UNCORRECTED PROOF

© 2002 Blackwell Science Ltd. http://www.blackwell-science.com/ddi 1

BIODIVERSITY RESEARCH

Paper no: 129 Pages: 18

Diversity and Distributions (2002) 8, 000–000

Blackwell Science Ltd

Taxonomic resolution of the genus bythotrephes Leydig using molecular markers and re-evaluation of its global distribution, with notes on factors affecting dispersal, establishment and abundanceTHOMAS W. THERRIAULT1*, IGOR A. GRIGOROVICH 1, MELANIA E. CRISTESCU 2, HENK A. M. KETELAARS 3 MARKKU VILJANEN 4, DANIEL D. HEATH1 and HUGH J. MACISAAC1 1Great Lakes Institute for Environmental Research, University of Windsor, Windsor, Ontario, Canada, N9B 3P4, *Corresponding author: E-mail: tomt@uwindsor.ca, 2Department of Zoology, University of Guelph, Guelph, Ontario, Canada, N1G 2W1, 3Water Storage Company Brabantse Biesbosch, PO Box 61, NL-4250 Werkendam, The Netherlands, 4Karelian Institute, University of Joensuu, PL 111, 80101 Joensuu, Finland

Abstract. Bythotrephes Leydig is a predatory,onychopod cladoceran native to Eurasia that typic-ally inhabits oligo- and mesotrophic water bodiesof the Palaearctic region. It recently invaded 66North American lakes, prompting a re-evaluationof the taxonomic status, global distribution, anddeterminants of local occurrence and abundance.European studies have reported two distinct species,B. longimanus, which lacks a kink on the caudalprocess, and B. cederstroemi, which possesses one.We employed sequencing of the mitochondrial genecytochrome c oxidase subunit I (COI) in additionto Directed Amplification of Minisatellite-regionDNA (DAMD) using polymerase chain reactions(PCRs) to assay the genetic nature of these ‘species’.These analyses revealed that the two taxa are con-sistent with a single, common species, Bythotrepheslongimanus Leydig 1860, thus resolving the nomen-clatural issue. Furthermore, a common haplotypewas identified between Lake Ontario and LakePuruvesi, Finland, suggesting a potential invasioncorridor via the nearby Baltic Sea. Statistical

analysis revealed that the two forms also occur insimilar habitats in Eurasia and North America.Habitat characteristics of water bodies in NorthernEurope, where both forms are found and occasion-ally co-occur, do not differ statistically. Similarly,no significant differences were detected betweencharacteristics of Eurasian habitats that supportlongimanus ‘forms’ and those of inland lakes inNorth America that support cederstroemi ‘forms’.Human activities have had a strong effect on thedistribution of Bythotrephes in both Europe andNorth America. Global and local distributionshave been affected by ballast water transfer and byboating and fishing activities, respectively. Culturaleutrophication, oligotrophication and acidificationalter the suitability of habitats for fish predators,and may indirectly influence Bythotrephes’ localoccurrence and abundance.

Key words. Biological invasions, distribution,DAMD, COI, DNA markers, non-indigenous species(NIS), fish predation, acidification, eutrophication.

INTRODUCTION

The spiny water flea Bythotrephes Leydig is anonychopod zooplankter whose native range encom-passes the Palaearctic region (Ischreyt, 1930;

Flössner, 1972; Grigorovich et al., 1998; Rivier &Grigorovich, 1999). Bythotrephes typically inhabitsdeep, oligotrophic lakes. It has been described asa ‘glacial relict’ by some but not all researchers(Ischreyt, 1930; Manuilova, 1964; Flössner, 1972).

UNCORRECTED PROOF

2 T. W. Therriault et al.

© 2002 Blackwell Science Ltd, Diversity and Distributions, 8, 000–000

Bythotrephes had a much broader geographicalrange during glacial and postglacial times, althoughsubsequent climatic changes have restricted itsnatural distribution (Rylov, 1921; Manuilova, 1964).Human-mediated activities in recent times havelargely supplanted natural changes in distribution.For example, Grigorovich et al. (1998) and Rivier& Grigorovich (1999) emphasized the importanceof human-mediated dispersal in shaping thecurrent range of this genus in the Commonwealthof Independent States (CIS; former USSR). Recentwork in North America also has implicated humandispersal vectors to inland lakes from the GreatLakes (Borbely, 2001).

Bythotrephes is an important component ofmany aquatic food webs. It almost exclusivelyfeeds on small-bodied zooplankters. Also, it feedson phytoplankton including diatoms, green algaeand peridineans (Flössner, 2000). Bythotrephespredation can prevent mid-summer increases of someherbivorous, crustacean zooplankton (Patalas, 1954;Bosselmann, 1974; Bürgi & Egli, 1984; Ketelaars &van Breemen, 1993; Yan & Pawson, 1997; Hoffmanet al., 2001). In turn, fish predation may affectBythotrephes’ abundance and reproductive pheno-logy, particularly in lakes where populationsexhibit vertical partitioning (e.g. Plankensteiner,1992; Yan et al., 2001). In shallow lakes lackingfish, Bythotrephes can reach maximum densitiesof 125 individuals m–3 (Vekhov, 1987), althoughdensities are typically much lower in systems withfish predators (Svärdson, 1976; Vekhov, 1982).Also, fish predation may influence and limitmorphological variation within populations ofBythotrephes. For example, variation in the abund-ance of different forms of B. longimanus in Russianwater bodies has been attributed to the presenceof fish that selectively feed on Bythotrephes,sometimes eliminating it from zooplanktoncommunities (Vekhov, 1987).

Bythotrephes was first observed in North Americain Lake Ontario in 1982 (Johannsson et al., 1991).This invasion is likely to have resulted fromballast water discharge by a transoceanic ship, theorigin of which was likely to be the Baltic Seanear the port of St Petersburg, Russia (Spruleset al., 1990; Berg et al., 2001). It has since spreadto all of the other Great Lakes and to at least 64inland water bodies in Ontario, Michigan, Ohioand Minnesota (Sprules et al., 1990; Johannssonet al., 1991; Yan et al., 1992; MacIsaac et al., 2000;

Borbely, 2001; Appendix 1). Bythotrephes also hasan extensive invasion history in western and easternEurope (Ketelaars & van Breemen, 1993; Ketelaars& Gille, 1994; Grigorovich et al., 1998).

Relatively few studies have used abiotic orbiotic factors to assess large-scale distributionalpatterns for Bythotrephes. Grigorovich et al. (1998)and Rivier & Grigorovich (1999) reported factorsthat influenced Bythotrephes in the CIS, whileMacIsaac et al. (2000) utilized physical, chemical,and biological parameters of European habitatsthat either supported or lacked Bythotrephes toforecast its distribution in Ontario. However, itremains unclear whether dispersal limitation orregional and local factors, including limnologicalvariables or fish predation, is more important indetermining its distribution at multiple spatialscales.

One problem in the determination of the globaldistribution of Bythotrephes is uncertainty regard-ing the number of species contained within thegenus. Researchers in Europe have reported at leasttwo morphologically distinct species, B. longimanusand B. cederstroemi (Mordukhai-Boltovskoi &Rivier, 1987; Grigorovich et al., 1998). The taxo-nomic status of these ‘forms’ is confounded byvariability in anatomical structures and body dimen-sions (Mordukhai-Boltovskaya, 1959; Zozulya &Mordukhai-Boltovskoi, 1975, 1977; Zozulya, 1977;Mordukhai-Boltovskoi & Rivier, 1987; Evans,1988; reviewed in Grigorovich et al., 1998). Berg& Garton (1994) and Berg et al. (2001) notedthat allozyme variation among sympatric popu-lations of B. longimanus and B. cederstroemi wasoften less pronounced than that among allopatricpopulations of either species, suggesting the exist-ence of only a single genetic species, B. longimanus.Also, Weider (1991) found low genetic variabilityamong Great Lakes’ populations of B. cederstroemiusing allozyme analyses. However, allozyme studiesare based on protein sequences, and owing tomultiple coding patterns, species-based variationin DNA sequences might not be detected. Thislimitation can be overcome through use of DirectedAmplification of Minisatellite-region DNA (DAMD)analysis. This approach uses a variable number oftandem repeats (VNTRs) as primers to amplifyfragments of genomic DNA that show little vari-ation within a species but considerable variationbetween species (e.g. Heath et al., 1993). Recently,Cristescu et al. (2001) used mitochondrial gene

UNCORRECTED PROOF

Bythotrephes taxonomy and distribution 3

© 2002 Blackwell Science Ltd, Diversity and Distributions, 8, 000–000

sequencing to determine divergence between theconfamiliar genera Bythotrephes and Cercopagis.This approach also is suitable for quantifyingthe divergence, if any, between longimanus andcederstroemi ‘forms’.

Records of Bythotrephes are scattered andincomplete for regions of continental Asia, Europeand the British Isles. In this study, we report anupdated distribution of Bythotrephes and providepossible mechanisms to explain observed distribu-tional patterns. We also assess relationships amongabiotic variables and distributions of longimanusand cederstroemi ‘forms’ in regions of Europewhere they co-occur, and between European andNorth American populations. Furthermore, weexplore the potential influences of biotic vari-ables on the distribution of Bythotrephes. In orderto determine whether morphological variationwithin the genus is genetically based, we employboth mitochondrial gene sequencing and DAMDmethodology to assess populations of longimanusand cederstroemi.

METHODS AND MATERIALS

Geographical and physicochemical data base

A data base was created from an extensive literaturesearch for reported occurrences of Bythotrephesin Eurasia. Whenever possible, information wasobtained for lake name, country of location, lakearea and volume. A second search was con-ducted using lake name and country of locationto collect basic physicochemical data includingtemperature, dissolved oxygen and nutrient (N, P)concentrations, and lake trophic status if thesedata were not available in the literature that reportedBythotrephes occurrence. Lake name, location andcitation data are provided in Appendix 1.

In order to test whether longimanus and ceder-stroemi ‘forms’ inhabit different types of waterbodies, we tested abiotic conditions of lakes, clas-sified by the presence of Bythotrephes, using non-parametric tests including Kruskal–Wallis

and Median test. This method was employed toensure that outlier values inherent in a data setcollected from multiple sources and times did notunduly influence statistical outcomes. Nonpara-metric tests are less sensitive to the impact of singleoutlier values than are comparable parametrictests (Zar, 1984). Lakes reported to contain

different Bythotrephes subspecies were classifiedat the species level.

Collection, identification and morphology of bythotrephes

Bythotrephes were collected within its native rangefrom Lake Puruvesi, Finland, by vertical haulsof a 450-µm plankton net on 2 August 2000 and2 August 2001. Additional collections were madefrom locations outside Bythotrephes’ native range,including the lower Volga River, Russia (27 July2000), Lake Tyson, Canada (25 July 2000), andthe St Clair River, Canada (September 1998).Specimens used for genetic sequencing werecollected from Finland (Lake Puruvesi), Canada(Lakes Ontario and Simcoe), the Netherlands(Petrusplaat near Dordrecht), Germany (SelenterSee and Post See), and Russia (Volgograd Reser-voir). Collections were made using standardplankton nets (253–450 µm mesh) and preservedin 95% ethanol to preserve DNA for geneticanalyses. Individual Bythotrephes were separatedfrom other plankton using light microscopy.

Measurements of several historically employedstructures were made for taxonomic identifica-tions (e.g. caudal process length, distance betweenarticular spines II and III, length of articularspine III, ratio of the distance between articularspines II and III to abdominal length, and ratioof articular spine III length to the diameter ofthe caudal process) following Grigorovich et al.(2000).

Traditional taxonomic classification of Bytho-trephes depends on a series of instar- and sex-specific characteristics (Manuilova, 1964; Zozulya& Mordukhai-Boltovskoi, 1975; Mordukhai-Boltovskoi & Rivier, 1987). Because instar IIIparthenogenetic females dominated all samples,we limited our analyses to these individuals.Morphology of the caudal process is the mostprominent distinguishing factor in the genus(Martin & Cash-Clark, 1995). However, the taxo-nomic utility of this structure is limited due toextensive ontogenetic, seasonal and geographicalvariability (Mordukhai-Boltovskaya, 1959; Zozulya,1977; Mordukhai-Boltovskoi & Rivier, 1987;Yurista, 1992; Burkhardt, 1994; Ketelaars et al.,1995). We used the ratio of the distance betweenarticular spines II and III to abdominal length,and the ratio of the length of articular spine III

UNCORRECTED PROOF

4 T. W. Therriault et al.

© 2002 Blackwell Science Ltd, Diversity and Distributions, 8, 000–000

to the diameter of the caudal process, to furthercharacterize populations. Morphological differ-ences between longimanus and cederstroemi ‘forms’were assessed using either or .

The ratio of the length of thoracopod I tototal body length, also used previously in assess-ments of Bythotrephes taxonomy (Manuilova,1964), was variable for instar III parthenogeneticfemales within each population and therefore didnot permit discrimination between ‘forms’ (F1,14 =0.01, P = 0.923). An additional index, consistingof the ratio of the caudal process length to totalbody length, also was incapable of discriminatingbetween ‘forms’ (F1,14 = 0.26, P = 0.619).

Based upon traditional taxonomic charac-teristics, we identified the ‘forms’ B. longimanusLeydig 1860, B. cederstroemi Schoedler 1863, andB. cederstroemi connectens Lilljeborg 1900 (Fig. 1).The cederstroemi ‘form’ was easily distinguishedfrom the longimanus ‘form’ owing to its posses-sion of a longer caudal process with a mid-lengthkink (Fig. 1A,C). Martin & Cash-Clark (1995)recognized a transitional form, B. cederstroemiconnectens, which possesses a variably reducedkink (Fig. 1B). Variability in the shape (bent tostraight) and amount of serration (two fields ofanteriorly pointed spines to individual posteriorlydirected spine-like scales) in the kink on the

Fig. 1 Instar III parthenogenetic females of (A) Bythotrephes longimanus from Lake Puruvesi, Finland,(B) Bythotrephes cederstroemi connectens from the Volga River, Ukraine, and (C) Bythotrephes cederstroemifrom Lake Puruvesi, Finland. Within this genus, local ‘varieties’ have been described (see Rivier, 1998),providing further evidence of polymorphism within Bythotrephes.

UNCORRECTED PROOF

Bythotrephes taxonomy and distribution 5

© 2002 Blackwell Science Ltd, Diversity and Distributions, 8, 000–000

caudal process of B. cederstroemi connectens hasbeen well documented (Mordukhai-Boltovskaya,1959).

Differences in morphological features wereused to confirm the identity of longimanus andcederstroemi ‘forms’. The cederstroemi ‘form’,including B. cederstroemi connectens, differedsignificantly from the longimanus ‘form’ withrespect to caudal process length (F1,14 = 18.23,P = 0.001), distance between articular spines IIand III (F1,10 = 8.81, P = 0.014), and length of arti-cular spine III (F1,10 = 9.41, P = 0.012); in all cases,these features were longer in the cederstroemi ‘form’(caudal process length: 4.14 mm vs. 6.90 mm;distance between articular spines II and III: 0.257 mmvs. 0.916 mm; and length of articular spine III:0.086 mm vs. 0.394 mm). These results did notchange when abdominal length or diameter of thecaudal process were used as covariates (,P < 0.02). Finally, the articular spines on thecaudal process were typically bent anteriorly in thecederstroemi ‘form’ but were almost straight orbent posteriorly in the longimanus ‘form’ (Fig. 1).

Genetic procedures

DNA was extracted from whole individuals ofBythotrephes using a modified salt–precipitationprotocol (Devlin et al., 1991) or a Wizard GenomicDNA Purification Kit (Promega) based on sim-ilar principles. DNA also was extracted fromCercopagis pengoi — another invader of the GreatLakes — collected from Lake Ontario, and usedas an out-group for genetic analyses. Extractedgenomic DNA was used as template DNA in mini-satellite repeat primed polymerase chain reac-tions (PCR), as described in Heath et al. (1993).Each reaction used one synthetic oligonucleotidedesigned from minisatellite repeat core sequence(see Heath et al., 1993) as a primer (NVB-3 = 5′-GGT GAA GCA NAG GTG-3′; or YNZ-22 =5′-CTC TGG GTG TCG TGC-3′). The balance ofthe PCR was double-distilled water, manufacture-supplied PCR buffer, MgCl2, dNTPs, and TaqDNA polymerase (Heath et al., 1993). Reactionswere run using a touchdown approach with thefirst 5 cycles decreasing 1 °C per cycle from a51 °C annealing temperature followed by 35 cyclesat a 46 °C annealing temperature (60 s). We alsoused a 72 °C extension cycle (90 s) and a 94 °Cdenaturing cycle (60 s) in a PTC-225 Programmable

Thermal Controller (MJ Research, Inc.). PCR pro-ducts were run on 2% agarose gels for 2–4 h at80–100 V and visualized using UV-transilluminationof ethidium bromide-stained gels. Photographs werestored as digitized images for subsequent analyses.Based on the total number of bands analysed(shared or not shared among ‘species’) a cluster-ing tree was generated using TFPGA software(Miller, 1997). We used Unweighted Pair GroupMethod based on Arithmetic Means ()clustering to measure the similarity betweenlongimanus and cederstroemi ‘forms’, with Cerco-pagis pengoi used as an out-group in the analyses.

Phylogenies also were constructed using themitochondrial gene cytochrome c oxidase subunitI (COI). We used standard DNA amplificationand sequencing procedures provided in Cristescuet al. (2001) using the primer pair LCO1490 andHCO2198 (Folmer et al., 1994) to amplify a 658base pair fragment of the COI gene. Bi-directionalsequencing was used only when ambiguous siteswere encountered. Sequences were aligned usingSequencher (Gene Codes Corporation).

was used to determine phylogenetic relationshipsamong ‘forms’. All sequences obtained in thisstudy have been submitted to GenBank. Accessionnumbers for Cercopagis pengoi are AF 320013(Black Sea) and AF 320014 (Caspian Sea).Accession numbers for Bythotrephes are AF435122 (Lake Simcoe, Lake Ontario, and LakePuruvesi), AF 435123 (Lake Puruvesi), AF 435124(Post See), AF 435125 (Post See), AF 435126(Selenter See), AF 435127 (Selenter See), AF435128 (Petrusplaat/Dordrecht), AF 435129 (Vol-gograd Reservoir), AF 435130 (Volgograd Reser-voir), and AF 435131 (Volgograd Reservoir).

RESULTS

Geographical distribution of Bythotrephes

The contemporary global distribution of Bytho-trephes extends throughout the Holarctic regionand includes the British Isles, mainland andnorthern Europe, Asia, and the Great Lakesregion of North America. This distribution includesregions to which it is native and others to whichit has been introduced. Recent invasions have beenwell documented, particularly for North America(MacIsaac et al., 2000; Berg et al., 2001), Belgiumand the Netherlands (Ketelaars & van Breemen,

UNCORRECTED PROOF

6 T. W. Therriault et al.

© 2002 Blackwell Science Ltd, Diversity and Distributions, 8, 000–000

1993; Ketelaars & Gille, 1994). However, availablehistorical information from Eurasia is insufficientto discriminate whether these populations arenative or introduced.

Bythotrephes has an extensive distribution innorthern Europe (Fig. 2). In Ireland, it has beenobserved in at least 21 water bodies (Appendix 1),including Lough Neagh (Northern Ireland) thelargest body of freshwater in the British Isles. Also,B. longimanus is a common component of mostplankton communities throughout the EnglishLake District (Goulden, 1964; Harmsworth, 1967;Macan, 1970). In Norway, B. longimanus has beenreported primarily in the south-western part ofthe country. Six of 20 identified water bodies thatcurrently support Bythotrephes are impoundments,suggesting that human activities have createdsuitable environments for the survival of thisspecies. Additional evidence supports the contentionthat human activities influence its occurrence inthis region. For example, Lake Mjøsa, recentlyreclaimed from a hypereutrophic state, nowsupports Bythotrephes (Næsje et al., 1987; Rognerud& Kjellberg, 1990). Likewise, it now occurs inLille Lungegårdsvann, a reclaimed lagoon that hasexperienced significant limnological change (Lysebo,1995). In Sweden, B. longimanus has been reportedfrom many water bodies in the south-eastern partof the country, as well as from those along thewestern border with Norway. Both ‘forms’ have

been reported to co-occur only in Lake Vänern, thecountry’s largest lake (Svärdson, 1976; Nilsson,1979). Both forms are, however, widely reportedfrom lakes in Finland (J. Sarvala, pers. comm.)and most lakes reported to support B. cederstroemialso contain B. longimanus.

The distribution of Bythotrephes in continentalEurope shows a distinct pattern of zonation, withmajor belts of occurrence in northern and southernregions of the continent (Fig. 2). For example,B. longimanus is found commonly in the lakedistricts of northern Poland and northern Germany,as well as in Denmark, the Netherlands andBelgium (Appendix 1). In both of the latter coun-ties, it is a non-indigenous species (NIS) that likelywas transported from north-eastern locations viacommercial ships (Ketelaars & van Breemen, 1993;Ketelaars & Gille, 1994). In southern Europe,B. longimanus occurs in lakes throughout thefoothills of the Alps in northern Italy, AustriAand Switzerland (Appendix 1). It also has beenreported farther south and east in Caucasianalpine lakes of Georgia (Kasymov, 1972), Armenia(Behning, 1941; Kasymov, 1972) and Turkey(Mordukhai-Boltovskoi & Rivier, 1987).

The Eurasian distribution of Bythotrephes isextensive but patchy, ranging throughout coun-tries of the former Soviet Union and into China(Fig. 2). The greatest concentration of waterbodies populated by Bythotrephes is in north-

Fig. 2 Biogeographical distribution of Bythotrephes in Eurasia.

UNCORRECTED PROOF

Bythotrephes taxonomy and distribution 7

© 2002 Blackwell Science Ltd, Diversity and Distributions, 8, 000–000

western Russia, extending from the Finnish borderto the Arctic. Both ‘forms’ have been reportedfrom all longitudes of Russia, including the AnadyrRiver drainage on the Pacific coast. Bythotrephesalso occurs in shallow ponds in the Arctic region(Vekhov, 1981, 1982, 1987). In eastern Europe, itsoccurrence is associated with large river systems(e.g. Dnieper, Don and Volga Rivers) that directlyor indirectly connect to the Caspian or Black-Azov Seas. In cool-temperate and subarctic Europe,Bythotrephes is located either in the north andassociated with large rivers flowing from thesouth (e.g. Pechora) or with waterbodies locatedbetween large river drainages (e.g. Ob and Yenisei).Bythotrephes has a patchy north–south distribu-tion in Kazakhstan and south-western Siberia(western Asia) but is frequently associated withriver drainages (Tseeb, 1940; Zhadin & Gerd, 1961;Lezin, 1982).

In Eurasia, Bythotrephes has been reportedfrom 40°N to 70°N. Its current North Americandistribution is considerably more confined, rang-ing from 40°N to 49°N. It is unlikely that itsnorthern distribution in North America willincrease appreciably, as human dispersal vectorsoperate primarily along east–west rather thannorth–south axes (Borbely, 2001). Natural dis-persal vectors such as migrating waterfowl couldoperate on a north–south axis but are not believedto be as important as human dispersal vectors indetermining range expansion for Bythotrephes. Therate of new invasions in Ontario is accelerating,a consequence of local diffusive dispersal and long-distance ‘jump’ dispersal (Borbely, 2001). Recentreports of Bythotrephes invasions in Belgium andthe Netherlands indicate that range expansion isnot confined to North America (e.g. Ketelaars &Gille, 1994).

Bythotrephes’ occurrence in relation to physicochemical properties

Abiotic conditions of waterbodies that supportB. longimanus and B. cederstroemi were exploredto determine whether these ‘forms’ occur in dif-ferent types of habitats. Because the former ‘form’has been reported from a much larger geographicalarea than the latter (Fig. 2), we excluded Euro-pean waterbodies from countries where only one‘form’ — typically longimanus — has been reported(i.e. we considered only lakes in regions where

‘forms’ co-occur). Included in our analysis werewaterbodies from Estonia, Finland, Kazakhstan,Russia, Sweden and the Ukraine. No significantdifferences were found between lake types sup-porting the two ‘forms’ with respect to lake area(÷2 = 1.10; d.f. = 1; P = 0.294), volume (÷2 = 0.14;d.f. = 1; P = 0.707), or maximum depth (÷2 = 0.001;d.f. = 1; P = 0.925). In addition, there were nosignificant differences in Secchi depth transparency(÷2 = 0.96; d.f. = 1; P = 0.329) or pH (÷2 = 1.39;d.f. = 1; P = 0.238) between waterbodies thatsupported longimanus and cederstroemi ‘forms’.However, mean depth differed significantly (÷2 =4.17; d.f. = 1; P = 0.041), with the cederstroemi‘form’ occurring in shallower waterbodies thanthe longimanus ‘form’.

As the previous analyses did not include water-bodies that supported both ‘forms’, the analyseswere repeated including these systems. In thiscase, significant differences were noted betweenwaterbodies supporting one ‘form’ or both ‘forms’only with respect to lake volume (÷2 = 9.03; d.f. = 2;P = 0.011). Lakes in which the cederstroemi ‘form’occurred alone tended to have greater volumesthan those in which longimanus occurred aloneor those in which the ‘forms’ were sympatric.

It is possible that several variables consideredtogether can resolve differences between habitatsutilized by longimanus or cederstroemi ‘forms’. Adiscriminant function analysis was developed usinginformation on lake area, volume, maximumdepth and mean depth to elucidate differencesbetween two co-occurring ‘forms’ in Europe. Therewas no significant difference between waterbodiesthat supported the two ‘forms’ (Wilks’ Lambda =0.88, F4,46 = 1.61, P = 0.188).

Only the cederstroemi ‘form’ is consistentlyreported in North America (Sprules et al., 1990;Johannsson et al., 1991; Yan et al., 1992; MacIsaacet al., 2000; Berg et al., 2001). Because physico-chemical conditions could influence which ‘form’occurs within a geographical region, we com-pared characteristics of European waterbodies inAustria, Belgium, Britain, Byelorussia, Denmark,Germany, Italy, Latvia, Lithuania, Netherlands,Norway, Poland and Switzerland that supportonly the longimanus ‘form’ with North Americansystems (Canada, United States) reported to sup-port the cederstroemi ‘form’. Significant differenceswere detected between regions with respect tolake area (÷2 = 5.64; d.f. = 1; P = 0.018), volume

1

UNCORRECTED PROOF

8 T. W. Therriault et al.

© 2002 Blackwell Science Ltd, Diversity and Distributions, 8, 000–000

(÷2 = 4.57; d.f. = 1; P = 0.032) and summer bottomwater temperature (÷2 = 7.43; d.f. = 1; P = 0.006).However, no differences were detected betweenEuropean and North American waterbodies formaximum depth (÷2 = 3.20; d.f. = 1; P = 0.074),mean depth (÷2 = 1.30; d.f. = 1; P = 0.255), Secchidisk depth (÷2 = 1.02; d.f. = 1; P = 0.312) or pH(÷2 = 0.01; d.f. = 1; P = 0.918). When each of thefive Great Lakes were removed from the analyses,no significant differences were detected betweenEuropean and North American waterbodies forlake area (÷2 = 1.93; d.f. = 1; P = 0.165), volume(÷2 = 2.40; d.f. = 1; P = 0.122) or summer bottomwater temperature (÷2 = 2.31; d.f. = 1; P = 0.129).

Bythotrephes is found in a wide variety of hab-itats in both Eurasia and North America (Fig. 3).In Eurasia, it is most common in lakes withsmall to intermediate surface areas, larger volume,intermediate mean and maximum depths, and

shallower Secchi depths (Fig. 3A). A similar trendwas apparent in North America (Fig. 3B). However,there are noticeable differences in the relativenumber of habitats that have larger lake surfacearea and larger volume, a clear influence of the GreatLakes in the North American analysis (Fig. 3B).

Genetic analysis of Bythotrephes

Sequencing of the mitochondrial gene COI showedthat the longimanus and cederstroemi ‘forms’represent a single species, B. longimanus. Thedivergence between allopatric populations oflongimanus and cederstroemi ‘forms’ was lessthan 0.6% while divergence between Bythotrephesand the confamiliar Cercopagis pengoi was 26.6%(Cristescu et al., 2001). Furthermore, longimanusand cederstroemi ‘forms’ from the VolgogradReservoir clustered at the same branch of the

Fig. 3 Distribution of environmental variables for waterbodies supporting Bythotrephes populations in(A) Eurasia and (B) North America.

UNCORRECTED PROOF

Bythotrephes taxonomy and distribution 9

© 2002 Blackwell Science Ltd, Diversity and Distributions, 8, 000–000

COI generated tree (Fig. 4). Detailed relationshipsamong longimanus, cederstroemi, and C. pengoi,are shown in Fig. 4. The observed DAMD band-ing patterns indicate that the longimanus andcederstroemi ‘forms’ belong to a single species,B. longimanus. Using Nei’s (1972) distance basedon 14 DAMD generated bands, the distancebetween longimanus and cederstroemi ‘forms’ was0.000 while that between Bythotrephes and C. pengoiwas 0.442. An example of an observed DAMDbanding pattern is shown in Fig. 5 (YNZ-22).

DISCUSSION

Clear differences exist with respect to morphologicalcharacteristics of longimanus and cederstroemi‘forms’. In addition, Bythotrephes exhibits mor-phological variability across both spatial andtemporal gradients, variability that has confoundedspecies identity in the genus. However, the geneticanalyses reported here indicate that different ‘forms’represent a common species, and, according totaxonomic nomenclature rules, the ancestral nameof Bythotrephes longimanus Leydig, 1860 should

apply. Our findings are consistent with a recentallozyme study that determined allopatric vari-ation within each ‘form’ exceeded that of sym-patric populations of the two ‘forms’ together(Berg et al., 2001). Moreover, breeding and hatchingstudies conducted in Europe and North Americahave produced either instar-based ‘forms’ orsome progeny of the other ‘form’ (Mordukhai-Boltovskaya, 1959; Zozulya, 1977; Yurista, 1992).Interestingly, a common haplotype is found inLake Ontario and Lake Puruvesi, Finland, pro-viding additional evidence that Bythotrephesintroductions to the Great Lakes originated froma population in or adjacent to the Baltic Sea.Previous studies suggested the port of St Petersburg,Russia, as the source of the original Great Lakespopulation (Sprules et al., 1990; Berg et al., 2001).Also of note, sympatric populations of longimanusand cederstroemi ‘forms’ exhibited no differenti-ation (Volgograd Reservoir, Fig. 4), though greatervariability was evident in populations from northernEurope and Russia (Fig. 4).

Yurista (1992) demonstrated that morphologicaldifferences exist among Bythotrephes individuals

Fig. 4 tree based on sequencing of the mitochondrial COI gene showing the divergence between thetwo ‘forms’ of Bythotrephes (longimanus and cederstroemi) and Cercopagis pengoi, used as an out-group, basedon p-distance.

UNCORRECTED PROOF

10 T. W. Therriault et al.

© 2002 Blackwell Science Ltd, Diversity and Distributions, 8, 000–000

hatched from resting stages and those producedby parthenogenetic reproduction. Likewise, studiesof Cercopagis revealed that ‘forms’ assignableto different morphological subgenera do notdiffer genetically, but rather appear to be the prod-uct of sexual vs. parthenogenetic reproduction(Makarewicz et al., 2001). Seasonality also mightbe important, as Evans (1988) and Rivier &Grigorovich (1999) demonstrated that the longi-manus ‘form’ was found in greater abundancesduring cold-water months (i.e. during spring andautumn), while the cederstroemi ‘form’ domin-ated during warm-water periods. Consequently,morphological variation observed within the genus

is neither unique nor temporally stable. However,the mechanism responsible for production ofdifferent ‘forms’ has yet to be deciphered.

Characterization of waterbodies supporting thelongimanus and cederstroemi ‘forms’ indicated asignificant difference among lakes only with respectto mean depth. The lack of differentiation amongthese basins provides further support for thehypothesis that the two ‘forms’ represent onespecies. To the extent that variation was revealed,the cederstroemi ‘form’ was associated withshallower waterbodies (median = 6.5 m) than thelongimanus ‘form’ (median = 6.9 m). However, itis unclear whether a difference of this magnitudecould contribute to distributional differences amongthese ‘forms’. It is more probable that other vari-ables have a greater influence on ‘form’ distributions(e.g. primary production or predation pressure).

Few differences in physicochemical conditionswere observed between Eurasian locations con-taining the longimanus ‘form’ and North Americanlocations containing the cederstroemi ‘form’. Thisfinding also provides corroborating evidence thatthe two ‘forms’ are variants of the same species.Furthermore, this finding suggests that abioticconditions are unlikely to limit colonization byBythotrephes in North American waterbodies ifappropriate dispersal vectors exist. This finding isconsistent with MacIsaac et al. (2000), who sug-gested that many inland lakes in Ontario couldsupport Bythotrephes but currently do not owingto lack of dispersal vectors. However, once Bytho-trephes has colonized a system, biotic variablesincluding predation could limit its establishmentor population size. To date, no study has success-fully separated the effects of dispersal limitationand abiotic or biotic limitation on the establish-ment of new populations of this invasive species.

Morphological variability observed among Bytho-trephes populations also could be due, in part, tocyclomorphosis. Cyclomorphosis is a defensivemechanism that can result in dramatic morpho-logical changes in cladocerans, including theformation of neckteeth, tailspines, or changesin head shape and length. Cladoceran cyclo-morphosis results from abiotic (e.g. watertemperature, turbulence and light) or biotic (e.g.presence of predators) factors (e.g. Hebert &Grewe, 1985; Straile & Hälbich, 2000). Forexample, the phantom midge Chaoborus releaseskairomone chemicals into the water that trigger

100

200

300

400

500

650

850

1000

1650

2000

B.c.B.c.B.l.B.l.

Fig. 5 Observed banding patterns for both Bytho-trephes ‘forms’ using the DAMD primer YNZ-22 forlongimanus ‘form’ (B.l .) and the cederstroemi ‘form’(B.c.). A 1Kb + ladder also is shown. Individuals werecollected from Lake Puruvesi, Finland in August 2001.

UNCORRECTED PROOF

Bythotrephes taxonomy and distribution 11

© 2002 Blackwell Science Ltd, Diversity and Distributions, 8, 000–000

cyclomorphosis in Daphnia ambigua (Hebert &Grewe, 1985). It is unclear whether fish predatorscan trigger cyclomorphosis in Bythotrephes,although they clearly influence body size in someEuropean systems (H. Ketelaars, pers. obs.).

Bythotrephes in the Great Lakes demonstrateconsiderable variation in morphology. For example,specimens collected from Lake Superior exhi-bited small body size and lack of a caudal kink(S.T. Jarnagin, pers. comm.), and resembled thelongimanus ‘form’ evenmthough only the ceder-stroemi ‘form’ is consistently reported in NorthAmerica. Sullivan & Lehman (1998) also reportedthat cederstroemi-type individuals from Lakes Erie,Huron and Michigan displayed considerable dif-ferences with respect to body allometry. However,the mechanism(s) responsible is not understood.

Various mechanisms exist to explain the large-scale, global distribution of Bythotrephes. As withmany cladoceran taxa, Bythotrephes can expandits range via natural dispersal of resting eggs bybirds (Charalambidou & Ketelaars, unpubl.), fish(Jarnagin et al., 2000) or by passive advection todownstream locations (Herzig, 1985a,b). Theseresting eggs, stored in sediment egg banks, alsoallow the continued persistence of Bythotrephes inhabitats under high predation pressure. Inaddition,human vectors are responsible for the increaseddistribution of Bythotrephes, including via ships’ballast water, canals or contaminated fishing gear.It now appears that natural dispersal is beingdramatically supplanted by human-mediated dis-persal. For example, human activities have increasedthe rate of spread and the geographical range ofBythotrephes faster during the latter half of thetwentieth century than at any time previously (seeMacisaac et al., 2000). Numerous mechanismsdetermine Bythotrephes’ distribution at smaller spa-tial scales including individual lakes and reservoirs.These factors include fish predation, eutrophica-tion, acidification and reservoir impoundment.

Fish predation effects

Fish predation is one mechanism that may clearlyregulate Bythotrephes abundance. Planktivorousfishes selectively consume Bythotrephes, espe-cially during summer months when it is mostabundant (e.g. Lindström, 1955; Nilsson, 1955;Berg & Grimaldi, 1966; Langeland, 1978; Millset al., 1992; Coulas et al., 1998). For example,

Coregonus spp. in Lago Maggiore, Italy had a higherpreference for B. longimanus than for any othercrustacean plankton (de Bernardi & Giussani,1975; de Bernardi et al., 1987). Bythotrephes cancompose > 90% of total gut contents at times ofhigh predation pressure (Nilsson, 1961, 1965;Fitzmaurice, 1979; Coulas et al., 1998).

A large body of literature has developed illus-trating that intense fish predation can be a potentmechanism limiting local occurrence or abund-ance of Bythotrephes (e.g. Huitfeldt-Kaas, 1935;Berg, 1951; Patalas, 1954; Stenson, 1978; Vekhov,1982). For example, Stenson (1972) and Langeland(1972) showed that B. longimanus was eliminatedor abundances reduced to non-detectable levelsfrom lakes with high fish predation intensity.Svärdson (1976) reported Bythotrephes in 37 of 41Swedish lakes without fish but in only four of eightlakes with fish. Langeland (1972, 1978) reportedthat B. longimanus had disappeared from threeNorwegian lakes between 1896 and 1971, likely dueto predation by Arctic char (Salvelinus alpinus).

It is important to note that resting eggs mayremain in lake sediments for some years, therebyallowing re-establishment of Bythotrephes follow-ing relaxation of predation pressure or changesin limnological conditions (e.g. recovery fromeutrophication or acid stress; see below). Never-theless, if Bythotrephes is found in fish gut con-tents but is not detectable in water columnsamples, it is likely to be experiencing intensepredation pressure (Stenson, 1978), the magni-tude of which might limit its ability to colonizeor persist in novel habitats. As evidence, Yan &Pawson (1997) and Yan et al. (2001) proposed thatBythotrephes gains refuge from planktivorous fishby living in regions of the water column notexploited by fishes. Yan et al. (2001) argued thatif the hypolimnion is deeper than twice the Secchidisk depth, a warm, dark layer exists whereinhypolimnetic, stenothermal planktivorous fishcould not see or prey on overlying Bythotrephes.Conversely, if Secchi disk depth exceeds thermo-cline depth, fish can see and exploit overlyingBythotrephes (Yan et al., 2001).

The caudal process affords some protectionagainst gape-limited planktivorous fish, but it isless likely to be effective against larger fishes. Ingeneral, larger-bodied planktivores can fully exploitBythotrephes, while smaller individuals may havetrouble handling or swallowing this species. Some

UNCORRECTED PROOF

12 T. W. Therriault et al.

© 2002 Blackwell Science Ltd, Diversity and Distributions, 8, 000–000

smaller fishes can exploit Bythotrephes, althoughpreferential selection is best described for old,large age classes (Klemetsen, 1973; Kogan, 1974;de Bernardi et al., 1987). Bythotrephes may respondto planktivory through adjustments in clutchsize, producing fewer, larger offspring, and byproducing longer caudal barb pairs; these defencesare apparently most effective against gape-limitedplanktivorous fishes (Straile & Hälbich, 2000).Thus, strong evidence exists to link fish predationto regulation of occurrence, abundance, life historycharacteristics and body allometry of Bythotrephes.

Eutrophication and acidification effects

Both cultural eutrophication and anthropogenicacidification operate on regional to local spatialscales. Lake eutrophication strongly affects zoo-plankton assemblages, including adverse effectson abundance of Bythotrephes in most instances.For example, Bythotrephes disappeared fromthe zooplankton community of Lake Ringsjön,Sweden sometime between 1883 and 1972–83 owingto eutrophication (Bergstrand, 1990) as well asfrom Lake Võrtsjärv, Estonia (Haberman, 1998).Bythotrephes longimanus is a relatively rare andunimportant member of the plankton communityin Lake Tyrifjord, Norway, a system that hasbeen affected by cultural eutrophication since the1930s (Langeland, 1974). This pattern has beenrepeated in many other systems across Europe(Goulden, 1964; Harmsworth, 1967; Smyly, 1972;Nilsson & Pejler, 1973; Jeppesen et al., 1996).

The effects of eutrophication on Bythotrephesabundance are, however, reversible. In Lake Mondsee,Austria, abundance of B. longimanus decreasedslightly during a period of rapid eutrophication,but later increased following implementation ofpollution control measures (Nauwerck, 1991).These effects were even more pronounced inLago Maggiore, Italy, where Bythotrephes declinedin abundance as the lake eutrophied, but sub-sequently recovered as phosphorus abatementwas implemented and the lake experienced oligo-trophication (Manca & Ruggiu, 1998). Virtuallythe same pattern was observed in Lake Lucerne,Switzerland as it experienced eutrophication andthen oligotrophication (Mookerji et al., 1998). InKortowskie Lake, Poland, an intensive restora-tion program resulted in improvement in waterquality and establishment of B. longimanus (Widuto,

1977). These observations also are consistent withsynoptic surveys conducted throughout the Maz-urian lake district of Poland, where Bythotrephesoccurs most commonly in less eutrophic water-bodies (Gieysztor, 1959; Patalas & Patalas, 1966).

Available evidence indicates that lake eutro-phication and the intensity of fish planktivory onBythotrephes are inextricably linked (see MacIsaacet al., 2000). Consequently, the disappearance ofBythotrephes from eutrophic lakes may be relatedmore to planktivory than to intolerance of physicalor chemical conditions. The best available evid-ence to explore this concept may be derived fromJeppesen et al.’s (1996) study of 30 Danish lakes.In these systems, planktivore biomass was posit-ively correlated with lake trophic condition, whilebody size of zooplankton was inversely related.

Acid stress is another important factor thataffects the distribution of fish and other aquatictaxa. Acid precipitation has affected many lakesin northern Europe, resulting in acid stress andreduced fish population densities. However, lackof fish in mildly acid-stressed or recoveringlakes may facilitate establishment or increasedpopulation densities of Bythotrephes. In Lake ÖstraNedsjön, Sweden, for example, B. longimanusdeclined as a limed lake slowly re-acidified, butthen increased after a second lime treatment raisedlake pH and, more importantly, caused the virtualloss of planktivorous vendace, C. albula (Ekström& Hörnström, 1995). Bythotrephes longimanusalso immigrated to Lake Gårdsjön, Sweden, afterliming increased its pH from 4.67 to 7.35 (Stenson& Svensson, 1995). It established in this lakedespite the presence of stocked brown trout (Salmotrutta) and rainbow trout (Onchorhynchus mykiss),however, Stenson & Svensson (1995) argued thatplanktivory on Bythotrephes was likely quitelow. Large brown trout preferentially selectedjuvenile rainbow trout whose diet was heavilydominated (> 99%) by daphnia, thereby allowingBythotrephes to avoid direct predation pressure.Other Swedish studies also have noted increasedimportance of B. longimanus to zooplankton com-munities in acid lakes with reduced predationpressure (Andersson, 1972; Hörnström et al., 1973).Thus, both lake oligotrophication and acidifica-tion may create conditions favourable to estab-lishment and survival of Bythotrephes. However,tolerance of Bythotrephes to acidification has notbeen determined.

2

UNCORRECTED PROOF

Bythotrephes taxonomy and distribution 13

© 2002 Blackwell Science Ltd, Diversity and Distributions, 8, 000–000

Reservoir impoundment and other human-mediated dispersal effects

Reservoir impoundment alters the hydrologicalregime of watersheds. Following impoundment,the water mass of the newly created reservoirassumes the characteristics of a lacustrine envir-onment (i.e. reduced water velocity), while theregulated outflow might have increased or decreasedwater velocity depending on operations. Zozulya

(1979) reported that Bythotrephes abundance wassignificantly higher near reservoir dams relativeto more riverine locations upstream. Its spreadthrough the Volga reservoir system of westernUkraine illustrates the effects human activitieshave had on the dispersal of this zooplankter inEurasia (Fig. 6). Originally confined to deep, lacu-strine habitats, Bythotrephes first appeared inthe Upper-Volga Reservoir in the early 1900s(Chugunov, 1922). Despite downstream drift of

Fig. 6 The Volga River drainage showing the reservoir system and year of impoundment for each reservoir.Spread of Bythotrephes through this system is indicated, as is year of establishment for each reservoirsuccessfully invaded.

3

UNCORRECTED PROOF

14 T. W. Therriault et al.

© 2002 Blackwell Science Ltd, Diversity and Distributions, 8, 000–000

planktonic stages or resting eggs, the species failedto establish at downstream locations before the1930s, likely owing to unfavourable flow condi-tions (Rivier & Dziuban, 1978). However, withreservoir impoundment on the Upper-Volga Riverin the 1930s and 1940s, suitable conditions werecreated for the establishment of new popula-tions (Mordukhai-Boltovskoi & Dziuban, 1976;Rivier & Dziuban, 1978). Rapid colonization byBythotrephes occurred as additional reservoirswere impounded farther downstream (Fig. 6).

Reservoir impoundment also affects fish popu-lations and, indirectly, Bythotrephes occurrence.For example, Jensen (1988) reported that sizeand predation pressure exerted by planktivorousbrown trout and Arctic char increased followingimpoundment of Nesjö Reservoir, Norway, resultingin increased predation on B. longimanus. Aass(1969) reported a similar pattern in AursjöenReservoir, Norway.

Canal creation is another human-mediatedactivity that has increased the spread of aquaticspecies. However, unlike reservoir impound-ment that creates favourable conditions for theestablishment of new species, canal constructionremoves barriers to dispersal at regional or largerspatial scales. For example, creation of the Volga-Don Canal in 1952 significantly increased com-mercial shipping activities, thereby providing amechanism for the spread of aquatic species inthe Ponto-Caspian region (Ricciardi & MacIsaac,2000; Cristescu et al., 2001). This and otherwaterways have created ‘invasion corridors’ thathave enhanced dispersal of Bythotrephes to theTsimlyansk Reservoir and Don River Drainage(Mordukhai-Boltovskoi & Rivier, 1987). Ship-ping activities also have been implicated in thespread of Bythotrephes to the Netherlands andBelgium (Ketelaars & van Breemen, 1993; Ketelaars& Gille, 1994). Moreover, Sprules et al. (1990)speculated that Bythotrephes was likely deliveredto the Great Lakes via a ship visiting from theLake Ladoga–Neva River–Baltic Sea region, apathway susequently corroborated by Berg et al.(2001) and by this study.

CONCLUSIONS

Results from the genetic analyses indicate thatlongimanus and cederstroemi ‘forms’ represent asingle species, B. longimanus. No significant dif-

ferences were detected in key habitat character-istics of waterbodies supporting the two ‘forms’.Furthermore, the observed global distribution pat-tern for Bythotrephes (longimanus in central andwestern Europe and cederstroemi in North America)provides additional evidence that Bythotrephes isa single, polymorphic species. Fish predation,cultural eutrophication, and lake acidificationdirectly or indirectly affect the local distribution,abundance, life history and body size of Bytho-trephes. Reservoir and canal construction andglobal shipping activity affect the regional andintercontinental distribution of Bythotrephes. Oncegeographical barriers to intercontinental dispersalhave been removed, other human mechanismsdisperse the species to novel local and regionalhabitats.

ACKNOWLEDGMENTS

Drs H. Brendelberger, R. de Bernardi, H.R.Bürgi, H. Dumont, C. Erkström, B.A. Faafeng,D. Flössner, S. Haertel, D. Harper, C. Heller,A. Hobæk, K. Irvine, J.G. Jones, W. Jurasz,P. Kasprzak, J. Lindgrin, C. Lingegaard, M. Manca,L. Neudahl, Norwegian Institute for WaterResearch (NIVA), B. Peljer, K. Pettersson, M.Pidgaiko, I. Rivier, J. Sarvala, I. Telesh, N. Walz,L. Weider, I. Winfield, N. Yalynskaya and J.O.Young provided data and insights from theirrespective European regions without which thiswork would not have been possible. Two anony-mous referees and D.M. Richardson providedvaluable comments on a previous version of thismanuscript. TWT was supported by an NSERCpostdoctoral fellowship, IAG was supported byNATO and GLIER postdoctoral fellowships,and MEC was supported by an OGS fellowship.Funding for this study was made possible bygrants to HJM and DH from NSERC (Canada).Dr P.D.N. Hebert provided funding for mito-chondrial gene sequencing.

SUPPLEMENTARY MATERIAL

The following material is available from http://www.blackwell-science.com/products/journals/suppmat/DDI/DDI132/DDI132sm.htm

Appendix 1 Eurasia and North America water bodieswith reported populations of Bythotrephes

UNCORRECTED PROOF

Bythotrephes taxonomy and distribution 15

© 2002 Blackwell Science Ltd, Diversity and Distributions, 8, 000–000

REFERENCES

Aass, P. (1969) Crustacea, especially Lepidurus arcticusPallas, as brown trout food in Norwegian moun-tain reservoirs. Institute for Freshwater Research,Drottningholm 49, 183–201.

Andersson, B. (1972) The diet of perch in acidifiedlakes in western Sweden. Institute for FreshwaterResearch, Drottningholm 17, 1–21.

Behning, A.L. (1941) Cladocera of the Caucasus.Gruzmedgiz, Tbilisi.

Berg, M. (1951) Om kosthaldet hos auren i Esped-alsvatnet og Sjusjöen. (On the feeding habits ofthe trout in Espedalsvatn and the Sjusjö). NyttMagasin Naturviteuskapene 88, 165–205.

Berg, D.J. & Garton, D.W. (1994) Genetic differenti-ation in North American and European populationsof the cladoceran Bythotrephes. Limnology andOceanography 39, 1503–1516.

Berg, D.J., Garton, D.W., MacIsaac, H.J., Panov, V.E.& Telesh, I.V. (2001) Changes in genetic structureof North American Bythotrephes populationsfollowing invasion from Lake Ladoga. Russia.Freshwater Biology, in press.

Berg, A. & Grimaldi, E. (1966) Ecological relation-ships between planktophagic fish species in the LagoMaggiore. Verhandlungen Internationale Vereinigungfür Theoretische und Angewandte Limnologie 18,1065–1073.

Bergstrand, E. (1990) Changes in the fish and zoo-plankton communities of Ringsjön, a Swedish lakeundergoing man-made eutrophication. Hydrobiologia191, 57–66.

de Bernardi, R. & Giussani, G. (1975) Populationdynamics of three cladocerans of Lago Maggiorerelated to predation pressure by a planktophagousfish. Verhandlungen Internationale Vereinigung fürTheoretische und Angewandte Limnologie 19,2906–2912.

de Bernardi, R., Giussani, G. & Manca, M. (1987)Cladocera: predators and prey. Hydrobiologia 145,225–243.

Borbely, J. (2001) Modelling the spread of the spinywaterflea (Bythotrephes longimanus) in inland lakesin Ontario using gravity models and GIS. MScThesis, University of Windsor, 149 pp.

Bosselmann, S. (1974) The crustacean plankton ofLake Esrom. Archiv für Hydrobiologie 74, 18–31.

Bürgi, H.R. & Egli, B. (1984) Crustacean-Planktondes Vierwaldstättersees. Grössenfraktionierung undAbundanzdynamik in Den Jahren 1976 Bis 1979.Schweizerische Zeitschrift für Hydrologie 47, 197–224.

Burkhardt, S. (1994) Seasonal size variation in thepredatory cladoceran Bythotrephes cederstroemii inLake Michigan. Freshwater Biology 31, 97–108.

Chugunov, N.L. (1922) On Bythotrephes cederstroemiiSchödl. from the lower reaches of the River Volga.Russkii Gidrobiologicheskii Zhurnal 1, 82–88.

Coulas, R., MacIsaac, H.J. & Dunlop, W. (1998)Selective predation by lake herring (Coregonusartedi ) on exotic zooplankton (Bythotrephes ced-erstroemi ) in Harp Lake, Ontario. FreshwaterBiology 40, 343–355.

Cristescu, M.E.A., Hebert, P.D.N., Witt, J.D.S.,MacIsaac, H.J. & Grigorovich, I.A. (2001) Aninvasion history for Cercopagis pengoi based onmitochondrial gene sequences. Limnology andOceanography 46, 234–239.

Devlin, R.H., McNeil, B.K., Groves, D.D. &Donalson, E.M. (1991) Isolation of a Y-chromosomalDNA probe capable of determining sex in Chinooksalmon (Oncorhynchus tshawytscha). CanadianJournal of Fisheries and Aquatic Sciences 48,1606–1612.

Evans, M.S. (1988) Bythotrephes cederstroemi: itsnew appearance in Lake Michigan. Journal of GreatLakes Research 14, 524–525.

Fitzmaurice, P. (1979) Selective predation on Cladoceraby brown trout Salmo trutta L. Journal of FishBiology 15, 521–525.

Flössner, D. (1972) Krebstiere, Crustacea. Kiemen- undBlattfüßer, Branchiopoda, Fischläuse, Branchiura.G. Fischer Verlag, Jena.

Flössner, D. (2000) Die Haplopoda und Cladocera(Ohne Bosminidae) Mitteleuropas. Backhuys Pub-lishers, Leiden.

Folmer, O., Black, M., Hoeh, W., Lutz, R. &Vrijenhoek, R. (1994) DNA primers for amplifica-tion of mitochondrial cytochrome c oxidase subunitI from diverse metazoan invertebrates. MolecularMarine Biology and Biotechnology 3, 294–299.

Gieysztor, M. (1959) On a continuous series of lakes.Polskie Archiwum Hydrobiologii 6, 175–187.

Goulden, C.E. (1964) The history of the cladoceranfauna of Estwaite Water (England) and its limno-logical significance. Archiv für Hydrobiologie 60,1–52.

Grigorovich, I.A., MacIsaac, H.J., Rivier, I.K.,Aladin, N.V. & Panov, V.E. (2000) Comparativebiology of the predatory cladoceran Cercopagispengoi from Lake Ontario. Baltic Sea and CaspianSea. Archiv für Hydrobiologie 149, 23–50.

Grigorovich, I.A., Pashkova, O.V., Gromova, Y.F. &van Overdijk, C.D.A. (1998) Bythotrephes longimanusin the Commonwealth of Independent States: vari-ability, distribution and ecology. Hydrobiologia379, 183–198.

Haberman, J. (1998) Zooplankton of Lake Võrtsjärv.Limnologica 28, 49–65.

Harmsworth, R.V. (1967) The developmental historyof Blelham Tarn (England) as shown by animalmicrofossils, with special reference to the Cladocera.Ecological Monographs 38, 223–241.

Heath, D.D., Iwama, G.K. & Devlin, R.H. (1993)PCR primed with VNTR core sequences yieldsspecies specific patterns and hypervariable probes.Nucleic Acids Research 21, 5782–5785.

4

UNCORRECTED PROOF

16 T. W. Therriault et al.

© 2002 Blackwell Science Ltd, Diversity and Distributions, 8, 000–000

Hebert, P.D.N. & Grewe, P.M. (1985) Chaoborus-induced shifts in the morphology of Daphnia ambigua.Limnology and Oceanography 30, 1291–1297.

Herzig, A. (1985a) Fischnahrtier-Almanach für denMondsee. Osterreichs Fischerer 38, 97–108.

Herzig, A. (1985b) Resting eggs — a significant stagein the life cycle of crustaceans, Leptodora kindtiiand Bythotrephes longimanus. Verhandlungen derinternationalen Vereinigung für theoretische undangewandte. Limnologie 22, 3088–3098.

Hoffman, J.C., Smith, M.E. & Lehman, J.T. (2001)Perch or plankton: top-down control of Daphniaby yellow perch (Perca flavescens) or Bythotrephescederstroemi in an inland lake? Freshwater Biology46, 759–775.

Hörnström, E., Ekström, C., Miller, U. & Dickson, W.(1973) Försurningens inverkan på västkustsjör.Institute for Freshwater Research, Drottningholm4, 1–97.

Huitfeldt-Kaas, H. (1935) Der Einfluss der Gewas-serregelungen auf den Fischbestand in Binnenseen.Nationaltrykkeriet, Oslo, 105 pp.

Ischreyt, G. (1930) Morphology and mode of life ofBythotrephes longimanus Leydig. Archiv für Hydro-biologie 21, 241–324.

Jarnagin, S.T., Swan, B.K. & Kerfoot, W.C.(2000) Fish as vectors in the dispersal ofBythotrephes cederstroemi: diapausing eggs sur-vive passage through the gut. Freshwater Biology43, 579–589.

Jensen, J.W. (1988) Crustacean plankton and fishduring the first decade of a subalpine, man-madereservoir. Nordic Journal of Freshwater Research64, 5–53.

Jeppesen, E., Madsen, E.A. & Jensen, J.P. (1996)Reconstructing the past density of planktivorousfish and trophic structure from sedimentary zoo-plankton fossils: a surface sediment calibrationdata set from shallow lakes. Freshwater Biology36, 115–127.

Johannsson, O.E., Mills, E.L. & O’Gorman, R.(1991) Changes in the nearshore and offshorezooplankton communities in Lake Ontario: 1981–88. Canadian Journal of Fisheries and AquaticSciences 48, 1546–1557.

Kasymov, A.G. (1972) Freshwater Fauna of the Caucasus.ELM Press, Baku.

Ketelaars, H.A.M. & Gille, L. (1994) Range extensionof the predatory cladoceran Bythotrephes longimanusLeydig 1860 (Crustacea, Onychopoda) in westernEurope. Netherlands Journal of Aquatic Ecology28, 175–180.

Ketelaars, H.A.M. & van Breemen, W.C.A. (1993) Theinvasion of the predatory cladoceran Bythotrepheslongimanus Leydig and its influence on the plank-ton communities in the Biesbosch Reservoirs.Verhandlungen der Internationalen Vereinigung fürTheoretische und Angewandte Limnologie 25, 1168–1175.

Ketelaars, H.A.M., Wagenvoort, A.J., Herbst, R.F.,van der Salm, P.A.W. & de Jonge-Pinkster, G.J.(1995) Life history characteristics and distributionof Bythotrephes longimanus Leidig (Crustacea,Onychopoda) in the Biesbosch reservoirs. Hydro-biologia 307, 239–251.

Klemetsen, A. (1973) Pelagic, plankton-eating perch.Astarte 6, 27–33.

Kogan, A.V. (1974) The role of predatory cladocerain the feeding of the Tyul’ka Clupeonella delicatulacaspia morpha tscharchalensis in Kuybyshev Res-ervoir. Voprosy Ikhtiologii 14, 110–114.

Langeland, A. (1972) A comparison of the zoo-plankton communities in seven mountain lakesnear Lillehammer, Norway (1896 and 1971).Norwegian Journal of Zoology 20, 213–226.

Langeland, A. (1974) Long-term changes in theplankton of Lake Tyrifjord. Norway. NorwegianJournal of Zoology 22, 207–219.

Langeland, A. (1978) Effect of fish (Salvelinus alpinus,arctic char) predation on the zooplankton in tenNorwegian lakes. Verhandlungen Internationale Verei-nigung für Theoretische und Angewandte Limnologie20, 2065–2069.

Lindström, T. (1955) On the relation fish size — food size.Institute for Freshwater Research, Drottningholm36, 133–147.

Lysebo, E.M. (1995) Behavioural and morphologicalchanges in polymorphic Daphnia related to differ-ent predation regimes. Hydrobiologia 307, 185–191.

Macan, T.T. (1970) Biological Studies of the EnglishLakes. American. Elsevier Co., Inc, New York.

MacIsaac, H.J., Ketelaars, H.A.M., Grigorovich, I.A.,Ramcharan, C.W. & Yan, N.D. (2000) ModelingBythotrephes longimanus invasions in the GreatLakes based on its European distribution. Archivfür Hydrobiologie 149, 1–21.

Makarewicz, J.C., Grigorovich, I.A., Mills, E.,Damaske, E., Cristescu, M.E., Pearsall, W.,LaVoie, M.J., Keats, R., Rudstam, L., Hebert, P.,Halbritter, H., Kelly, T., Matkovich, C. &MacIsaac, H.J. (2001) Distribution, fecundity,and genetics of Cercopagis pengoi (Ostroumov)(Crustacea, Cladocera) in Lake Ontario. Journalof Great Lakes Research 27, 19–32.

Manca, M. & Ruggiu, D. (1998) Consequences ofpelagic food-web changes during a long-term lakeoligotrophication process. Limnology and Oceano-graphy 43, 1368–1373.

Manuilova, E.F. (1964) Cladocerans of the Fauna ofthe USSR. Keys to the Fauna of the USSR pub-lished by the Zoological Institute of the AcademySciences of the USSR, no. 88, p. 328. Nauka,Moscow–Leningrad.

Martin, J.W. & Cash-Clark, C.E. (1995) The externalmorphology of the onychopod ‘cladoceran’ genusBythotrephes (Crustacea, Branchiopoda, Onychopoda,Cercopagididae), with notes on the morphology

UNCORRECTED PROOF

Bythotrephes taxonomy and distribution 17

© 2002 Blackwell Science Ltd, Diversity and Distributions, 8, 000–000

and phylogeny of the order Onychopoda. ZoologicaScripta 24, 61–90.

Miller, M.P. (1997) Tools for population genetic ana-lyses (TFPGA) 1.3: a windows program for theanalysis of allozyme and molecular populationgenetic data. E-mail: MPM2@jan.ucc.nau.edu.

Mills, E.L., O’Gorman, R., DeGisi, J., Heberger, R.& House, R. (1992) Food of the alewife (Alosapseudoharengus) in Lake Ontario before and afterthe establishment of Bythotrephes cederstroemi.Canadian Journal of Fisheries and Aquatic Sci-ences 48, 140–145.

Mookerji, N., Heller, C., Meng, H.J., Bürgi, H.R. &Müller, R. (1998) Diel and seasonal patterns offood intake and prey selection by Coregonus sp.in re-oligotrophicated Lake Lucerne, Switzerland.Journal of Fish Biology 52, 443–457.

Mordukhai-Boltovskaya, E.D. (1959) On the system-atics of the genus Bythotrephes Leydig (Cladocera).Byulleten instituta Biologii vodokhranilishch akademiinauk SSSR 4, 29–32.

Mordukhai-Boltovskoi, F.D. & Dziuban, N.A. (1976)Alterations in composition and distribution of theVolga fauna as a result of anthropogenic influence.Biological Production Processes in the Volga Basin(ed. by G.V. Nikolskii & N.V. Butorin), pp. 67–82.Nauka, Leningrad.

Mordukhai-Boltovskoi, F.D. & Rivier, I.K. (1987)Predatory cladocerans Podonidae, Polyphemidae,Cercopagidae, and Leptodoridae of the world fauna.Keys to the Fauna of the USSR published by theZoological Institute of the Academy Sciences of theUSSR, no. 148, p. 183. Nauka, Leningrad.

Næsje, T.F., Jonsson, B., Klyve, L. & Sandlund, O.T.(1987) Food and growth of age-0 smelts, Osmeruseperlanus, in a Norwegian fjord lake. Journal ofFish Biology 30, 119–126.

Nauwerck, A. (1991) Zooplankton changes in Mondsee.Verhandlungen Internationale Vereinigung für The-oretische und Angewandte Limnologie 24, 974–979.

Nei, M. (1972) Genetic distance between populations.The American Naturalist 106, 283–292.

Nilsson, N.-A. (1955) Studies on the feeding habitsof trout and char in north Swedish lakes. Institutefor Freshwater Research, Drottningholm 36, 163–225.

Nilsson, N.-A. (1961) The effect of water-level fluc-tuations on the feeding habits of trout and char inthe Lakes Blåsjön and Jormsjön, north Sweden.Institute for Freshwater Research, Drottningholm42, 238–261.

Nilsson, N.-A. (1965) Food segregation betweensalmonid species in north Sweden. Institute forFreshwater Research, Drottningholm 46, 58–73.

Nilsson, N.-A. (1979) Food an habitat of the fishcommunity of the offshore region of Lake Vänern,Sweden. Institute for Freshwater Research, Drot-tningholm 58, 126–139.

Nilsson, N.-A. & Pejler, B. (1973) On the relationbetween fish fauna and zooplankton compositionin North Swedish Lakes. Report to the Institutefor Freshwater Research, Drottningholm 53, 51–77.

Patalas, K. (1954) Pelagic crustacean complex of 28Pommeranian lakes. Ekologia Polska 2, 61–92.

Patalas, J. & Patalas, K. (1966) The crustaceanplankton communities in Polish lakes. Verhand-lungen Internationale Vereinigung für Theoretischeund Angewandte Limnologie 16, 432–440.

Plankensteiner, C. (1992) Produktionsbiologie undBionomie von Bythotrephes longimanus (Leydig)im Mondsee, O·. Diplomarbeit zur Erlangung desMagistergrades an der NaturwissenschaftlichenFakultät der Universität Salzburg.

Ricciardi, A. & MacIsaac, H.J. (2000) Recent massinvasion of the North American Great Lakes byPonto-Caspian species. Trends in Ecology andEvolution 15, 62–65.

Rivier, I.K. (1998) The predatory Cladocera (Ony-chopoda: Podonidae, Polyphemidae, Cercopagidae)and Leptodorida of the world. Guides to the Identi-fication of the Microinvertebrates of the ContinentalWaters of the World 13 (ed. by H.J. Dumont),Backhuys Publishers, Leiden.

Rivier, I.K. & Dziuban, N.A. (1978) Zooplankton. TheVolga River and its Life (ed. by F.D. Mordukhai-Boltovskoi), pp. 153–179. Nauka, Leningrad.

Rivier, I.K. & Grigorovich, I.A. (1999) Biology ofBythotrephes Leydig (Crustacea, Cladocera, Ony-chopoda): Summary of Research. GidrobiologicheskiiZhurnal 35, 13–40.

Rognerud, S. & Kjellberg, G. (1990) Long-termdynamics of the zooplankton community in LakeMjøsa, the largest lake in Norway. VerhandlungenInternationale Vereinigung für theoretische undangewandte Limnologie 24, 580–585.

Rylov, V.M. (1921) On the glacial relicts. IzvestiyaRossiiskogo Gidrologicheskogo Instituta 1–3, 97–111.

Smyly, W.J.P. (1972) The crustacean zooplankton,past and present, in Esthwaite Water. Verhandlun-gen Internationale Vereinigung für theoretische undangewandte Limnologie 18, 320–326.

Sprules, W.G., Riessen, H.P. & Jin, E.H. (1990)Dynamics of the Bythotrephes invasion of theSt. Lawrence Great Lakes. Journal of Great LakesResearch 16, 346–351.

Stenson, J.A.E. (1972) Fish predation effects on spe-cies composition of the zooplankton communityin eight small forest lakes. Institute for FreshwaterResearch, Drottningholm 52, 132–148.

Stenson, J.A.E. (1978) Relations between vertebrateand invertebrate zooplankton predators in somearctic lakes. Astarte 11, 21–26.

Stenson, J.A.E. & Svensson, J.-E. (1995) Changes ofplanktivore fauna and development of zooplanktonafter liming of the acidified Lake Gårdsjön. WaterAir, and Soil Pollution 85, 979–984.

5

6

UNCORRECTED PROOF

18 T. W. Therriault et al.

© 2002 Blackwell Science Ltd, Diversity and Distributions, 8, 000–000

Straile, D. & Hälbich, A. (2000) Life history andmultiple antipredator defenses of an invertebratepelagic predator, Bythotrephes longimanus. Ecology81, 150–163.

Sullivan, C.A. & Lehman, J.T. (1998) Character vari-ation and evidence for spine length selection inthe invertebrate predator Bythotrephes (Crustacea:Cladocera) from Lakes Michigan, Huron, andErie. Archiv für Hydrobiologie 142, 35–52.

Svärdson, G. (1976) Interspecific population domin-ance in fish communities of Scandinavian lakes.Institute for Freshwater Research, Drottningholm55, 144–171.

Tseeb, Y.Y. (1940) Contribution to the fauna of lakesof northern Kazakhstan in relation to their fisheryestimation. Uchenye Zapiski Sankt–PeterburgskogoGosudarstrennogo Universiteta 1, 37–125.

Vekhov, N.V. (1981) Some features of Bythotrepheslongimanus (Cladocera, Cercopagidae) biology intundra lakes. Vestnik Zoologii 5, 72–75.

Vekhov, N.V. (1982) Distribution and features ofecology of some Cladocera in the subarctic waterbodies of the European part of the USSR. VestnikZoologii 4, 43–48.

Vekhov, N.V. (1987) Notes on the distribution, bio-logy and morphological variability of Bythotrepheslongimanus (Leydig) s. lat. in the European Subarctic.Nauchnye doclady vysshei shkoly. BiologicheskieNauki 2, 27–35.

Weider, L.J. (1991) Allozymic variation in Bytho-trephes cederstroemi; a recent invader of the GreatLakes. Journal of Great Lakes Research 17, 141–143.

Widuto, J. (1977) Zooplankton in the KortowskieLake after 18 years of the experimental restora-tion of the lake. Polskie Archiwum Hydrobiologii24, 91–106.

Yan, N.D., Blukacz, A., Sprules, W.G., Kindy, P.K.,Hackett, D., Girard, R.E. & Clark, B.J. (2001)Changes in zooplankton and the phenology of thespiny water flea, Bythotrephes, following its inva-sion of Harp Lake, Ontario, Canada. CanadianJournal of Fisheries and Aquatic Sciences, in press.

Yan, N.D., Dunlop, W.I., Pawson, T.W. & MacKay, L.E.(1992) Bythotrephes cederstroemi (Schoedler) inMuskoka lakes: first records of the European invaderin inland lakes in Canada. Canadian Journal ofFisheries and Aquatic Sciences 49, 422–426.

Yan, N.D. & Pawson, T.W. (1997) Changes in thecrustacean zooplankton community of Harp Lake,Canada, following invasion by Bythotrephes ceder-stroemi. Freshwater Biology 37, 409–425.

Yurista, P.M. (1992) Embryonic and postembryonicdevelopment in Bythotrephes cederstroemii. Cana-dian Journal of Fisheries and Aquatic Sciences 49,1118–1125.

Zar, J.H. (1984) Biostatistical Analysis. Prentice Hall,Inc., Englewood Cliffs, N.J.

Zhadin, V.I. & Gerd, S.V. (1961) Fauna and flora ofthe rivers, lakes, and reservoirs of the USSR.Uchpedgiz Min. Prosveshchenia RSFSR, Moscow.

Zozulya, S.S. (1977) Peculiarities of the first Bytho-trephes generation hatched from the latent eggs.Informatsionnyi Byulleten. Biologia vnutrennikhvod 33, 34–38.

Zozulya, S.S. & Mordukhai-Boltovskoi, F.D. (1975)On the morphology and systematics of the genusBythotrephes (Cladocera). Informatsionnyi Byulleten.Biologia vnutrennikh vod 27, 24–28.

Zozulya, S.S. & Mordukhai-Boltovskoi, F.D. (1977)Seasonal variability of Bythotrephes longimanusLeydig (Crustacea, Cladocera). Doclady akademiinauk SSSR 232, 493–495.

7

Author Query Form

Journal: Diversity and Distributions

Article: 129

Dear Author,

During the preparation of your manuscript for publication, the questions listed below have arisen. Please attend to these matters and return this form with your proof.

Many thanks for your assistance.

Query Refs.

Query Remarks

1 Lezin 1982 has not been included in the list

2 Ekström & Hörnström, 1995 has not been found in list.

3 Zozulya 1979 has not been included in the list

4 Berg et al. 2001: Still in press?

5 Plankensteiner, C. (1992): Please could you clarify publisher and place of publication.

6 Rivier, I.K. (1998): Is this a contribution or an entire work? Please give page range if former.

7 Yan, N.D. et al. (2001): Still in press.