeastern Naturalist, Volume 8, Number 2 (2009): 305–316

Full-text pdf (Accessible only to subscribers.To subscribe click here.)


Site by Bennett Web & Design Co.
2009 SOUTHEASTERN NATURALIST 8(2):305–316 Spatial Distribution of Epiphytic Diatoms on Lotic Bryophytes Jessica M. Knapp1,2,3,* and Rex L. Lowe1,2 Abstract - In stream ecosystems, bryophytes greatly increase substrate heterogeneity and support a high density and diversity of lotic primary producers, such as epiphytic algae. However, there is little information about how the spatial distribution and density of epiphytic diatoms varies with respect to bryophyte morphology. This study examined epiphytic diatom communities from the contrasting bryophyte morphologies of mosses and liverworts. We predicted that mosses, with morphologies that create more crevices, would have a higher density of epiphytic diatoms than liverworts, with leaves highly exposed to the turbulence of the stream current. Six species of bryophytes (two mosses and four liverworts) were collected from streams in the Great Smoky Mountains National Park, and 37 species of epiphytic diatoms were identified on these bryophytes. Diatom density was significantly higher on the adaxial leaf surface of mosses compared to the abaxial leaf area (ANOVA, df = 29, P < 0.001). There was no difference in diatom density on either the adaxial or abaxial leaf surfaces of liverworts, and these diatom densities were statistically identical to the density observed on the abaxial surface of moss leaves. The findings of our study support our hypothesis that the morphology of mosses, comprised of leafy whorls, provides a greater level of protection from disturbance than the open, fl at nature of leafy liverworts. These findings emphasize that differences in microscale habitats can result in varying diatom distribution and density that may be critical to food-web interactions, such as grazing. Introduction Diatoms are key components of primary production in stream ecosystems, and the structural complexity of the available substrata is pivotal in the determination of diatom density and community structure (Eminson and Moss 1980). Bergey (1999) demonstrated that diatoms growing on etched glass rods were present in higher densities in the etched crevices than any other part of the glass rod. Diatoms have also been shown to inhabit crevices on rock surfaces and sand grains in higher abundance than on the flat, more exposed regions of the substrata (Krejci and Lowe 1986, Round 1981). These studies suggest that different structural features of microscale habitats, such as those of bryophytes, may influence diatom community dynamics and growth patterns, especially in streams where available substrata can vary considerably. 1Department of Biological Sciences, Bowling Green State University, Bowling Green, OH 43403. 2University of Michigan Biological Station, Pellston, MI 49769. 3Current address - Large Pelagics Research Center, University of New Hampshire, Spaulding/ Rudman Halls, 46 College Road, Durham, NH 03824 - 2618. *Corresponding author - jessie.knapp@unh.edu. 306 Southeastern Naturalist Vol. 8, No. 2 Without a quiescent microhabitat, diatoms are susceptible to a variety of disturbances, such as grazing, desiccation, scour, or fl ooding (Bergey 1999, Biggs 1996, Krejci and Lowe 1986), and aquatic bryophytes have the potential to provide diatoms with the refugia required for proliferation despite disturbances (Resh et al. 1988). Although bryophytes are subjected to abrasion from mobile substrata, such as sand and rocks, they are seldom directly removed from the substratum by fast currents or fl ood events and have been reported to provide protection for loosely attached organisms such as diatoms (Power and Stewart 1987; Suren 1991, 1996; Suren and Winterbourn 1992; Suren et al. 2000). Water velocity is decreased within the bryophyte thallus, creating a protected habitat for epiphytic diatoms (Sand-Jensen and Mebus 1996, Suren 1991, Suren et al. 2000). To date, little work has been done directly examining how the variability in the structure of bryophytes, i.e., leafy liverworts versus whorled mosses, shapes the distribution and density of diatoms within and between the different morphologies. Mosses are characterized by a radially symmetric, leafy gametophyte (Crum and Anderson 1981, Schofield 1985). Leaf phyllotaxy is spiral, and the leaves are arranged around the stems in 3 or more rows with broad insertion. Leaves are 3 to many ranked and not lobed. Most leaves possess a midrib, and all leaves on one plant are alike. In contrast, the liverwort plant body growth form can be thalloid, or fl attened, or can consist of a stem with oppositely or alternately arranged leaves. The leaves vary in shape, never have a midrib, and are only 1 cell layer in thickness (Ammons 1940). These contrasting morphologies are likely to provide varying degrees of protection and refugia for diatoms and thus may contribute to the structure and community dynamics of the base of the food web in stream ecosystems. This study explored how diatom densities varied with differing bryophyte morphology in 3 streams in the Great Smoky Mountains National Park. The density of diatoms was compared on the abaxial and adaxial surfaces of both mosses and leafy liverworts. We predicted that mosses, with morphologies that create more crevices, would have a higher density of epiphytic diatoms than liverworts, with leaves highly exposed to the turbulence of the stream current. Methods Study site The Great Smoky Mountains National Park (GSNP) straddles the border between North Carolina and Tennessee and contains the largest old-growth forest in the eastern United States (Fig. 1). Algal samples were collected in the spring (19–20 May 2004) and fall (22–23 October 2004) from 3 streams: an unnamed tributary herein referred to as Stream A, the Little Pigeon River, and the Little River. Streams were chosen based on accessibility and presence of bryophytes. 2009 J.M. Knapp and R.L. Lowe 307 Table 1.Information about collections made at Stream A (SA), Little Pigeon River (LPR), and the Little River (LR) across all sampling dates. “Submerged bryophyte coverage” and “exposed bryophyte coverage” are the average estimated coverages of bryophytes across both sampling dates, below and above the water surface, respectively. “Samples collected” is the number of species of bryophyte samples collected at each river. % % bryophyte Samples Elev. Stream overstory coverage collected River Date GPS (m) order coverage Submerged Exposed May Oct SA 05/19/04, N 35°42.398, 1150 1 80 80 20 2 2 10/22/04 W 83°19.536 LPR 05/19/04, N 35°42.409, 1150 4 15 50 50 2 3 10/22/05 W 83°19.457 LR 05/20/04, N 35°36.966, 650 4 70 30 70 3 3 10/23/06 W 83°39.662 Stream A is a first-order tributary of the Little Pigeon River located on the Ramsay Cascades Trail. The Little Pigeon River is a fourth-order stream, and the sampling location was located 2 km upstream from Stream A on the Ramsay Cascades Trail. The Little River collection site was located in a fourth-order stream off of the Middle Prong Trail (Table 1). Sampling protocol At each collection site, GPS coordinates and elevation were recorded using a Garmin eTrex Legend GPS unit (Garmin International Inc., Olathe, KS). Temperature was recorded using a digital thermometer, and pH was measured using EMD colorphast pH-indicator strips (range = 2.5 to 10 pH units, sensitivity of 0.2 to 0.3 pH units; EMD Chemicals Inc., Figure 1. Location of the Great Smoky Mountains National Park in the southeastern United States , and the position of the Little River, the Little Pigeon River and its tributary Stream A. 308 Southeastern Naturalist Vol. 8, No. 2 Gibbstown, NJ). Bryophyte coverage and overstory coverage were visually estimated (Table 1). Stream current velocity was not measured, as this study focuses on the microhabitat between bryophyte leaves and not on the general stream velocity. To characterize the general water chemistry of each site, 500-mL water samples were collected 10–15 cm below the water’s surface (Csuros 1994). Samples were frozen within 24 hours of collection and were sent to the University of Michigan Biological Station for analysis. Water samples were filtered and tested for total fixed nitrogen (NO3 -N) and soluble reactive phosphorus (PO4 3-P) using the cadmium reduction method and the ascorbic acid reduction method, respectively (Eaton et al. 1995). At each stream, 2 types of samples were collected for each bryophyte species. An undisturbed/epiphytic sample (herein referred to as the epiphytic sample) was collected to examine the natural distribution of diatoms on the bryophyte. For this sample, 3 representative samples of each bryophyte species were collected randomly from the stream. To preserve the spatial arrangement of the diatoms on bryophyte leaves, the epiphytic samples were placed into bags and transported upright in a box to minimize agitation. An epipelic sample was also collected to determine which diatom taxa were actually associated with the sediment rather than with the bryophyte leaves. For the epipelic sample, a composite of each bryophyte species was collected by sampling an area of 6 cm2 from 3 to 5 randomly selected rocks covered with the target bryophyte species. The sample bag was then gently shaken to remove the sediment loosely attached to the bottom of the bryophyte patch sampled. The bryophyte was removed after the sediment settled and the remaining sediment was kept for analysis. Sample processing and analysis Bryophyte samples were preserved with 50% glutaraldehyde within 24 hours of collection. Bryophytes were identified to species (Crum and Anderson 1981) at the University of Tennessee. The epiphytic samples were assigned an arbitrary code to ensure blind analysis, and a small section was randomly selected and removed from the thallus for examination with a Hitachi S-2700 scanning electron microscope (SEM; Hitachi Ltd., Tokyo, Japan). Epiphytic samples were dehydrated with an ethanol series using a Samdri 780A critical point dryer (Tousimis Research Corp., Rockville, MD), mounted on aluminum stubs, and sputter coated with 10 nm of AuPd (Postek et al. 1980). General bryophyte morphology was digitally recorded at 35–70 times magnification (Fig. 2). Spatial arrangement and densities of diatoms were determined from 10 images each of abaxial and adaxial surfaces of leaves chosen randomly and taken at 1000x magnification. In order to determine the surface area of images at different tilts, an image of a 10-μm grid was taken at 1000x with a tilt 2009 J.M. Knapp and R.L. Lowe 309 of zero degrees. From this image, the surface area of each image was calculated using basic trigonometry. Leaves did not always lie parallel to the SEM stub, so there was some error (both over- and underestimations) associated with the calculated surface areas. Diatoms were subsampled from the epipelic samples by homogenizing and then removing 25 mL of the diatom/sediment slurry, which was boiled Figure 2. Scanning electron micrographs documenting the differences in plant morphology of each bryophyte species. Mosses are Platyhypnidium riparioides (a) and Fontinalis dalecarlic (b). Liverworts are Jubula pennsylvanic (c), Scapania undulata (d), Porella pinnata (e), and Marsupella emarginata (f). Note the midrib on the mosses and the broad fl at leaves on the liverworts. The adaxial surface of (a) is marked with an arrow and abaxial is marked with a double arrow. Bars are 500 μm. 310 Southeastern Naturalist Vol. 8, No. 2 with nitric acid to remove organic matter (Round et al. 1990). Permanent slides were made with Naphrax® for examination using light microscopy. Diatoms were identified to species (Krammer and Lange-Bertalot 1986, 1988, 1991a, 1991b) using differential interference contrast on an Olympus BX51 light microscope (LM; Olympus, Melville, NY) at 1000x magnification under oil immersion. Samples were counted until the relative frequencies of the dominant taxa did not change with counting additional fields of view (about 300–700 valves). Statistical analyses All statistical analyses were run using JMP (5.1, SAS Institute Inc., Cary, NC). Analysis of variance (ANOVA; α = 0.05) was used to examine the relationship between bryophyte surface and diatom density while considering whether the bryophyte was a leafy liverwort or a moss (herein referred to as bryophyte type). Individual plants were nested within bryophyte type, and bryophyte type and leaf surface were crossed factors. Upon determination of significance, t-tests were used to compare leaf surface at each bryophyte type (α = 0.05). Since there was only one comparison at the second level, there is no need for a Bonferroni correction. Results Stream conditions Elevation at the 3 sites ranged from 650 to 1150 m (Table 1). Submerged bryophyte coverage, defined as bryophytes on rocks below the water surface, was variable across the 3 sampling sites, ranging from 30 to 80% coverage of the streambed. Exposed bryophyte coverage, defined as bryophytes on rocks above the water surface, was also variable, ranging from 20 to 70% coverage. Temperatures were variable between sites and sampling dates, ranging from 11 to 15 °C with the lowest recordings occurring in May (Table 2). The pH levels in the streams were consistently acidic throughout the study. Stream nutrient levels were variable both between streams and between collection dates with Stream A having the lowest total fixed nitrogen (NO3 -N) and soluble reactive phosphorus (PO4 3-P; SRP) levels. In the May collections, NO3 -N ranged from 18.0 μg/L to 300 μg/L, and SRP ranged from 1.1 μg/L to 13.0 μg/L. In the Table 2. Temperature, pH, and water chemistry data from Stream A (SA), Little Pigeon River (LPR), and Little River (LR) collection sites. The pH measurement reported is the level that was recorded at both sampling events. Temp (°C) NO3 - (ppb) PO4 3- (ppb) River May October pH May October May October SA 13.0 14.0 5 18.0 3.7 13.0 11.3 LPR 11.1 12.0 5 300.0 131.3 1.1 0.8 LR 15.0 14.0 5 126.0 3.0 3.9 2.9 2009 J.M. Knapp and R.L. Lowe 311 October collections, the NO3 -N ranged from 3.0 μg/L to 131.3 μg/L, and SRP ranged from 0.8 μg/L to 11.3 μg/L (Table 2). Community composition and distribution Six species of bryophytes were identified: 2 mosses—Platyhypnidium riparioides (Hedwig) Dixon and Fontinalis dalecarlica B.S.G; and 4 leafy liverworts—Scapania undulata (L.) Dumortier, Jubula pennsylvanica (Stephani) Evans, Porella pinnata L., and Marsupella emarginata (Ehrhart) Dumortier. The bryophyte morphology documented with SEM (Fig. 2) was variable in a manner consistent with known leaf arrangement of both leafy liverworts and mosses (Schofield 1985). Using LM and SEM, 19 genera and 37 species of epiphytic diatoms were identified. Most species observed were prostrate forms (fl at against the substrate), such as Cocconeis placentula. There was little variation in the diatom species identified between bryophyte types and between leaf surfaces. Although many of the taxa observed growing directly on the bryophyte leaves were also observed in the epipelic samples, additional diatom species were found only associated with the bryophyte sediments (e.g., Frustulia rhomboides (Ehrenberg) De Toni). Overall, the epiphytic diatom density varied between 0 and 231 diatoms·mm-2 with Eunotia rhomboidea, present in the highest density (Table 3). Diatom densities were significantly different with respect to bryophyte type and leaf surface (ANOVA, df = 29, P < 0.001; Fig. 3, Table 4). Diatom density was significantly higher on the adaxial surface of the moss leaves than on both the abaxial leaf surface of mosses and either leaf surface of liverworts (Fig. 4, Table 4). Discussion As predicted, the mosses (whorled leaves with more crevices) had a significantly higher density of epiphytic diatoms than the leafy liverworts (fl at Table 3. Epiphytic diatom densities according to diatom species. Species reported here occurred on more than one bryophyte sample. Total diatom density for the abaxial and adaxial surfaces for each species was calculated by dividing the total number of diatoms for each surface by the total area captured for the equivalent surface (mm2). Total density Diatom species Adaxial Abaxial Achnanthidium appalachianum Camburn et. Lowe 39 29 Achnanthidium minutissimum (Kützing) Czarnecki 4 4 Cocconeis placentula Ehrenberg 15 1 Cymbella sp. 2 0 Decussata placenta (Ehrenb.) Lange-Bert. et Metzeltin 2 0 Diadesmis sp. 23 19 Eunotia praerupta Ehrenberg 1 1 Eunotia rhomboidea Hustedt 231 144 Meridion alansmithii Brant 13 6 Planothidium lanceolatum (Brébisson) Lange-Bertalot 47 16 312 Southeastern Naturalist Vol. 8, No. 2 exposed leaves). Within mosses, diatom densities were significantly higher on the more protected adaxial surface of the moss than the more exposed abaxial surfaces. Results from this study suggest that the significant difference in diatom density between mosses and liverworts is partially due to the protected environment created by moss morphology. The adaxial surface of moss leaves is positioned close to the stem forming a shelter in which epiphytic diatoms Table 4. ANOVA table for examining the relationship between diatom density and leaf surface while considering bryophyte type. Effects test table is below the ANOVA table showing significant differences between diatom density on leaf surfaces with respect to bryophyte type. An individual is each piece of moss that was examined using SEM. Bryophyte type is nested within individual, and leaf surface and bryophyte type are crossed factors. a. ANOVA Source df Sum of squares Mean square F Ratio Prob > F Model 29 0.00005012 0.0000017 7.6792 <0.0001 Error 270 0.00006076 2.25E–07 Total 299 0.00011088 b. Effects test Source df Sum of squares F Ratio Prob > F Bryophyte type 1 0.00000135 5.9863 0.0151 Adaxial vs. abaxial 1 0.00000283 12.5715 0.0005 Individual (bryophyte type) 13 0.00003921 13.4030 <0.0001 Adaxial vs. abaxial*bryophyte type 1 0.00000189 8.4100 0.0040 Adaxial vs. abaxial*individual (bryophyte type) 13 0.00000513 1.7519 0.0508 Figure 3. Diatom density (least squares means) on the adaxial and abaxial leaf surfaces of mosses (diamond) and leafy liverworts (square). Diatom density on the adaxial leaf surface on mosses is significantly higher than that on the abaxial leaf surface on mosses and either leaf surface on liverworts. 2009 J.M. Knapp and R.L. Lowe 313 are protected from disturbances (e.g., scour, desiccation, etc.). This shelter is not provided by the abaxial surface of moss leaves or by either leaf surface on leafy liverworts. Furthermore, epiphytic diatoms on bryophytes are also protected because bryophytes increase substrate stability by decreasing the drag of the rocks on which they are growing (Suren et al. 2000). This contrasting distribution of diatom densities between different leaf surfaces and different bryophyte types was consistent across streams, regardless of the variability in stream conditions. The elevated nitrogen levels during the May collection, relative to the October collection are consistent with nitrogen peaks observed as a result of the spring snowmelt (Campbell et al. 2000) and did not appear to affect diatom distribution or community structure. Diatom community composition similarities between leaf surfaces within and across streams may reflect the influences of grazing pressure Figure 4. Scanning electron micrographs depicting the high variation in diatom density between the adaxial leaf surface (top) and the abaxial leaf surface (bottom) of mosses. Both images captured from the moss Fontinalis dalecarlica. Bars are 20 μm. 314 Southeastern Naturalist Vol. 8, No. 2 in addition to the acidic nature of streams in the GSMNP. The most abundant taxon, E. rhomboidea, is acidophilic and often associated with mosses. Many of the other identified taxa are acidophilic and/or grazer resistant (Lowe 1974). Bryophytes harbor higher densities of invertebrates than any other stream substratum (Brusven et al. 1990, Suren 1991), which may explain the abundance of grazer-resistant algal taxa observed in this study. High densities of epiphytic diatoms on bryophytes may influence food-web dynamics, especially in the bryophyte-rich streams in the GSMNP, where nutrients are low and bryophytes shape much of the lotic landscape. Bryophyte-associated differences in diatom community structure, density, and spatial distribution may result in different degrees of food resources available to grazers. Conditions such as pollution or disturbance that result in a decline or shift in the types of bryophytes would directly affect the periphyton population density and thus food-web interactions. Further studies should include experimental disturbances in controlled environments to determine the effect of different disturbances on diatom spatial distribution and composition. Acknowledgments We thank Mike Grant and the University of Michigan Biological Station for water chemistry data, David Smith and the University of Tennessee at Knoxville for bryophyte identification, and the field assistants who helped with sample collection. This work was funded by a US NSF grant (0315979) to R.L. Lowe. Portions of this manuscript were completed while J.M. Knapp was supported by Bowling Green State University. Literature Cited Ammons, N., 1940. A manual of the liverworts of West Virginia. American Midland Naturalist 23:3–164. Bergey, E.A., 1999. Crevices as refugia for stream diatoms: Effect of crevice size on abraded substrates. Limnology and Oceanography 44:1522–1529. Biggs, B.J.F., 1996. Patterns in benthic algae of streams. Pp. 31–56, In R.J. Stevenson, M.L. Bothwell, and R.L. Lowe (Eds.). Algal Ecology: Freshwater Benthic Ecosystems. Academic Press, San Diego, CA. Brusven, M.A., W.R. Meehan, and R.C. Biggam, 1990. The role of aquatic moss on community composition and drift of fish-food organisms. Hydrobiologia 196:39–50. Campbell, J.L., J.W. Hornbeck, W.H. McDowell, D.C. Buso, J.B. Shanley, and G.E. Likens. 2000. Dissolved organic nitrogen budgets for upland, forested ecosystems in New England. Biogeochemistry 49:123–142. Crum, H.A., and L.E. Anderson, 1981. Mosses of Eastern North America. Columbia University Press, New York, NY. Csuros, M., 1994. Environmental Sampling and Analysis for Technicians. Lewis Publishers, Boca Raton, FL. 2009 J.M. Knapp and R.L. Lowe 315 Eaton, A.D., L.S. Clesceri, and A.E. Greenberg (Eds.). 1995. Standard Methods for the Examination of Water and Wastewater, 19th Edition. American Public Health Association, American Water Works Association, Water Environment Federation, Washington DC. Eminson, D., and B. Moss. 1980. The composition and ecology of periphyton communities in freshwaters. 1. The infl uence of host type and external environment on community composition. British Phycological Journal 15:429–446. Krammer, K., and H. Lange-Bertalot. 1986. Bacillariophyceae. Volume 1: Naviculaceae. In H. Ettl et al. (Eds.). Süßwasserfl ora von Mitteleuropa 2/1, Gustav Fischer Verlag, Stuttgart, Germany. Krammer, K., and H. Lange-Bertalot. 1988. Bacillariophyceae. Volume 2: Bacillariaceae, Epithemiaceae, Surirellaceae. In H. Ettl et al. (Eds.). Süßwasserfl ora von Mitteleuropa 2/2. Gustav Fischer Verlag, Stuttgart, Germany. Krammer, K., and H. Lange-Bertalot. 1991a. Bacillariophyceae. Volume 3: Centrales, Fragilariaceae, Eunotiaceae. In H. Ettl et al. (Eds.). Süßwasserfl ora von Mitteleuropa 2/3. Gustav Fischer Verlag, Stuttgart, Germany. Krammer, K., and H. Lange-Bertalot. 1991b. Bacillariophyceae. Volume 4: Achnanthaceae. In H. Ettl et al. (Eds.). Süßwasserfl ora von Mitteleuropa 2/4. Gustav Fischer Verlag, Stuttgart, Germany. Krejci, M.E., and R.L. Lowe. 1986. Importance of sand grain mineralogy and topography in determining micro-spatial distribution of epipsammic diatoms. Journal of the North American Benthological Society 5:211–220. Lowe, R.L., 1974. Environmental requirements and pollution tolerance of freshwater diatoms. US Environmental Protection Agency, Cincinnati, OH. Postek, M.T, K.S. Howard, A.H. Johnson, and K.L. McMichael. 1980. Scanning Electron Microscopy: A Student’s Handbook. Ladd Research Industries, Williston, VT. Power, M.E., and A.J. Stewart, 1987. Disturbance and recovery of an algal assemblage following fl ooding in an Oklahoma stream. American Midland Naturalist 117:333–345. Resh, V.H., A.V. Brown, A.P. Covich, M.E. Gurtz, H.W. Li, G.W. Minshall, S.R. Reice, A.L. Sheldon, J.B. Wallace, and R.C. Wissmar. 1988. The role of disturbance in stream ecology. Journal of the North American Benthological Society 7:433–455. Round, F.E., 1981. The Ecology of Algae. Cambridge University Press, New York, NY. Round, F.E., R.M. Crawford, and D.G. Mann, 1990. The Diatoms: Biology and Morphology of the Genera. Cambridge University Press, New York, NY. Sand-Jensen, K., and J.R. Mebus, 1996. Fine-scale patterns of water velocity within macrophyte patches in streams. Oikos 48:271–273. Schofield, W.B., 1985. Introduction to Bryology. Macmillan Publishing Company, New York, NY. Suren, A.M., 1991. Bryophytes as invertebrate habitat in two New Zealand alpine streams. Freshwater Biology 26:399–418. Suren, A.M., 1996. Bryophyte distribution patterns in relation to macro-, meso-, and micro-scale variables in South Island, New Zealand streams. New Zealand Journal of Marine and Freshwater Research 30:501–523. 316 Southeastern Naturalist Vol. 8, No. 2 Suren, A.M., and M.J. Winterbourn, 1992. The infl uence of periphyton, detritus and shelter on invertebrate colonization of aquatic bryophytes. Freshwater Biology 27:327–339. Suren, A.M., G.M. Smart, R.A. Smith, and S.L.R. Brown, 2000. Drag coefficients of stream bryophytes: Experimental determinations and ecological significance. Freshwater Biology 45:309–317.