NETHERLANDSJOURNALOFAQUATICECOLOGY29(3-4)359-368(1995)

GEOGRAPHICALVARIATION IN THE DISTRIBUTIONS OF MACROALGAEIN ESTUARIES

MARTIN WILKINSON, TREVORTELFER1 and SARAH GRUNDY

KEYWORDS:estuaries; seaweeds; macroalgae; distributions; biogeography; multivariate analysis.

ABSTRACT Assemblages of macroalgal species present in the upper reaches of estuaries were compared using published species lists for the North Atlantic, covering a north-south gradient in Europe from Iceland to northern Spain, and in North America between the St. Lawrence and Virginia. Patterns were sought using cluster analysis and ordination. Three groupings of estuaries were noted: Icelandic, mainland European and American. There was a slight trend of increase in the ratio of red to brown algae with decreasing latitude in Europe, though not as marked as for open coastal algae. This might be associated with the effect of temperature on salinity tolerance. Despite this biogeographic trend, some of the commonest species appear to be widely spread through estuaries, possibly including southern Atlantic and Pacific locations. Confirmation of the trends suggested requires further data from southern European estuaries.

INTRODUCTION WILKINSON(1980) has summarised the general pattern of distribution of intertidal algae into estuaries as follows: (i) colonisation is almost wholly by marine species; (ii) reduction in species number occurs going upstream as a result of the selective attenuation, firstly of red algae and then of brown algae, with green algae present throughout; (iii) a very few brackish-water species occur in the mid and upper reaches. He pointed out that this general pattern occurred in many estuaries over a wide latitude range in different continents. An alternative way of viewing this distribution pattern on temperate European shores is shown in Fig. 1. Three zones are present. Zone A is the relatively sheltered open coast at the mouth of the estuary, which being marine has a relatively large number of species and a high degree of coverage by fucoid algae. Zone B is the lower estuary and has an impoverished version ofthe sheltered open coast flora with domination still by fucoids. Zone C is the inner half of the estuary where species number is severely 359

reduced, and the flora is characterised by small, mat-forming, opportunist species, mainly green, blue-green and xanthophycean, although occasional fucoids may be present, especially the brackishwater adapted Fucus ceranoides. On the basis of the species present in this inner zone C of 15 estuaries, WILKINSONet al. (1976) divided British estuaries into three categories: type A dominated by green algae and Vaucheria spp; type B dominated by dense mats of the filamentous diatom Melosira nummuloides in addition to those algae in type A; and type C dominated by gelatinous bluegreens e.g. Rivularia in addition to those algae in type A. But after examination of a further 30 estuaries, WILKINSON(1980) and WILKINSONand RENDALL (1985) concluded that most UK estuaries had a similar flora in the upper reaches, with apparent differences between estuaries being due to which species was dominant out of a common pool of species. The low number of upper estuarine species was attributed to the harshness of the physical environment resulting in only a few widely tolerant species being present (28 species in a common pool

360

WILKINSON, TELFER and GRUNDY

so tolerant and universal that there is little difference in species composition at different latitudes in the North Atlantic compared with well-documented differences in the open coast flora.

fucoids

I J opportunists species number

C Fig. 1. Diagram to show variation in algal type and species richness over three zones A - C along the length of an estuary (for further explanation see text).

according to WILKINSONand RENDALL,1985). This is a contrast to the open coast where substantial differences occur in the macroalgal community depending on a wide range of biotic and abiotic factors. On 35 open coastal shores examined in the British Isles, WILKINSONand RENDALL(1985) reported an average of 90 seaweed species present per shore out of the total of 619 species recorded in the British Isles (SOUTHand TITTLEY,1986). Out of that same 619 species only about 28 are found regularly in the upper reaches of estuaries. This contrast in species number is especially well represented in a survey of one single area, the estuary and Firth of Forth in Scotland (WILKINSON et al., 1987). Of 230 species found in the Forth system, only 6 to 13 were found at each site in the upper estuary. The dominant zone-forming upper estuarine algae listed for the Forth include species of very wide distribution e.g.

Rhizoclonium tortuosum, Blidingia minima, Monostroma oxyspermum, Enteromorpha intestinalis, Phormidium spp. and Vaucheriaspp. So wide is the geographical distribution that these same taxa are even recorded from brackish-water situations in the southern hemisphere, such as the Patos Lagoon estuary, Brazil (COUTHINOand SEELIGER,1984). Therefore the question posed in this paper is whether or not the widely tolerant, species-poor, algal vegetation of the upper reaches of estuaries is

METHODS For this paper estuarine macroalgae have been defined as those benthic or attached red (Rhodophyta), green (Chlorophyta), brown (Phaeophyta) and xanthophycean seaweeds and freshwater algae found in estuaries, together with those benthic bluegreens (Cyanobacteria) that are present as macroscopically visible mats, and those few filamentous or chain-forming diatoms (Bacillariophyceae) that can form masses indistinguishable with the naked eye from filamentous brown algae e.g. Melosira spp. Using published lists taken from the scientific literature, a table was compiled of the distribution between estuaries of algae at the upper stations in the estuaries. The range of estuaries used (Table 1) extended from Iceland to Spain in Europe, from the St Lawrence to Virginia in the eastern USA, together with a single west coast USA site from San Francisco Bay. Most records of estuarine algae in the literature are from north America and Europe of which these represent a selection covering two biogeographic areas (warm temperate and cold temperate; LUNING,1990). The range of estuaries used is smaller than is desirable and there are more northern European sites than southern European ones. This is an unfortunate consequence of the nature of the published species lists. It was necessary to reject lists which had neglected particular algal groups, and also those which had amalgamated lists for many sites in an estuary so that the marine records and the truly estuarine ones were indistinguishable. To avoid a bias towards British estuaries we ignored our own unpublished records spanning many years. We only used those earlier records of our own which had been published. This was also done to avoid, as far as possible, using long-term data sets that would be biassed in comparison with single surveys in the literature. Selection of sites in the upper reaches of estuaries was a problem. It would be convenient to be able to select sites which were upstream of a particular critical salinity such as the 10%o CI( = 18 %o S) downstream limit of the mesohaline zone in the Venice system for classification of brackish waters (DENRARTOG,1960) but there are several problems with this. Salinity data are not given in many papers, and where they are given they

Geographic distribution of estuarine algae

361

Table 1. List of the estuarinesystemsfor which speciespresenceat upperestuarinesiteswas takenfrom the referencesgiven.The letters

showthe coding usedto identifyestuariesin Figs. 2 and 3, and in Tables2 and 3. Estuary Identification

Nameof Estuary

Latitude oN

Reference

England." A B C

Tees Tyne Wear

55 55 55

D E

Tweed Thames

56 51

Edwards,1972; HardyetaL, 1993 Edwards,1972; HardyetaL, 1993 Edwards,1972; Hardyet aL, 1993; Wilkinson, 1973b Norton, 1976 littley, 1985;Tittleyand Price,1977

Scotlan~ F G H I J K

Nodh Esk Add Deeand Tarff Fleet Urr Clyde

57 56 55 55 55 56

Wilkinson, 1979 Wilkinson, and Roberts1974 Wilkinson, 1975 Wilkinson, 1975 Wilkinson, 1975 Wilkinson, 1973a

Holland. L

Rhine, Meuseand Scheldt

52

Nienhuis, 1980

France. M

Orne

49

Lepailleur,1971

Norway: N

Hardangerfjord

60

Jordeand Klavestad,1963

Spain: 0 P

Ria de Arosa Ria de Lires

42 43

Donze,1968 Perez-Cireraand Pacheco,1985

Iceland. O R S T

Botn, Oyrafjordur Hvammur,Dyrafjordur Stadara,Steingrimsfjordur Reydarfjordur

66 66 66 65

Munda, 1972 Munda, 1972 Munda, 1972 Munda, 1972

USAand Canada: U V W X Y Z

GloucesterPoint, York River,Virginia GreatBay system.New Hampshire St. Lawrence Merrimack Hampton-Seabrook San FranciscoBay

38 43 47 42 42 38

Wulff and Webb.1969 Mathiesonet aL, 1981 Cardinaland Villalard,1971 Mathiesonand Fralick,1973 Mathiesonand Fralick,1972 Josselynand West, 1985

are not necessarily in the same style, e.g. long and short term averages and ranges, and it is not clear which style is biologically relevant. Den Hartog has criticised the Venice system on grounds that it should be based on organism distributions rather than on physical measurements. There can be an element of circular argument in that approach as far as this paper is concerned. When comparing algal distributions at different latitudes, species may penetrate estuaries to different extents because of the change of salinity tolerance with temperature

(WILKINSON, 1980). There is no easy resolution to this difficulty with the way data are presented in the literature. Therefore the stations chosen as upper estuarine are to some extent subjective. Where possible, stations were chosen to be upstream of an average salinity of about 18%o, or to be from reaches described by the authors as. mesohaline, oligohaline, or tidal freshwater. Otherwise species were taken which occurred at stations upstream of the outer limit of two species which are particularly characteristic brackish-water forms, at

362

WILKINSON,

and G R U N D Y

TELFER

T a b l e 2. The o c c u r r e n c e of algal species in the u p p e r reaches of the estuaries. A - Z (see Table 1 ). T h e presence of a '1' in the table indicates the presence of the species in the estuanJ indicated by the c o l u m n heading.

ESTUARY:

A

B

C

D

I

I

I

I

E

F

G

H

I

I

I

I

I

I

J

K

L

M

I

I

I

I

I

I

I

I

I

N

0

P

O

R

S

T

U

V

I

I

. . . . . . . .

I

I

W

X

I

I

Y

Z

CHLOROPHYTA

Blidingia marginata Blidingia minima 8ryopsis spp. Capsosiphon fulvescens Chaetomorpha capillaris Chaetomorpha linum Chaetomorpha melagonium Cladophora sp. Cladophora albida Cladophora fracta Cladophora glomerata Cladophora rupestris Cladophora sericea Enteromorpha ahlneriana Enteromorpha clathrata Enteromorpha compressa Enteromorpha erecta Enteromorpha flexuosa Enteromorpha groenlandica Enteromorpha intestinalis Enteromorpha linza Enteromorpha prolifera Enteromorpha torta Monostroma grevillei Monostroma oxyspermum Ochlochaete ferox Percusaria percursa Prasiola stipitata Rhizoclonium tortuosum Rosenvingiella polyrhiza Ulothrix sp. Ulothrix flacca Ulothrix pseudoflacca Ulothrix subflaccida Ulva curvata Ulva lactuca Ulvaria obscura Urospora peniciliformis

I I

I .

.

.

.

. . I

.

.

.

. . . . . . . . . . . . . . . .

I

.

.

.

.

.

.

.

. . . . . . . . . . . . . . . . . . . . . .

I

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

I

.

.

I

.

.

.

.

.

.

.

.

.

.

.

.

. .

.

.

.

.

.

.

.

.

.

I .

.

I

. I

.

.

I

I

.

. I

I .

.

.

I

.

.

.

.

I .

.

.

.

.

.

.

.

.

I

.

.

.

.

.

1

I

.

.

.

.

.

I .

.

. .

I

.

I

I

I

I

. . . . . . .

I

I

I

. . . . . . .

I

I

I

I

I

I

I

I

I

I

I

.

I I

I

I

I

I .

.

I

I

I

.

.

I

. .

.

.

I

.

.

.

.

.

.

I

I

I

I .

.

.

.

.

.

.

. .

. . . . . .

I

I

.

I

I

.

.

.

.

. . . .

I I

I

.

.

.

.

.

.

. .

.

.

I

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

I

I

I

I

I

.

.

I

.

.

1 . .

I

I

. I . . .

I

.

.

I

I I

.

I I

I

.

I .

I

.

I I

I

.

.

I

I I

I

I

.

I

I

I .

I I

I I

.

.

I I

I I

.

.

I

I

.

.

I I

I

I

.

I

I

I

.

I

I I

I

.

I

. I

. . . . .

. .

I

.

I

I . . . . .

.

. .

I

.

. . . . .

I

I .

I

. . . . . . . . . .

.

I

I

I

.

I

.

. .

I

I

I

. . . .

. . . . . . . . . .

I

I

I

. . . . . . . . .

.

.

I .

I

.

I

I

.

I

. . . . . . . . I . . . . . .

.

.

.

I I

I

I

.

I

.

.

I

1

I

. .

. . . . . . . I I I

I

.

. . . . . .

.

I

. . . . . . . . . . . . .

.

.

I

I

.

I

I

. I

.

I I

I

.

I

I

.

.

I .

.

I I

I

I

I

. .

I I

I

.

.

I

I

I .

.

I

I

I

. .

I I

I

I

I

.

I

. . . . . . . I

.

. I

I

. . . . . I . . . . . I . . . . . . . .

I

I

I

I

. I

. . I

I

.

. .

.

.

I

I

I

. . . . . . .

I . . . . . . . . . . . .

.

1

. . . . . . .

.

.

.

I

.

.

I .

I

.

.

.

I

I

I .

.

. I

I

I

I .

.

. I

. . . . . I I I .

. I

. . . . . .

I

. . . . . . . . . . . .

.

I

I

.

I . . . . . . .

. . . . . . .

.

.

I

. . . . . .

I

. .

I

I

I . . . . . . . . . . . . . .

I

. . . . . . . . . . .

I

I

.

I

. I

. . . . . . . . . . . . . . . . .

I

.

.

. . . . . . . . . . . .

I

I

I

I

. I

I

1

I

I

I I

.

.

.

I

.

.

.

.

.

.

.

.

.

. . . . . . . . . . . . .

I

.

.

.

.

.

.

.

.

.

.

I

. . . . . . . . . . . . .

I

.

.

.

.

.

.

.

.

.

.

I

. . . . . . . . . . . . .

I

. . . . . . . .

. . . . . . . . . . . . . . . . . . . . I

I

. . . . . . . . I I

I

I

I

PHAEOPHYTA

Ascophyllum nodosum Chorda ilium Chorda tomentosa Chordaria flagelliformis Oesmarestia aculeata Oictyosiphon foeniculaceus Ectocarpus sp. Ectocarpus silliculosus Elachista fucicola Oictyosiphon chordaria Fucus ceranoides Fucus distichus distichus Fucus distichus edentatus Fucus distichus evanescens

I

. . . . .

I

I

I

. . . . . . . . . .

9I .

.

. . .

. . . . . . . . . . .

I

. . . . . . . . . . . . I I

I I I

I . . . . . . I

I

I

. . . . . . . .

.

.

.

I

I

. .

. .

. .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.

.

.

I

I

I

I

.

.

I .

I .

.

.

.

I

I

I I

I I

I

.

I .

I I

I

.

.

.

I

.

.

.

I

.

.

. I

I

I

.

.

.

I

I

.

.

.

I

. . . . .

. I

I

. I

I

. I

.

I

I

I

.

I

I

I

. . . . . . . . . . . .

I

I

Geographic

Table

distribution

of estuarine

algae

363

2. ( c o n t i n u e d

ESTUARY

A

Fucus serratus Fucus spiralis Fucus vesiculosus Pelvetia canaliculata Petalonia fascia Pilayella littoralis Ralfsia verrucosa Scytosiphon Iomentaria Sphacelaria britannica Sphacelaria radicans

. . . . . . I I . . . . . . I I .

B

C

.

D

E

F

G

H

I

J

. . . I . . . . . . I I I . I I . . . I I . . . . . . . . I I I I I .

.

.

.

.

.

.

.

.

.

.

K

L

M

N

. . . I I I I I I I . . . . I . . . . . . .

.

I I I

0

P

Q

R

. . . . I . . . I I I I I . . . . . . I I

.

I

.

S

T

U

V

. . . . . . . . . . . I I I I I . . . . . . . . . . . I

.

.

.

.

I .

.

.

I I

I I

I

I

.

.

.

.

.

.

.

.

.

.

.

.

I

I

.

.

.

.

.

.

.

.

.

.

.

.

.

I

.

.

.

.

.

.

.

.

.

.

.

.

y

I

I I

.

Z

I I I

.

.

X

.

. . . . . . . . . . . . . . . . . . . . . .

W

. .

.

.

.

.

I

RHODOPHYTA

Ahnfeltia plicata Audouinella floridula Audouinella purpurea Audouinella virgatula Bangia atropurpurea Bostrychia scorpioides Callithamnion corymbosum Ceramium rubrum Ceramium strictum Chondrus crispus Dumontia incrassata Hildenbrandia rubra Mastocarpus stellatus Phyllophora brodiaei Phyllophora pseudoceranoides Polyides rotundus Polysiphonia denudata Polysiphonia lanosa Polysiphonia nigrescens Polysiphonia urceolata Porphyra laciniata Porphyra leucosticta Porphyra linearis Porphyra umbillicalis

. . . . . . . . . . . . . I . I . . . . . . . I . . I . . . . . . .

.

.

.

.

I

.

.

.

.

.

.

.

.

.

.

.

.

.

. . . . . . . . . .

.

.

.

.

.

.

I I

I

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

I

.

.

.

.

.

.

.

I

I

.

.

.

.

.

.

I

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

. . . . . .

.

I

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

. .

.

. .

I

.

.

.

I

. .

.

.

.

. .

.

. .

.

.

.

.

.

.

.

I

.

.

.

.

.

.

I

.

.

I I

I

I

.

.

.

.

.

.

.

.

.

.

I

.

.

1

1

I

I

I

I

.

.

.

.

.

.

.

I

.

.

.

.

.

I

.

.

.

.

.

.

.

I

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

I

.

I

. .

.

.

I

.

.

.

.

.

.

.

.

. .

I

. .

.

.

.

. .

I

.

I

. . . . I

.

.

.

.

.

.

. .

. .

I

I . I

.

. .

I

I . I

.

. .

.

I

.

. . . . . . . . . . . .

.

.

I

I .

.

.

.

I

.

.

.

I

.

.

.

I

.

.

.

I

.

.

.

I I

.

.

.

. . . .

I

.

.

.

I

. . . . .

.

I

.

. . . . . . . . . . . . . . . . I . . . . . . I .

I

.

.

. .

.

I

I .

.

I

.

.

.

.

.

.

.

I

I

. . . . . . .

. .

. .

I

.

.

. . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . I I . . . . . . . . . . . . . . . . . . I

I

.

.

I

I I I

.

.

.

I

.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

I

I I

I

. .

.

I

I

I

I

I

XANTHOPHYCEAE

Vaucheriaspp.

11

1 1 1 1 1

1 1 1

I

. . . . . . .

I

BACILLARIOPHYCEAE

Melosira spp.

I

I

I

I

9

,

I

.

I

I

.

.

.

.

.

.

.

.

.

.

.

.

.

.

CYANOBACTERIA

Calothrix spp. Entophysalis s p p . Hydrocoleum s p p . Lyngbya spp. Microcoleus spp. Oscillatoria spp. Phormidium spp. Rivularia spp. Schizothrix s p p .

.

.

.

.

.

.

.

.

.

.

. . . . . . .

.

.

.

.

.

.

. . . . . I

I

.

.

. . . . . . I

,

I

.

.

.

.

.

I

.

I

I

I

. . . . . I

I

. . . .

I

. . . . . .

I

I

. . . . .

I

. . . . .

I

. . . . .

I

. . . . .

I

.

.

.

.

.

.

.

.

.

.

I

I

I

.

.

.

.

.

.

.

.

.

.

I

I

I

. . . . .

I

.

I 1

I

. . . . . . . I

I

.

I

I

. . . . .

I

. . . . . . . . . .

. I

.

.

.

.

.

I .

.

.

.

. . . . . . . I

.

I

. . . . .

.

.

.

.

.

.

.

.

.

.

. . . . . .

I

.

. . . . .

I

.

.

.

.

. . . . .

I

364

WILKINSON, TELFER and GRUNDY

0.13

0+34 Jic~ld'!

0,06

0,70

Coefficient

Fig. 2. Dendrogram of the cluster analysis of estuaries data. Major group divisions are numbered; the light shading indicates the major divisions used and the darker shading the subdivisions within. These groups are used in the ordination plot in Fig. 3a. AX

least in the UK, namely Fucus ceranoides and Monostroma oxyspermum. Species presence data were subjected to cluster analysis using Jaccard's Coefficient of Similarity to seek geographically meaningful groupings of estuaries based on their species assemblages. The data were also ordinated by detrended correspondence analysis (DECORANA) to show major trends in the data which may be related to community parameters from geographic localities. The robustness of the classification from the cluster analysis can be confirmed by superimposing the groupings on the ordination plot. By overlaying the species ordination plot on the plot for the ordination of estuaries it was possible to see if any groups of estuaries could be attributed to the presence of any particular species types. Spearman's rank correlation coefficient was used to seek correlations between ordination scores for estuaries and the totals of species, both overall and in particular taxonomic groupings. RESULTS

I)

45

90

135

180

225

270

.9 9 9 1 4 9 1 4 9A A 4 ~

"

1

|149

9

%9

AXIS1

B

9

~9

9~ l e 9

*

i

0 9 .0"

I i

0 ~ i9 leo

-120 -140

200

i b 9 9

AXIS 1

)

Fig. 3. Plot of the two most significant trends of ordination by DECORANA of: (A.) estuaries (the divisions and subdivisions within shown by the cluster analysis are indicated; the light shading indicates the major divisions and the darker shading the subdivisions), (e.) species ( circles, Chlorophyta ; triangles, Phaeophyta ; squares, Rhodophyta ; circle/spot, Xanthophyceae; square/spot, 8acilliarophyceae ; slars, Cyanobacteria ). The eigen values for Axis1 = 0.414 and for Axis 2 = 0.248.

The species records extracted for upper stations in each estuary are given in Table 2. Where possible the names were revised in accordance with the taxonomic nomenclature of SOUTHand TIFILEY (1986) to overcome the problem of different taxonomic usage in different locations. Nonetheless there were isolated records of species which could not be identified in that checklist which might represent idiosyncracies of individual authors. To overcome this problem, records of species occurring in only one estuary were deleted. Because of the taxonomic confusion or difficulty, blue-greens and Vaucheria were only given to genus level. To resolve possible bias arising from large numbers of species within single genera, some of which may be conspecific, being recorded differently by different authors, the cluster analysis and ordination were also performed with all records amalgamated into genera only. Because some authors also appeared to neglect blue-greens, the genus only analyses were also carried out both with and without the blue-green data being included. No substantial difference occurred between analyses based on species records and those based on genera (both with and without blue-greens). Therefore only the results are presented for analysis with species including blue-greens to genus level. Three major groupings, labelled 1, 2 and 3, are shown by the cluster analysis in Fig. 2. Group 1 is made up of the 4 Icelandic sites. Three of these,

Geographic distribution of estuarine algae

365

Table 3. The total and numbers of taxa for the major systematic divisions of algaefor each estuary along with their Rhodophyta -

Phaeophyta(R:P) ratio.The lettersfor eachestuarycorrespondto thosein Table1. Estuary A B C D E F G H

Chlorophyta 10 11 15 16 8 8 10 8

Phaeophyta Rhodophyta Xanthophyceae Bacillariophyceae 4 1 1 1 5 1 1 0 3 1 0 1 2 2 0 0 1 1 1 1 1 0 1 0 7 5 1 1 7 1 1 0

Cyanobacteria TOTAL 2 19 1 19 1 21 0 20 2 14 2 12 4 28 0 17

I

6

1

0

1

0

0

8

J K L M N 0 P O R S T U V W X Y Z

4 11 15 10 14 17 8 6 5 5 9 9 17 16 5 18 9

4 6 4 1 14 8 5 6 0 5 7 1 8 11 2 15 1

0 4 2 0 14 8 2 0 1 0 1 6 14 7 1 11 1

0 1 1 1 0 0 1 0 0 0 0 0 0 0 1 0 0

0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

1 3 6 6 0 7 0 0 0 0 0 5 0 0 0 0 0

9 26 28 18 42 40 16 12 6 10 17 21 39 34 9 44 11

labelled la, are closely related and the fourth one is more closely related to Icelandic estuaries than to any others. Group 2 consists of all the North American sites, except Merrimack River, and also the Norwegian Hardangerfjord. The two sites that do not fall into the tighter subcluster 2a, Gloucester Point (Virginia) and San Francisco Bay, are the two southernmost sites in this survey, both at about 38~ but surprisingly on opposite coasts of the USA. Group 3 is a large cluster, containing as an outlier the remaining North American site, Merrimack River, and as two subclusters, all the estuaries from Britain, Holland, France and Spain. Broadly the same picture emerges from the ordination as from the cluster analysis. The ordination of sites, or estuaries, in Fig. 3a allows the classification of the cluster analysis to form distinct groups. The groups 1 and 2, Icelandic and American estuaries, are clearly distinguished from the European estuaries along the principal axis 1. The Icelandic and American estuaries are separated less distinctly along axis 2. The eigen value, which indicates the strength of the trends shown by the axes is much stronger for axis 1 (0.414) than for axis 2 (0.248). This difference illustrates that the separation shown along axis 1 is much more significant

R:P ratio 0.25 0.20 0.33 1.00 1.00 0.00 0.71 0.14 0.00 0.00 0.67 0.50 0.00 1.00 1.00 0.40 0.00 oo 0.00 0.14 6.00 1.75 0.64 0.50 0.73 1.00

than along the second axis. Table 3 shows totals of species extracted from Table 2. The strongest correlation with axis 1 was the R:P ratio (significant at p = 0.05), a conventional measure in algal biogeography believed to relate to latitude (FELDMANN,1937). Overlaying the ordination of species (Fig. 3b) on the site ordination substantiates the trend of algal colour groups, with greater preponderance of red algae in the American sites and brown algae in the Icelandic sites. It also suggests that the small European subcluster 3a is characterised by a greater preponderance of blue-greens than in the other European sites.

DISCUSSION

The aim of this work was to determine whether or not geographic trends existed in the upper estuarine flora principally of the North Atlantic. Some geographic differences were noted, particularly the change in the proportion of red to brown algal species, and some geographic grouping of estuaries was seen by use of ordination and cluster analysis. However the validity of these groupings and their possible causal factors should be examined.

366

WILKINSON,TELFERand GRUNDY

Broadly similar results were obtained using the ordination and cluster analysis which, at first sight, suggests robustness in the conclusions. However the apportionment of estuaries to groupings on the ordination diagram (Fig. 3a) was done on the basis of the clusters in Fig. 2. This gives rise to an arguable grouping for a few sites on the ordination diagram. For example groupings 1 and 2 have been delimited on Fig. 3a because they were identified by cluster analysis. Yet they lie so close to each other that one might argue that estuaries W (St. Lawrence) and Q (Dyrafjordur), placed in the separate groupings, are closer to each other on both axes 1 and 2 than are, for example, W and Z (San Francisco Bay) which are in the same grouping. The separate identity of groupings 1 and 2 is therefore arguable. A clearer separation exists between grouping 3 on the one hand and groupings 1 and 2 on the other. The difference is mainly between mainland Europe and the other sites (American and Icelandic). The data used were not ideal but efforts were made, as described in the methods section, to avoid using estuaries with markedly different levels of sampling, or for which the species data were clearly incomplete. It cannot be assumed that any trend over these estuaries is due to a major geographic factor such as temperature. For example, study sites may differ in the type of substratum or level of nutrients. Such factors are not uniformly presented in the papers used and so cannot be analysed along with the species data. Our own observations on British estuaries suggest that the upper estuarine algal flora is not so seasonally variable as the open coast one, so that the time of year at which the estuaries were sampled would not be a serious problem. In order to obtain a good geographic spread of estuaries it was necessary to include ones of different sizes. The second axis of ordination in Fig. 3a showed a significant correlation with the total number of species within the estuary (at p = 0.05). The species total differed considerably between the estuaries according to the size of the estuary and the number of sites presented, e.g. the two extremes were at Gloucester Point in the York River, Virginia, for which data were given for only one site and the St Lawrence estuary for which records were amalgamated for 11 upper estuarine sites. It is logical to expect more species in the larger estuaries, not merely because more stations could be used, but because the salinity variation at a single station might not be so great as in a small estuary where there could be fluctuations from full seawater to

freshwater over a single tidal cycle. DEN HARTOG (1967) showed that more species of marine algae penetrated the Baltic Sea to lower salinity limits because of the more stable salinities. It is therefore important that the first axis of ordination, the strongest trend in Fig. 3a, was independent of species number, but correlated with the R:P ratio. This ratio was first proposed by FELDMANN (1937) who suggested that this ratio varied in a graded fashion from about 1 in polar waters to greater than 4 in tropical waters. GARBARY(1987) has pointed out that there are more modern and useful approaches to marine algal biogeography. Nonetheless he concedes that the R:P ratio is useful in a descriptive sense although it only works well in the North Atlantic. Because of the high proportion of green algae in all estuaries it would be less useful to consider other ratios that include green totals such as the Cheney index which relates brown algal numbers to the combined total of reds and greens (CHENEY,1977). The altered flora in estuaries means that the actual values of the R:P ratios for the estuaries in this survey are not the same as Feldmann would have predicted for their particular localities. Nonetheless the existence of a north-south trend in the ratio for the estuaries suggests that there may be some geographical difference, however slight, in the estuarine floras. The trend towards a greater proportion of brown species in the estuarine flora at more polar latitudes may be a logical result of the combined salinity and temperature tolerance. DEN HARTOG (1971) suggested that salinity and temperature interacted to control penetration of marine algal species into estuaries. He suggested that the large temperature fluctuations in west European estuaries were the main obstacle to the penetration of euryhaline marine organisms towards freshwater. Brown seaweeds are regarded as more tolerant of lower temperature than red ones. The lower R:P ratio of sites in Iceland is particularly emphasised by a shore profile presented by MUNDA(1972) for the innermost part of the Dyrafjordur showing the normally marine brown seaweed OicO/osiphon as a dominant organism intermingled with more typically brackish species. The results so far discussed suggest that there may be a weak geographic trend with latitude in the upper estuarine flora. This trend is hard to establish on the present data for the following reasons. Firstly, all the estuaries considered belong to only two of the already postulated biogeographic regions, both of them temperate, as pointed out earlier. Secondly, there are likely to be differences in

Geographic distribution of estuarine algae the extent to which different authors, accustomed to working with open coastal communities, have included identifications of the algal groups that are important in estuaries but less so on the open coast, especially blue-greens and xanthophytes. Thirdly, the nature of the estuarine habitats used are not the same at each latitude. It is recognized that a possible criticism of this survey could be that like is not being compared with like. Such are the data that the present inventory offers the best comparison that can be made. It is conventional to assume that algal phytogeography dictates that there will be geographic differences. This idea is based on a long history of subjective assessments by many workers. However a recent study of open coast floras of the North Atlantic (TIi-rLEY et aL, 1990) using more objective computer-aided techniques, such as ordination (the same type of ordination as in the present work) and an ecocladistic method, starting from a null hypothesis that algal biogeographic provinces are proven not to exist, suggested that the North Atlantic may be a single biogeographic province (except for the Azores). The gradual change in R:P ratio with latitude in the present paper may reflect a continuum of change within a single province. Another interesting comparison is the percentage occurrence of species in the estuaries. The ten most common species, with the percentage of estuaries in which they were recorded, are as follows: Enteromorpha intestinalis (85%), Blidingia minima (77%), Rhizoclonium tortuosum (73%), Fucus

367

vesiculosus (65%), Enteromorpha prolifera (65%), Monostroma oxyspermum (62%), Fucus ceranoides (58%), Pilayella littoralis (54%), Vaucheriaspp. (46%), and Ulothrix flacca (46%). Of these ten species, five were also found at a southern hemisphere estuarine site in warm waters at 32~ in the Patos Lagoon Estuary in Brazil (COUTHINO and SEELIGEB, 1984). This find indicates that some of the commonest estuarine algae may have widespread distributions as suggested in the introduction. While a slight geographic trend may have been suggested by some of the results analysed in this paper there is the possibility that there are some widely distributed tolerant estuarine algae, perhaps with estuarine algae forming a continuum over a greater range than just even a single North Atlantic phytogeographic province. There is a need for more data on algae in estuaries in southern Europe in order to substantiate the tentative ideas raised for the North Atlantic and from elsewhere in the world to study how universal is the upper estuarine flora. ACKNOWLEDGEMENTS During this work T. Telfer and S. Grundy were in receipt of Marine Technology Research Studentships from the (UK) Engineering and Physical Science Research Council. lan Tittley (Natural History Museum, London) is thanked for helpful discussion.

REFERENCES CARDINAL, A. and M. VILLALARD, 1971. Inventaire des algues marines benthiques de I'estuaire du Saint-Laurent (Quebec). Naturaliste Can., 98: 887-904. CHENEY,O.F., 1977. R and C / P - a new and improved ratio for comparing seaweedfloras. J. Phycol., 13 (Suppl.): 12. COUTHINO,R. and U. SEELIGER,1984. The horizontal distribution of the benthic algal flora in the Patos Lagoon estuary, Brazil, in relation to salinity, substratum and wave exposure. J. Exp. Mar. Biol. Ecol., 80: 247-257. DEN HARTOG,C., 1960. Comments on the Venice system for the classification of brackish waters. Int. Revueges. Hydrobiol., 45: 481-485. DEN I-IARTOG,C., 1967. Brackish water as an environmentfor algae. Blumea, 15: 31-43. DEN HARTOG,C., 1971. The border environment betweenthe sea and the freshwater with special referenceto the estuary. Vie Milieu, 22 (Suppl.): 739-751. DONZE, M., 1968. The algal vegetation of the Ria de Arosa (N.W. Spain). Blumea, 16: 159-192. EDWARDS, P., 1972. Benthic algae in polluted estuaries. Mar. Pollut. Biol., 3: 55-60. FELDMANN,J., 1937. Recherchessur la v~getationde la Mediterran6e. La cotes des Alberes. Rev.Algol., 10: 1-339. GARBARY, D., 1987. A critique of traditional approachesto seaweeddistribution in light of the developmentof viacariancebiogeography. Helg. Meeresunt.,41: 235-244. HARDY, F.G., S.M. EVANSand M. TREMAYNE,1993. Long-term changes in the marine macroalgaeof three polluted estuaries in north-east England. J. Exp. Mar. Biol. Ecol., 172: 81-92. JORDE, I. and N. KLAVESTAD,1963. The natural history of the Hardangerfjord.4. The benthonic algal vegetation. Sarsia,9: 1-99. JOSSELYN, M.N. and J.A. WEST, 1985. The distribution and temporal dynamics of the estuarine macroalgalcommunity of San Francisco Bay. Hydrobiologia, 129: 139-152. LEPAILLEUR, H., 1971. Contribution a I'~tude de la v~getationalgale dans I'estuaire de rOrne (France). Vie Milieu, 22 (Suppl.): 234-241.

368

WILKINSON, TELFERand GRUNDY

LUNING, K. 1990. Seaweeds- Their environment, biogeographyand ecophysiology.Wiley, New York. MATHIESON,A.C. and R.A. FRALICK, 1972. Investigationsof New England marine algae.V. The algal vegetation of the Hampton-Seabrook estuary and the open coast near Hampton, New Hampshire. Rhodora,74: 406-435. MATHIESON, A.C. and R.A. FRALICK, 1973. Benthic algae and vascular plants of the lower Merrimack River and adjacent shoreline. Rhodora, 75: 52-64. MATHIESON,A.C., N.B. REYNOLDSand E.J. HEHRE,1981. Investigationsof New England marine algae I1: The species composition, distribution and zonation of seaweedsin the Great Bay estuary system and the adjacent open coast of New Hampshire. BotanicaMarina, 24: 533-545. MUNDA, I., 1972. Generalfeatures of the benthic algal zonation around the Icelandic coast. Acta Naturalia Islandica, 21: 1-34. NIENHUIS, P.H., 1980. The epilithic algal vegetation of the SW Netherlands. Nova Hedwigia,33: 1-94. NORTON,T.A., 1976. The marine algae of the eastern border counties of Scotland. Br. Phycol. J., 11: 19-27. PEREZ-CIRERA,J.L. and J. PACHECO,1985. Zonacion y distribucion geograficade la vegetacionbentonica de la Ria de Lires (No. Espana). Trabajos Compostelanosde Biologia, 12:153-183. SOUTH, G.R. and I. TITTLEY, 1986. A checklist and distributional index of the benthic marine algae of the North Atlantic Ocean. Huntsman Marine Laboratory and British Museum (Natural History), St. Andrews and London. TII-ILEY, I. 1985. Zonation and seasonality of estuarine benthic algae: artificial embankments in the River Thames. Botanica Marina, 28: 1- 8. TITILEY, I., G.LJ. PATERSON,P.J.D. LAMBSHEADand G.R. SOUTH, 1990. Algal provinces in the North Atlantic - do they exist? In: O.J. Garbary and G.R. South, Eds., Evolutionary Biogeographyof the North Atlantic. NATOASI Series Vol. G 22, Springer-Verlag,Berlin: 291-322. TITiLEY, I. and J.H. PRICE, 1977. The marine algae of the tidal Thames. London Naturalist, 56: 10-17. WILKINSON, M., 1973a. A preliminary survey of the intertidal benthic algae of the Clydeestuary. Western Naturalist, 2: 59-69. WILKINSON, M., 1973b. The distribution of attached intertidal algae in estuaries with particular reference to the River Wear. Vasculum, 58: 22-28. WILKINSON, M., 1975. Intertidal algae of some estuaries in Galloway.Western Naturalist, 4: 42-50. WILKINSON, M., 1979. Marine algae of the Grampian region of Scotland. Br. Phycol. J., 14: 33-41. WILKINSON, M., 1980. Estuarine benthic algaeand their environment: A review. In: J.H. Price, D.E.G. Irvine and W.F. Farnham, Eds., The Shore Environment, Vol 2. Ecosystems.Academic Press, London and New York: 425-486. WILKINSON, M., A.R. HENDERSONand C. ROBERTS,1976. Distribution of attached algae in estuaries. Mar. Pollut. Bull., 7: 183-184. WILKINSON, M. and D.A. RENDALL, 1985. The role of benthic algae in estuarine pollution assessment. In: J.G. Wilson and W. Halcrow, Eds. Estuarine managementand quality assessment. Plenum, New York: 71-81. WILKINSON, M. and C. ROBERTS,1974. Intertidal algae of the estuary of the River Add, Argyllshire. Western Naturalist, 3: 73-82. WILKINSON, M., C.M. SCANLANand I. TITILEY, 1987. The attached algae of the estuary and Firth of Forth, Scotland. Proc. Roy. Soc. Edinb., 93B: 343-354. WULFF, B.L. and K.L. WEBB, 1969. Intertidal zonation of marine algaeat Gloucester Point, Virginia. ChesapeakeScience, 10: 29-35.

ADDRESSOF THE AUTHORS:

Biological Sciences, Heriot-Watt University, Edinburgh, EH14 4AS, Scotland. 1) Presentaddress: Institute o1 Aquaculture, Stirling University, Stifling, FK9 4LA, Scotland.