nena masthead
NENA Home Staff & Editors For Readers For Authors

Rediscovery of the Freshwater Brown Alga Heribaudiella in Connecticut After 100 Years
John D. Wehr, Sarah E. Steirer, and Robin S. Sleith

Northeastern Naturalist, Volume 26, Issue 2 (2019): 343–361

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

 

Access Journal Content

Open access browsing of table of contents and abstract pages. Full text pdfs available for download for subscribers.



Current Issue: Vol. 30 (3)
NENA 30(3)

Check out NENA's latest Monograph:

Monograph 22
NENA monograph 22

All Regular Issues

Monographs

Special Issues

 

submit

 

subscribe

 

JSTOR logoClarivate logoWeb of science logoBioOne logo EbscoHOST logoProQuest logo

Northeastern Naturalist Vol. 26, No. 2 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 343 2019 NORTHEASTERN NATURALIST 26(2):343–361 Rediscovery of the Freshwater Brown Alga Heribaudiella in Connecticut After 100 Years John D. Wehr1,*, Sarah E. Steirer1, and Robin S. Sleith2 Abstract - Heribaudiella fluviatilis is a freshwater species in the predominantly marine class of brown algae (Phaeophyceae). The first reported North American population was collected in 1898 from Island Brook, CT. Here we confirm that the species was once present in Island Brook but has been extirpated from that location. Our 2016 survey rediscovered Heribaudiella in the New England flora, in 6 streams in western Connecticut ~70 km inland from marine water. Ecological data indicate these streams are deeper and have large-grained sediments, but lower specific conductance, dissolved NO3 -, and inorganic P as soluble reactive phosphorous (SRP) than nearby streams lacking this alga. We ran a multivariate, boosted regression tree (BRT) analysis, which confirmed that the niche of Heribaudiella in Connecticut is limited to minimally disturbed streams with greater pH, a high percentage of streambed boulders, and lower concentrations of dissolved NO 3 - and SRP. Introduction Brown algae, comprising the class Phaeophyceae, form a group of roughly 2000 species, nearly all of which occupy marine waters. This group of golden-brown, photosynthetic organisms vary in size and morphology from microscopic filaments to giant kelps many meters in length, with complex reproductive structures and life cycles (Graham et al. 2016). A very small number of brown algal species—perhaps 6 or 7 in total—occur in freshwater environments (Wehr 2015). Most of these taxa are known from very few localities worldwide, leading some to suggest their global rarity, as compared with most species in the freshwater algal flora (Wehr 2015). Researchers have observed that some populations of freshwater phaeophytes occur near coastal waters and have suggested that these representatives may have invaded freshwater habitats relatively recently (Israelsson 1938, Waern 1952, Wilce 1966). Thus far, molecular data (rbcL chloroplast gene) have indicated that most of the known freshwater species are members of separate clades within the class (Mc- Cauley and Wehr 2007), although their tolerance of saline water has not been tested (Wehr 2015). Of those few freshwater species, Heribaudiella fluviatilis (Aresch.) Sved. is the most widely reported, with roughly 30 records from North America, nearly all of which are in western states and provinces, with 1 putative population from Tennessee (Johansen et al. 2007, Wehr 2015, Wehr and Stein 1983). The alga forms conspicuous brown crusts on rocks and can become a major part of the benthic 1Louis Calder Center–Biological Field Station and Department of Biological Sciences, Fordham University, Armonk, NY 10504. 2Lewis B. and Dorothy Cullman Program for Molecular Systematics, The New York Botanical Garden, Bronx, NY 10458. *Corresponding author - wehr@fordham.edu. Manuscript Editor: Hunter Carrick Northeastern Naturalist 344 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 Vol. 26, No. 2 algal community in streams (Holmes and Whitton 1975, Wehr and Stein 1985). H. fluviatilis typically colonizes rocky, clear streams with high current velocity and relatively low concentrations of dissolved phosphorus (Holmes and Whitton 1975, Schneider and Lindstrøm 2011, Wehr 2015, Wehr and Perrone 2003). It has been reported from rivers and streams in subarctic to temperate locations in Europe, Japan, China, and Russia, but thus far has not been observed in subtropical or tropical locations, or in any streams in the Southern Hemisphere (W ehr 2016). The species was first described as Lithoderma fluviatile by Areschoug (1875; the basionym), but later transferred to a new genus, Heribaudiella, by Svedelius (1930), as Lithoderma is a marine genus with different reproductive structures. Notably, the very first collection of Heribaudiella from North America (as Lithoderma fluviatile Aresch.) was by Isaac Holden in 1898 from Island Brook in Connecticut, and was documented in the exsiccatae of algae Phycotheca Boreali-Americana (Collins et al. 1898). Those specimens are deposited in several major herbaria in the US. That single record stood for decades, until Smith (1950) in Freshwater Algae of the United States, cast doubt on the identity of Holden’s 1898 collection from Island Brook, due to its proximity to the high tide, and suggested it was likely a different marine species. Later reports from British Columbia and elsewhere (e.g., Pueschel and Stein 1983, Wehr and Stein 1985) resurrected the species as part of the North American flora, but the identity of the Connecticut population remained a mystery. Given the apparent rarity of Heribaudiella globally and especially in the eastern US, we endeavored to determine whether Holden’s collections were properly identified and if the population still existed in its original location, examine streams more widely for its presence within Connecticut, and if present, compare its morphology and ecological niche, based on chemical and geophysical properties, with that known for this species in other regions. Field-site Description We conducted the field portion of this study in mid- and late summer 2016 in western Connecticut, in river basins at varying distances from Island Brook in Bridgeport, CT, the first reported collection of H. fluviatilis All sites are located within the Western Uplands physiographic province, which has a complex bedrock geology. Streambeds consist of gneiss, quartzite, and schist, mixed with varying amounts of sandstone, dolomitic limestone, and shale. The northernmost sites drain watersheds along the Housatonic Mountains. Suitable sites were based on similarity to the physical properties of streams in Europe and the western US and Canada in which Heribaudiella has been reported (Holmes and Whitton 1975, Wehr 2015, Wehr and Perrone 2003, Wehr and Stein 1983). Key properties included a rocky substratum, at least moderate current velocity (>10 cm s-1), and low turbidity (less than 10 NTU). Based on these criteria, a reconnaissance survey determined that none of 5 sites along Island Brook were suitable habitat for the species (samples also proved negative), likely due to loss of solid substratum and highly turbid water. We identified sites near and more distant from the original location to create a list of 43 candidate streams in western Connecticut to sample. All had rocky substrata, averaged 5.4 m in width Northeastern Naturalist Vol. 26, No. 2 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 345 (min–max = 0.8–13.3 m), 23 cm depth (min–max = 7–42 cm), and 75% riparian canopy cover (min–max = 21–93%). Details are in Appendices 1–3. Methods We obtained herbarium specimens of Holden’s 1898 material (as Lithoderma fluviatile Aresch.) originally designated as item 536 in Fascicle IX of Phycotheca Boreali-Americana (Collins et al. (1898) with permission from the University of Michigan (MICH: 636207), Trinity College, CT (courtesy of C. Schneider), and New York Botanical Garden (NY: 02137512, 02137513). The material was originally dried onto mica slides and stored inside paper folders. We removed from each specimen a small (less than 1 mm) fragment that we placed in a sterile 1.5-mL microcentrifuge tube for transport to the laboratory. We sampled streams in western Connecticut between June 2016 and October 2016 following standardized methods used by our laboratory (e.g., Grubaugh and Wehr 2017, O’Brien and Wehr 2010). In each stream, we designated a 30-m reach that included at least 3 riffle–pool associations. At each site, we removed and inspected at least 20 rocks (where feasible) for the presence of any macroalgae (sensu Holmes and Whitton 1975), and specifically, any obviously dark brown crusts that might be later identified as Heribaudiella. When necessary, we used a field microscope (Swift FM-31LWD; Swift Instruments, San Jose, CA) for confirmation. We scraped algal material from at least 3 rocks into 15-mL or 50-mL centrifuge tubes and placed them in an ice chest until return to the laboratory. We retained 1 cobblesized rock (~10–20 cm) as a voucher specimen. At each site, we made a visual estimate of the percentage of sizes of streambed substrata based on the Wentworth scale (Cummins 1962). At locations where we collected rock samples, we measured light availability as percent canopy cover using a Model C spherical crown densiometer (Forestry Suppliers, Jackson, MS), current velocity using a Model 2000 Flo-Mate flow meter (Marsh–McBirney, Frederick, MD), stream width (m) with a tape measure, and maximum depth (cm) with a meter stick. We measured water temperature, pH, specific conductance, turbidity, and dissolved oxygen in situ using a YSI ProDSS water quality meter (YSI, Yellow Springs, OH). We collected a set of four 9-mL water chemistry samples for later chemical analysis after syringefiltration (0.2-μm pore-size) and preservation to pH less than 2.0 as pe r USEPA (1987). In the laboratory, we prepared algal material (dried herbarium sub-samples and freshly collected material) for microscopy as described previously (Wehr 2015). We rehydrated the dried material with deionized water on microscope slides for ~5 min prior to making observations. We compared the morphology and cell size of the historical specimens with values published in recent floras (Eloranta et al. 2011, Wehr 2015), and modern-day material. We examined samples using a Nikon Eclipse E600 interference microscope with Plan-Apo objectives, fitted with a DSFi2 (5 megapixel) digital camera (Nikon Instruments, Melville, NY). We analyzed water samples for dissolved organic C (DOC) using a Shimadzu TOC-L analyzer (Shimadzu, Columbia, MD), dissolved N and P with an Astoria A2 analyzer (Astoria, Clackamas, OR), and dissolved Ca and Mg using a Perkin-Elmer 1100B Northeastern Naturalist 346 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 Vol. 26, No. 2 atomic absorption spectrophotometer in flame mode, following the manufacturer’s guidelines(Waltham, MA). We followed EPA method 415.3 to measure DOC as non-purgeable organic C (Potter and Wimsatt 2005). We measured dissolved N as NH4 + using the phenol–hypochlorite method, and NO3 - by the sulfanilamide- NNED method after reduction of NO3 - to NO2 - in a Cd–Cu column (APHA 1985). We measured inorganic P as soluble-reactive phosphorus (SRP) using the antimony- ascorbate-molybdate method (APHA 1985). We assembled the physical and chemical data in a spreadsheet and imported these data into SYSTAT (v. 13; SYSTAT Software, Inc., Chicago, IL). Our primary aim was to compare ecological conditions among stream sites and test for differences in streams with and without Heribaudiella present. We tested all variables for normality and homogeneity of variances prior to statistical analyses. We analyzed those variables that conformed to these assumptions using a Student’s t-test; otherwise we employed non-parametric tests (Mann-Whitney U; Sokal and Rohlf 2012). We set the a priori Type-I significance level at P = 0.05. We used boosted regression tree (BRT) analysis, a multivariate approach, to describe the ecological niche of Heribaudiella in streams where it occurred. BRT analysis is a model-free approach that employs decision trees to estimate complex, non-linear effects of multiple predictors. We implemented a boosted regression tree analysis (Elith et al. 2008), using the gbm.step function of the “dismo” package (Hijmans et al. 2017). We employed an ad-hoc approach to tuning parameters to identify a parameter set that produced low cross-validation deviance. This approach led to a learning rate of 0.01, tree complexity of 10, and a bagging fraction of 0.9. We averaged results from 100 runs with 10-fold cross validation. Results Examination of Holden’s 1898 collections Microscopic examination of specimens from Island Brook revealed many densely packed, dichotomously branched filaments, although plastids were not easily distinguishable and pigments were faded (Fig. 1). Nonetheless, the general morphology closely resembled that of H. fluviatilis with many remnants of the multiply dichotomous-branched prostrate form (Fig. 1A, B). The thick-walled, tightly packed vertical filaments that characterize the species were common in all specimens examined (Fig. 1C). As noted by Holden on the original exsiccatae label, we observed a few terminal unilocular sporangia (Fig. 1D, E); no plurilocular sporangia were observed. Cells forming the vertical filaments were rectangular to quadrate in shape and varied from 8 μm to 15 μm in diameter, comparing closely with those we measured in live material (see details below). Distribution and morphology of Heribaudiella in Connecticut We surveyed streams in 4 western counties of Connecticut (Litchfield, Fairfield, Hartford, New Haven), and discovered populations of Heribaudiella in 6 of 43 candidate streams, all located within westernmost part of Litchfield County (Fig. 2). We observed no populations in streams within the Island Brook–Pequonnock River watershed, from which the original (now extirpated) population was collected in Northeastern Naturalist Vol. 26, No. 2 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 347 1898. All streams in which we found Heribaudiella occurred in upland locations more than 70 km from marine water. In the field, thalli occurred as circular, dark reddish-brown crusts (~5–50 mm diameter) on boulders and cobbles (Fig. 3A, B). Where it occurred, boulders typically had multiple colonies, but we observed none on smaller stones. On boulders larger than 100 cm, several colonies commonly expanded and coalesced to form more extensive patches. We found that not all dark-colored crusts sampled were Heribaudiella. Several smaller dark crusts were formed by the cyanobacterium Chamaesiphon geitleri Luther or Ch. cf. polonicus (Rost.) Hansg., or by the green alga Gongrosira fluminensis Fritsch. Although some crusts may be mistaken for Heribaudiella, cellular morphology and pigmentation of this alga was distinctive. Figure 1. Micrographs of Heribaudiella fluviatilis prepared from dried specimens collected in 1898 by Isaac Holden from Island Brook, CT. (A, B): prostrate series of cells forming a crust, (C): vertical series of tightly-packed upright filaments, (D, E): vertical filaments with terminal (empty) unilocular sporangia. Herbarium sources: A and C from MICH; B from Trinity College, CT; D and E from NY (all scale bars = 10 μm). Northeastern Naturalist 348 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 Vol. 26, No. 2 Microscopically, the morphology of Heribaudiella in Connecticut streams consisted of a basal, broadly spreading prostrate form (Fig. 3C), which gave rise to a vertical series of tightly packed upright filaments, some of which produced terminal unilocular sporangia (Fig. 3D, E). We observed both the prostrate and vertical forms in all collections. Sporangia released multiple zoospores (not shown). We observed no plurilocular sporangia in any of the populations we collected. Cells differed in size and shape between the 2 growth habits. The prostrate form consisted of multiply-branched (dichotomously), thick-walled filaments composed of quadrate, rectangular, or polygonal cells. Cell size varied widely: ~10–50 μm diameter x 15–40 μm long, with cell walls 1–3 μm in thickness. The quadrate or rectangular cells in the vertical series were more consistent in size (10–15 μm x 9–15 μm), but the filaments themselves varied from 2 or 3 cells to more than 20 cells in length and branched less frequently. Individually, cells contained numerous discoid goldenbrown chloroplasts, and physodes, which are refractive storage bodies (Fig. 3E). Ecological conditions Geomorphological properties of the streams with and without Heribaudiella differed in several respects. Those in which Heribaudiella occurred had a significantly Figure 2. Map showing locations of stream sites in western Connecticut sampled for Heribaudiella fluviatilis; filled circles = Heribaudiella present; open circles = Heribaudiella not observed; asterisk (*) = original location of collection by Holden, now extirpated. Details of site locations are given in Appendix 1. Northeastern Naturalist Vol. 26, No. 2 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 349 greater percentage of boulders in the streambed and greater depth, but significantly lower specific conductance, than those in which Heribaudiella was not found (Fig. 4). We detected no significant differences in streambed % cobble, % gravel, Figure 3. Images of Heribaudiella fluviatilis collected from contemporary populations in western Connecticut. (A): Brown macroscopic crusts on a large boulder in Gunn Creek (scale bar = 5 cm); (B): close-up view of crusts on a rock from Macedonia Brook (scale bar = 2 cm); (C): microscopic appearance of prostrate form with densely arranged, dichotomously- branched filaments; (D): series of vertically arranged, tightly packed filaments with terminal unilocular sporangia; (E): details of cells in vertical filaments with multiple golden-brown, disc-shaped chloroplasts (scale bars C–E = 10 μm). Northeastern Naturalist 350 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 Vol. 26, No. 2 stream width, canopy cover, current velocity, or water temperature on the dates we sampled. Nearly all streams were relatively small (mean width = 5.4 m ± 0.4 [SE]), and well-shaded in summer (mean max depth = 75% ± 2.8 [SE]). Water chemistry conditions differed among the streams in several key variables (Fig. 5). Streams with Heribaudiella had significantly greater average pH (+ 0.40 units), 20% lower average dissolved NO3 - (-14 mg/L), and 60% lower average SRP Figure 4. Physical and geomorphological characteristics of sampled streams with ( c r o s s - h a t c h e d bars) and without (open bars) Heribaudiella fluviatilis. (A): percentage of streambed substrata; (B): stream width, maximum depth, and % canopy cover; and (C): current velocity, temperature, and specific conductance (details in Appendix 1). Significant differences based on Mann–Whitney U test. Northeastern Naturalist Vol. 26, No. 2 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 351 (-17 mg/L) than in nearby streams lacking this species. We found no significant differences in concentrations of dissolved Ca, DOC, NH4 + (Fig. 4), or dissolved O2 (not shown). The complete set of data is given in Appendices 1–3. BRT results were well supported, with an average AUC of 0.99. BRT identified the variables pH, boulder percentage, dissolved NO3 -, and SRP to have the highest relative influence on the presence/absence of this species (Table 1). From these results, we characterized the niche of Heribaudiella in this region as streams with an elevated pH, a high percentage of boulders in the streambed, and low concentrations of dissolved NO3 - Figure 5. Chemical characteristics of sampled streams with (cross-hatched bars) and without (open bars) Heribaudiella fluviatilis. (A): streamwater pH, dissolved calcium, and dissolved organic carbon; (B): streamwater dissolved NH4 +, NO3 -, and SRP (soluble-reactive phosphorus) (details in Appendix 1). Significant differences based on Mann–Whitney U test. Northeastern Naturalist 352 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 Vol. 26, No. 2 and SRP (Fig. 6). However, while there was a significant difference in the depth of streams with and without Heribaudiella (Fig. 4), this variable was not a significant factor in the BRT model (Table 1). Figure 6. Partial dependence plots, based on boosted regression tree (BRT) analysis for the 4 variables with the highest relative influence (indicated as a percentage below each plot) of each variable on the response, after accounting for the average effects of all other variables in the model. Higher values of the fitted function indicate more suitable habitat for Heribaudiella fluviatilis. Table 1. The relative percentage contribution of different predictor variables identified by the boosted regression tree (BRT) model used to predict the main factors shaping the niche of Heribaudiella fluviatilis in Connecticut rivers (11 most-important variables are listed in order from greatest to least; values rounded to 2 decimal places). Variable Percent contribution pH 40.69 Percent boulder 27.44 Dissolved NO3 - 17.97 Dissolved SRP 10.00 Maximum depth 1.06 Percent riparian canopy cover 0.75 Specific conductance 0.66 Dissolved Ca 0.53 Current velocity 0.50 Dissolved NH4 + 0.32 Temperature 0.07 Northeastern Naturalist Vol. 26, No. 2 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 353 Discussion The rediscovery of H. fluviatilis in Connecticut streams is notable because the very first record of this alga on the North American continent was from a stream in Island Brook, in Bridgeport, CT. This specimen was also part of an historically important collection of algae—the Phycotheca Boreali-Americana (Collins et al. 1898)—now housed in several major herbaria (Sayre 1969, University and Jepson Herbaria 2009). The specimens of H. fluviatilis represented the first and only record of any freshwater member of the brown algae (Phaeophyceae) reported from North America for more than 70 y. Eventually, two other phaeophyte species, Pleurocladia lacustris A. Braun (Wilce 1962) and Sphacelaria lacustris Schloesser and Blum (1980) were discovered, but no additional populations of Heribaudiella were verified. That single record of H. fluviatilis in Connecticut stood for decades, until Smith (1950) in Freshwater Algae of the United States, cast doubt on the identity of Holden’s 1898 collection from Island Brook, due to its proximity to the high tide, and suggested it was a different species from the marine flora. But later, collections in western US and Canada in the 1980s, reignited interest in the species (Pueschel and Stein 1983, Wehr and Stein 1985). After recent re-examination of dried material from several exsiccatae, we determined (from co-occurring diatoms) that Holden’s collections were from a true freshwater habitat, and that the original identification was plausible (Wehr 2015). However, our survey, conducted 118 years later, found that suitable habitats and extant populations no longer exist in Island Brook. In the meantime, several populations of Heribaudiella have been uncovered in western US states and Canadian provinces (Wehr 2015, Wehr and Stein 1985). One putative eastern population was detected in a stream in Smoky Mountains National Park (Johansen et al. 2007), but until the present study, evidence to support its inclusion in the flora for New England remained unsettled. Our study now confirms that H. fluviatilis occurs in at least 6 streams in western Connecticut, and thalli from these locations are morphologically indistinguishable from populations in the western US and Europe. Our ecological data suggest that streams in this region in which Heribaudiella occurs are relatively pristine, rocky systems with lower concentrations of dissolved nutrients in the same region. Values measured in the present study are similar to those found for Heribaudiella streams in British Columbia (Wehr and Stein 1985). Average data in Connecticut compare favorably with the western streams with regard to pH (CT: 7.9 ± 0.1, BC: 7.9 ± 0.4), dissolved calcium (CT: 33 ± 4, BC: 33 ± 13), SRP (CT: 11.5 ± 1.4, BC: 9.5 ± 5.2), and specific conductance (CT: 203 ± 26, BC: 143 ± 94). Nitrate concentrations in Connecticut streams averaged somewhat higher (CT: 60.3 ± 1.9, BC: 39.4± 28.2). The Connecticut data are also in agreement with a large-scale bioassessment study in Norway that assigned H. fluviatilis a periphyton index of trophic status (PIT) score of 4.98, which classified it as indicative of relatively low total phosphorus concentration (Schneider and Lindstrøm 2011). In a prior study, a niche analysis based on a canonical correspondence analysis (CCA) of benthic macro algae in Austrian mountain streams similarly indicated that H. fluviatilis was typical of higher pH, combined with lower NH4 +, NO3 -, Northeastern Naturalist 354 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 Vol. 26, No. 2 and SRP concentrations (Rott and Wehr 2016). The BRT analysis largely agreed with previously published ecological descriptions, and together this indicates that H. fluviatilis in Connecticut has a niche similar to that described for populations found across the world. This information makes it possible to identify potential new sites to survey for this apparently rare species. Based on water chemistry variables, specific areas of the Northeast may be suitable for H. fluviatilis (e.g., Catskills, western New York, regions of Maine), where there is a mix of igneous and carbonate-rich geology. While widespread globally, the species remains rare. The local extirpation of H. fluviatilis from the Island Brook–Pequonnock River watershed, and its presence on some European Red Lists for conservation status suggest it may remain so (Temniskova et al. 2008, Wehr 2015). In addition, our present data indicating a regional preference for less-disturbed streams with lower nutrient levels suggest that the continued presence of Heribaudiella fluviatilis in New England will depend on limiting future levels of anthropogenic disturbance to streams in the region. Acknowledgments We thank Natalie Lynch and Yibing Zhou for assistance with field sampling and Kam Truhn for assistance with chemical analyses. We thank the NY Botanical Garden Herbarium (Kenneth G. Karol and Barbara M. Thiers), the University of Michigan Herbarium (Michael Wynn, Richard Rabeler), and Trinity College (Craig Schneider) for loans of Collins’ specimens from the Phycotheca Boreali-Americana. This work was supported in part from a Faculty Research Grant from Fordham University awarded to J.D . Wehr. Literature Cited Areschoug, J.E. 1875. Observationes phycologicae. Particula tertia. De algis nonnullis scandinavicis et de conjuctione Phaeozoosporarum Dictyosiphonis hippuroidis. Nova Acta Regiae Societatis Scientiarum Upsaliensis, Series 3 10(1): 1–36. American Public Health Association (APHA). 1985. Standard Methods for the Analysis of Water and Wastewater, 16th Edition. American Water Works Association, and Water Environment Federation, Washington, DC. 1268 pp. Collins, F.S., I. Holden, and W.A. Setchell. 1898. Phycotheca Boreali–Americana. Fascicle XI, Malden, MA. 536 pp. Cummins, K.A. 1962. An evaluation of some techniques for the collection and analysis of benthic samples with special emphasis on lotic waters. American Midland Naturalist 67:477–504. Elith, J., Leathwick, J.R., and T. Hastie. 2008. A working guide to boosted regression trees. Journal of Animal Ecology 77:802–813. Eloranta, P., J. Kwandrans, and E. Kusel–Fetzmann. 2011. Rhodophyta and Phaeophyceae. P. 155, In K. Krammer and H. Lange-Bertalot. Süßwasserflora von Mitteleuropa, 7. Spektrum Akademischer Verlag, Heidelberg, Germany. 311 pp. Graham, L.E., J.M. Graham, and M.E. Cook. 2016. Algae, 3rd Edition. LJLM Press, Madison, WI. 720 pp. Grubaugh, C.R., and J.D. Wehr. 2017. Suitability of total nitrogen as a predictor of protein content in stream periphyton. Freshwater Science 36:533–541. Northeastern Naturalist Vol. 26, No. 2 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 355 Hijmans, R.J., Phillips, S., Leathwick, J., and J. Elith. 2017. Package ‘dismo’. Species distribution modeling. R package version 0.8–11. Available online at https://cran.rproject. org/web/packages/dismo/dismo.pdf. Accessed 23 July 2018. Holmes, N.T.H., and B.A. Whitton. 1975. Notes on some macroscopic algae new or seldom recorded for Britain: Nostoc parmelioides, Heribaudiella fluviatilis, Cladophora aegagropila, Monostroma bullosum, Rhodoplax schinzii. Vasculum 60:47–55. Israelsson, G., 1938. Über die Süsswasserphaeophycéen Schwedens. Botanisk Notiser 1938:113–128. Johansen, J.R., R. Lowe, S.R. Gomez, J.P. Kociolek, and S.A. Makosky. 2007. New algal species records for the Great Smoky Mountains National Park, US A, with an annotated checklist of all reported algal species for the park. Southeastern Naturalist 6(Special Issue 1):99–134. McCauley, L.A., and J.D. Wehr. 2007. Taxonomic reappraisal of the freshwater brown algae Bodanella, Ectocarpus, Heribaudiella, and Pleurocladia (Phaeophyceae) on the basis of rbcL sequences and morphological characters. Phycologia 46:429–439. O’Brien, P.J., and J.D. Wehr. 2010. Periphyton biomass and ecological stoichiometry in streams within an urban to rural land-use gradient. Hydrobiolog ia 657:89–105. Potter, B.B., and J.C. Wimsatt. 2005. Determination of total organic carbon and specific UV absorbance at 254 nm in source water and drinking water. Method 415.3. US Environmental Protection Agency, Cincinnati OH. 56 pp. Pueschel C.M., and J.R. Stein. 1983. Ultrastructure of a freshwater brown alga from western Canada. Journal of Phycology 19:209–215. Rott, E., and J.D. Wehr. 2016. The spatio–temporal development of macroalgae in rivers. Pp. 159–195, In O. Necchi Jr. (Ed.). River Algae. Springer, New York, NY. 279 pp. Sayre, G. 1969. Cryptogamae exsiccatae. An annotated bibliography of published exsiccatae of algae, lichenes, Hepaticae and Musci. Introduction, I. General cryptogams, II. Algae, III. Lichenes. Memoirs of the New York Botanical Garden 19:1–174. Schloesser, R.E., and J.L. Blum. 1980. Sphacelaria lacustris sp. nov., a freshwater brown alga from Lake Michigan. Journal of Phycology 16:201–207. Schneider, S.C., and E.A. Lindstrøm. 2011. The periphyton index of trophic status PIT: A new eutrophication metric based on non–diatomaceous benthic algae in Nordic rivers. Hydrobiologia 665:143–155. Smith, G.M. 1950. Fresh-Water Algae of the United States. McGraw–Hill, New York, NY. 719 pp. Sokal, R.R., and F.J. Rohlf. 2012. Biometry, 4th Edition. W.H. Freeman, New Nork, NY. 937 pp. Svedelius, N. 1930. Über die sogenannten Süsswasser–Lithoderman. Zeitschrift für Botanik 23:891–918. Temniskova, D., Stoyneva, M.P., and I.K. Kirjakov,. 2008. Red List of the Bulgarian algae. I. Macroalgae. Phytologia Balcanica 14:193–206. US Environmental Protection Agency (USEPA). 1987. Handbook of Methods for Acid Deposition Studies: Laboratory Analysis for Surface Water Chemistry. EPA 600/4–87/026. US Environmental Protection Agency, Washington, DC. 168 pp. University and Jepson Herbaria, University of California, Berkeley. 2009. Phycotheca Boreali–Americana. Available online at: http://ucjeps.berkeley.edu/ina/pba/pba_main. html. Accessed 10 July 2018. Waern, M. 1952. Rocky-shore algae in the Öregund Archipelago. Acta Phytogeographica Suecica 30:1–298. Northeastern Naturalist 356 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 Vol. 26, No. 2 Wehr, J.D. 2015. Brown algae. Pp. 851–871, In J.D. Wehr, R.G. Sheath, and J.P. Kociolek (Eds.). Freshwater Algae of North America: Ecology and Classification. Elsevier, San Diego, CA. 1050 pp. Wehr, J.D. 2016. Brown algae (Phaeophyceae) in rivers. Pp. 129–151, In O. Necchi Jr. (Ed.) River Algae. Springer International Publishing, Basel, Switzerland. 2 79 pp. Wehr, J.D., and A.A. Perrone. 2003. A new record of Heribaudiella fluviatilis, a freshwater brown alga (Phaeophyceae), from Oregon. Western North American Naturalist 63:517–523. Wehr, J.D., and J.R Stein. 1985. Studies on the biogeography and ecology of the freshwater phaeophycean alga Heribaudiella fluviatilis. Journal of Phycology 21:81–93. Wilce, R.T., 1966. Pleurocladia lacustris in arctic America. Journal of Phycology 2:57–66 Northeastern Naturalist Vol. 26, No. 2 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 357 Appendix 1. Geomorphological properties of sampled streams in western Con necticut. H. f. = whether H. fluviatilis was present. % % % % % % Stream name H. f. Date Lat. (°N) Long. (°W) bedrock boulder cobble gravel sand silt Beacon Hill Brook No 14 July 2016 41.46882 73.04569 0 5 40 50 5 0 Bee Brook No 11 July 2016 41.68164 73.33296 0 0 40 10 0 50 Booth Hill Brook No 10 October 2016 41.24654 73.18161 0 5 80 10 0 5 Butternut Brook No 11 July 2016 41.75528 73.22434 0 80 10 3 3 4 Carse Brook Yes 30 June 2016 41.85562 73.37635 0 55 30 10 5 0 Clapboard Oak Brook No 23 June 2016 41.53053 73.38544 0 70 20 10 0 0 East Aspetuck River No 23 June 2016 41.59795 73.41533 0 5 90 5 0 0 East Br Naugatuck No 22 June 2016 41.83000 73.12003 1 69 10 5 15 0 East Br Silvermine Brook No 9 June 2016 41.94266 73.39066 0 1 4 85 8 2 East Spring Brook No 1 July 2016 41.61205 73.17574 0 10 60 20 5 5 Eight Mile Brook No 10 October 16 41.38903 73.16442 1 19 50 20 5 5 Finch Brook No 14 July 2016 41.56112 72.98357 0 30 50 10 0 10 Fullingmill Brook No 14 July 2016 41.50603 73.02987 0 20 70 5 5 0 Furnace Brook Yes 12 July 2016 41.81835 73.36811 5 60 20 10 5 0 Guinea Brook Yes 30 June 2016 41.82433 73.42987 0 45 50 5 0 5 Gunn Brook Yes 17 June 2016 41.80347 73.38287 5 80 5 5 5 0 Hancock Brook No 14 July 2016 41.65227 73.00767 0 40 0 40 10 10 Hop Brook No 14 July 2016 41.53447 73.10584 0 60 20 10 3 7 Horse Tavern Brook No 10 October 2016 41.21189 73.22380 0 0 70 20 10 0 Kent Falls Brook Yes 17 June 2016 41.76504 73.40873 0 50 40 5 5 0 Kirby Brook No 1 July 2016 41.62188 73.31631 0 10 80 5 0 5 Leadmine Brook No 22 June 2016 41.75358 73.06310 1 69 20 7 0 3 Little River No 9 June 2016 41.18326 73.21562 0 5 80 10 5 0 Long Meadow Brook No 14 July 2016 41.49757 73.09827 0 5 25 40 20 10 Lovers Lane Brook No 11 July 2016 41.81759 73.15599 5 70 15 5 0 5 Macedonia Brook Yes 17 June 2016 41.76024 73.49367 1 59 20 10 5 5 Merryall Brook No 12 July 2016 41.67541 73.44402 0 70 15 15 0 0 Northeastern Naturalist 358 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 Vol. 26, No. 2 % % % % % % Stream name H. f. Date Lat. (°N) Long. (°W) bedrock boulder cobble gravel sand silt Mill Brook No 12 July 2016 41.84949 73.47830 0 1 60 19 10 10 Miry Brook No 15 June 2016 41.36363 73.50088 0 1 79 10 0 10 Mopus Brook No 15 June 2016 41.33410 73.54162 0 0 70 20 10 0 Moulthrop Brook No 11 July 2016 41.74864 73.20752 0 40 40 15 0 5 Nepaug River No 22 June 2016 41.83875 73.02312 0 20 50 20 10 0 Norwalk River No 9 June 2016 41.29886 73.27209 0 20 70 0 5 5 Padanaram Brook No 15 June 2016 41.43031 73.47976 10 30 50 5 5 0 Saugatuck River No 9 June 2016 41.19277 73.26066 0 70 15 5 5 5 Second Hill Brook No 23 June 2016 41.55312 73.34426 0 80 10 5 5 0 Spruce Brook No 22 June 2016 41.76634 73.15157 50 40 2 3 5 0 Still River No 15 June 2016 41.38914 73.47976 5 25 60 5 5 0 Sucker Brook No 11 July 2016 41.70274 73.34662 0 30 40 15 0 15 Tollgate Brook No 23 June 2016 41.57426 73.49972 0 15 60 15 10 0 West Aspetuck River No 12 July 2016 41.67159 73.39678 0 20 40 20 0 20 Wachocastinook Brook No 30 June 2016 41.98058 73.42306 0 1 90 10 0 0 West Redding Brook No 9 June 2016 41.33278 73.44255 0 10 30 50 5 5 Northeastern Naturalist Vol. 26, No. 2 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 359 Appendix 2. Physical-ecological properties of sampled streams in western Connecticut. H. f. = whether H. fluviatilis was present. Stream Max Canopy Current width depth cover velocity Temp. Turbidity Stream name H. f. (m) (cm) (%) (m/s) (°C) (NTU) Beacon Hill Brook No 8 22 57 0.55 19.5 0.9 Bee Brook No 6 12 86 0.47 19.7 3.5 Booth Hill Brook No 3 16 64 0.46 12.1 1.8 Butternut Brook No 7 26 85 0.47 17.5 2.9 Carse Brook Yes 4 30 86 0.70 19.4 3.2 Clapboard Oak Brook No 3 23 78 0.64 16.9 0.3 East Aspetuck River No 11 21 68 1.43 17.7 1.0 East Branch Naugatuck No 9 21 78 1.17 18.2 0.9 East Branch Silvermine Brook No 1 15 87 0.30 12.8 0.5 East Spring Brook No 7 37 76 1.04 18.4 1.0 Eight Mile Brook No 8 32 55 0.40 11.2 0.9 Finch Brook No 5 11 84 0.37 20.2 2.5 Fullingmill Brook No 13 27 83 1.20 19.9 1.2 Furnace Brook Yes 6 42 85 1.27 18.6 1.4 Guinea Brook Yes 5 17 28 0.50 22.5 2.4 Gunn Brook Yes 3 41 92 0.84 14.9 1.8 Hancock Brook No 5 39 72 0.52 22.3 5.1 Hop Brook No 8 23 21 1.07 21.0 1.5 Horse Tavern Brook No 6 17 72 0.25 11.6 1.9 Kent Falls Brook Yes 5 28 82 0.60 16.3 0.8 Kirby Brook No 4 16 76 0.59 16.9 0.8 Leadmine Brook No 5 31 80 0.60 16.5 0.7 Little River No 5 22 76 0.70 15.6 1.2 Long Meadow Brook No 4 20 86 0.25 22.7 1.2 Lovers Lane Brook No 4 24 93 0.70 17.3 1.0 Macedonia Brook Yes 5 25 91 1.11 15.5 5.5 Merryall Brook No 3 8 77 0.40 15.0 1.1 Mill Brook No 5 19 85 0.57 19.0 0.9 Miry Brook No 4 10 28 0.16 17.1 5.1 Mopus Brook No 3 13 68 0.46 21.2 2.6 Moulthrop Brook No 6 7 92 0.33 20.7 3.9 Nepaug River No 5 36 73 0.70 18.0 1.4 Norwalk River No 7 36 87 0.81 17.6 1.6 Padanaram Brook No 5 30 89 0.36 17.2 0.3 Saugatuck River No 8 29 74 0.68 17.6 2.1 Second Hill Brook No 4 14 84 0.48 14.9 0.8 Spruce Brook No 3 20 91 0.67 17.3 1.0 Still River No 8 32 86 1.08 18.9 2.3 Sucker Brook No 6 28 80 0.76 18.9 0.8 Tollgate Brook No 5 10 71 0.59 18.8 0.6 West Aspetuck River No 4 23 80 0.64 20.6 5.3 Wachocastinook Brook No 4 20 58 0.60 18.1 0.9 West Redding Brook No 2 15 74 0.46 16.9 1.6 Northeastern Naturalist 360 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 Vol. 26, No. 2 Appendix 3. Chemical properties of sampled streams in western Connecticut . H. f. = whether H. fluviatilis was present. Conductance Ca Mg NH4 + NO3 - SRP DOC Stream name H. f. (μS/cm) pH (mg/L) (mg/L) (μg N/L) (μg N/L) (μg/L) (mg P/L) Beacon Hill Brook No 326.3 7.23 31.3 4.0 40.8 119.1 6.0 4.9 Bee Brook No 297.8 7.52 39.1 8.7 60.0 52.4 26.1 12.0 Booth Hill Brook No 273.0 6.91 15.6 4.1 13.6 64.0 7.4 15.9 Butternut Brook No 222.4 7.82 38.2 9.9 31.8 67.9 29.3 9.7 Carse Brook Yes 256.5 8.06 44.3 11.7 19.6 61.1 11.0 7.0 Clapboard Oak Brook No 379.6 7.65 61.0 16.2 11.3 89.2 2.1 3.8 East Aspetuck River No 384.0 8.21 52.2 14.7 27.0 89.7 21.9 5.0 East Branch Naugatuck No 237.5 7.37 25.7 5.1 28.9 69.8 6.4 5.0 East Branch Silvermine Brook No 447.1 7.43 68.2 19.2 91.4 87.7 37.4 7.5 East Spring Brook No 220.4 7.39 23.4 5.4 47.5 72.6 36.6 6.9 Eight Mile Brook No 215.9 7.29 18.7 4.7 8.5 61.1 6.5 12.3 Finch Brook No 397.6 7.47 43.2 4.9 10.2 137.3 29.7 19.7 Fullingmill Brook No 307.6 7.40 34.3 5.4 41.9 116.4 10.9 8.0 Furnace Brook Yes 288.6 8.16 45.4 11.4 55.3 55.4 16.7 23.0 Guinea Brook Yes 127.8 7.68 24.7 5.4 51.7 55.8 8.2 10.0 Gunn Brook Yes 232.2 7.85 33.7 10.4 52.9 66.1 15.0 8.0 Hancock Brook No 149.4 6.44 12.2 1.7 55.6 51.0 1.3 23.3 Hop Brook No 328.4 7.05 30.2 6.3 69.7 80.6 22.8 10.7 Horse Tavern Brook No 234.8 6.82 14.5 3.3 10.2 88.7 15.1 9.5 Kent Falls Brook Yes 152.1 7.68 22.7 5.6 29.5 65.4 8.9 3.9 Kirby Brook No 433.9 7.49 49.5 9.7 28.3 98.2 23.8 22.7 Leadmine Brook No 377.1 7.44 31.2 7.4 51.0 69.0 6.1 8.4 Little River No 200.5 7.16 22.0 4.8 52.1 62.6 8.3 5.5 Long Meadow Brook No 283.0 7.26 20.9 3.6 76.9 68.0 13.8 6.8 Lovers Lane Brook No 366.0 7.66 40.1 11.4 21.1 84.1 91.9 5.4 Macedonia Brook Yes 163.0 7.70 29.0 7.8 39.3 58.0 9.4 3.8 Merryall Brook No 309.4 7.95 54.3 15.6 15.6 95.9 32.0 7.4 Northeastern Naturalist Vol. 26, No. 2 J.D. Wehr, S.E. Steirer, and R.S. Sleith 2019 361 Conductance Ca Mg NH4 + NO3 - SRP DOC Stream name H. f. (μS/cm) pH (mg/L) (mg/L) (μg N/L) (μg N/L) (μg/L) (mg P/L) Mill Brook No 455.2 8.13 80.2 23.4 15.1 69.3 15.3 43.2 Miry Brook No 444.9 7.25 69.8 14.6 158.7 65.3 17.1 6.7 Mopus Brook No 357.9 7.70 66.6 14.8 135.5 77.2 35.3 17.1 Moulthrop Brook No 224.0 7.39 34.4 8.1 79.9 51.1 11.6 6.5 Nepaug River No 137.0 7.50 16.1 3.0 34.8 75.5 19.0 8.2 Norwalk River No 651.0 7.64 84.0 21.0 79.9 62.6 192.1 11.2 Padanaram Brook No 495.0 7.61 65.4 14.0 52.5 81.6 15.1 11.2 Saugatuck River No 297.1 7.45 44.8 10.6 37.6 51.7 20.6 8.3 Second Hill Brook No 223.7 7.34 27.3 6.6 110.7 81.2 29.3 5.5 Spruce Brook No 150.6 7.44 17.8 4.3 21.6 65.0 4.7 6.6 Still River No 585.0 7.68 64.8 14.8 126.2 63.4 18.9 9.8 Sucker Brook No 287.7 7.91 36.2 8.4 32.5 68.6 38.4 7.6 Tollgate Brook No 369.4 7.75 46.3 10.2 64.4 75.3 139.8 6.9 West Aspetuck River No 254.6 7.06 41.5 9.6 41.8 19.7 33.2 21.7 Wachocastinook Brook No 92.3 7.63 19.2 3.2 18.7 65.1 4.6 3.5 West Redding Brook No 180.0 7.35 31.1 5.7 127.7 54.6 17.4 16.1