From: The Ecological Strategies of Aquatic Ranunculus Species 

"Water Crow-foot hath tender branches trailing far abroad, whereupon grow leaves under the water most finely cut and jagged like those of Cammomill. Those above the water are somewhat round, indented about the edges, in form not unlike the tender leaves of the mallow but lesser: among which do grow the floures, small, and white of colour, made of five little leaves, with some yellowness in the middle like the floures of the straw-berry, and of a sweet smell: after which these come round rough and prickly knaps like those of the field Crowfoot. The roots be very small hairy strings."

Gerard (1633)
 

1.1 Riverine Ranunculus Species

Ranunculus subgenus Batrachium (DC.) A. Gray species are the dominant plants of many British streams and rivers, especially in lowland streams subjected to regular cutting. Their flowers are similar in appearance to meadow buttercups (R. repens, R. acris), except that their petals are white rather than yellow. Some Ranunculus plants growing in rivers may reach increasing the risk of flooding (see for example Murphy et al. 1990). Dawson (1989) has estimated that the cost of aquatic plant management in the U.K. could be as high as £100 million, with a large proportion of this being devoted to the control of R. fluitans and R. penicillatus.

The plants appear to have little current or historical positive medicinal or agricultural uses although Pultney (1798) described how farmers in Hampshire fed cattle, horses and pigs on 'R. fluitans' from the River Avon, in some cases the Ranunculus forming almost the entire diet of the animals. The cattle were reported to find it so palatable that they had to be restrained to eating 11 - 14 kg each day. Cattle in Dorset still graze R. penicillatus when given the opportunity (pers. obsv.) and a local farmer has described how when his cattle are let into a field adjacent to the River Frome for the first time in the Spring they ignore the fresh grass and instead eat the R. penicillatus in the river.

Along with the majority of aquatic plants, Ranunculus species tend to be dismissed as 'water weeds', though as well as the common name of Water Crowfoot, Lodewort and Rams Foot have been historically used as names for the subgenus (Gerard 1633). Grigson (1955) reports a number of regional names including Bacon and Eggs (Somerset), Cow-Weed (Hampshire), Eel Weed (Donegal), Rait, Pickerel Weed (East Anglia - a pickerel is a young pike), and Rawheads (Shropshire). Scots Gaelic names include fleann uissage (water-follower) and lion na h'aibhue (the river flax) (Cameron 1883).

The taxonomy of the group is difficult and complex due to extreme phenotypicplasticity together with morphological reduction (Webster 1984). Over 300synonyms have been applied to the taxa in the group and it was not untilCook's 1966 monograph that the situation began to be resolved (althoughit is still far from clear: Webster 1991). As Babington (1855) put it 'thegreat difficulty of the subject necessarily weighs heavily on the mind'.More recently Wiegleb (1988) wrote that 'in Batrachium, all kinds of hybrids,intermediates and mysterious forms occur which should not puzzle the observer'.The taxonomic uncertainly has led to it being quite unclear which speciesare being referred to in some earlier ecological work, with associatedimprecision in distribution maps, etc.

1.1.1 Ranunculus omiophyllus Ten. and Ranunculus hederaceus L.

These two species are distinct from the rest of the sub-genus due tothe absence of submerged, dissected leaves (Cook 1963). It is thought thatthey have evolved from amphibious ancestors and have lost the ability todevelop submerged divided leaves (Cook 1966b, 1970). R. hederaceus, asthe name implies, has leaves shaped somewhat like the ivy, with the lobes widest at the base, whereas R. omiophyllus has more rounded lobes which are narrowest at the base. Although the leaf shapes are fairly distinct (Fig. 1.1), they can be quite variable (Pearsall 1929) which has given rise to problems of identification in the past (Babington 1855), though the leaves do retain the characteristic form of each species (Cook 1966a).

When in flower there is no problem distinguishing between the two species; R. omiophyllus has petals which are at least twice as long as the (often recurved) sepals, whereas the petals in R. hederaceus are much smaller (usually 2.5-3.5 mm) (Holmes 1979; Webster 1988b).

R. hederaceus grows throughout the British Isles with the exception of part of the central and northern Scottish Highlands and a somewhat scattered distribution in southern and eastern England (Biological Records Centre, personal communication 1991; Perring & Walters 1976). The common name, Mud Crowfoot, reflects its habitat in that it tends to be found in wet mud at the edges of pools, ditches, etc (Webster 1988b). A study in the Netherlands has shown that R. hederaceus is less common there but occupies similar habitats to those it is found in in Britain. It appears to be in decline due to changes in the hydrological stability of its sites rather than actual habitat destruction (Van Diggelen & Klooker 1990).

R. omiophyllus has a more Western distribution than R. hederaceus; in a recent phytogeographical study Arts & Den Hartog (1990) described it as a characteristic Atlantic species. In Britain it is only found in regions where the mean August rainfall is greater than 75 mm (Webster 1988) and so it is less common than R. hederaceus and is absent from much of Scotland, Ireland and eastern England (Biological Records Centre, personal communication 1991; Perring & Walters 1976).

Salisbury (1934) considered that R. hederaceus grows in mineral waters whereas R. omiophyllus is 'invariably' found in more peaty waters, and this is still thought to be a good generalisation (Webster 1988b). Newbold and Palmer (1979) have ranked all the British aquatic species according to the nutrient ('trophic') status of the waters in which they grow. By this ranking R. omiophyllus has a trophic rank of 19 (oligotrophic) and R. hederaceus has a trophic rank of 150 (eutrophic).

In Britain R. omiophyllus appears to be a calcifuge (Clapham, Tutin & Moore 1987), but Cook (1966a) has observed it growing on a calcareous substrate in Italy. In cultivation British material of R. omiophyllus and R. hederaceus grows equally well on calcareus and non-calcareus substrates (Cook 1966b). Both tend to grow in somewhat open and disturbed habitats (Webster 1988). Occasionally the two species grow together (Salisbury 1934; see Mill Lawn Brook in Appendix A), but they do not hybridize (Cook 1970). R. omiophyllus forms a natural hybrid with R. peltatus (Webster 1984, 1986; see Avon Water in Appendix A), known at R. hiltonii Groves & Groves after Mr T. Hilton who discovered it in 1896 (Groves & Groves 1901). This plant is highly unusual as it overwinters in the heterophyllous state. It is also a British endemic (Stace 1991). R. omiophyllus also forms a (Webster, 1990) which to a large extent replaces the parent species in the New Forest (Cook 1975).

1.1.2 Ranunculus aquatilis L.

Linnaeus (1762) recognised just two species in what is now the sub-genus Batrachium; R. hederaceus and R. aquatilis. As recently as 1951 Willis described the other taxa as 'so-called species', preferring to keep R. aquatilis to include the whole sub-genus. Although this position finds little support today, the decision as to how to separate the various taxa, and at what hierarchical level the separations should lie would appear to be far from settled, in spite of recent photochemical and morphological studies (Webster 1984, 1991). In this article the nomenclature and classification of Webster (1988b) is followed.

R. aquatilis sensu stricto has a number of characteristics that enable it to be distinguished from other members of the sub-genus. It has both dissected and floating leaves, the latter tending to have rather more crenate (i.e. less rounded) lobes than R. peltatus - which it otherwise closely resembles. However in flower it is easily distinguished by the circular shape of the nectar pit (though this characteristic must be used with care as the shape undergoes a developmental sequence, so that it is only reliable on mature petals). R. aquatilis also forms intermediate leaves with a proximal capillary portion (see Fig 1.1). In common with all the species with dissected leaves it also grows in a terrestrial form, in which it is indistinguishable from the other species, even when in flower (Webster 1988b).

R. aquatilis has been used in a variety of studies on the mechanisms underlying the control of heterophylly (for example Askenasy 1870, Bostrack & Millington 1962, Davis and Heywood 1963; Cook 1969). Nielsen & Sand-Jensen (1989) found that it had an intermediate photosynthetic rate compared with thirteen other aquatic species studied (4.92 mg O2g-1h-1at pH 8.3) though a higher than average dark respiration rate (1.20 mg O2g-1h-1). It was found to have one of the highest laboratory growth rates of the species studied (0.097 day-1). This growth rate is very similar to that found in field experiments (0.092 day-1) by Nørgaard (1989) for R. peltatus.

Litav & Agami (1976) & Agami et al. (1976) found that it was one of the species that disappeared from the River Yarkon over a 25-30 year period, during which time a variety of pollutants (especially detergents) increased in concentration in the river.

R. aquatilis is found in scattered localities throughout much of the British Isles and is the commonest species that grows in still water (Butcher 1960) but it is absent from much of Scotland and Ireland (Biological Records Centre, personal communication 1991; Perring & Walters 1976).

Crowder et al. (1977) found that it grows over a wide range of substrate-types, although Haslam (1978) found it tended to grow on harder rock-types than R. peltatus. Especially in the lowlands, it is more commonly found in ponds than streams. It is frequently found in farm ponds (Pip 1979, Webster 1988b), but if these are very enriched it is displaced by other species. It appears to be able to tolerate a moderate amount of disturbance in the form of drought and grazing (NCC 1989). Newbold and Palmer (1979) placed it at trophic rank 70.

The National Vegetation Classification of Aquatic Communities (NCC 1988) found that R. aquatilis was most frequently associated with Callitriche species (especially C. stagnalis and C. obtusangula) and Glyceria fluitans. It grows in very similar habitats to R. peltatus - Cook (1966a) reports that the two species do not grow together, although natural hybrids have been reported (Stace 1991) and R. trichophyllus is found growing intermingled with both of them.

R. aquatilis forms several natural hybrids (Cook 1975, Webster 1990, tripartitus (no name).

1.1.3 Ranunculus peltatus Schrank

R. peltatus is morphologically and ecologically similar to R. aquatilis, especially when not flowering (although Butcher (1960) considered it to be quite different physiologically). When in flower, any confusion which arises is more likely to be with R. penicillatus subsp. penicillatus as it has similarly sized petals and the same shaped nectary pit as well as similar floating leaves. The most reliable character to separate them is that in R. peltatus the dissected leaves are distinctly shorter than the internodes, whereas they are longer in R. subsp penicillatus (Webster 1984; Holmes pers. comm. 1991).

R. peltatus has a similar distribution in the British Isles to R. aquatilis, but it is found in a greater number of localities (Biological Records Centre, personal communication 1991).

Although it has a closely similar carbon extraction capacity to R. aquatilis (Madsen & Sand-Jensen 1991), R. peltatus seems to be a little more stress-tolerant than R. aquatilis, growing in a little deeper water (Holmes 1979) and with the widest physical range of habitats amongst the sub-genus (Haslam 1978). It has a lower trophic rank than R. aquatilis (48; Newbold and Palmer 1979). However, Monschau-Dusenhausen (1982) found that it was the most pollution-sensitive species of a range of aquatic macrophytes.

It can tolerate greater fluctuating levels of water and is the most frequent Ranunculus species in reservoirs (Grime, Hodgson & Hunt 1988). Arts et al. (1990) found that it was found in waters in The Netherlands with a mean pH of 6.8 and a mean alkalinity of 0.7 (meq l-1). At pH values of less than 5.7 damage occurred to the plants. Dawson & Kern-Hansen (1979) showed that there was a linear relationship between the amount of artificial shading applied to a R. peltatus community and its standing crop.

Cook (1966a) described it as being characteristic of temporary or disturbed habitats, and Ladle & Bass (1981) have described how R. peltatus displaced R. var. when a chalk stream suffered severe disturbance as it dried up in a period of drought. The colonization of R. peltatus was by means of seedlings germinating on the dry stream bed. R. peltatus is frequently found at the head of winterbournes (see in Appendix A the Rivers Bourne & Wylye). Valane et al. (1982) have shown that R. peltatus has a lower degree of intercellular structural adaptation to being submerged than R. baudotii. R. peltatus has epidermal cells overlaying chloroplast cells (a characteristic of terrestrial plants) and the ultrastructure of its chloroplasts are more similar to those normally found in land plants than aquatics.

Cook (1966a) stated that as the water 'matures' in a pond R. peltatus (as well as R. aquatilis and R. trichophyllus) is replaced by rhizomatous species such as Potamogeton pectinatus and P. crispus. However if the water is susceptible to regular disturbance R. peltatus can persist (NCC 1988). The species is susceptible to frost if not submerged but can withstand being frozen in ice (Cook 1966a).

Natural hybrids are formed with R. omiophyllus and R. baudotii (Webster 1986), R. trichophyllus, R. aquatilis (Stace 1991), and with R. fluitans Stace 1991) which has spread extensively along the River Welland (Cook rise to R. penicillatus subsp. penicillatus.

Babington (1855) considered R. peltatus ('R. floribundus') to be 'the most beautiful of our species; its large white flowers being so numerous as to cover the places that it inhabits with a sheet of bloom'.

1.1.4 Ranunculus baudotii Godron

R. baudotii is an uncommon species, occurring at only 170 sites in Britain. It is named after the botanist de Baudot of Saarburg (Barnhart 1965).

R. baudotii can be difficult to distinguish from R. aquatilis (Blackmore 1985), but although it has similar morphology its distribution is heavily influenced by the fact that (as Babington (1855) put it) it 'appears to delight in slightly brackish water'. Thus it is always found in coastal habitats, with the exception of a few inland records from Cambridgeshire brick pits (Biological Records Centre, personal communication 1991; Perring & Walters 1976). The most reliable character to distinguish it from R. aquatilis or other taxa is the shape of the receptacle which elongates in fruit. Other corroborative characteristics are the lunate nectar pit, blue-tipped sepals and winged fruits. The latter two of these characters do not always occur. R. baudotii shows considerable phenotypic plasticity. Another characteristic which is not constant is the formation of floating ben elevated to specific rank; see Moss 1914) do not form these leaves; for example the R. baudotii at the site at Worth Matravers, Dorset (Appendix A). Luther (1947) considered 'form marinus'to be a locally induced habitat form.

Although only found in brackish water, Cook (1966a) showed that it will grow with no loss of vigour in cultivation in fresh water. Cook (1966a) has reported it growing naturally at a site which is covered by the sea at high tide, and in cultivation it will survive in 100% sea water. Kautsky (1991) has found it in sites of up to 1% salinity. Van Viersen & Verhoeven (1983) came to the conclusion that R. baudotii was apparently not as salt tolerant as Zannichellia pedunculata or Potamogeton pectinatus.

Although in the UK it normally grows in shallow water of less than 0.3 m depth (Cook 1966a) it is found in deeper water occasionally (for example at 0.54m at Porth Oer, Gwynedd, see Appendix B) and Kautsky (1991) has recorded it growing at 3 m depth near Stockholm, Sweden. Kiørboe (1980) has observed a standing crop of 3.5 g dry weight R. baudotii m-1 in a water depth of 0.6 m in a Fiord in Denmark. It has a trophic rank of 133 (Newbold and Palmer 1979).

R. baudotii is characteristic of open and disturbed habitats (Webster 1988b). In view of this and its preference for shallow water it is significant that Van Viersen & Verhoeven (1983) considered its ability to survive desiccation a decisive factor in its ecology. Kirboe (1980) has reported wildfowl grazing on R. baudotii but concluded that because most of the grazing occurred outside of the plants'main growing season, the damage done would be negligible. The main wildfowl were Cygnus olor, Anas platyrhynchos, A. penelope and Fulica atra. Crivelli (1983) has described a negative effect of carp (Cyprinus carpio) on R. baudotii, which appeared to be caused more by uprooting than by an increase in turbidity or actually eating the plants. Reproduction of R. baudotii appears to be mainly by vegetative means (axillary buds) (Kautsky 1990).

Kautsky (1991) has shown that in an experimental situation R. baudotii grows better in a 50/50 mix of sand and mud than either pure mud or sand. In a de Wit replacement competition experiment Kautsky (1991) found that the effects of interspecific competition between R. baudotii and Potamogeton filiformis were less than the effects of intraspecific competition when R. baudotii was grown in monoculture. When R. baudotii was grown with P. pectinatus there was greater interspecific than intraspecific competition.

The National Vegetational Classification of Aquatic Communities (NCC 1988) described vegetation dominated by stands of R. baudotii but found there was no other species constantly associated with it. In standing or sluggish brackish waters Ceratophyllum submersum and Potamogeton pectinatus are associated with R. baudotii (see Van Viersen and Verhoeven 1983), and as the waters become more saline Zannichellia and Ruppia spiralis become more frequent. These communities appear to be in decline, so that the 170 sites referred to above may be an over-estimate.

R. baudotii forms natural hybrids with R. aquatilis, R. peltatus and (Cook 1975, Stace 1991).

1.1.5 Ranunculus trichophyllus Chaix

R. trichophyllus (as with all the species below) forms no floating leaves, just submerged, dissected leaves (hence the name, which means 'hair-like' Gledhill 1989). It may be distinguished from the other species which do not form floating leaves by its small (< 7mm) petals and lunate nectary pit (although R. circinatus has those characteristics its distinctive leaf-shape makes confusion unlikely). However, when it is not in flower it may easily be confused with R. aquatilis (Holmes 1979). Several of the sites visited in Lothian, whilst carrying out the work described in the thesis from which this introduction is taken, had old records of R. trichophyllus populations, but turned out to have floating-leaved species present.

R. trichophyllus is found in sites scattered throughout the British Isles, but with a pronounced south-easterly bias (Biological Records Centre, personal communication 1991; Perring & Walters 1976). On a global scale it is probably the most widely distributed Batrachian Ranunculus species (Drew 1936, Cook 1966a). It grows from sea-level to 2500 m in the Alps (Arber 1920). It has a trophic rank of 75 (Newbold and Palmer 1979). All the populations in Britain may be assigned to subspecies trichophyllus (Cook 1966a).

In temporary waters it behaves as an annual, but if conditions allow it to persist, it grows as a perennial (Holmes 1979). R. trichophyllus is more frequently found in eutrophic than oligotrophic waters (Cook 1966a). It is rarely found in swift currents (Holmes 1979) and is the only riverine Ranunculus species that has not been included in any of the work described in the thesis from which this introduction is taken.

In spite of its wide distribution, relatively few studies appear to have been carried out on this species. Pond (1905) found that it grew better when rooted than when suspended in tap water. Dale and Miller (1978) observed that it decreased in abundance in a lake that was subjected to sewage and mining discharges over a thirty year period. Murphy & Pearce (1987) described the use of diquat alginate to control the growth R. trichophyllus in a Scottish loch used as a salmonid fishery. Lorch and Ottow (1988) have described the bacteria and diatoms that are epiphytic upon R. trichophyllus.

R. trichophyllus forms hybrids with R. baudotii, R.aquatilis, R. peltatus, pools in West Suffolk - it is rare as the parents are not often found together (Cook 1966a, 1975).

1.1.6 Ranunculus circinatus Sibth.

R. circinatus is instantly recognisable by its distinctive fan-shaped leaves, with radiating dividing spokes in one plane (Fig. 1.2). 'circinatus'literally means 'curled round'(Gledhill 1989).

It has two distinct growth states; in the winter it is prostrate and branching and in the summer erect and simple. The winter state may persist in the summer if the plants are stressed. Both states can co-exist in the same individual (Cook 1966a).

The species does not grow in either fast currents or base-poor waters (Holmes 1979) and is thus virtually absent from upland regions of Britain (Biological Records Centre, personal communication 1991; Perring & Walters 1976). It is usually found in deep water - Luther (1951) recorded it at 5 m deep in Finland. Although it can behave as an annual (Salisbury 1960) it normally has a perennial life-history (Cook 1966a).

R. circinatus is usually found in relatively nutrient rich waters (Holmes 1979), and has a tropic rank of 98 (Newbold & Palmer 1979). It has occasionally been recorded in brackish waters (Olsen 1950) and waters with saline incursions (personal observation 1990, see Old Bedford River site in Appendix B). In a survey of Welsh lakes Seddon (1972) found R. circinatus in lakes with a conductivity of > 10 mS m-1and a hardness ratio of >2.5. Bernatowicz (1965) observed the expansion of the range of a R. circinatus population in a Polish lake following cutting of the Phragmites stand in which it was growing.

Forsberg (1964) found that R. circinatus was unable to root in the soft sediments of a Swedish lake, in common with other macrophytes. In a survey of soft-water lakes in The Netherlands Arts & Leuven (1988) found R. circinatus occurring in a plant communities characterised by the absence of isoetids and the presence of the moss Fontinalis antipyretica. This is a little surprising as although F. antipyretica is often found in rivers and lakes it is usually associated with shallow water (Watson 1955) - though

R. circinatus forms a hybrid with R. fluitans which grows in the Berwickshire Blackadder (Cook 1975).

1.1.7 Ranunculus penicillatus (Dumort.) Bab.

This taxon has been surrounded by what Holmes (1980) described as 'understandable confusion'. It was not until 1966 that Cook was able to show that it is a collection of segmental amphiploids resulting from hybridization (i.e. speciation has resulted from polyploidy occurring after hybridization). It is believed to have been formed as a result of hybridization of R. fluitans with R. aquatilis, R. trichophyllus and possibly R. peltatus (Stace 1975). Taxonomic opinion is still unresolved as to whether taxa within penicillatus should be divided from each other at the species level (Haslam & Wolseley 1981; Murrell and Sell 1990), subspecies (Webster 1988a, Stace 1991) or variety (Holmes 1980; Clapham Tutin & Moore 1987). In his 1855 monograph on the Batrachian Ranunculi of Britain, Babington observed that 'we have no good definition of a species... and that it is hard or even impossible to apply those which we possess'. One wonders if he might come to the same conclusion with penicillatus today, certainly modern biochemical techniques that have proved so useful in separating other taxa have as yet proved inadequate (Webster 1991).

R. penicillatus subsp. penicillatus is separated from R. penicillatus subsp. (= R. calcareus) by the formation of floating laminar leaves in the former. Together with the disjunct distribution and differing ecology, Webster (1988a) considered this difference too great to allow separation only at the varietal level, but as this is the only difference she does not consider that the taxa merit specific rank. As there is morphological and geographical continuity between var. and var. vertumnus Webster (1988a) assigned these two to varieties within the subspecies .

Among the 'unfortunate circumstances'(Webster 1988a) surrounding the nomenclature of the group is the widespread use of the term R. calcareus for R. penicillatus subsp. pseudofluitans. Webster (1988a) has convincingly shown that is the earliest epithet to be applied to this taxon, and its use is adopted in this thesis (this term is also used in the most recent British Flora, Stace 1991, and the second edition thereof, 1997).

R. penicillatus subsp. penicillatus

As indicated above R. penicillatus subsp. penicillatus is separated from subsp. by the presence of laminar leaves, which it forms after flowering (Holmes 1980). 'penicillatus' refers to the shape of the dissected leaves and means brush-like (Gledhill 1989). R. penicillatus may be confused with R. peltatus when flowering; R. peltatus has capillary leaves that are usually shorter than the internodes and they tend to be more rigid than R. penicillatus (Webster 1988b).

R. penicillatus subsp. penicillatus is confined to 41 sites in the west of England and Wales and 25 scattered localities in Ireland (Biological Records Centre, personal communication 1991; Perring & Walters 1976). Palmer & Newbold (1983) identified the taxon as in need of protection. In Britain it is found mostly in base-poor fast flowing rivers where R. fluitans is absent or rare (Holmes 1980) but in Ireland it is found in more calcareus habitats (Webster 1988a). It has a trophic rank of 69 (Newbold and Palmer 1979). Decamps (1985) found that its germination was slightly depressed in waters with a high calcium content.

In a survey of macrophytes in the River Suir in Ireland, Caffrey (1985) placed R. penicillatus subsp. penicillatus in the group most sensitive to pollution. In a survey of 52 Irish rivers Caffrey (1990b) showed that R. penicillatus subsp. penicillatus was associated with communities at relatively shallow depths, low conductivities and moderate flow-rates.

R. penicillatus subsp. pseudofluitans var. (Syme) S.Webster

R. penicillatus subsp. pseudofluitans var. (= R. calcareus, referred to as below as var. ) is by far the most abundant Batrachian Ranunculus species in Britain, dominating and forming extensive stands in many stretches of rivers and streams. However, it is probably less common than it was previously (Grime et al. 1988). As with the rest of the group, it is a morphologically plastic species. Some populations have leaves as long as those of R. fluitans (up to 385 mm; the 'holmes' and the flowers have a densely hairy receptacle (Webster 1988a). Separation from var. vertumnus is discussed under that taxon. The variety may occasionally be confused with R. trichophyllus; when in flower the species can be separated by the shape of the nectary pit (lunate in R. trichophyllus, pear-shaped in R. penicillatus, Webster 1988b).

Variety is found in rivers and streams throughout England and Wales, though it is sparse in Scotland and absent from Ireland except for one possible site in Derry (Webster 1988a). It has a wider ecological amplitude in terms of calcium requirement than its alternative name of R. calcareus would imply. However, it is usually found in base-rich rivers, with alkalinities greater than 100 mg l-1 CaCO3 and a high conductivity (Webster 1988a). Holmes (1983) recorded var. in all four of his types of river communities, and Newbold & Palmer (1979) considered var. to be distributed in waters of a trophic status from mesotrophic to eutrophic (trophic rank 99). Merry et al. (1981) also found var. to be present in sites with a wide range of conductivity, pH, calcium and altitude values. The overlap of its habitat requirements (ecological niche) with other Batrachian Ranunculus species is illustrated by the fact that Gehu and Meriaux (1983) place it in the same phytosociological community as R. fluitans.

Due to its importance in lowland rivers, especially chalk streams, a considerable amount of work (including some early physiological studies) has been carried out on this taxon. Studies concerning the effects of shading are reviewed in the introduction to Chapter Three of the thesis from which this introduction is taken and those concerning the response of the plant to cutting are reviewed in Chapter Four of that thesis.

One of the factors that appears to determine the maximum standing crop formed by the plant is the rate of water flow in the spring; there is a positive correlation between these two factors in the River Wye over several years (Edwards & Brooker 1982). Other studies have found similar results. For example in a survey of river sites dominated by Ranunculus species in northern England and Scotland, Spink et al. (1990) found discharge to be an important factor for determining community type. However particularly in chalk streams where discharge is less variable and cutting is frequent other factors such as the cutting regime, turbidity and insolation may be more important than discharge (Westlake & Dawson 1982).

Marshall and Westlake (1990) measured the water velocities around and within var. clumps. and found that although the water in a stream had a velocity of more than 0.5 m s-1 the current velocity inside the Ranunculus clump was less than 0.1 m s-1. In still or slow waters one of the most important limiting factors for plant growth is the availability of carbon (either as bicarbonate or as dissolved carbon dioxide) (Black et al. 1981; Hough & Fornwall 1988). Westlake (1966) showed that the rate of photosynthesis in var. was primarily limited by the rate of diffusion of carbon dioxide to the plant's leaves, which in turn is determined by the rate of flow of the water (faster flows reduce boundary layer resistance).

Var. shows a marked increase in growth with increase in water velocity (Westlake 1967) and Ham et al. (1981) found that the increase in area of var. in the spring in the River Lambourn (a chalk stream in southern England) was correlated with the mean discharge at that time. Percentage cover was lowest in March and the area then rapidly increased to reach a maximum in early summer. During August and September (after flowering) the plants declined (peak biomass is about a month after flowering, Dawson 1976) and continued to be washed-out during the winter (some above-ground material usually remains and photosynthesis occurs, Sculthorpe 1967). This phenology has been observed in many other studies (Ladle & Casey 1971, Ladle & Bass 1981, Wright et al. 1982, Ham et al. 1982, Spink et al. 1990).

In the light of the above it would appear unlikely that increased nutrient (NPK) supply has caused these growths, contrary to the opinion of many anglers. Large Ranunculus standing crops have been observed in rivers before the advent of artificial fertilisers (Hutton 1930) and the concentration of major plant nutrients in chalk streams where var. is dominant is far greater than the nutrient requirements of the plant (Ladle & Casey, 1971). Casey & Downing (1976) found that on the whole there was no correlation between nutrient concentrations in var. shoots taken from eight sites and the water chemistry, with the exception of phosphorus (the Ranunculus appearing to exhibit luxury consumption of very high phosphorus concentrations in the water). However during periods of rapid growth in small streams measurable amounts of nutrients are removed from the water (Casey & Westlake, 1974), indicating that nutrients et al. (1993) showed experimentally that eutrophication can cause more competitive species such as Potamogeton pectinatus to out-compete and displace Ranunculus in chalk streams.

The growth of R. penicillatus subsp. pseudofluitans in many rivers is so abundant as to have a very significant impact on the ecology of the river. Westlake et al. (1972) recorded a biomass of 630 g dry weight m-1 in a chalk stream dominated by var. . Although this is at the upper end of the productivity recorded for submerged plants in temperate regions it is still considerably less than the maximum productivity of emergent or some terrestrial plants (Westlake 1975). The 'world record' for a submerged macrophyte standing crop is claimed to be 3518 g dry weight m-1 of Lagriosiphon major in New Zealand (Clayton 1982). More typical values of 130-260 g m-1 were measured by Westlake (1968) in the River Frome. The growth of var. can be very variable from one year to the 1977 as in 1976.

If a large growth occurs and then decays due to drought conditions this can cause almost complete deoxygenation of the river in combination with other factors (Edwards & Brooker, 1982). After the var. was cut in a stream in Dorset, Westlake (1968) found that the night minimum oxygen concentration was raised from 40-60% saturation to 60-75% and the day maximum was lowered from 130-160% to 90-100%. The growth of var. stands can lead to increased water depth with an increased risk of flooding. In many lowland rivers this has led to long-term management by cutting

If var. is introduced to a suitable river it can rapidly grow to dominate the flora (Holmes & Whitton 1977a, 1977b). However if the conditions change to make the river less suitable, the cover of var. will decrease. Ladle & Bass (1981) described how following a drought R. peltatus competitively replaced var. in a winterbourne in Dorset. Var. has since re-established itself (personal observation 1989-90). Brookes (1986) observed the effects of sedimentation on var. . In the River Wylye sedimentation had little effect as the sediment remained in suspension, but in Wallop Brook the sediment was deposited onto the plant clumps, the Ranunculus appeared to be unable to vary its rooting level in response to this and a year later the Ranunculus cover was only 5% of that expected. In natural conditions var. clumps act as sediment traps with a consequent build up of sediment around its roots, so that (in contrast to R. fluitans), var. clumps are rarely in the same place one year as they were the previous year (Westlake 1968, 1973, Casey & Westlake 1974, Furse 1977). The tendancy of R. penicillatus subsp. pseudofluitans clumps to move with time is also caused by the effects of competition with species such as Berula erecta and Nasturtium officinale. Sedimentaion may also be another way that water flow rates influence the growth of var. as Ham et al. (1981) found a negative correlation between discharge and sedimentation rates in a chalk stream.

Several studies have been carried out which describe the invertebrate communities associated with var. ; see for example Gunn (1985). The epiphytic bacteria associated with var. have been detailed by Hossell & Baker (1979) and Baker & Orr (1986). Protozoa associated with var. in chalk streams have been described by Baldock et al. (1983).

R. penicillatus subsp. pseudofluitans var. vertumnus C.D.K. Cook

This taxon is separated from var. by having leaves usually shorter than the internodes, 30 - 70 mm long (in summer) and with many (100-400(900)) divergent segments (Webster 1988a). Whereas var. leaves are obconical (Figure 1.2), var. vertumnus leaves are globose. The variety was newly described by Cook (1966a). 'vertumnus'is derived from the Latin vertere (to change). vertumnus was a Roman agricultural god who assumed various disguises and was venerated with the god of the River Tiber, the course of which he was supposed to have altered.

Within the British Isles it only occurs in 41 sites in England and one site in Wales (Biological Records Centre, personal communication 1991). Cook (1966a) states that it requires clear rather than flowing water; 16% of its records are from canals (Webster 1988a). It is not found in rivers subject to frequent flooding (Cook 1966a).

Presumably as a consequence of the rarity of the taxa, little experimental work has been carried out on it. Fox & Murphy (1986) showed that the herbicide diquat alginate was effective in killing back var. vertumnus plants so completely in a small river in northern England that the root stock did not re-grow at all later in the season.

1.1.8 Ranunculus fluitans Lam

R. fluitans is the largest of the Ranunculus species, frequently 2-4 identification clear, no other species has leaves more than 80 mm and £ means floating on water (Gledhill 1989).

R. fluitans is found in rivers throughout much of Britain, but is virtually absent from Ireland and Western Scotland (Biological Records Centre, personal communication 1991; Perring & Walters 1976). It is found in 'rivers with a decided current'(Pearsall 1929), which Butcher (1933) quantified to being between 0.4 and 1.0 ms-1. Cook (1966a, 1967) reports that it is occasionally found in stationary waters, where it rarely flowers. Haslam (1978) found that it appeared to tolerate spates better than consistently fast flows and that deeper water (often > 1.0 m) is preferred. R. fluitans is most often found in the lower reaches of a river (Haslam 1982), for example Holmes & Whitton (1981) only recorded R. fluitans at the most downstream of their sampling sites on the River Tees, and Dethiox (1982) found that R. fluitans grows in broader and deeper rivers than R. penicillatus.

R. fluitans tends to be less common in limestone areas (Cook 1966a), though it does grow in chalk streams (for example the River Rye in Yorkshire, Appendices A & B) and Decamps (1985) found that it had a higher germination rate in water from a river with a high calcium content than from a river with a low calcium content. It has a trophic rank of 45 (Newbold and Palmer 1979).

The nature of the substrate appears to be important in limiting the distribution of R. fluitans. In the River Wye there is a close correlation between the occurrence of R. fluitans and the presence of old fords and collapsed bridges (Brian 1983). Cook (1966a) carried out transplant experiments which indicated that its absence from limestone streams appeared to be due to the lack of stable smooth pebbles on the river bed rather than calcium concentration. This limitation may be due to the fact that R. fluitans (like R. circinatus, see above) has a winter and a summer mode of growth. In the summer R. fluitans has a very rapid growth rate, and is able to flower but does not root, whereas in the winter it is slower growing but is able to root (Cook 1966a), so that a clump will tend to remain in the same place in the river bed for several years (Cook 1966, pers. obsv. 1989-1991 in River Rye, Yorkshire). This growth-form may not be constant for all populations as Whitton & Buckmaster (1970) reported that R. fluitans clumps in the River Wear and the River Tees do change their size, shape and position quite rapidly.

Several studies have been carried out to ascertain the effects of pollution on R. fluitans. Cook (1966a) stated that it was fairly tolerant of pollution, as long as the water remained clear, and attributed the decline in this species in the English Midlands to pollution causing increased turbidity in the water in those sites. Harding (1979, 1980) noted that R. fluitans was replaced by Potamogeton pectinatus in conditions of increased salinity and nutrient supply. Whitton & Buckmaster (1970) observed that although R. fluitans was abundant in the relatively unpolluted River Tees it was much scarcer in the nearby River Wear which had suffered toxic discharges from the coal and coke industries for a number of years. Some R. fluitans was transplanted into the Wear, but few plants grew. However following that survey, the pollution ceased and on re-surveying the river Holmes & Whitton (1977) found that R. fluitans had increased both its range and abundance. Janauer (1981, 1982) has qunatified the uptake and storage of chemical compounds by R. fluitans shoots.

Workers in Continental Europe have found R. fluitans populations there to be rather less susceptible to pollution damage. Ska & Vander Borght (1986) found that there was a positive correlation between increased algal blooms, R. fluitans growth and increased eutrophication in the River Semois. R. fluitans in Germany has been found to be less susceptible to experimentally elevated ammonium concentrations than other species (Von Glänzer et al. (1977). Eichenberger & Weilmann (1982) grew R. fluitans from the River Rhine in artificial rivers. They found that elevating the phosphorous concentration to 300 micro-gPO4-P l-1 had little effect, as did adding domestic sewage from Zurich, but the two added in combination caused a significant increase in growth. It is thought that increased domestic sewage discharges to rivers such as the Rhine may be causing increased growth of R. fluitans at those sites. Monschau-Dusenhausen (1982) found R. fluitans to be associated with polluted sites. The differences observed in the effects of pollution on R. fluitans in Britain and the rest of Europe may either be due to genetically different populations or (probably more likely), the different nature of the pollutants studied.

In the above study Eichenberger and Weilmann (1982) found that herbivory by the crustacean Gammarus sp. had a 'devastating'effect on the R. fluitans grown in their artificial rivers, devouring stout plants in a couple of days. A similar effect has been observed on R. penicillatus subsp. pseudofluitans growing in the artificial rivers described in Chapter 3 (pers. obsv. 1989). Although Eichenberger & Weilmann (1982) concluded that Gammarus may play a role in limiting the distribution of R. fluitans, Harrod (1964) found that Gammarus rarely occurred on R. fluitans, even though it was present on other plants in the same river (the Test, in Hampshire). She found that R. fluitans had a fairly varied flora (which was attributed to the divided leaves), with Simulium ornatum larvae particularly abundant.

R. fluitans often grows in a monoculture (NCC 1988), but it does grow with R. penicillatus - both subsp. penicillatus and and both var. and var. vertumnus (Appendix A & B). Callitriche stagnalis and Potamogeton pectinatus are occasionally associated with R. fluitans (NCC 1988).