The effect of fetch on periphyton spatial variation

Hydrobiologia 206: 1-10, 1990.© 1990 Kluwer Academic Publishers. Printed in Belgium. 1

The effect of fetch on periphyton spatial variation1

Antonella CattaneoDepartment of Biology, McGill University, 1205 Avenue Docteur Penfield, Montreal, Quebec, Canada,H3A lB1; Present address: Dipartement de sciences biologiques, UniversitJ de Montreal, C.P. 6128Succursale 'A', Montreal, Quebec, H3C 3J7, Canada

Received 8 March 1989; in revised form 10 September 1989; accepted 12 November 1989

Abstract

The marked spatial variation in periphyton could reflect differences in exposure, grazing or substratum.To determine if any of these factors was significant, I studied the relationship between the degree ofexposure to waves (measured as fetch), grazing intensity (measured as invertebrate biomass) and spatialvariation in periphyton biomass in sites of similar substratum along an island in the central mesotrophicbasin of Lake Memphremagog (Quebec). In spring, there was a positive relationship between fetch andperiphyton biomass both on stones and on artificial substrata. This effect was especially strong fordiatoms, for algae between 100-1000 m3, for planktonic forms and for filamentous and long-stalkedspecies. In spring, water renewal encouraged growth of these forms, but the effect disappeared in summer,coincident with silica depletion. Grazers, which were not important in the spatial variation observed inspring, probably contributed to the sudden decline of periphyton in the exposed sites in summer.Observations in nearby lakes indicate that these patterns may be general.

Introduction

Many authors have remarked on the patchiness ofperiphyton along the shore of a lake (Pieczynska,1965; Brown & Austin, 1973; Cox, 1988), andtemporal variation is also high (Cattaneo & Kalff,1978). Nutrients that are effective in explainingvariation in phytoplankton biomass (Dillon &Rigler, 1974) have proven less successful in treat-ing periphyton biomass in the same lakes(Cattaneo, 1987), probably because periphytonabundance is also affected by physical factors,like substratum type, slope, and wave exposure,and biological factors, like grazing. Grazing con-tributes to seasonal decline in periphytonbiomass, likely after nutrient limitation has

depressed growth rates (Cattaneo, 1983). Expo-sure to waves affects macrophytes (Keddy, 1982;Duarte & Kalff, 1986; Chambers, 1987), andinvertebrates (Barton & Carter, 1982), but itsimportance to periphyton has never been quanti-fied. Because wave action can detach part of theattached communities and decreased periphytonbiomass has been observed after storms (Young,1945; Fox etal., 1969; Kairesalo, 1983), wavesare thought to depress periphyton biomass. Lessheavy wave action could have a positive effect,the stimulation of periphyton growth throughincreased water renewal (Tanimazu et al., 1981;Reuter et al., 1986).

Wave action is a difficult variable to explorebecause it demands simultaneous measurements

Contribution of the Lake Memphremagog Project, Limnology Research Centre.

2

over the season at many sites. These limitationsimply that even if extraordinary efforts weremade, the results would be difficult for otherworkers to apply because the instrumentation isprohibitively expensive. Because fetch is propor-tional to wave height and frequency (U.S. BeachErosion Board, 1982), it can be used as a simplealternative index of wave regime at various sites.

To test the importance of exposure to waves onperiphyton spatial variation, I studied thebiomass, life forms, size structure, and taxonomiccomposition of communities growing in protectedand exposed sites along the shores of an island inthe central basin of Lake Memphremagog(Qu6bec-Vermont). These shores are relativelyuniform in slope and substratum composition, butdiffer in exposure to the winds. I hypothesizedthat, within seasons, algal development woulddepend on exposure, but that differences betweenseasons would reflect grazing intensity, measuredas the biomass of grazers. The generality of thepatterns in periphyton biomass with exposure andgrazing obtained in Lake Memphremagog werethen compared to data collected previously inother lakes (Cattaneo, 1987). The potential effectsof substratum on these patterns was examinedwith artificial substrata.

Methods

Molson Island is an uninhabited island located inthe central mesotrophic basin of lakeMemphremagog (76° 16' W; 45° 06' N). Ninesites, differently exposed to the winds (Fig. 1),were sampled five times during June-July 1984.At each sampling, 5-10 stones, at 1 m depth, wererandomly selected and sampled underwater witha brush-syringe sampler (Loeb, 1981). This devicescraped periphyton material from a constant areaof 5.7 cm2 and collected it in a small volume ofwater. This material was then filtered ontoGelman type A-E glass fiber filters and keptfrozen until analysis. Plant pigments were sub-sequently extracted with 96 % ethanol for 24 hand chlorophyll a concentration determined in aBaush & Lomb spectrophotometer (Bergman &

Fig. 1. Molson Island and the sampling sites in LakeMemphremagog (Qu6bec-Vermont).

Peters, 1980). In summer 1985, six of the siteswere sampled four times. This time, together withthe chlorophyll samples, collected and treated asin the previous year, five stones (surface50-250 cm2) were collected at each site. Allperiphyton was dislodged with a plastic brush,fixed with 2% formalin, and subsequently usedfor microscopical examination of algae and in-vertebrates.

To examine the effect of exposure on differentperiphytic taxa, life forms and size classes, algaewere identified and counted in random fields of anannoplankton chamber at 400 x and 160 x mag-nification. Large, rare algae were counted in theentire slide. Cell volumes were calculated byapproximation to solids of known volume.Volumes of entire colonies and filaments wererecorded rather than the volume of the single cells(Cattaneo, 1987).

All invertebrates whose diet consist at leastpartially of algae, according to Pennak (1978),were counted and measured at 25 x magnifica-tion, and biomasses were calculated using pub-lished length-weight relationships in Smock(1980) for aquatic insect larvae, except chiro-nomids, and McCauley (1984) for cladoceransand copepods, except Sida. For chironomids,Sida, and all other invertebrates, weights wereobtained by weighing dried samples (60 C) on aCahn electrobalance. For snails, shell-free weightwas estimated as 20% of total weight.

Although stones were rather uniform in size

3

and composition at all sites, the possibility ofsome substratum effect was tested by offeringartificial substrata which differed markedly fromthe natural rock. These simply consisted of stand-ard glass slides (75 x 25 mm) held horizontally orvertically by aluminum frames, anchored tobricks. On June 11 1985, 24 slides were deployedat depth of 1 m at each of six sites where naturalstones were sampled. On June 25, five replicatehorizontal and vertical slides were collected ateach site, frozen and subsequently extracted with96% ethanol, to measure chlorophyll a concen-tration. If substratum effects produced theobserved spatial variation, the periphytic develop-ment on these artificial substrata should havebeen similar at all sites.

To monitor nutrient concentrations, subsurfacesamples for total phosphorus were collected nearthe sampling sites in 1984 and 1985, and analyzedusing the ascorbic acid modification of themolybdenum blue technique (Strickland &Parsons, 1972) preceded by digestion under pres-sure with potassium persulfate (Menzel &Corwin, 1965). Integrated tube samples (1-2 mdepth) were collected weekly in 1985 about 400 moffshore from Molson Island in the central basinof the lake and used to determine dissolved silicaconcentrations. The samples were filteredthrough 0.45 m Millipore filters held in plasticfiltering system, and the filtrate analyzed for dis-solved silica concentration (Strickland &Parsons, 1972).

Wave height and frequency are proportional tofetch and wind (U.S. Beach Erosion Board,1972). For each site I measured the effective fetchfor the eight cardinal compass bearings on a topo-graphical map. A transparency with one centralline with seven lines radiating at 6 intervals oneach side was placed over each sampling site.With the central line directed toward each of theeight cardinal bearings in turn, the distance alongthe lines to the opposite shoreline was measured(xi). Each length measurement was multiplied bythe cosine of the angle between the radial and thewind direction (cos yi). Effective fetch for eachcompass bearing (F) was then calculated accord-ing to the formula (Hakanson, 1981):

F = Xi cosyicosyi

At each site, the values for effective fetch calcu-lated for the eight compass bearings wereaveraged to obtain the average fetch (Fa), whilethe largest of the effective fetch values measuredrepresented the maximum fetch (Fm). Thesemeasures of fetch were correlated (r2 = 0.95), butsubsequent regression analysis of biologicalresponse on fetch showed that maximum fetchwas most effective for data from Molson Island,while data from other lakes were better describedby average fetch. More complex exposure indicesthat also account for wind direction and speed(measured in a weather station 40 km west ofMolson Island), did not improve the correlationand therefore are not discussed.

The data were analyzed using least squareregression technique, a commercial statisticalpackage (Statistix), and an IBM-PC. Prior toanalysis, the data were logarithmically trans-formed to stabilize the variance.

Results and discussion

Periphyton development

Because Molson Island is composed almostentirely of slate, and because it lies close to theshore, it provides a range of exposures, measuredas fetch, but minimizes differences in substratumtype and in bottom slope (Table 1). Sites wereclassified as 'exposed' if their maximum fetch waslonger than 1 Km and as 'protected' otherwise. Inspring (June) of both years of study, periphytonbiomass in the exposed sites along the shores ofMolson Island was three to four times higher thanthat in the protected sites (Fig. 2). By mid summerperiphyton in the exposed sites had declined tovalues even lower than in the protected ones. Thispattern whereby periphyton biomass rises to amaximum in late June and then declines rapidlyto a summer low has been consistently observedin Lake Memphremagog and in other lakes ofsimilar trophic status (Cattaneo & Kalff, 1978;

4

Table 1. Characteristics of the sites sampled along Molson Island. Slope was measured by recording depths at 1 m intervalsalong a transect perpendicular to the shore. Substrate diameter is the median of the diameters of thirty randomly collected stones.Total phosphorus values (TP) are the range observed during the study.

Site Effective fetch (m) Slope Stone diameter TP

% (cm) (g 1 )Maximum Average

* 1 2590 774 8.7 4.5 10.9-11.4**2 3248 1165 14.1 5.5 8.6- 9.1*3 2427 914 12.5 8.0 10.9

** 4 1960 530 15.7 5.0 8.0-12.1* 5 1166 351 21.4 6.0 7.8*6 615 219 18.9 4.0 9.1-12.9

** 7 463 218 21.6 5.0 7.7-10.9** 8 809 326 19.6 6.0 7.1-10.3**9 1938 702 15.8 8.0 7.7- 9.5

* Sampled in 1984. ** Sampled in 1984 and 1985.

150 -

120 -

90-

60-

30-

150

120

90

60

30

Fig. 2. Seasonal changes of periphyton biomass (as chloro-phyll) in two consecutive years in protected (maximum effec-tive fetch < 1 Km) and exposed sites. Points represent themean, and vertical lines the standard errors (n = 24-30 in

1984; n = 15 in 1985).

Cattaneo, 1987). This consistency suggests thatthe development in June must be separated fromsubsequent summer patterns. The effectiveness ofsuch division is also suggested by marked dif-ferences between 'spring' and 'summer' in regres-sions (Fig. 3) between periphytic chlorophyll andmaximum fetch (Fm). In spring, there is a highlysignificant (P < 0.00001), positive relation:

(1)Chl-spring = 0.241 x Fm0 76

n = 35 r2 = 0.41 F= 23

The scatter is further reduced when springaverages are considered instead of point values:

Chl-spring Average = 0.214 x Fm0 7 8

n = 15 r2 = 0.67 F = 27(2)

There is no significant relation betweenperiphyton chlorophyll and exposure in summer.

The contribution of different groups of algae tothese seasonal and spatial patterns was examinedby dividing the algal community according totaxa, size and life form. The biomasses of greenand bluegreen algae did not change significantlywith season or exposure (t-test n = 12), but thebiomass of diatoms was higher in spring and in

1984

- Exposed

Protected

E

E

c-

0

0

_C

I

-I

I-----

5

* springo summer

00

a0

0 *

g 1 °00

00

3-

500

0

S

0~~

· 0

0 0

o 8

S0o o

0

1000 2d00 4000

Maximum Fetch (m)Fig. 3. Plot of periphyton biomass (as chlorophyll) versus maximum fetch in Molson Island. Solid circles represent spring (June)values and open are summer (July and August) values observed in 1984 and 1985. The line represents the regression between

periphyton chlorophyll and fetch calculated with spring values only (equation 1 in the text).

exposed sites (Fig. 4). The total diatom volume, inspring, was also significantly, positively corre-lated with fetch, but green and bluegreen volumeswere not (Table 2). A significant, but slight, nega-tive correlation of diatom volume with fetchappeared in the summer data (r = - 0.65). Thusthe differences in periphytic development reflectchanges in the response of diatoms.

Algal life forms in the periphyton were dividedinto three groups, according to the degree ofattachment and proximity to the substratum.1. 'Tightly attached algae' include forms that growappressed to the substratum (Cocconeis, Eunotia,Gloeotrichia), and those attached with a shortstalk (Achnanthes minutissima). 2. 'Filamentousand long stalked algae' are also attached to thesubstratum but extend outward to form a mat.The dominant genus in this group was Cymbella,

which occurred both as a large cell (> 5000 am3 )held on the tip of long mucilaginous stalks, and asa smaller one (300-500 #m 3) in tube. 3. 'Motile-planktonic algae' include motile forms (mainlyNavicula and Nitzschia), and those suspended(Synedra ulna), or perhaps trapped in the mat(Asterionella formosa, Synedra acus). Motile-planktonic and filamentous-long stalked algaehad a strong positive response to exposure in thespring, but not in the summer, while forms moreappressed to the substratum had no significantcorrelation with fetch at any time (Table 2).

When algae are divided into logarithmicallyincreasing size classes, spatial and seasonalchanges were observed only in algae > 100 #m3

(Fig. 5), with the class 100-1000 #um3 showing thestrongest correlation with fetch (Table 2). Smallalgae (<100 im 3), dominated by the diatom

300-

c'

E

0)E

-C

00L)U

100-

30-

10-

I I

v

i

6

Taxa

500

400

300

200

100 -

100-

Life Forms

Protected Exposed Protected Exposed

Fig. 4. Biomass of the various algal taxa (left panel) and life forms (right panel) in spring (June) and summer (July) 1985 atprotected (maximum fetch < 1 Km) and exposed sites along Molson Island. Each value is the mean of six samples. Standard

errors were 10-25% of the mean.

Achnanthes minutissima, were independent of sea-son and exposure (Fig. 5).

Table 2. Correlation coefficients observed in spring betweenthe maximum fetch (Fm) and the volumes of different taxa,life forms or size classes (n = 12). In summer all correlationswere insignificant, except a negative correlation betweendiatom volume and fetch (r = - 0.65*).

TaxaDiatomsGreensBlue-greens

Life FormsTightly attachedFilamentous-long stalkedMotile-planktonic

Size Classes< 100 /m3

100-1000 Am3

1000-10000 #m3

> 10 000 #m3

r0.80 **0.42

- 0.05

0.190.78 **0.81 **

0.530.86 **0.68 *0.74 **

*P<0.05; ** P < 0.01.

Water renewal

Wave action in lake littoral generates currentswhose speed is proportional to wave length, waveheight, and water depth (Komar, 1976). Calcu-lations based on these equations, and confirmedby current meter readings in a site in LakeMemphremagog with similar fetch (E. Petticrew;Dept. of Biology, McGill University; pers.comm.), show that in the exposed sites of MolsonIsland, at 1 m depth, currents above 10 cm s- could be generated by winds of only 15 km h- '.These currents are of the same magnitude of thosethat increase periphyton growth in artificial chan-nels (McIntire, 1966) and rivers (Horner &Welch, 1981). Whitford (1960) related the positiveeffect of current to faster nutrient renewal aroundthe cell. He calculated that a current of 15 cm s - 'was necessary to produce this effect, but currentsas low as 5 cm s- ' have been shown to stimulatephosphorus uptake by a thin algal film (Lock &John, 1979). In this study, the effect of fetch wasmore significant for medium to large diatoms,

E0

or

E

-o

0-9

Spring

W diatoms

I greens

e blue-greens

J , .

Summer

~////////////.// ///////////////

NNRRENll~Irrr

7

EIo00

5Ea

(n0E

-1 2 3 4>4 1 2 3 4>4

Log10 Size (urn3 )

Fig. 5. Biomass of logarithmic size classes of periphyticalgae in spring (June) and summer (July) 1985 at protected(maximum fetch < 1 Km) and exposed sites along MolsonIsland. Histograms represent the mean of six samples and

lines represent standard errors.

which form mats extending outward or areentangled in such mats (Fig. 4 and Table 2).Because these forms are only loosely attached tothe substratum, they should be the more vulnera-ble to detachment by waves (Hoagland, 1983),but they can also become seriously limited if freshnutrients and light (Tanimizu et al., 1981) are notprovided by water movement. The lack ofresponse to waves of small and tightly attachedalgae (Figs. 4 and 5) could either indicate thateven low currents are sufficient for such forms orit may suggest that they do not experience changesin current due to their size and proximity to thesubstratum (Silvester & Sleigh, 1985).

Faster water renewal associated with wavescould therefore be responsible for the spatial pat-tern observed in June. Because this pattern dis-appeared in July and August, even if wind fre-quencies and directions remained similar to thoseobserved in June (Environmental CanadaWeather report, Sherbrooke Airport), additionalfactors must be postulated. This differential effectof similar current regime may be explained byseasonal changes in nutrient concentration.Although total phosphorus remained constant(Table 1), silica sharply declined at the end of

June (Fig. 7) and remained low (<0.5 mgSiO2 1 - ') for the rest of the study. Such silicaconcentration are often critical for diatom growth(Reynolds, 1984; Lund, 1964). This interpretationis strengthened because diatoms responded mostto exposure and declined most sharply at the endof June, whereas other groups were little affected(Fig. 4).

Grazers

The most important grazers on the shores ofMolson Island were oligochaetes, cladocerans,ostracods, chironomids, mayflies, and snails.Such grazers can reduce attached algae biomass(Lamberti & Resh, 1983; Cattaneo, 1983) and aresometimes more abundant in protected locations(Barton & Carter, 1982). Thus grazing in pro-tected sites could account for the positive effect ofexposure on algal biomass. To examine thishypothesis, I assumed that community grazingrates would be proportional to the biomass ofgrazers associated with periphyton at the varioussites along Molson Island. Total biomass of thiscommunity was highly variable, but tended to behigher in exposed sites in June and July (Fig. 6).Even considering the uncertainty inherent in suchan estimation of grazing pressure, it appearsunlikely that the reduced algal biomass observed

IE0)E 400-

0)co

E0. 200-

a)N

(9

June July Aug

Fig. 6. Seasonal changes of grazer biomass in 1985 atprotected (maximum fetch < 1 Km) and exposed sites alongMolson Island. Points are the mean and vertical lines the

standard errors (n = 15).

[ J

I I~~~~~~~~~~~~~~~~~~~~~~~

I

8

2

E

v9

1.5 -

1-

0.5-

June July August

Fig. 7. Seasonal changes of dissolved silica concentrationoffshore Molson Island in 1985.

in spring in protected sites can be attributed tograzing. Grazers could however be important inthe sharp seasonal decline in the exposed sites.This decline mirrors the timing and patternobserved in previous years for epiphytes onaquatic plants in Lake Memphremagog. Thatdecline has been linked to invertebrate grazing(Cattaneo, 1983).

Artificial substrata

Stones were similar in size and chemical com-position at all stations, however, there wasnoticeably more silt on stones in the protectedsites. Silt could interfere with algal growth bydistabilizing the substratum and by shading thealgae. To determine if some unrecognized dif-ferences in the natural substrata may be responsi-ble for the patterns observed in exposed andprotected sites, I measured the periphyton growthon horizontal and vertical glass slides. The dif-ferent orientation of these substrata was intendedto induce different sedimentation and lightregimes (Liaw & MacCrimmon, 1978). Thebiomass on these slides was much lower than onnearby stones, reflecting the short colonizationperiod and the regularity of the smooth surface.On both horizontal and vertical artificial sub-strata, chlorophyll was higher in exposed than inprotected sites, repeating the pattern observed onstones (Table 3). In exposed sites, the ratio

Table3. Periphyton chlorophyll (mgm 2) on horizontaland vertical substrata immersed for two weeks at varioussites along Molson Island. The values represent the mean +standard error (n = 5). Sites are ordered from lower to higherfetch.

Site Horizontal Vertical Horizontal

Vertical

7 10.1 + 0.6 3.3 + 0.5 3.16 9.7 + 0.6 4.0 + 0.7 2.48 7.5+ 0.3 5.2 + 0.2 1.49 21.8 + 1.2 14.9 + 0.9 1.54 23.7 + 3.5 22.5 + 0.9 1.12 17.8 + 2.1 19.9 + 0.8 0.9

between the biomasses on horizontal and verticalslides was close to unity, while in protected sites,horizontal slides supported about three timesmore biomass than vertical ones. Thereforeincreased supply of organic matter and algal sedi-mentation appear to outweigh any negative effectsof siltation. In lakes, horizontal substrata oftensupport more biomass than do vertical ones(Newcombe, 1950; Castenholz, 1961), while thistrend is not observed in rivers, possibly becausesedimentation is reduced in flowing waters (Liaw& MacCrimmon, 1978).

Other lakes

The generality of the patterns observed in MolsonIsland was tested using data from a previousstudy of 13 other lakes of the Eastern Townshipsof Quebec (Cattaneo, 1987). These lakes, locatedwithin a 40 km radius of Lake Memphremagog,differ in area (1-100 km2) and nutrient concen-tration (TP 4-72 /g 1- ; TN 200-800 ,g - ').Chlorophyll average concentrations from May-June samplings fitted the spring model based onMolson Island data (Fig. 8): the regressionbetween observed and predicted values was sig-nificant (r2 = 0.58; P < 0.001), the slope (1.002)was not significantly different from 1, and theintercept (- 0.078) did not differ from 0. LakeBrome, an outlier in relation between phosphorusand periphyton chlorophyll (Cattaneo, 1987), did

I

best predictor, but data from other lakes werebetter desqcrihed hv a model using average fetch

(Fa). Despite the large range of trophy of theselakes, neither the introduction of total P nor totalNin a multivariate model improved the predictionof periphyton spring chlorophyll and diatom vol-ume based on fetch alone. In these lakes, the

_ Lr A1- _ -.. L__

1oo00

0S

0~~~~~~~~~~~0~~~~~~

S~~~~~~~~~~

idoo

Average Fetch (m)

Fig. 8. Plots of the values of chlorophyll (top panel) and ofdiatom volume (bottom panel) versus average effective fetch(Fa) observed in other lakes of the Eastern Townships(Qu6bec). Lines represent the models calculated fromMolson Island data. The model for chlorophyll isChl = 0.29 x Fa"8 6 ; the model for diatom volume is

Diat = 0.24 x Fa'-'4 . The outlier, Lake Brome, is repre-sented with an open circle in the top panel.

not fit this relationship either. When observeddiatom biomass in these lakes and in thesemonths is compared to prediction from theequation from Molson Island, Lake Brome wasno longer an outlier (Fig. 8). The relation betweendiatom volume and fetch in these lakes was betterdescribed by a curvilinear model, although therewas a significant correlation between diatom vol-umes observed here and those predicted from theMolson Island model (r2 = 0.56; P< 0.001;slope = 1.023; intercept = -0.1762). For theMolson Island data, maximum fetch (Fm) was the

posLUve eect 01 IeLcn again alsappeareu mn mIIsummer, when biomass declined after a Junemaximum in the eight oligotrophic and meso-trophic sites. The remaining five eutrophic sitessupported a different community dominated byfilamentous greens (Cattaneo, 1987).

Conclusions

In different years (Fig. 2) and in different lakes(Fig. 8) spring periphyton development is corre-lated with fetch. Fetch is a rather crude estimateof wave regime at the various sites and this isreflected in the low coefficient of determinationobserved in the regressions (Figs. 3 and 8), eventhough they were highly significant. Although themodels in this study were not improved by theaddition of wind data collected some distanceoverland, this could be a route to better models,but it would scarcely be practical to install remotedevices at each site. The potential effect of otherphysical factors, like slope and substratum type,on periphyton spatial variation could not be testedin this study, where their ranges were too limited(Table 1). Grazing did not account for the spatialvariations, but accelerated the seasonal changesbrought about by silica depletion. Although thisstudy does not provide a tool for precise predic-tion of periphyton, it clearly shows that fetch canhave a strong effect on attached algal biomass insmall to medium-sized lakes, and can account formuch of their spatial variation at the time of peakdevelopment.

Acknowledgements

This study was funded by grants of the DonnerCanadian Fondation, of the Qu6bec Ministry of

160-

80-

40-

EE

JX-a-20

U

oOL)

9

20-0

250 500

800-

nE

uo

E

400-

200-

100-

50-

250 56o0

4

Education, and of the National Sciences andEngineering Research Council of Canada to theLimnology Research Centre. D. Brumelis assistedin the field and with the chemical analysis. J. Kalffand R. Peters generously provided advice andencouragement. E. Petticrew kindly supplied herunpublished data on near shore currents.

References

Barton, D. R. & J. C. H. Carter, 1982. Shallow-water epilithicinvertebrate communities of eastern Georgian Bay,Ontario, in relation to exposure to wave action. Can. J.Zool. 60: 984-993.

Beach Erosion Board, 1972. Waves in inland reservoirs. U.S.Army Corps Eng., Beach Erosion Bd Tech. Memo 132.

Bergman, M. & R. H. Peters, 1980. A simple reflectancemethod for the measurement of particulate pigment in lakewater, and its application to phosphorus-chlorophyll-seston relationships. Can. J. Fish. aquat. Sci. 37: 111-114.

Brown, S. D. & A. P. Austin, 1973. Spatial and temporalvariation in periphyton and physico-chemical conditions inthe littoral of a lake. Arch. Hydrobiol. 71: 181-232.

Castenholz, R. W., 1961. An evaluation of a submerged glassmethod of estimating production of attached algae. Verh.int. Ver. Limnol. 14: 155-159.

Cattaneo, A., 1983. Grazing on epiphytes. Limnol Oceanogr.28: 124-132.

Cattaneo, A., 1987. Periphyton in lakes of different trophy.Can. J. Fish. aquat. Sci. 44: 296-303.

Cattaneo, A. & J. Kalff, 1978. Seasonal changes in theepiphyte community of natural and artificial macrophytesin Lake Memphremagog (Que. & Vt.). Hydrobiologia 60:135-144.

Chambers, P. A., 1987. Nearshore occurrence of submersedaquatic macrophytes in relation to wave action. Can. J.Fish. aquat. Sci. 44: 1666-1669.

Cox, E. J., 1988. Has the role of substratum been under-estimated for algal distribution patterns in freshwater eco-systems. Biofouling 1: 49-63.

Dillon, P. J. & F. H. Rigler, 1974. The phosphorus-chloro-phyll relationship in lakes. Limnol. Oceanogr. 19: 767-773.

Duarte, C. M. & J. Kalff, 1986. Littoral slope as a predictorof the maximum biomass of submerged macrophyte com-munities. Limnol. Oceanogr. 31: 1072-1080.

Fox, J. L., T. O. Olaug & T. A. Olson, 1969. The ecology ofperiphyton in Western Lake Superior. 1. Taxonomy anddistribution. Bull. Water Resour. Res. Cent. Univ. Minn.14: 1-97.

Hakanson, L., 1981. A manual of lake morphometry.Springer-Verlag. New York, NY, 78 pp.

Hoagland, K. H., 1983. Shot-term standing crop and diver-sity of periphytic diatoms in a eutrophic reservoir. J.Phycol. 19: 30-38.

Horner, R. R. & E. B. Welch, 1981. Stream periphyton devel-opment in relation to current velocity and nutrients. Can.J. Fish. aquat. Sci. 38: 449-457.

Kairesalo, T., 1983. Dynamics of epiphytic communities onEquisetumfluviatile L.: Response to short-term variation inenvironmental conditions. In R. G. Wetzel (ed.), Periphy-

ton of freshwater ecosystems. Hydrobiologia/Dev. Hydro-biol. 17: 153-160.

Keddy, P. A., 1982. Quantifying within-lake gradients ofwave energy: Interrelationships of wave energy, substrateparticle size and shoreline plants in Axe Lake, Ontario.Aquat. Bot. 14: 41-55.

Komar, P. D., 1976. Beach processes and sedimentation.Prentice Hall, Englewood Cliffs, N.J. 429 pp.

Lamberti, G. A/ & V. H. Resh, 1983. Stream periphyton andinsect herbivores: an experimental study of grazing by acaddisfly population. Ecology 64: 1124-1135.

Liaw, W. K. & H. R. MacCrimmon, 1978. Assessing changesin biomass of riverbed periphyton. Int. Revue ges. Hydro-biol. 63: 155-171.

Lock, M. A. & P. H. John, 1979. The effect of flow patternon uptake of phosphorus by river periphyton. Limnol.Oceanogr. 24: 376-383.

Loeb, S. L., 1981. An in situ method for measuring theprimary productivity and standing crop of the epilithicperiphyton community in lentic systems. Limnol.Oceanogr. 26: 394-399.

Lund, J. W., 1964. Primary production and periodicity ofphytoplankton. Verh. int. Ver. Limnol. 15: 37-56.

McCauley, E., 1984. The estimation of the abundance andbiomass of zooplankton in samples. In J. A. Downing andF. H. Rigler (eds.), A manual on methods for the assess-ment of secondary productivity in freshwater. BlackwellScientific Publications, Oxford: 228-265.

McIntire, C. D., 1966. Some effect of current velocity onperiphyton communities in laboratory streams. Hydro-biologia 27: 559-570.

Menzel, D. W. & H. Corwin, 1965. The measurement of totalphosphorus based on the liberation of organically boundfractions by persulfate oxidation. Limnol. Oceanogr. 10:280-282.

Newcombe, C. L., 1950. A quantitative study of attachmentmaterials in Sodon Lake, Michigan. Ecology 31: 204-215.

Pennak, R. W., 1978. Fresh-water invertebrates of theUnited States. J. Wiley & Sons. New York. 803 pp.

Pieczynska, E., 1965. Variation in the primary productivityof plankton and periphyton in the littoral zone of lakes.Bull. Acad. Pol. Sci. 13: 219-225.

Reuter, J. E., S. L. Loeb & C. R. Goldman, 1986. The physio-logical ecology of nuisance algae in an oligotrophic lake. InL. V. Evans & K. D. Hoagland (eds.). Algal Biofouling.Elsevier Press, Amsterdam: 115-127.

Reynolds, C. S., 1984. The ecology of freshwater phyto-plankton. Cambridge University Press, Cambridge.384 pp.

Silvester, N. R. & M. A. Sleigh, 1985. The forces on micro-organisms at surfaces in flowing water. Freshwat. Biol. 15:433-448.

Smock, L. A., 1980. Relationships between body size andbiomass of aquatic insects. Freshwat. Biol. 10: 375-383.

Strickland, J. D. H. & T. R. Parsons, 1972. A practicalhandbook of seawater analysis. 2nd. ed. Bull. Fish. Res. BdCan. 167: 310 pp.

Tanimizu, K., T. Miura & M. Higashi, 1981. Effect of watermovement on the photosynthetic rate of an algal com-munity attached to reed stems. Verh. int. Ver. Limnol. 21:584-589.

Whitford, L. A., 1960. The current effect and growth of fresh-water algae. Trans. am. Microsc. Soc. 79: 302-309.

Young, O. W., 1945. A limnological investigation of periphy-ton in Douglas Lake, Michigan. Trans. am. Microsc. Soc.64: 1-20.