UBC Community, Partners, and Alumni Publications

Fishes in databases and ecosystems: Proceedings of the 2006 FishBase Symposium. Palomares, Maria Lourdes D.; Stergiou, Konstantinos I.; Pauly, Daniel 2006-03-26

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
37052-FCRR_2006_14-4.pdf [ 5.35MB ]
Metadata
JSON: 37052-1.0074760.json
JSON-LD: 37052-1.0074760-ld.json
RDF/XML (Pretty): 37052-1.0074760-rdf.xml
RDF/JSON: 37052-1.0074760-rdf.json
Turtle: 37052-1.0074760-turtle.txt
N-Triples: 37052-1.0074760-rdf-ntriples.txt
Original Record: 37052-1.0074760-source.json
Full Text
37052-1.0074760-fulltext.txt
Citation
37052-1.0074760.ris

Full Text

  ISSN 1198-6727  Fisheries Centre Research Reports  2006 Volume 14 Number 4    Fishes in Databases and Ecosystems   Proceedings of the 2006 FishBase Symposium    Fisheries Centre, University of British Columbia, Canada   Fishes in Databases and Ecosystems     edited by Maria Lourdes D. Palomares, Konstantinos I. Stergiou and Daniel Pauly                    Fisheries Centre Research Reports 14(4) 95 pages © published 2006 by  The Fisheries Centre, University of British Columbia  2202 Main Mall Vancouver, B.C., Canada, V6T 1Z4       ISSN 1198-6727   Fisheries Centre Research Reports 14(4) 2006  FISHES IN DATABASES AND ECOSYSTEMS Proceedings of the 2006 FishBase Symposium  edited by Maria Lourdes D. Palomares, Konstantinos I. Stergiou and Daniel Pauly  CONTENTS   Page DIRECTOR’S FOREWORD ...................................................................................................................................... 1 PAPERS PRESENTED AT THE FISHBASE MINI-SYMPOSIUM Scientific impact of FishBase: a citation analysis Konstantinos I. Stergiou and Athanassios C. Tsikliras ........................................................................... 2 Analysis of common names of Brazilian freshwater fishes Kátia M.F. Freire ...................................................................................................................................... 7 Prominence trend in maximum lengths recorded for fishes: a preliminary analysis  Nicolas Bailly.......................................................................................................................................... 12 Age and growth of Mediterranean marine fishes Konstantinos I. Stergiou, Athanassios C. Tsikliras, Charalambos A. Apostolidis ................................ 18 Trophic levels of north Aegean Sea fishes and comparisons with those from FishBase Paraskevi K. Karachle, Konstantinos I. Stergiou...................................................................................22 Distribution ranges of commercial fishes and invertebrates Chris Close, William Cheung, Sally Hodgson, Vicky Lam, Reg Watson, Daniel Pauly ........................ 27 RELICT PAPERS A preliminary list of English common names for as yet unnamed fish families Maria Lourdes D. Palomares, Nicolas Bailly, Rainer Froese, Daniel Pauly..........................................38 Growth, reproduction and food of the mudskipper, Periophthalmus barbarus  on mudflats of Freetown, Sierra Leone Ibrahim Turay, J. Michael Vakily, Maria Lourdes D. Palomares, Daniel Pauly...................................49 Note on the weight of body parts, including fins, of the smalltooth sawfish Pristis pectinata Daniel Pauly ........................................................................................................................................... 55 An overview of biological data related to anchovy and sardine stocks in Greek waters Stylianos Somarakis, Dimitrios E. Tsianis, Athanassios Machias, Konstantinos I. Stergiou............... 56 A comparison of growth parameters of Australian marine fishes north and south of 28° South Claire Andersen, Daniel Pauly ............................................................................................................... 65 Growth parameters and length-length relationships of Greek freshwater fishes Platonas K. Kleanthidis, Konstantinos I. Stergiou ................................................................................69 Assessment of growth and apparent population trends in Grand Canyon native fishes  from tag-recapture data Carl Walters, Michael Douglas, William R. Persons, Richard A. Valdez.............................................. 78 Effects of lake and pond aeration on fish growth and related processes Daniel Pauly ...........................................................................................................................................89   A Research Report from the Fisheries Centre at UBC 95 pages © Fisheries Centre, University of British Columbia, 2006   FISHERIES CENTRE RESEARCH REPORTS ARE ABSTRACTED IN THE FAO AQUATIC SCIENCES AND FISHERIES ABSTRACTS (ASFA) ISSN 1198-6727   Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 1 DIRECTOR’S FOREWORD This report was assembled for the 4th Annual FishBase Symposium, and also celebrates the 7th meeting of the FishBase Consortium, gathered for the second time outside of Europe (the first being at Los Baños, Philippines in 2003). Last year, the Consortium met at the Aristotle University of Thessaloniki, Greece, hosted by Consortium member K.I. Stergiou and his team, and the proceedings of that symposium were published two days before it was over. Here we tried to emulate this, but failed: this report was published two weeks after the event. This report consists of 6 papers that were presented at the Symposium and 8 ‘relict’ papers, i.e., papers which, for various reasons, did not find their way into print right after they were originally written. Typically, these papers, which here contain facts on the life history of fishes (growth, size at maturity, etc.) and/or parameter estimates that would be useful for FishBase and its user community, languish in the drawers of middle-aged scientists after rejection from a prestigious journal (“we don’t publish local studies”), or because they were just about, but never completely, finished. Such unpublished manuscripts, turned into ‘relict papers’ and published, are useful not only because they make available to the community a body of knowledge, acquired at great cost, which otherwise would be lost, but also because this knowledge refers to historical states of fish population or ecosystems, and thus can serve as baseline. Thus, relict papers can help counter the effects of shifting baselines. Also, relict papers represent much of the personal knowledge of authors, a type of knowledge that is often lost upon their retirement. This loss has been identified in connection with taxonomists. It also happens, however, with other students of applied ichthyology, e.g., with stock assessment scientists, who usually know much more field biology than may be inferred from their equation-ridden papers. Conventional peer-reviewed journals often have problems with the subject matter that would be typical of relict paper: they often cover topics viewed as pedestrian, such as age and growth studies of fish. Such studies, however, are the motor that drives comparative studies, meta-analysis and biodiversity studies, and evaluation of the impact of global change. Hence, this compilation of relict papers, if the first, is not the last to be published as a Fisheries Centre Research Reports.  Daniel Pauly Director Fisheries Centre, UBC 02 September 2006      Scientific impact of FishBase, Stergiou, K.I., Tsikliras, A.C. 2 SCIENTIFIC IMPACT OF FISHBASE: A CITATION ANALYSIS1 Konstantinos I. Stergiou and Athanassios C. Tsikliras Aristotle University of Thessaloniki, School of Biology, Department of Zoology, UP Box 134, 54 124 Thessaloniki, Greece; Email: kstergio@bio.auth.gr ABSTRACT Since its creation in the late 1980s FishBase has evolved into a highly dynamic and versatile ecological tool. A citation analysis based on Scopus, mainly for citations in journals, and Google Books, for citations in books, revealed that it has penetrated into the primary aquatic and general literature, review literature, and aquatic and general books and textbooks. With a cumulative number of citations of 653 during 1995- 2006, it belongs to the 0.11 % of the highly-cited items published during 1900-2005, irrespective of discipline. INTRODUCTION FishBase (www.fishbase.org) is a global information system on fishes useful for research, for education at all levels, as an information source, and for the sensitization of the public at large (Froese and Pauly, 2000; Stergiou, 2004, 2005; Nauen, 2006). It includes a plethora of data, covering all levels of biological organization, for the known 29,400 fish species (as of August 2006). These data are derived from over 37,000 published sources (gray literature, books, journals, symposia proceedings, reports, etc.). FishBase, which was developed in the late 1980s (Froese and Pauly, 2000), and another ecological tool, Ecopath (www.ecopath.org; Christensen et al., 2000), which was also developed during the same period, widened the scope of fisheries science. This is because these two tools, in a synergetic fashion, led to global studies (e.g., Pauly, 1998; Pauly et al., 1998; Froese and Pauly, 2000; Froese and Binohlan, 2001, 2003; Christensen et al., 2003; Froese et al., 2005; Froese, 2006) in which previously-reported pieces of information on local knowledge were transformed into global knowledge, thus providing the framework for answering ‘mega-questions’ (i.e., questions pertinent to large spatial and temporal scales, and many species; see Stergiou and Karpouzi, 2002; CIESM, 2003; Stergiou, 2003, 2004). The success of the FishBase website is demonstrated by the large number of ‘hits’ (about 30 million hits per month, with number of hits/month increasing exponentially with time), coming from all continents and from a variety of users (i.e., individuals, universities, museums, research institutes, NGOs) (Nauen, 2006; Froese, unpubl. data). In this report, we show that this success is also true in terms of the scientific impact of FishBase, when impact is evaluated based on ‘traditional’ bibliometric indices (i.e., citation analysis). SCIENTIFIC IMPACT: ESTABLISHING CRITERIA Visualizing the scientific impact of a work requires the establishment of measures of impact. “Science employs a knowledge filter that slowly separates the wheat from the chaft” (see Chapter 3 of Bauer, 1992). Such a filter acts at different steps (Bauer, 1992): • A scientific finding is subjected to peer-review; • If peers find it useful then it gets published in the primary literature; • If other scientists also find it useful, it is cited;                                                  1 Cite as: Stergiou, K.I., Tsikliras, A.C. 2006. Scientific impact of FishBase: a citation analysis. In: Palomares, M.L.D., Stergiou, K.I., Pauly, D. (eds.), Fishes in Databases and Ecosystems. Fisheries Centre Research Reports 14(4), pp. 2-6. Fisheries Centre, University of British Columbia [ISSN 1198-6727].  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 3 • If it is cited a lot it gets into review articles/monographs/books; and eventually • It is cited into university textbooks. In addition, there is a strong gap between terrestrial and aquatic ecologists (Stergiou and Browman, 2005): they read, cite, and publish in different journals. Thus, two other indices of the impact of an ecological work, which measure the exchange of ideas between ecologists and the education of ecologists, are (Stergiou and Browman, 2005): • Its penetration into the primary ‘general ecological’ literature; and • Its penetration into ‘general ecology’ textbooks (in which the percentage of aquatic references is less than 15%). SCIENTIFIC IMPACT: THE CITATION SOURCE Until recently, Thomson’s ISI Web of Science was the only citation source available. However, in recent years other bibliographic services have become available, such as the Google Scholar (scholar.google.com) (Butler, 2005) and Scopus (www.scopus.com). These alternative tools perform as well as ISI (e.g., Google Scholar: Pauly and Stergiou, 2005), and moreover they are more flexible in terms of options for analyses that they provide (Scopus). In addition, these alternatives do not distort the scientific output of countries and institutions as ISI does through its limited use of sources, because they cover a much wider range of sources (sensu Stergiou and Tsikliras, 2006). For our analysis we selected Scopus for citations in scientific journals and Google Books (http://books.google.com) for locating citations in books. Scopus is an abstract and citation database covering more than 15,000 peer-reviewed series from more than 4,000 international publishers, including coverage of 500 Open Access journals, 700 Conference Proceedings, 600 Trade Publications, 125 Book Series, 28 million abstract records, and 245 million references (going back to 1996), added to all abstracts, and 200 million quality web sources, with more than 60% of the titles covered being from countries other than the US (copied from http://www.info.scopus.com/detail/what/).  FISHBASE: CITATION ANALYSIS FishBase was not subjected to formal peer- review in the sense that journal articles do. However, since its development in the late 1980s, it has undergone several reviews by experts and in response is constantly adapted to meet suggestions and new needs (Froese and Pauly, 2000). citations and implies an annual mean rate of exponentially during this period from 1 in 1995 to 155 citations in 2005 and 51 for the first half of 2006 (Figure 1). The 580 Scopus citations occurred in 199 A citation analysis with Scopus (on 5 July 2006) using ‘FishBase’ as the keyword in all fields revealed 580 citations for 1995-2006, whereas a search in Google Books revealed citations in 73 books. This adds up to 653 about 57. A cumulative citation rate of 653 puts FishBase into a very small group of highly-cited published items. This is because from the ca 38 million items that have been published since 1900, half have not been cited at all. From the remaining half that has been cited at least once, only 21,200 items (0.11%) have been cited more than 500 times (Garfield, 2005). The number of Scopus citations per year increased y = 1.2e^0.47x n=11, r2 = 0.96, p<0.01140 20 100 120 160 ita tio n s 40 60 80 N u m b er  o f c 0 1995 1997 1999 2001 2003 2005 Year  Figure 1 Annual number of citations to FishBase (source: Scopus, www.scopus.com, accessed on 5 July 2006).  Scientific impact of FishBase, Stergiou, K.I., Tsikliras, A.C. 4 different journals. Fourteen out of the 199 journals (i.e., Systematic Parasitology, Journal of Fish Biology, Folia Parasitologica, Marine Ecology Progress Series, Fisheries Research, Acta Parasitologica, Journal of Parasitology, Journal of Applied Ichthyology, ICES Journal of Marine Science, Reviews in Fish Biology and Fisheries, Canadian Journal of Fisheries and Aquatic Sciences, Aquatic Living Resources, Bulletin of Marine Science, and Ecological Modelling), covering parasitology, fish and fisheries, and aquatic ecology, each cited FishBase more than 9 times and cumulatively accounted for 223 citations (38.4%). The 199 journals covered different fields, from agricultural and biological sciences to energy, business management and accounting (Figure 2). 0 50 100 150 200 250 300 350 400 450 500   Agricultural & Biological Sciences   Environmental Science   Immunology & Microbiology   Earth & Planetary Sciences   Biochemistry, Genetics & Molecular Biology   Medicine   Social Sciences   Engineering   Multidisciplinary   Chemistry   Veterinary   Computer Science   Pharmacology, Toxicology and Pharmaceutics   Economics, Econometrics and Finance   Neuroscience   Arts and Humanities Number of citations   Energy   Chemical Engineering   Mathematics   Business, Management and Accounting  Figure 2 Number of citations to FishBase per field of journals (many journals cover more than one field) (source: Scopus, www.scopus.com, accessed on the 5 July 2006).  rnals (there were also 5 citations in conference proceedings). Thus, more than half of the citations to FishBase and of the sources of such itations were in ‘general’ journals. This clearly indicates that FishBase had a very good penetration into the primary ‘ecological’ (and other general) literature. FishBase also had a good penetration into the review literature: 46 (8%) out of the 580 citations were in journals specializing in reviews. as also cited in 48 aquatic and 18 general books as well as in three general (e.g., Lévêque and ounolou, 2003) and four aquatic textbooks (e.g., Jennings et al., 2001; Walters and Martell, 2003) (Figure 3). However, it is not yet cited in recent general ecology textbooks (e.g., Smith and Smith, 2003; In total, 121 (60.8%) out of the 199 journals were general journals and 74 (37.2%) were aquatic journals (there were also 4, 2%, conference proceedings). In terms of citations, 299 (51.6%) out of the 580 citations occurred in general journals and 276 citations (47.6%) in aquatic jou c FishBase w M Townsend et al., 2003; Odum and Barrett, 2005; Begon et al., 2006).   Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 5 general  textbooks, 3, 4% aquatic textbooks, 4, 5% general books, 48, 66% Fig s (left) and percentages (right) of citations in books (source: Google Books, accessed 7 To sum up, the analysis presented here shows that FishBase is also very successful in terms of scientific cit hBase is much larger than the one presented here, since it is most probably cited in many (e. echnical reports, and technical papers, if online). In addition, the analyses of citations part of the public knowledge infrastructure that people from all walks of life refer to it (Cornelia Nauen, iel, Germany) for his comments and versity of Illinois Press. Ch ., Guénette, S., Heymans, J.J., Walters, C., Watson, R., Zeller, D., Pauly, D., 2003. Hundred-year CIE rranean Biological Time Series. CIESM Workshop Monograph Series 22. CIESM publications,  yield per recruit in fishes, with a simple method to evaluate length frequency data. J. Fish Biol. 56, 758-773. Fro inohlan, C., 2003. Simple methods to obtain preliminary growth estimates for fishes. J. Appl. Ichthyol. 19, 376-379. 18, 25% aquatic books,  ure 3 Number of citation July 2006).  impact, when the latter is evaluated based on ‘traditional’ bibliometric indices. Naturally, the number of ations to Fis other scholarly publication types not covered by Scopus, some of which are covered by Google Scholar g., theses, t especially in non-peer reviewed, ‘popular’ items (e.g., general public publications, newsletters, newspaper articles, thematic maps, websites) will also be useful for measuring whether FishBase has become such a European Commission, Brussels, pers. comm.). ACKNOWLEDGEMENT The authors would like to thank Dr Rainer Froese (University of K suggestions. REFERENCES Bauer, H.H., 1992. Scientific Literacy and the Myth of the Scientific Method. Uni Begon, M., Harper, J.L., Townsend, C.R., 2006. Ecology: Individuals, Populations and Communities. Blackwell Science Ltd, USA. Butler, D., 2005. Science searches shift up a gear as Google starts Scholar engine. Nature 432, 423. ristensen, V decline of North Atlantic predatory fishes. Fish and Fisheries 4, 1-24. Christensen, V., Walters, C.J., Pauly, D., 2000. Ecopath with Ecosim: A user’s guide. Fisheries Centre, University of British Columbia, Vancouver, Canada and ICLARM, Penang, Malaysia. SM, 2003. Medite Monaco (available online at www.ciesm.org/publications/split03.pdf). Froese, R., 2006. Life-history strategies of recent fishes: a meta-analysis. Habilitationsschrift, Cristian-Albrecht Universität zu Kiel, Kiel. Froese, R., Binohlan, C., 2001. Empirical relationships to estimate asymptotic length, length at first maturity, and length at maximum ese, R., B  Scientific impact of FishBase, Stergiou, K.I., Tsikliras, A.C. 6 Froese, R., Pauly, D. (eds.), 2000. FishBase 2000: Concepts, Design and Data Sources. ICLARM, Los Baños, Philippines. Froese, R., Garthe, S., Piatkowski, U., Pauly, D., 2005. Trophic signatures of marine organisms in the Mediterranean arfield, E., 2005. The Agony and the Ecstasy—The History and Meaning of the Journal Impact Factor. International Congress on Peer Review And Biomedical Publication, Chicago, September 16, 2005. Jennings, S., Kaiser, M.J., Reynolds, J.D., 2001. Marine Fisheries Ecology. Blackwell Science, London. Lévêque, J-C., Mounolou, C. 2003. Biodiversity. Wiley Publishers. London. Nauen, C., 2006. Implementing the WSSD decision of restoring marine ecosystems by 2015—scientific information support in the public domain. Mar. Policy 30, 455-461. Odum, E.P., Barrett, G.W., 2005. Fundamentals of Ecology. Belmont, CA, Thomson Brooks/Cole. Pauly, D., 1998. Tropical fishes: patterns and propensities. J. Fish Biol. 53, 1-17. Pauly, D., Stergiou, K.I., 2005. Equivalence of results from two citation analyses: Thomson ISI’s Citation Index and Google’s Scholar service. Ethics Sci. Env. Polit. 2005, 33-35. Pauly, D., Christensen, V., Dalsgaard, J., Froese, R., Torres, F. Jr., 1998. Fishing down marine food webs. Science 279, 860-863. Smith, R.L., Smith, T.M., 2003. Elements of Ecology. Benjamin Cummings, San Francisco, CA. Stergiou, K.I., 2003. The balance and conservation of the North Atlantic ecosystems? Book review of Pauly D. and J. MacLean’s ‘In a perfect ocean – the state of fisheries and ecosystems in the North Atlantic Ocean’. Rev. Fish Biol. Fish. 13, 455-457. Stergiou, K.I., 2004. FishBase: The Global Information System on fishes. Greek Fish. News 06-04, 141-144. Stergiou K.I., 2005. Time-lagged scientific dialogues. Book review of D. Pauly’s ‘Darwin’s Fishes – An Encyclopaedia of Ichthyology, Ecology and Evolution’ (Cambridge University Press). Conserv. Biol. 19 (3), 983-985. Stergiou, K.I., Browman, H.I. (eds.), 2005. Bridging the gap between aquatic and terrestrial ecology. Mar. Ecol. Progr. Ser. 304, 271-307. Stergiou, K.I., Karpouzi, V.S., 2002. Feeding habits and trophic levels of Mediterranean fish. Rev. Fish Biol. Fish. 11, 217-254. Stergiou, K.I., Tsikliras, A.C., 2006. Under-representation of regional ecological research output by bibliometric indices. Ethics Sci. Env. Polit. [in press]. Townsend, C.R., Begon, M., Harper, J.L., 2003. Essentials of Ecology. Blackwell Science Ltd., USA. Walters, C.J., Martell, S.J., 2003. Fisheries Ecology and Management. Princeton University Press. Princeton.  as compared with other ecosystems. Belg. J. Zool. 135 (2), 139-143. G  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 7 ANALYSIS OF COMMON NAMES OF BRAZILIAN FRESHWATER FISHES1 Kátia M.F. Freire Universidade Estadual de Santa Cruz, Departamento de Ciências Exatas e Tecnológicas, Rodovia Ilhéus-Itabuna, km 16, Ilhéus, Bahia, Brazil CEP: 45650-000; Email: kmffreire@uesc.br ABSTRACT A database of 2230 common names for 769 Brazilian freshwater fishes was compiled based on fifteen sources. An average of three names per species was found, with 361 species associated with only one common name. However, each of these names may be associated with other species. This is the case for cascudo, which is associated with 46 species. This does not cause much problem for analysis of catch statistics as cascudo catches are very small (408 tonnes·year-1). On the other hand, curimatã catches are the highest for Brazilian freshwaters (28,700 tonnes·year-1) and the correspondence between common and scientific name is not well-understood. Most of the common names originated from Ameridian languages, followed by Latin names. These names represent mainly primary lexemes and mostly describe the morphology or colour pattern of each species. INTRODUCTION There is a growing concern with the biodiversity loss caused by different factors, particularly anthropogenic ones. Fishing is one of these factors. Fishing causes differential mortalities for different segments of a population, depending on its sex, age, and size, besides the total removal. Depending on the effect on these different segments, the impact may be more or less serious. In some countries, e.g., the USA and Canada, catch statistics are recorded by common name. In some cases, there is enough knowledge about the correspondence between common and scientific names for each species based on extensive work that has been done since the late 1940s and has culminated in the most recent volume, authored by Nelson et al. (2005). In other countries, this correspondence is far from well-understood. In Brazil, for example, Freire and Pauly (2005) analyzed the diversity of marine and brackish fishes, and found that each species is associated with six names in average, and that each name may be used for different species, even from different families. Such a large scale analysis was lacking for Brazilian freshwater fishes and is presented here. MATERIALS AND METHODS A total of 2230 common names referring to 769 freshwater species from Brazil, representing about 35% of the species recorded for Brazil in FishBase (www.fishbase.org), were compiled in this study. Fifteen references were included in this compilation: Mutti (1971), Santos (1981), Nomura (1984), FUEM- NUPÉLIA (1987), Godoy (1987), Komissarov (1988), Begossi and Garavello (1990), Begossi et al. (1999), Ferreira et al. (1999), Toledo-Piza (2002), Godinho and Godinho (2003), CBPDS(2004), IBAMA (2004a), PNDPA (2006), and Becker et al. (2006). The states associated with the occurrence of each name were recorded. Each name was translated from Portuguese to English and the origin of the name was identified. The core of the name and its modifiers were classified according to Palomares and Pauly (2000) as associated with behaviour, color pattern, habitat/ecology, inanimate object, locality/area, abundance, size, morphology, non-fish animal, other, person (generic), person (specific), plant, primary lexeme, or taste/smell. Catch data were obtained from the national statistics provided by the Brazilian Institute for the Environment and Renewable Resources (www.ibama.gov.br): IBAMA (2001, 2003, 2004b, 2005).                                                  1 Cite as: Freire, K., 2006. Analysis of common names of Brazilian freshwater fishes. In: Palomares, M.L.D., Stergiou, K.I., Pauly, D. (eds.), Fishes in Databases and Ecosystems. Fisheries Centre Research Reports 14(4), pp. 7-11. Fisheries Centre, University of British Columbia [ISSN 1198-6727].  Common names of Brazilian freshwater fishes, Freire, K. 8 RESULTS AND DISCUSSION A total of 361 species are associated with only one common name, 147 species with two common names, 78 with three names and 183 with four or more names (Figure 1). Trachelyopterus galeatus is an extreme case, associated with 30 common names, some of them representing only different spellings of the same word: Anduiá, Anojado, Anuiá, Anujá, Cabeça de ferro, Cachorro, Cachorrinho, Cachorrinho de padre, Cachorro de padre, Cangatá, Cangatí, Cangati, Capadinho, Carataí, Chorão, Chorãozinho, Cumbá, Cumbáca, Cumbaca, Jauzinho, Mandi, Mandí cumbá, Mandi cumbá, Mandí sapo, Mandizinho, Pacamão, Pacu, Peixe cachorro, Pocomão, and Ronaca ronca (Figure 1). On the average, each species was associated with three common names. For marine fishes, Freire and Pauly (2005) found an average of six common names per species, with an analysis based on a more extended database. 0 100 200 300 400 0 10 20 30 4 Number of common names per species 0 N um be r o f s pe ci es Trachelyopterus galeatus  Figure 1 Richness of names of Brazilian freshwater fishes represented by the frequency of scientific species that have one to thirty common names. Cascudo Ancistrus brevipinnis (1) Ancistrus formoso (0) Ancistrus stigmaticus (0) Harttia rhombocephala (0) Hemiancistrus chlorostictus (1) Hemiodontichthys acipenserinus (0) Hemipsilichthys cameroni (0) Hemipsilichthys garbei (0) Hemipsilichthys gobio (8) Hemipsilichthys mutuca (0) Hoplosternum littorale (6) Hypostomus alatus (2) Hypostomus albopunctatus (1) Hypostomus auroguttatus (0) Hypostomus carvalhoi (0) Hypostomus commersonii (18) Hypostomus derbyi (0) Hypostomus fluviatilis (0) Hypostomus garmani (0) Hypostomus gomesi (0) Hypostomus jaguribensis (0) Hypostomus lima (0) Hypostomus luetkeni (0) Hypostomus macrops (0) Hypostomus nudiventris (0) Hypostomus obtusirostris (0) Hypostomus papariae (0) Hypostomus plecostomus (12) Hypostomus pusarum (0) Hypostomus regani (1) Hypostomus scabriceps (13) Hypostomus unae (0) Hypostomus vaillanti (0) Hypostomus variipictus (0) Hypostomus wuchereri (0) Lithodoras dorsalis (6) Loricaria parnahybae (0) Megalechis thoracata (9) Pogonopomoides parahybae (0) Pseudancistrus luderwaldti (0) Pterygoplichthys anisiti (0) Pterygoplichthys etentaculatus (10) Rineloricaria cubataonis (0) Rineloricaria kronei (0) Rineloricaria lanceolata (1) Rineloricaria latirostris (1) Rineloricaria lima (10)  Figure 2 Scientific names associated with the common name cascudo. Numbers in parentheses represent other names besides cascudo that each species is associated with.  Even though almost 50% of these species are associated with only one common name, that name may be associated with one or more species. Cascudo, for example, is a common name used in association with 46 species (Figure 2). Thirty-one of these species are not associated with any other common name, seven are associated with one more name besides cascudo, one with 2 names, two with 6 or 10 names, one with 8, 9, 12, 13, and 18 other common names. Most of these species belong to the family Loricariidae, but two belong to the family Calichthydae (Hoplosternum littorale and Megalechis thoracata) and one to the family Doradidae (Lithodoras dorsalis). In order to better understand the intricate naming system, all the  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 9 additional names (besides cascudo) for each species were presented together with the number of additional species the names are associated with (Figure 3). Thus, Hypostomus scabriceps is associated with the names bacu puá, cascudo, cascudo cachichô, cascudo prêto, iarú urá, jaru itaquara, jaruitacoara, pirá tatu, uacarí, uará urá, and yaru itacura, all exclusively used for the species presented in Figure 3. Acarí, cascudo comum, and guacarí are associated with H. scabriceps as well, but also with four, one and one other not shown species, respectively. Catches for cascudo are low in Brazil corresponding to only 408 tonnes on average for the period 2001-2004 (in Maranhão, Paraná, Santa Catarina, Rio Grande do Sul and, more recently, in Minas Gerais states).      anto and Goiás states, all dourada catches originate from northern Brazil as is the case for piramutaba. IBAMA associates dourada to Brachyplatystoma flavicans, but two other species are also associated with this name (Brach according to the National Plan for the Development one is to compare catches originating from commerc , this difference should be considered. Piramutaba represents Brachyplatystoma va ording to 005 is considered an overexploited resource (Anonymous, 005). Even thou is is not ue association (as this common name was also related rus mucosus in Acr e; Begos al., Cascudo Ancistrus Cascudo de espinhos Anhã (3) Cascudinho (7) Cascudo espinho (1) Cascudo barbado (0) Cascudo pardo (0) Hemipsilichthys Cascudo piririca (0) Hoplosternum Couraçado (0) Mãe de anhã (1)  Figure 3 Scientific names associated with the common name cascudo and additional names for each species. Numbers in parentheses represent other species associated with each name.  The highest catches originating from Brazilian continental waters are for curimatã, dourada and piramutaba, which account, together, for about 30% of the total catch of freshwater fishes (Figure 4). IBAMA does not mention in the national bulletins the species associated with the name curimatã, which alone is responsible for annual catches of about 28,700 tonnes (2000-2004). According to the database compiled here, this name is associated with five species: Prochilodus brevis, P. costatus, P. lacustris, P. nigricans, and P. vimboides. As this name is associated with catches in all 26 Brazilian states, where they may represent different species, it is critical to establish the correct correspondence. With the exception of a small catch in Espírito S yplatystoma rousseauxii and Pellona castelnaeana)  of Recreational Fisheries in Brazil (PNDPA, 2006). If ial and recreational fisheries illantii acc  IBAMA (2 ) and 1999; Ruffino, 2 gh th a uniq  to Platysilu e stat si et Hypostomus alatus Hypostomus Hypostomus Guacarí (1) Cascudo prêto (0) Cascudo comum (1) Cascudo trepa pau (0) Acarí (4) Bacu puá (0) Cascudo asa branca Hypostomus plecostomus Acari (4) Cascudo chitão (0) Cascudo preto (3) Cascudo barata (0) Hypostomus regani Hypostomus scabriceps Cascudo cachichô (0) Vieja (10 Lithodoras Pirá tatu (0) Megalechis thoracata Uacarí (0) Yaru itacura (0) Iarú urá (0) Jaruitaquara (0) Cascudo cinzento (0) Pterygoplichthys Viola (4) Rineloricaria lanceolata Uará urá (0) Rineloricaria Rineloricaria lima Cascudo viola (6) Atipa (0) Camboatá (4) Curite (0) Tamboatá (1) Tamoatá (3) Tamuatá (1) Daqueiro (1) Bacu (4) Pacu (9) Bacu de pedra Vacu (3) Bacu pedra (0) Cascudo espada (5) Cascudo chinelo (2) Cascudo lima (0) Acarí lima (0) Acary lima (0) Aperta galha (0) Barbado (2) Cascudo barbudo (0) Guacari (0) Acarí pedral (0) Armadilho (0) Bode (1) Bode de Igarapé (0) Bodó (2) Vacarí (0) Yau urá (0) Panaque (0) Cascudo amarelo Carinhanha (0) Coroncho (0) Acarí amarelo (0) Acari amarelo Acarí juba (0) Acarí roncador (1) Acarijuba (0) Jaruitacoara (0) Jaru itaquara (0)  Common names of Brazilian freshwater fishes, Freire, K. 10 1999), it does not cause any problem to catch s piramutaba in this state. The origin of several names could not be identified using Tibiriçá (1984), Bueno (1998) and Ferreira (1999). Among those that could be identified, 61 % originated from Ameridian languages, represented mainly by Tupi and Guarani; these were followed by Latin names (26.3%), Brazilianisms (3.5%), African names (0.9%), and others (8.1%), including French, Greek, Spanish, Italian, German, English, and Arabic. For marine fishes, Freire and Pauly (2003) found the opposite, with most of the names originating from Latin through t to the inexistence of catch records for Portuguese (40%) and less influence from primary lexemes or are associated with the morphology or color pattern of the species (Table 1). or scriptors related to freshwater species is rather small as only seven states are associated with more than 5o names each: Amazonas, Acre, Tocantins, Minas Gerais, Paraná, Santa Catarina, and Rio Grande do Sul. do Sul, and Espírito Santo. This is the first database of names ever compiled for the analysis of Brazilian  been e next DGEMENTS I would like to thank Juarez Rodrigues for encoding part of the common names compiled here; Carolina Minte-Vera, Fernando Becker, Juarez bo Anonymous, 1999. V reunião do grupo permanente de estudos sobre a piramutaba. Série Estudos Pesca, n. 26. nd Beg ichthyology of fishermen from the Tocantins River (Brazil). Acta atistics due Table 1 Descriptors used in the co nd in the nd f the common names of Brazilian ater and marine fishes. Descriptor Fre er Marine re, a first a second modifiers o freshw shwat Primary lexeme 857 1793 Morphology 461 557 Color pattern 408 444 Inanimate object 297 314 2  Modification for size 199 232 124 162 106 228 Non-fish animal 206 419 Behaviour 02 158 Plant Other Person (generic) 92 143 Habitat/ecology 74 138 Locality/area 24 32 Modification for abundance 21 5 Person (specific) 19 46 Taste/smell 8 21 Ameridian languages (24%). This may be due to Amerindian influence being stronger in the interior of the country. Most of the descriptors used in the names of Brazilian freshwater and marine fishes represent A test of independence between the descriptor and med and the null hypothesis of independence was inanimate objects were more frequently used to describe freshwater than marine species. For the latter, the use of descriptors associated with non-fish animals was more common. The spatial coverage of the name database for the environment (fresh and marine water) was perf rejected (χ2 = 212.8; df = 14). De  For the remaining 19 states, there are less than 50 names and none were recorded for Amapá, Roraima, Rondônia, Mato Grosso freshwater species. Ten other sources have selected to be included in the database in th future. ACKNOWLE Rodrigues, Maurício Cetra, Paulo Chaves, and Projeto PróVárzea/IBAMA for providing lists of names and oks, including unpublished materials. REFERENCES Becker, F. G., Grosser, K.M., Milani, P.C.C., Braun, A.S., 2006. Peixes. Chapter 19. In: Becker, F.G.; Ramos, R. A. a Moura, L. A. (eds.), Biodiversidade. Regiões da Lagoa do Casamento e dos Butiazais de Tapes, Planície Costeira do Rio Grande do Sul. Brasília, Ministério do Meio Ambiente, pp. 262-275. ossi, A., Garavello, J.C., 1990. Notes on the ethno Amazonica 20(único), 341-351. 0 2000 2001 2002 2003 2004  ure 4 Catches from Brazilian continental waters for 5 10C 15 20 ch  ( to nn es ; Dourada 25 1 0 3 ) 30 at Curimatã Piramutaba Fig 35 the three main species (2000-2004).  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 11 Begossi, A., Silvano, R.A.M., Amaral, B.D., Oyakawa, O.T., 1999. Uses of fish and game by inhabitants of an extractive reserve (Upper Juruá, Acre, Brazil). Environ. Dev. Sustain. 1, 73-93. Bueno, F. S., 1998. Vocabulário Tupi-Guarani - Português. Éfeta Editora, São Paulo, Brazil. DS, 2004. ConfederCBP ação Brasileira de Pesca e Desportos Aquáticos [www.antares.com.br/cbpds/html/nova.htm], Fre rine fishes. In: Haggan, N., Brignall, ine fishes and its effect on catch statistics. FU o projeto ‘Ictiofauna e Biologia Go cisco das Minas Gerais. Belo Horizonte, Go ditora da UFSC. IBAMA,  Instituto Brasileiro do IBA de 2004. Anexo 1.  desenhos e aquarelas, 160 p.; V3 Florence, desenhos e aquarelas). Nelson Crossman, E.J., Espinosa-Pérez, H., Findley, L.T., Gilbert, C.R., Lea, R.N., Williams, J.D., 2004.  Special Publication 29, Bethesda, Maryland. alomares, M.L., Pauly, D., 2000. The COMMON NAMES Table. In: Froese, R., Pauly, D. (eds.), FishBase 2000: Concepts, Design and Data Sources. ICLARM, Los Baños, Laguna, Philipinnes, pp. 85-99. PNDPA, 2006. Plano Nacional de Desenvolvimento da Pesca Amadora [www.pescaamadora.com.br], accessed in August 20, 2006. Ruffino, M.L., 2005. Gestão do Uso dos Recursos Pesqueiros na Amazônia. Ibama/PróVárzea, Manaus. Santos, E., 1981. Peixes da Água Doce (Vida e costumes dos peixes do Brasil). Belo Horizonte, Brasil, Editora Itatiaia Limitada. Tibiriçá, L.C., 1984. Dicionário Tupi-Português - Com esboço de gramática de tupi antigo. Traço Editora, Santos, Brasil. Toledo-Piza, M., 2002. Peixes do Rio Negro [Fishes of the Rio Negro]: Alfred Russel Wallace (1980-1952). Editora da Universidade de São Paulo, Imprensa Oficial do Estado, S’ao Paulo.  accessed in August 20, 2006. Ferreira, A.B.H., 999. Dicionário Aurélio Eletrônico - Século XXI. Rio de Janeiro, Brasil, Lexicon Informática Ltda. ire, K.M.F., Pauly, D., 2003. What’s in there? Common names of Brazilian ma C., Wood, L. (eds.), Putting Fishers' Knowledge to Work. Fisheries Centre Research Reports 11(1), pp. 439-444. Fisheries Centre, University of British Columbia. Freire, K.M.F., Pauly, D., 2005. Richness of common names of Brazilian mar J. Ethnobiol. 25(2), 279-296. EM-NUPÉLIA/SUREHMA/ITAIPU BINACIONAL, 1987. Relatório anual d Pesqueira’. Nupélia, Universidade Estadual de Maringá. Maringá, Universidade Estadual de Maringá. dinho, H.P., Godinho, A. L., 2003. Águas, Peixes e Pescadores do São Fran PUC Minas. doy, M.P. , 1987. Peixes do Estado de Santa Catarina. Florianópolis, Brazil, E IBAMA, 2001. Estatística da Pesca do Brasil. 2000. Grandes regiões e unidades da federação. Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis/CEPNE, Tamandaré-PE.  2003. Estatística da Pesca do Brasil. 2001. Grandes regiões e unidades da federação. Meio Ambiente e dos Recursos Naturais Renováveis/CEPENE, Tamandaré-PE. MA, 2004a. Instrução normativa no. 5 de 21 de maio IBAMA, 2004b. Estatística da Pesca do Brasil. 2002. Grandes regiões e unidades da federação. Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis/CEPENE, Tamandaré-PE. IBAMA, 2004c. Estatística da Pesca do Brasil. 2003. Grandes regiões e unidades da federação. Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis/CGREP, Brasília-DF. IBAMA, 2005. Estatística da Pesca do Brasil. 2004. Grandes regiões e unidades da federação. Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis/CGREP, Brasília-DF. Komissarov, B., 1988. Expedição Langsdorff ao Brasil, 1821-1829. Iconografia do Arquivo da Academia de Ciências da União Soviética. Edições Alumbramento/Livroarte Editora, Rio de Janeiro, 3 volumes. (V1. Rugendas, desenhos e aquarelas; V2. Taunay, Mutti Pedreira, J. M. , 1971, Dicionário de Peixes de Couro do Brasil. Belém, SUDAM, Assessoria de Programação e Coordenação Divisão de Documentação. , J.S., Common and Scientific Names of Fishes from the United States, Canada, and Mexico. 6th ed., American Fisheries Society, Nomura, H., 1984. Nomes científicos dos peixes e seus correspondentes nomes vulgares. Dicionário dos peixes do Brasil. H. Nomura. Brasília, Brasil, Editerra, pp. 27-63. P  Prominence trend in maximum lengths, Bailly, N. 12 PROMINENCE TREND IN MAXIMUM LENGTHS RECORDED FOR FISHES: A PRELIMINARY ANALYSIS1 Nicolas Bailly WorldFish Center, Philippine Office, Natural Resources Management, IRRI College, Khush Hall, Los Baños, Laguna 4031, Philippines; Email: n.bailly@cgiar.org ABSTRACT FishBase, an information system on all finfishes of the world, records observed maximum lengths needed for growth studies and ecosystem modelling, among others. A plot of the number of species and subspecies against the maximum length by each centimetre from 1 cm to 20 m showed a bias related to Albers’ theory on prominence in the decimal system which defines the most prominent numbers as ‘spontaneous numbers’. The explanation lies in the origin of the data (mainly synthetic documents like FAO catalogues, regional faunas and check-lists), where maximum lengths are often rounded to the highest ten or hundred. Further analyses are suggested to check the impact of this bias in global trend analysis as well as possible methods to overcome this impact. INTRODUCTION The form of living organisms as the combination of shape and size (Thompson, 1917; Rohlf and Marcus, 1993) tells us about their position/place in ecosystems and is thus a powerful synthetic indicator of their life history traits. But we lack simple mathematical estimators to describe the form. The size is expressed as numbers which allows all mathematic operations when the shape is expressed as textual items (‘elongated’, ‘compressed’, etc.), hence the use of size alone as a form estimator. Volume is the best measure of the size in our three-dimensional world, but is difficult to measure. The centroid size used in multivariate morphometrics (see various papers in Rohlf and Bookstein, 1990) is a better estimate, but requires several measurements along three perpendicular axes for a good approximation. For the finfishes of the five classes of Craniata (Myxini, Cephalaspidomorphi, Elasmobranchii, Actinopterygii, Sarcopterygii), body length measured from the tip of the snout to the end of the caudal fin (TL: total length), or to the end of caudal peduncle (SL: standard length), is the most currently applied among various size estimators; the width of the disc (WD) is used for Rajoidei (skates and rays), head length (HL) for Macrouridae (grenadiers), etc. They all require only one measurement and are the size estimators used in fisheries management for growth and ecosystem models, yield per recruit assessment, and various relationships being used to link the length to the volume (Beverton and Holt, 1957; von Bertalanffy, 1960). Maximum length is an important parameter to record for these models. That is why FishBase, the major information system on finfishes in the world, available freely on the web (www.fishbase.org), gathers this information for the about 30,000 species and subspecies already described (see fishbase.sinica.edu.tw/report/MissingDataList.cfm?what = maxlength) for a list of species and subspecies with no maximum length recorded yet; we are looking for references for these species and calling on the community to send us references to complete the dataset. Plotting the number of species and subspecies against the maximum length for each centimetre rounded to the nearest centimetre (Figures 1-3) showed that there is a general bias in the data. We explore why this bias occurs, and if it can affect inferences built on correlations with classes of maximum lengths. Also, methods are suggested for further testing to overcome this bias when performing global trend analyses or creating size classes.                                                  1 Cite as: Bailly, N. 2006. Prominence trend in maximum lengths recorded for fishes: a preliminary assessment. In: Palomares, M.L.D., Stergiou, K.I., Pauly, D. (eds.), Fishes in Databases and Ecosystems. Fisheries Centre Research Reports 14(4), pp. 12-17. Fisheries Centre, University of British Columbia [ISSN 1198-6727].  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 13 MATERIALS AND METHODS The maximum lengths were extracted from FishBase, version of 18 Aug. 2006 (see Binohlan and Pauly, 2000). FishBase records information at species and subspecies levels; when subspecies are recognized as valid, information is recorded at subspecies level but not at species level. In the text below, the word ‘taxa’ is used for ‘species and subspecies’. In the version used, there were 29,497 taxa; 281 subspecies (i.e., not the nominal subspecies) were recognized for 193 species (i.e., information recorded for the nominal subspecies) making 29,216 species in total (Table 1). Five major types of length are used in FishBase: Total length, Standard length, Fork length, Disk length, Width of Disk; some rarely used (like head length) are noted ‘other’; it may be ‘not given’. Maximum lengths are recorded for adults ‘male/unsexed’ and/or ‘female’. Table 1 Number of taxa (species and subspecies) with information in FishBase (as of 18 Aug. 2006). Category Number Species (excl. species with subspecies) 29,023 Nominal subspecies 193 Subspecies 281 The maximum lengths used here are computed as the maximum between ‘male/unsexed’ and ‘female’, whatever the length type is. For instance, in the case of dwarf parasitic males in the family Oneirodidae, the maximum length for female is selected. RESULTS The numbers of taxa are computed for each centimetre rounded to the nearest centimetre. On the total of 29,497 taxa, 4,693 had no maximum length recorded for adults (see Table 2 for a gap analysis and other statistics). The minimum length is 0.84 cm TL (Perciformes: Schindleriidae: Schindleria brevipinguis, Watson and Walker, 2004) and the maximum 20 m (Orectolobiformes: Rhincodontidae: Rhincodon typus, Smith, 1828). The maximum number of taxa is 1,334 for 5 cm maximum length. Table 2 Statistics on maximum lengths (Lmax) in FishBase (as of 18 Aug. 2006). Taxa with Lmax Number All taxa n = 29,497 % With Lmax n = 24,804 % For one of or both sexes 24,804 84.1 100.0 For both sexes 1,082 3.7 4.4 For one of sexes 23,722 80.4 95.6 For male/unsexed only 23,395 79.3 94.3 For female only 327 1.1 1.3 Missing for both sexes 4,693 15.9 18.9 Missing for male/unsexed 5,020 17.0 20.2 Missing for female 28,088 95.2 113.2 Where Female > Male/unsexed 544 1.8 2.2 >= 100 cm 1,136 3.9 4.6 < 100 cm 23,668 80.2 95.4 >= 80 cm 1,568 5.3 6.3 < 80 cm 23,236 78.8 93.7 Figures 1 to 3 present the following characteristics: • up to 80 cm, there are more species with maximum length expressed primarily as tens of centimetres; and secondarily as fives of centimetres; • from 80 cm to 500 cm, there are more species with maximum length expressed in tens of centimetres except for 190, 290, 460, 480, and 490 cm; fives of centimetres are more frequently between 80 and 100 cm, 160 and 180 cm, and 300 and 320 cm, but not as much as below 80 cm; beyond this, there are more tens or fives but they pertain only to one taxon; • from 100 cm up to 500 cm, there are more species with maximum length expressed primarily for hundreds of centimetres; and secondarily for 250, 320 (higher than for 350 cm), and 430 cm (higher than for 450 cm); • beyond 500 cm, there are only 18 species, 13 measured in tens of centimetres, 3 in hundreds (800, 1,100, and 2,000 cm, the latter maybe as thousand), and 2 with other final digits (549 and 656 cm).  Prominence trend in maximum lengths, Bailly, N. 14  0 500 1000 1500 0 10 20 30 40 50 60 70 80 90 100 Maximal lengths (cm) N u m be r of  t ax a  Figure 1 Number of taxa (species and subspecies) for each centimetre of maximum length from 0 to 100 cm. 0 50 100 80 100 120 140 160 180 200 220 240 260 280 300 Maximal lengths (cm) N u m be r of  t ax a 100 cm: 217 taxa Figure 2 Number of taxa (species and subspecies) per each centimetre of maximum length from 80 to 300 cm. 0 1 2 3 4 5 6 7 300 350 400 450 500 550 Maximal lengths (cm) Nu m ber of  ta xa 300 cm: 27  Fig 3 Numbe of taxa (s ies and subspecies) per each centimetre of maximum length from 300 to 550  ure r pec  cm.  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 15 We have also ranked the unit digits 0-9 from 1 to 10 in each group of ten according to the decreasing number of species with the same unit digit of maximum length. Table 3 presents the mean position between 1 to 10 in each tenth between 11 and 79. If the rank was equiprobable for each digit, the mean position should not be significantly different from 5.5 for all. As evidenced by Table 3, this is not the case. Tens followed by digits 5, 1 and 2, and digit 9 is definitely the least-represented digit. es progressively when moving per tens but the same tendency n Figures 1 to 3 are consistent with the theory of prominence in the decimal system where Albers (1997) proposes that a set S called ‘spontaneous numbers,’ defined as  is ours] ( p s ca gge r out e s ly t ter r 7 a the m isson distribu e results (in  a strong tendency of authors to round to the highest ten or hundred, which confirms the its 0, 5, 1, and 2 between 11 and 99; and 120, 250, 310, (not all shown on Benford (1938): first digit law sing number of species grouped by tens, and by hundreds (Table 4), as a logarithmic istribution p(i) = log10(1+1/i) for i = 1 to 9 (our number fits the logarithmic tendency, but not exactly the alues computed from the equation). The Poisson distribution of our data explains why the distribution of Table 3 Me nk of di -9 amon ens 10-70 en ordered down by number of species with the same unit an ra gits 0 g t  wh digit of maximum length. SE = standard error. Digit Rank  Mean SE Min Max 0 1.0 0.0 1 1 5 2.0 0.0 2 2 1 4.1 0.8 3 5 2 4.3 1.0 3 6 6 5.3 1.8 3 8 3 6.3 1.4 4 9 4 6.6 1.3 4 8 8 7.3 1.9 3 10 7 8.4 1.0 7 10 9.7 0.4 9 10 are dominant The signal fad towards the up remains strong for tens, digits 5 and 9, and to a lesser degree for digits 1 and 2; digits 6, 3, 4, 8 9 and 7 are more variable. DISCUSSION Fitting laws Albers and Albers (1983): Prominence in the decimal system The results shown i {s*10i | s∈{1, 1.5, 2, 3, 5, 7} | i∈ℵ} [the notation preferentially chosen by persons for estimations (of 100, 150, 200, 300, 500 are prominent in our result not prominent. Two complementary explanations without measurements, and does not completely app length distribution among fishes (approximately a Po particular 1 and 2). Hertwig et al., 1999), for numbers rices in Albers’ work). Indeed 5, 10, 15, 20, 30, 50, (Figure 4). Nevertheless, 1 and 2, and 7 and 70, are n be su sted: the theo y is ab stimation o the lat  case (fo tion) strongly nd 70); and  constrains th maximu A series including 8, 40, 60, 120, 250, 400, 430 constitutes a second line of prominent numbers, but non-spontaneous numbers (Figure 4), more or less mixed with the first one up to 150. This indicates Table 4 Taxa n an ximum len s inumbers by r ges of ma gth  ten  0 t , and eds from  to 99 by tens Num of t by s Number of taxa s from o 99 cm by hundr  100 9 cm. Range ber axa Range  hundred 0-9 8 23,668 ,305 0-99 10-19 6 868 20-29 3,213 200-299 164 30-39 2,017 300-399 55 40-49 1,254 400-499 24 50-59 774 500-599 11 ,637 100-199 strong constraint of the data distribution. For the second digit, the results are similar: dig 320, 350, 430, 450 beyond Figure 4). Again digit 7 is not prominent in our results. Consistently, the 9 digit is under-represented at all levels, showing again the strong tendency of authors to round to the highest ten or hundred. 60-69 604 600-699 6 70-79 435 700-799 4 80-89 231 800-899 1 90-99 198 900-999 1 The distributions of the first digit between 0 and 99, and 100 and 999, follow Benford’s law (1938), which predicts a decrea d v the first digit fits with Benford’s law. A more rigorous demonstration can be found in Hill (1996).  Prominence trend in maximum lengths, Bailly, N. 16 430 500400 5 10 15 100 0 100 200 300 400 500 Maximal lengths (cm) 3 i ncoding in FishBase. Data are encoded as published. However, values with tenth of millimetres for lengths beyond 50 cm are rounded to In general, in publications that first describe a species, the length is not rounded because only few usually measured, and the maximum length is extracted by the FishBase encoders and maximum length reported in FishBase).  sta lthough 6 is syntheses, such as the FAO ts and revisions to a lesser 20 30 60 50 40 300 250 200 150120 N u m be r of  t ax a  2 1 [l og (n  + 1) ]  Figure 4 Logarithm of number of taxa (species and subspecies) as log10(n+1) per each centimetre of maximum length from 0 to 500 cm.  It could be suggested that the full step numbers F, defined as {f*10  | s’∈{1, 2, 5} | i∈ℵ} (Albers, 2001) fit both Albers’ theory and Benford’s law. This is consistent for instance with the choice of coins and notes for the European currency (1, 2, 5, 10, 20, 50 Cents, and 1, 2 Euros for coins; 10, 20, 50, 100, 200, 500 Euros for notes). When and how the bias occurs We can reject the hypothesis that the bias is generated during data e the nearest millimetre. As mentioned above, authors tend to round at the highest tens or hundreds. In which publications? specimens are entered as such. This is confirmed by a plot (not shown) similar to Figure 4 for the taxa described from 2000 onwards that were encoded mainly from the original description (1,780 taxa of which 1,348 have a  In this case, the prominent numbers are 6, 19, 25, 30, 41, 58, 69, 78 (for species described from 1996 onwards, 13, 25, 30, 44, 50, 58, 80. The spontaneous number signal rts to appear for species described from 1990 onwards: 25, 30, 40, 50, 70, 80, and 120, a still the highest value). For species described before 2000, the FishBase Team used mainly large catalogues, regional accounts, identification sheets, regional faunas and checklis degree (about 89% on 23,238 taxa before 2000 with maximum length; 90% on 21,205 taxa before 1990  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 17 with maximum length). It is in these publications that authors round maximum lengths. Others few cases olved unit conversion (incinv hes in cm, for instance). Co usu ove tural resources, the lengths recorded in nature tend to decrease, and the reported and document databases, mber of taxa: n-2, n-1, ore complex solution  below 5 cm or by umming them (i.e., power or logarithmic fit). These solutions are to be tested thoroughly and eventually mathematical solutions proposed to avoid biasing global trend analyses by an over-estimation of maximum lengths. ACKNOWLEDGEMENTS I thank D. Pauly, Fisheries Centre, University of British Columbia, who pointed out the prominence theory and spontaneous numbers, and without whom this matter would have remained ‘une aimable curiosité’. REFERENCES Albers, W., 1997. Foundations of a theory of prominence in the decimal system (Parts I-V). Bielefeld (Germany): Bielefeld University, Institute of Mathematical Economics, Working Papers Nos. 265-271. Albers, W., 2001. Prominence theory as a tool to model boundedly rational decisions. In: Gigerenzer G, Selten R (eds.), Bounded Rationality: The Adaptive Toolbox. Cambridge (MA, USA): MIT Press, pp. 297–317. Albers, W., Albers G., 1983. On the prominence structure of the decimal system. In: Scholz, R.W. (ed.), Decision Making Under Uncertainty. Amsterdam: Elsevier, pp. 271-287. Benford, F., 1938. The law of anomalous numbers. Proc. Amer. Phil. Soc. 78, 551-572. Bertalanffy, L. von, 1960. Principles and theory of growth. In: Nowinski, W.W. (ed.), Fundamental Aspects of Normal and Malignant Growth. Elsevier, Amsterdam, pp. 137-259. Beverton, R.J.H., Holt, S.J., 1957. On the dynamics of exploited fish populations. Fish. Invest. Minist. Agric. Fish. Food U.K. (2. Sea Fish.), 19. Binohlan, C., Pauly, D., 2000. The POPCHAR Table. In: Froese, R., Pauly, D. (eds.), FishBase: Concepts, Design and Data Sources. ICLARM, Los Baños, Laguna, Philippines, pp. 130-131. Hertwig, R., Hoffrage, U., Martignon, L., 1999. Quick estimation: letting the environment do the work. In: Gigerenzer, G., Todd, P.M., ABC Research Group (eds.), Simple Heuristics that Make Us Smart. New York: Oxford University Press, pp. 209–234. Hill, T.P. , 1996. A statistical derivation of the significant-digit law. Stat. Sci. 10, 354-363. Rohlf, F.J., Bookstein, F.L., 1990. Proceedings of the Michigan Morphometrics Workshop. Michigan (USA): Univ. Michigan Museum Zoology Spec. Publ. no. 2. Rohlf, F.J., Marcus, L., 1993. A revolution in morphometrics. Trends Ecol. Evol. 8, 129-132. Thompson, D’A., 1917. On Growth and Form. Cambridge (UK): Cambridge University Press.  nclusions We demonstrated that, overall, FishBase may over-estimate maximum lengths. In synthetic works, it is ally not possible to track back the origin of the information used by authors. Unfortunately, with rexploitation of na maximum lengths cannot be validated in the field. Species, museum collection, associated within information system networks like GBIF, should help to reappraise and re-evidence the maximum lengths from specimens and primary literature. They should not be rounded then. Pending this, one simple solution is to compute a moving average on 4 values of nu n, n+1 for a given maximum length. This however generates other biases. Another m is to fit a distribution analytically, either a Poisson distribution, or by ignoring lengths s  Age and growth of Mediterranean fishes, Stergiou, K.I. et al. 18 AGE AND GROWTH OF MEDITERRANEAN MARINE FISHES1 Konstantinos I. Stergiou, Athanassios C. Tsikliras, Charalambos A. Apostolidis Aristotle University of Thessaloniki, School of Biology, Department of Zoology, UP Box 134, 541 24 Thessaloniki, Greece; E-mail: kstergio@bio.auth.gr ABSTRACT Age and growth data have been so far collected from the literature on 383 Mediterranean marine fish stocks, most of which (77 %) are at present not included in FishBase. The most intensively studied species were highly commercial species. Maximum length ranged between 4.2 cm (Aphia minuta) to 225 cm (Xiphias gladius), and maximum lifespans between 0.66 year (A. minuta) and 30 years (Helicolenus dactylopterus). INTRODUCTION Age and growth, including maximum length, which is related to many other biological, ecological and population dynamic parameters, are the cornerstones of fish biology, ecology and fisheries management (Campana, 2001; Froese and Pauly, 2000). The longevity and the maximum size which the individuals of a given species are capable of reaching, as well as the growth rate (i.e., change in size with time) are determined mainly by the environment and the genotype. Under the most favourable conditions, the individuals of a population may reach a characteristic maximum length, which is specific for each species, i.e., there are no sardine which reach 1 m (Bond, 1996). The interaction of growth with food availability and reproduction determines other crucial parameters of a species (e.g., size at first maturity, fecundity, mortality) and hence its biomass and stock composition in space and time. The present work was motivated by the need to update FishBase (Froese and Pauly, 2000; www.fishbase.org) with information on the age, growth and maximum length of Mediterranean marine fishes, and expands on earlier compilations of such data on Greek fishes (Stergiou et al., 1997). This will allow us to identify patterns and propensities (sensu Pauly, 1998) in the age and growth of fishes in this semi-enclosed basin, which has been subjected to fishing for thousands of years, and eventually to test various hypotheses (e.g., nanism in the eastern Mediterranean, see Stergiou et al., 1997). MATERIALS AND METHODS We collected data on: (i) maximum length (Lmax, cm; mainly total length) and age (tmax, year); (ii) growth parameters, i.e., asymptotic length L∞ (cm), the rate at which L∞ is approached, K (year-1), and the theoretical age at zero length, t0, (year). We also tabulated auxiliary information such as study area and year, frequency of sampling, sampling gear, sample size, method used for the estimation of growth parameters, and skeletal structure used for age determination. All sources as well as the data collected will shortly be incorporated into FishBase. RESULTS AND DISCUSSION Overall, we collected to date data for 383 records (or stocks: species-sex-area-year combinations) belonging to 86 Mediterranean fish species (Figure 1). From these records only 88 (23 %) are at present included in FishBase. We point out that so far all our records are derived from journals listed in the Science Citation Index and from other international or local journals, whereas we have not yet included records appearing in the grey literature (i.e., technical reports, conference proceedings and theses, with the exception of the CIESM proceedings), which accounts for more than 35 % of all published items on                                                  1 Cite as: Stergiou, K.I., Tsikliras, A.C., Apostolidis, C.A., 2006. Age and growth of Mediterranean marine fishes. In: Palomares, M.L.D., Stergiou, K.I., Pauly, D. (eds.), Fishes in Databases and Ecosystems. Fisheries Centre Research Reports 14(4), pp. 18-21. Fisheries Centre, University of British Columbia [ISSN 1198-6727].  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 19 Mediterranean ecological research (Stergiou and Tsikliras, 2006). These items will be also covered during the next six months. The most intensively-studied species were all highly commercial species: the common pandora (Pagellus erythrinus), European anchovy (Engraulis encrasicolus), European hake (Merluccius merluccius), red mullet (Mullus barbatus), picarel (Spicara smaris), annular seabream (Diplodus annularis) and bogue (Boops boops), each represented by more than ten stocks (Figure 2). The vast majority of the stocks belonged to families Sparidae (59 stocks), Mullidae (23), Clupeidae (22), and Gadidae, Scophthalmidae and Scorpaenidae (20). Spain France Marocco Algeria Tunisia Egypt Lebanon Syria Turkey Greece Croatia Italy Yugoslavia  Figure 1 Map of the Mediterranean Sea showing approximate locations where data on at least one aspect of fish age and growth are available (total: 383 stocks). The red-bubble size is proportional to the number of stocks (small = 1, medium = 5, large = 10 stocks). The bottom bar indicates depth and altitude (in m). Map from Università degli Studi di Pavia (Centro Interdisciplinare di Bioacustica e Ricerche Ambientali; downloadable at http://www.unipv.it/web cib/edu-Mediterraneo-uk.html).  Age and growth data were available for the waters of Turkey (105 stocks), Greece (58), Italy (53), Spain (40), Tunisia (38), Croatia (33), Algeria (13), France (13), Egypt (8), Portugal (7) Lebanon (4), Morocco (4), Cyprus (2) and Yugoslavia (2). Our preliminary analysis showed that the eastern (179) and western (201) Mediterranean studies are so far relatively balanced (Figure 1). However, this is not true for southern when compared to northern Mediterranean (Figure 1). 0 5 10 15 20 Pagellus erythrinus Engraulis encrasicolus Merluccius merluccius Mullus barbatus Spicara smaris Diplodus annularis Boops boops S p ec ie s Number of stocks  Figure 2 The most intensively-studied species in the Mediterranean Sea.  Age and growth of Mediterranean fishes, Stergiou, K.I. et al. 20 Overall, tmax was available for 285 stocks and Lmax for 266 stocks. L∞ and K were provided by the original authors for 340 stocks, and t0 for 330 stocks. In 40 (11.5 %) out of the 340 stocks, the growth parameters were estimated from length frequency distributions using ELEFAN/FiSAT and other length-based methods. For the remaining 300 stocks (88.5 %), the growth parameters were estimated using age-at-length data derived from skeletal structures (otoliths: 204 stocks, 68 %; scales: 75 stocks, 25 %; otoliths and scales: 4 stocks, 1.3 %; spines: 8 stocks, 2.7 %; vertebrae: 9 stocks, 3 %). Maximum length ranged between 4.2 cm (Aphia minuta, Balearic Islands, Spain) and 215 cm (Xiphias gladius, southern Aegean Sea, Greece) and had a mean value of 37.6 cm, while L∞ ranged between 4.7 and 220.1 cm for the same species (mean = 42.8 cm). The dimensionless Lmax/L∞ ratio ranged between 0.40 (Helicolenus dactylopterus, eastern Ligurian Sea, Italy) and 1.37 (Diplodus vulgaris, Algarve coast, Portugal), with a mean value of 0.89. The relationship between logL∞ and log Lmax (Figure 3) had a slope (0.984) similar to that reported by Froese and Binohlan (2001), which was based on 551 data pairs. y = 0.99x + 0.075 r2 = 0.93 n=265 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0.0 0.5 1.0 1.5 2.0 2.5 Maximum recorded length (Lmax, log, cm) A sy m p to ti c le n g th  ( L ∞ , l o g , c m )  Figure 3 The relationship between asymptotic length (L∞, log, cm) and maximum recorded length (Lmax, log, cm) for 265 Mediterranean marine fish stocks. The maximum lifespan recorded was 30 years (Helicolenus dactylopterus, Alboran Sea, Spain) and the minimum one 0.66 year (Aphia minuta, Balearic Islands, Spain). The von Bertalanffy K coefficient ranged between 0.028 year-1 (Epinephelus guaza, Gulf of Gabes, Tunisia) and 2.76 year-1 (Aphia minuta, Balearic Islands, Spain), with a mean value of 0.32 year-1. The double logarithmic relationship between K and L∞ (Figure 4) is described by the following equation: logK = -0.36·logL∞ + 0.07 (r2 = 0.13, n = 340). The ‘outlier’ stocks noted on this graph are Macrouridae, Coryphaena hippurus and d from -5.36 year to 0.96 year (mean = -1.12 year), with very large negative values most probably indicating unreliable estimates. The relationship between K and L∞ (as well as other life-history relationships) will be examined separately for the main families as well as for the eastern and western Mediterranean. Lepidopus caudatus. Finally, the theoretical age at zero length, t0, range    -2.0 0.0 0.5 1.0 1.5 2.0 2.5 -1.5 -1.0 -0.5 G ro w th  c oe ff ic ie nt  (K Lepidopus caudatus Macrouridae 0.0 Asymptotic length (L∞, log, cm) , l 0.5 1.0 og , 1 /y ) Coryphaena hippurus  Figure 4 The relationship between the growth efficient (K, log, year-1) and asymptotic length (L∞, log, cm) for 340 Mediterranean marine fish stocks. The utliers are circled. co o  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 21 REFERENCES Bond, C.E., 1996. Biology of Fishes. Saunders College Publishing, USA. Campana, S.E., 2001. Accuracy, precision and quality control in age determination, including a review of the use and abuse of age validation methods. J. Fish Biol. 59: 197-242. Froese, R., Binohlan, C., 2001. Empirical relationships to estimate asymptotic length, length at first maturity, and length at maximum yield per recruit in fishes, with a simple method to evaluate length frequency data. J. Fish Biol. 56: 758-773. Froese, R., Pauly, D. (eds.), 2000. FishBase 2000: Concepts, Design and Data Sources. ICLARM, Los Baños, Philippines. Pauly, D., 1998. Tropical fishes: patterns and propensities. J. Fish Biol. 53: 1-17. Stergiou, K.I., Tsikliras, A.C., 2006. Under-representation of regional ecological research output by bibliometric indices. Ethics Sci. Env. Polit. 2006, 15-17. Stergiou, K.I., Christou, E.D., Georgopoulos, D., Zenetos, A., Souvermezoglou, C., 1997. The Hellenic seas: physics, chemistry, biology and fisheries. Ocean. Mar. Biol. Ann. Rev. 35: 415-538.   Trophic levels of North Aegean sea fishes, Karachle, P.K., Stergiou, K.I. 22 TROPHIC LEVELS OF NORTH AEGEAN SEA FISHES AND COMPARISONS WITH THOSE FROM FISHBASE1 Paraskevi K. Karachle, Konstantinos I. Stergiou Aristotle University of Thessaloniki, School of Biology, Department of Zoology, UP Box 134, 541 24 Thessaloniki, Greece; Email:pkarachl@bio.auth.gr ABSTRACT We estimated trophic levels (TROPHs) for 76 species from the north Aegean Sea using diet composition, and compared the estimated TROPHs with those reported in FishBase (TROPHFB). For 41 and 14 out of the 76 species, there is no such information from the Aegean and the Mediterranean Seas, respectively. North Aegean TROPHs were linearly related to TROPHFB (TROPH = 1.24+0.65TROPHFB, R2 = 0.54, p<0.01). INTRODUCTION Although food composition and feeding habits of fishes have been a favored research field for more than a century (e.g., Gerking, 1994), it was only in the early 1940s that the trophic level concept was introduced in ecology (Pauly et al., 2000a). Lindeman (1942) introduced quantitative food webs as a tool in understanding temporal change in aquatic ecosystems in a paper that was initially rejected by two reviewers of the journal Ecology (Sobczak, 2005). Nowadays, the concept of the fractional trophic level (TROPH) is widely used, being of high importance to ecological research (e.g., for estimating ‘primary production required’ to support fisheries, see Pauly and Christensen, 1995; Tudela, 2000; for identifying the ‘fishing down the food webs’ process, see Pauly et al., 1998; for comparative community analysis and construction of trophic signatures, see Pauly et al., 2000b, Froese et al., 2005; and for constructing trophic signatures for fishing gears, see Stergiou et al., 2006; see also CIESM, 2000; Stergiou and Karpouzi, 2002 and references therein). The success of the application of TROPH in ecological research is also demonstrated by the fact that the Marine Trophic Index (MTI), which actually refers to the mean TROPH of the landings for fish with TROPH higher than a cut-off value, has been selected by the Convention of Biological Diversity as one of the eight indices to be tested for use as an indicator of biodiversity changes (Pauly and Watson, 2005). In order for this to be realized, accurate area-specific TROPH estimates for all or most important fish species in the ecosystem must be readily available. However, this is not always the case, as diet composition data are rarely available for the area under study. Thus, one has to use general TROPH values (e.g., for the same species from different areas or from another species of the same genus). Such general values are extracted from various sources, among which FishBase (www.fishbase.org; Froese and Pauly, 2005), the largest electronic encyclopedia for fish, is the most important (see Stergiou et al., 2006). In this report, we estimated TROPH for 76 species from the north Aegean Sea using diet composition. Data were collected within the framework of a project on the feeding habits and TROPHs of fishes in the north Aegean Sea. These data were then compared to those reported in FishBase (TROPHFB). MATERIALS AND METHODS Samples were collected on a seasonal basis, from spring 2001 to winter 2006, using commercial fishing vessels (i.e., trawlers, purse-seiners, and gill-netters), and preserved in 10% formalin. At the laboratory, stomach contents were identified to the lowest possible taxonomic level and weighted (wet weight) to the                                                  1 Cite as: Karachle, P.K., Stergiou, K.I., 2006. Trophic levels of north Aegean Sea fishes and comparisons with those from FishBase. In: Palomares, M.L.D., Stergiou, K.I., Pauly, D. (eds.), Fishes in Databases and Ecosystems. Fisheries Centre Research Reports 14(4), pp. 22-26. Fisheries Centre, University of British Columbia [ISSN 1198-6727].  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 23 nearest 0.01 g (expressed as a percentage of the total stomach content weight) (Hyslop, 1980). For stomachless species, the anterior half of the digestive tract was used for the analysis (e.g., Bell and Harmelin-Vivien, 1983). From the diet content, we estimated TROPH values per species, using the routine for quantitative data of TrophLab (Pauly et al., 2000c), a stand-alone application for the estimation of TROPHs from the contribution to the diet and the TROPH of the prey organisms. When a prey item was an ‘identified fish’ species, then the TROPH of this species (as estimated in this study) was used for the estimation of predator’s TROPH. The diet composition data will be incoporated into FishBase. RESULTS Overall, we studied and estimated feeding habits and TROPHs for 76 fish species (7,124 individuals; Table 1). No information on feeding habits existed for 41 out of the 76 species for the Aegean Sea and for 14 out of the 76 species for the Mediterranean Sea. Finally, no information on the feeding habits of Monochirus hispidus is included in FishBase (date: 4/7/2006; Table 1). The number of individuals examined ranged from 1 for very rare species (Chelidonichthys lastoviza, Dasyatis pastinaca, Dipturus oxyrinchus, Fistularia commersonii and Labrus viridis) to 759 (Engraulis encrasicolus) (Table 1). Table 1 Fractional trophic levels (TROPH) and their standard error (SE) for 76 fishes, north Aegean Sea, Greece. N: number of individuals; TROPHFB: TROPH reported in FishBase (www.fishbase.org; Froese and Pauly, 2005). Species N Length range TROPH TROPHFB Alosa fallax 1,2 27 15.0-46.8 4.32±0.48 3.60±0.60 Anthias anthias 1,2 9 12.7-16.6 3.54±0.52 3.80±0.58 Apogon imberbis 37 8.0-11.5 3.54±0.56 3.90±0.64 Arnoglossus laterna 1 212 4.5-16.9 4.35±0.74 3.60±0.54 Arnoglossus thori 1 3 9.1-11.2 3.61±0.57 3.30±0.53 Belone belone 1,2 69 27.2-53.5 3.48±0.45 4.20±0.74 Blennius ocellaris 1,2 23 7.0-13.7 3.26±0.46 3.50±0.43 Boops boops 106 11.2-19.9 3.52±0.52 3.00±0.12 Bothus podas 1 22 11.3-17.2 3.39±0.53 3.40±0.49 Caranx rhonchus 1,2 16 18.0-19.8 4.50±0.80 3.60±0.59 Cepola macrophthalma 195 13.2-54.9 3.13±0.31 3.20±0.30 Chelidonichthys lastoviza 1 16.9 3.32±0.49 3.40±0.50 Chelidonichthys lucernus 1 15 6.0-21.6 3.64±0.63 3.70±0.61 Chromis chromis 1 97 8.6-13.3 3.25±0.37 4.10±0.70 Citharus linguatula 170 3.9-24.3 4.34±0.69 4.00±0.65 Conger conger 1 31 34.1-99.8 4.18±0.58 4.30±0.75 Coris julis 78 11.3-18.2 3.42±0.53 3.20±0.45 Dalatias licha 1 2 38.0-40.2 4.50±0.39 4.20±0.66 Dasyatis pastinaca 1 1 50.1 3.46±0.53 4.10±0.63 Dentex dentex 10 11.7-15.3 4.49±0.80 4.50±0.70 Diplodus annularis 427 6.1-17.5 3.20±0.43 3.40±0.44 Diplodus vulgaris 1 50 9.0-16.7 3.08±0.28 3.20±0.37 Dipturus oxyrinchus 1,2 1 81.2 3.60±0.59 3.50±0.37 Engraulis encrasicolus 1 759 5.8-14.0 3.38±0.44 3.10±0.51 Eutrigla gurnardus 1 10 6.3-14.8 3.58±0.58 3.60±0.57 Fistularia commersonii 1 92 4.50±0.80 4.30±0.74 Gaidropsarus biscayensis 1 65 9.0-15.3 3.93±0.67 3.60±0.54 Gaidropsarus mediterraneus 1 15 8.5-14.5 3.95±0.61 3.40±0.53 Galeus melastomus 1 3 23.3-54.0 4.50±0.41 4.20±0.58 Labrus viridis 1 1 19.5 3.29±0.51 3.80±0.64 Lepidotrigla cavillone 4 8.8-11.7 3.50±0.50 3.20±0.43 Lesueurigobius suerii 1,2 141 5.8-9.4 3.35±0.43 3.60±0.50 Lophius budegassa 45 5.0-38.4 4.54±0.60 4.50±0.76 Lophius piscatorius 6 7.7-12.7 4.48±0.54 4.50±0.76 Merlangius merlangus 1,2 41 14.1-29.1 4.38±0.73 4.40±0.77 Merluccius merluccius 21 11.7-37.0 4.45±0.74 4.40±0.78 Microchirus variegatus 1,2 3 9.1-10.6 3.06±0.26 3.30±0.45 Micromesistius poutassou 77 9.2-24.0 4.18±0.66 4.00±0.68 (Continued on next page)        Trophic levels of North Aegean sea fishes, Karachle, P.K., Stergiou, K.I. 24 Table 1 (continued)     Species N Length range TROPH TROPHFB Monochirus hispidus 1,2,3 24 9.2-12.8 3.19±0.32 - Mullus surmuletus 55 9.1-23.1 3.19±0.37 3.40±0.51 Oblada melanura 1 56 12.6-22.7 3.11±0.42 3.00±0.12 Pagellus acarne    r t   63 10.5-19.2 3.84±0.55 3.50±0.45 Pagellus bogaraveo 72 9.3-23.1 4.43±0.76 3.50±0.46 Pagellus erythrinus 59 8.4-16.4 3.30±0.39 3.40±0.47 Pagrus pagrus 10 10.2-15.5 3.36±0.34 3.70±0.61 Parablennius gattorugine 4 13.4-17.9 2.11±0.09 2.90±0.29 Phycis blennoides 1 20 8.1-37.4 3.55±0.59 3.70±0.58 Pomatomus saltatrix 1,2 6 13.1-18.5 4.50±0.80 4.50±0.55 Raja clavata 1,2 7 25.6-46.5 3.90±0.67 3.80±0.59 Raja miraletus 1 3 22.6-33.9 3.82±0.54 3.80±0.74 Raja radula 3 21.8-32.0 3.97±0.69 3.70±0.54 Sardina pilchardus 752 7.6-16.7 3.14±0.29 2.80±0.23 Sardinella aurita 230 8.4-23.9 3.20±0.32 3.00±0.00 Sarpa salpa 1 25 11.7-19.5 2.00±0.00 2.00±0.00 Sciaena umbra 1 11 12.2-16.0 3.53±0.54 3.70±0.65 Scomber japonicus 1,2 371 8.8-26.8 3.99±0.57 3.10±0.43 Scomber scombrus 204 13.3-27.4 4.37±0.54 3.70±0.56 Sco paena no ata1 42 8.3-17.8 3.60±0.62 3.50±0.50 Scorpaena porcus 96 8.2-26.4 3.90±0.69 3.90±0.65 Scyliorhinus canicula 1 34 24.1-45.1 4.41±0.58 3.70±0.55 Serranus cabrilla 34 9.5-23.1 3.90±0.67 3.40±0.47 Serranus hepatus 99 5.7-13.1 3.77±0.63 3.50±0.56 Serranus scriba 81 10.6-23.6 3.94±0.66 3.80±0. 62 Sphyraena sphyraena 1,2 104 21.6-45.1 4.30±0.46 4.00±0.51 Spicara maena 282 9.0-20.2 3.24±0.34 4.20±0.70 Spicara smaris 118 7.0-18.5 3.49±0.46 3.00±0.04 Spondyliosoma cantharus 1 82 9.7-14.0 3.41±0.46 3.30±0.43 Symphodus tinca 221 11.1-22.0 2.95±0.25 3.10±0.45 Symphurus nigrescens 1 10 6.4-11.9 3.35±0.51 3.30±0.43 Torpedo marmorata 1 118 8.8-37.3 4.39±0.67 4.50±0.80 Trachinus draco 1 25 15.0-30.5 4.19±0.66 4.20±0.71 Trachurus mediterraneus 627 7.0-25.8 4.01±0.64 3.60±0.58 Trachurus trachurus 133 6.3-3.9 3.58±0.50 3.60±0.58 Trisopterus minutes 167 5.7-24.5 4.13±0.64 3.80±0.53 Uranoscopus scaber 70 8.7-26.9 4.43±0.75 4.40±0.70 Xyrichtys novacula 1 12 12.3-17.1 3.37±0.51 3.10±0.32 1 no TROPH estimates available from the Aegean Sea; 2 no TROPH estimates available from the Mediterranean; 3 no TROPH estimates available in FishBase (date: 4/7/2005).  TROPH values ranged from 2.00 (±0.00) for Sarpa salpa, which feeds exclusively on algae, to 4.54 (±0.60) for Lophius budegassa, which preys on fish (Table 1). Differences between TROPHs and TROPHFB (Table 1; Figure 1a) ranged from 0.00 (for Pomatomus saltatrix, Sarpa salpa and Scorpaena porcus) to 0.96 (for Spicara maena), with the mean difference being 0.29±0.03 units. In addition, the difference for more than 75% of the species was smaller than 0.40 (Figure 1a). According to the functional trophic groups identified by Stergiou and Karpouzi (2002) for the Mediterranean, approximately one-third of the TROPHs estimated in this study classified the corresponding species in a different functional group than using TROPHFB. Our TROPHs were linearly related to TROPHFB (R2 = 0.54, p<0.01; Figure 1b). However, no relationship was found (p = 0.16) between the number of stomachs examined and the TROPH-TROPHFB difference for the studied species (Figure 2).   Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 25 0.0 25.0 50.0 0-0.2 0.2-0.4 0.4-0.6 0.6-0.8 0.8-1.0 |TROPH-TROPHFB| %  n u m b er  o f sp ec ie s 1.50 2.50 3.50 4.50 1.50 2.50 3.50 4.50 TROPHFB T R O P H TROPH = 1.2392+0.6459TROPHFB N=75, R2 = 0.54, SEb=0.09, p<0.01 (a) (b)42.7 (32) 8.0 (6) 6.7 (5) 9.3 (7)%  n u m b er  o f sp ec ie s T R O P H TROPH-TROPHFB TROPHFB TR O P H %  n um be r of  s pe ci es %  n u m b er  o f sp ec ie s T R O P H %  n u m b er  o f sp ec ie s T R O P H TR O P H %  n um be r of  s pe ci es  Figure 1 Troph of north Aegean Sea fishes: (a) Frequency distribution of the differences between the fractional trophic levels (TROPHs) estimated in this study for 75 fishes compared with TROPH values reported in FishBase (TROPHFB; www.fishbase.org). The percentages and the number of species (in parentheses) are also shown. (b) Relationship between TROPH and TROPHFB for 75 species (the dashed line indicates the 1:1 relationship). DISCUSSION Our results indicate that differences between area-specific TROPHs from the general ones reported in FishBase are generally small, with a mean difference of only 0.29 units; for more than 40% of the species studied the difference was smaller than 0.2 units (Figure 1a). Yet, in a few cases the difference can be as high as one full trophic level. Still, this indicates that in the absence of regional estimates, those from FishBase are the best estimates available. Observed differences between our TROPHs and TROPHFB could be attributed mainly to: (a) spatio-temporal differences in feeding habits, reflecting variations of prey abundance in different ecosystems (e.g., Scomber japonicus, Scyliorhinus canicula); (b) different length ranges studied (e.g., Pagrus pagrus); and/or (c) different methods of estimating prey contribution (e.g., qualitative: numerical, frequency of occurrence; quantitative: gravimetric, volumetric) (e.g., Arnoglossus laterna, Scomber scombrus). Although small sample size might also be responsible, the fact that there was no relationship between TROPH-TROPHFB and number of stomachs examined indicates that this is not probable. In addition, the absence of such a relationship (Figure 2) also indicates that small stomach number is not an important impediment for a preliminary TROPH estimation. The high difference for Spicara maena should be attributed to the fact that the species in the Mediterranean, and hence in the Aegean, feeds mainly on copepods (54.2% present study; 6-100%, Stergiou and Karpouzi, 2002). For this species the p between TROPH and TROPH  for species for which regional use TROPHFB est TROPHs are available should be reported diet in FishBase, from which TROPH is estimated, is composed mainly of zooplankton (Pinnegar and Polunin, 2000) and bony fish (Khoury, 1984). The relationshi FB TROPH values are available could be d to refine the general values for species for which no regional imates are available (e.g., rare species). To generate more accurate predictions, the number of species for which regional increased. Number of stomachs (log) N=75, R2 = 0. 03, p=0.161.20 0.00 0.40 0 1 2 3 R O P H -T 0.80 R O P H F B | |T R O P H -T R O P H F B | |T R O P H -T R O P H F B | |T R O P H F B | R O P H -T  |T Figure 2 Relationship between the number of stomachs examined and the difference of fractional trophic levels (TROPH) estimated in this study from the FishBase ones (TROPHFB).  Trophic levels of North Aegean sea fishes, Karachle, P.K., Stergiou, K.I. 26 RE an Posidonia oceanica seagrass meadows. 2. CIE 2000. Fishing down the Mediterranean food webs? CIESM Workshop Series, 12. FERENCES Bell, J.D., Harmelin-Vivien, M.L., 1983. Fish fauna of French Mediterrane Feeding habits. Tethys 11, 1-14. SM, (www.ciesm.org/online/monographs/Corfou.html). ese, R., Pauly, D. (eds.), 2005. FishBase. World Wide Web electronic publication. URL: Fro www.fishbase.org. Froese, R., Garthe, S., Piatkowski, U., Pauly, D., 2005. Trophic signatures of ma as compared with other ecosystems. Belg. J. Zool. 135 (2), 139-143. rine organisms in the Mediterranean s. In: Boudouresque C.F., Jeudy de Grissac A. and Olivier J. (eds.), International Workshop Posidonia Pau 74, 255–257. oc. (Ser. B) 360, 415-423. , W.V., 2005. Lindeman’s trophic dynamic aspect of ecology: ‘Will you still need me when I’m 64?’ Limn. Ocean. Bull. 14 (3), 53-57. Stergiou, K.I., Karpouzi, V.S., 2002. Feeding habits and trophic levels of Mediterranean fish. Rev. Fish Biol. Fish. 11, 217-254. Stergiou, K.I., Moutopoulos, D.K, Casal, J.A.H., Erzini K., 2006. Trophic signatures of small-scale fishing gears and their implications for conservation and management. Mar. Ecol. Prog. Ser. [in press]. Tudela, S., 2000. Assessment of the ecological footprint of fishing in the Catalan central coast (NE Spain). In: Briand, F. (ed.), Fishing Down the Mediterranean Food Webs?. CIESM Workshop Series 12, pp. 79–82.  Gerking, S.D., 1994. Feeding Ecology of Fish. Academic Press, San Diego. Hyslop, E.J., 1980. Stomach contents analysis – a review of the methods and their application. J. Fish Biol. 17, 411- 429. Khoury, C., 1984. Ethologies alimentaires de quelques espèces de poisons de l'herbier de Posidonies du Parc National de Port-Cro oceanica Beds, GIS Posidonie Publications, France 1, pp. 335-347 Lindeman, R.E., 1942. Trophic-dynamic aspect of ecology. Ecology 23, 399-418. ly, D., Christensen, V., 1995. Primary production required to sustain global fisheries. Nature 3 Pauly, D., Watson, R., 2005. Background and interpretation of the ‘Marine Trophic Index’ as a measure of biodiversity. Phil. Trans. R. S Pauly, D., Christensen, V., Dalsgaard, J., Froese, R., Torres, F. Jr., 1998. Fishing down marine food webs. Science 279, 860–863. Pauly, D., Christensen, V., Froese, R., Palomares, M.L., 2000a. Fishing down aquatic food webs. Am. Scient. 88, 46-51. Pauly, D., Froese, R., Rius, M.J.F.D., 2000b. Trophic pyramids. In: Froese, R. and Pauly, D. (eds.), FishBase 2000: Concepts, Design and Data Sources. ICLARM, Los Baños, Laguna, Philippines, pp. 203–204. Pauly, D., Froese, R., Sa-a, P., Palomares, M.L., Christensen, V., Rius, J., 2000c. TrophLab manual. ICLARM, Los Baños, Laguna, Philippines. Pinnegar, J.K., Polunin, N.V., 2000. Contributions of stable-isotope data to elucidating food webs of Mediterranean rocky littoral fishes. Oecologia 122, 399-409. Sobczak  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 27 DISTRIBUTION RANGES OF COMMERCIAL FISHES AND INVERTEBRATES1 Chris Close, William Cheung, Sally Hodgson, Vicky Lam, Reg Watson, Daniel Pauly The Sea Around Us Project, Fisheries Centre, University of British Columbia, 2202 Main Mall, Vancouver, BC V6T 1Z4 Canada; Email: c.close@fisheries.ubc.ca ABSTRACT Distribution ranges of commercial fish and invertebrates are required by the Sea Around Us Project for mapping of global fisheries catches. However, published ranges exist for only a small fraction of the 1231 taxa, composed of 923 species, 161 genera and 147 higher groups used in the latest version of the mapping process (Version 3.1, representative of catches from 1950 to 2003). This paper summarizes the methods employed by the Sea Around Us Project to reduce potentially global distributions to realistic ranges by identifying key ecological information for each of the 1231 commercial taxa, specifically: (i) presence in FAO area(s); (ii) latitudinal range; (iii) range-limiting polygons; (iv) depth range; and (v) habitat preferences. Furthermore, this paper presents an additional filter that outlines how (ii) and (iv) are used to correct the depth range for the effect of ‘equatorial submergence.’ Several examples are used to illustrate this process, notably the Florida pompano (Trachinotus carolinus) and the Silver hake (Merluccius bilinearis). Throughout this paper, the data sources emphasized include FishBase, other fish and invertebrate databases, and online information where applicable. In addition, simple heuristics are used to replace ecological information that is unavailable or missing. It should be noted that the Sea Around Us Project does not explicitly use temperature and primary production for any of the procedures discussed in this paper. The purpose of this is to allow for subsequent analyses of distribution ranges using these variables. INTRODUCTION The Sea Around Us Project, hosted at the Fisheries Centre, University of British Columbia, is a research initiative devoted to documenting the effects of fisheries on marine ecosystems worldwide and to propose methods to mitigate these impacts. One of the key elements of this work is mapping of marine fisheries catches onto the ecosystems from which they were extracted. The approach used therein is documented in Watson et al. (2004) and its results, regularly updated, are available on the project website (www.seaaroundus.org). This mapping approach depends crucially on the availability of distribution ranges for all taxa (species, genera, etc) reported in marine fisheries catch statistics. Previous mapping of catches relied on distributions constructed from a mixed set of ecological information that resulted in varying degrees of accuracy. This paper, therefore, documents a major revision of all commercial distribution ranges (totaling 1231 for the time period 1950 – 2003) using a set of rigorously applied filters that markedly improved the accuracy and appearance of the Sea Around Us Project maps and other products. These filters include: (i) presence in FAO area(s); (ii) latitudinal range; (iii) range-limiting polygons; (iv) depth range; (v) habitat preferences; and (vi) accounting for the effect of ‘equatorial submergence’ (Ekman, 1967) Two sample taxa are used to illustrate the results of the filter process, the Florida pompano (Trachinotus carolinus) and the Silver hake (Merluccius bilinearis), each representing pelagic and demersal species, respectively. Other species are used to illustrate specific aspects of this filter process, and are referred to in the appropriate section.                                                  1 Cite as: Close, C., Cheung, W., Hodgson, S., Lam, V., Watson, R., Pauly, D., 2006. Distribution ranges of commercial fishes and invertebrates. In: Palomares, M.L.D., Stergiou, K.I., Pauly, D. (eds.), Fishes in Databases and Ecosystems. Fisheries Centre Research Reports 14(4), pp. 27-37. Fisheries Centre, University of British Columbia [ISSN 1198-6727].  Distribution of commercial fishes and invertebrates, Close, C. et al. 28 The procedures presented here avoid use of temperature and primary production to define or refine distribution ranges for any of the taxa. This was done in order to allow for subsequent analyses of distribution ranges to be legitimately performed using these variables. This differs from previous construction methods of distribution maps that used primary production to distinguish area of low vs. high abundance within a taxon’s distribution range (Watson et al., 2004). MATERIAL AND METHODS The ‘filters’ used here are listed in the order that they are applied; each filter is documented with a figure and a short description of major sources for the information required at that level. Prior to the ‘filter’ approach presented below, the identity and nomenclature of each taxon was verified using FishBase (www.fishbase.org) and other sources, and the English common names and scientific names were updated. Filter 1: FAO Area The United Nations Food and Agriculture Organization (FAO) has divided the world’s oceans into 18 areas for statistical reporting purposes (Figure 1). Information on the occurrence of commercial taxa within these areas is available primarily through: (a) FAO publications and the FAO website (www.fao.org) and (b) FishBase. Figures 2a and 3a illustrate FAO area occurrence of Silver hake and Florida pompano respectively. Filter 2: Latitudinal range The second filter applied in this process is latitudinal ranges. Charles Darwin, after reviewing literature on the distribution of marine organisms, concluded that “latitude is a more important element than longitude” (see Pauly 2004, p. 125, for the sources of this and the quote below). This does not mean, however, that longitude and other factors do not play a role in determining a taxon’s distribution. Still, in the following quote, Darwin illustrates how latitude provides the key to understanding the composition of certain fauna: “Sir J. Richardson says the Fish of the cooler temperate parts of the S. Hemisphere present a much stronger analogy to the fish of the same latitudes in the North, than do the strictly Arctic forms to the Antarctic.” Latitudinal range is defined as a taxon’s northernmost and southernmost latitudes of what is considered their ‘normal’ distribution range and can be found in FishBase for most fishes. For other fishes and invertebrates, latitudes were inferred from the latitudinal range of countries that reported them, and/or from occurrence records in the Ocean Biogeographic Information System website (OBIS; www.iobis.org). A further refinement of a taxon’s latitude range can be defined by its relative occurrence throughout its latitudinal range. From first principles, a taxon can be assumed to be most abundant at the center of its range (McCall, 1990). In cases of distributions confined to either of the two hemispheres, this is approximated by a symmetrical triangular distribution peaking at the mean of the northernmost and southernmost latitudes. For distributions that straddle the equator, it is assumed that a taxon’s range can be broken into three parts – the outer two thirds and the inner or middle third. If the equator falls within one of the outer thirds of the latitudinal range, then the abundance is assumed to be the same as above, and thus the symmetrical triangular distribution can be applied. If, however, the equator falls in the middle third of the range, then the abundance distribution is assumed to be flat in the middle third and decreasing to the poles for the remainder of the distributions range. Figures 2b and 3b illustrate the result of the FAO and Latitudinal filter combined. Both the Silver hake and the Florida pompano follow the symmetrical triangular distribution as noted above.  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 29  71 61 37 88 81 34 2167 31 4741 81 18 88 71 87 48 18 57 51 61 58 27 77  FAO #  FAO Name FAO # FAO Name FAO # FAO Name 21 Atlantic, North West  48 Antarctic (South Atlantic) 87 Pacific, South East  27 Atlantic, North East 51 Indian, West  18 Arctic Sea 31 Atlantic, West Central  57 Indian, East 61 Pacific, North West   Antarctic (South Indian) 71 Pacific, West Central  Pacific, North East 81 Pacific, South West   that the United Nations Food and Agriculture Organization (FAO) uses for statistical reporting purposes. tably those of FAO (species catalogues, species identification sheets, guides to the commercial species of various countries or regions), and in various online sources. For taxon without published polygons, the filters described in this paper were used to generate range maps from which polygons were then drawn. In the case of many invertebrates, however, this procedure was reversed, whereby the countries that reported the taxon are used as the taxon’s occurrence. In these instances, particular emphasis was given to the FAO statistics, where countries that reported the taxon in their catch were used as occurrence. However this method was not used if the taxon was caught by the country’s distant water fleet. 34 Atlantic, East Central  58 37 Mediterranean & Black Sea 67 41 Atlantic, South West  77 Pacific, East Central  88 Antarctic (South Pacific) 47 Atlantic, South East        Figure 1 The 18 areas of the world’s oceans  Filter 3: Range-limiting polygons The third filter in the distribution process is the use of range-limiting polygons.  Range-limiting polygons help to confine species in areas where they are known to occur and also to prevent occurrence in semi- enclosed seas (e.g., of low salinity) where the taxon does not occur, but which are otherwise located within its FAO areas, latitude and depth ranges. Polygonal distributions for a vast number of species of commercial fish and invertebrates can be found in various publications, most no  Distribution of commercial fishes and invertebrates, Close, C. et al. 30 In addition to the above polygonal methods, faunistic works that cover the high-latitude end of continents and/or semi-enclosed coastal seas with depauperate faunas (e.g., Hudson Bay, or the Baltic) were used to avoid, where appropriate, distributions reaching into these extreme habitats. Polygons were then drawn resembling those published for similar species, i.e., at similar distances from coastlines.2  Figure 2 Sequence of filters used for deriving the species distribution range of the Silver hake (Merluccius bilinearis): (a) illustrates the Silver hake’s presence in FAO areas 21 and 31; (b) illustrates the result of applying the FAO and latitudinal range (24°S to 62°N); (c) shows the result of applying the FAO, latitudinal, and the range-limiting polygon; and (d) illustrates the final result after the application of the four filters. All available polygons, whether available from a publication or newly drawn, were digitized used ESRI’s ArcGIS and stored in the Sea Around Us Project’s database, along with the latitude ranges derived from them, which were then used for inferences on equatorial submergence (see below). Figures 2c and 3c illustrate the result of the combination of the first three filters, i.e., FAO, latitude and range-limiting polygons. These parameters and polygons will be revised periodically, as our knowledge of the species in question increases. Habitat parameters for higher taxa It should be noted that, because the Sea Around Us Project mapping process only deals with commercially-caught species, the distribution ranges for higher level taxa (genera, families, etc) were generated using the combination of range polygons from the taxa level below it. Thus, the range polygons for genera were built using the range polygons of the commercial species that fall within them. Similarly, family-level polygons were generated from genus-level polygons, and so on. Latitude ranges, depth ranges and habitat preferences were expanded in the same manner. While this procedure does not mimic the true distribution of the genera in question, which usually consists of more species than are reported in catch statistics, it is likely that the generic names in the catch statistics refer to the very commercial species that are used to generate the distribution ranges, as these taxa are frequently more abundant. However, to avoid misunderstandings, the number of species used in generating such generic distribution ranges will be made visible where appropriate, and the maps will be referred to as catch distributions, rather than taxon range distributions.                                                  2 Some of these polygons were obtained by making our GIS system (see below) ‘buffer’ the distribution ranges resulting from Filter 1, 2 and 4. This yielded polygons slightly different in appearance from the others, but which met our needs, nevertheless.  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 31 Filter 4: Depth range Similar to the latitudinal range, the ‘depth ned as “[t]he depth range (in fication mercial OBIS where, in some cases, the deepest range is its relative abundance within the ater column. Based on Alverson et al. (1964), Pauly and Chua (1988), Zeller and Pauly (2001) and other sources, it was assumed that the abundance of a taxon within the water column follows a triangular distribution, whereby a taxon’s maximum abundance, approximated by a scalene triangle, occurs in the top one-third of its depth range. Note that with full implementation of ‘equatorial submergence’ (described below as ‘Filter 6’), the depth of maximum abundance will vary with latitude. Filter 5: Habitat preference Habitat preference is an important factor affecting the distribution of marine taxa. Thus the aim of this filter is to enhance the predictions of a taxon’s distribution based on its association with different habitats. etermined mber of habitats that a taxon associates with in that same cell, and by ow far the association effect will extend from that habitat. The latter is assumed to be a function of the taxon’s body size (maximum length) and its habitat ‘versatility’. Thus a large species that inhabits a wide range’, i.e., “[t]he depth (in m) reported for juveniles and adults (but not larvae), from the most shallow to the deepest [water]”, is available from FishBase for most fish species, along with the common depth, defi m) where juveniles and adults are most often found. This range may be calculated as the range within which approximately 95% of the biomass occurs” (Froese et al., 2000). When the depth range for a taxon was not available, it was obtained from FAO (species catalogues, species identi sheets, and guides to the com species of various countries or regions), or online sources. One of these sources was record was taken to estimate a taxon’s maximum depth. Where no information was available, the depth range of a similar species was applied. A further refinement of a taxon’s depth  Figure 3 Sequence of filters used for deriving the species distribution range of the Florida pompano (Trachinotus carolinus): (a) illustrates the Florida pompano’s presence in FAO areas 21, 31, and 41; (b) illustrates the result of applying the FAO and latitudinal range (43°S to -9°N); (c) shows the result of applying the FAO, latitudinal, and the range-limiting polygon; and (d) illustrates the final result after the application of the four filters. w In this context, it is assumed that the relative abundance of a taxon in a spatial cell is, in part, d by the fraction derived from the nu h range of habitats is more likely to occur far from the habitat(s) with which it is associated, than a small species of low habitat versatility (Kramer and Chapman, 1999).   Distribution of commercial fishes and invertebrates, Close, C. et al. 32 B 0.0 0.2 0.4 0.6 0 50 100 150 200 250 Maximum length (cm) D eg re e of  m em be rs hi p Small Medium Large 0.8 1.0 r eMediuSma l A D eg re e of  m em be rs hi p   0.0 0.2 0.4 0.6 0 0.2 0.4 0.6 0.8 1 Versatility D eg re e of  m em b Low Moderate High 0.8 1.0 er sh ip HiMod tew D eg re e of  m em be rs hi p  Figure 4 Fuzzy membership functions for the three categories of (A) maximum versatility. Habitat versatility is defined as ratio of number of habitat types in which a tax  length and (B) taxon’s on occurs to the total number of defined habitat types.  Table 1 Habitat categories used here, and for which global maps are available in the Sea Around Us Project, with some of the terms typically associated with them (in FishBase and other sources). Categories Specifications of global map Terms often used Estuary Alder (2003) Estuaries, mangroves, river mouth Coral UNEP World Cons. Monit. Cent. (2005) Coral reef, coral, atoll, reef slope Seagrass Not yet available* Seagrass bed Seamounts Kitchingman and Lai (2004) Seamounts – Muddy/sandy/rocky bottom Other habitats Continental shelf NOAA (2004) Continental shelf, shelf Continental slope NOAA (2004) Continental slope, upper/lower slope Abyssal NOAA (2004) Away from shelf and slope Inshore NOAA (2004) Shore, inshore, coastal, along shoreline Offshore NOAA (2004) Offshore, oceanic * The Sea Around Us Project is currently developing a global map of seagrass which will be applied when available.  The maximum length and versatility of a taxon are classified o p  a ned membership functions (Figure  (membership ranges from 0 to 1). Th e a a Around Us Project da e, v d is define e ta tats (Table r B el nd inshore/offshore), the h the versatility of Striped bass is classified as low to m 0.48 and 0.52 respectively. Determining habitat association Based on qualitative descriptions from the published sources s such as FishBase and/or through personal communications from experts, each taxon’s degree of asso ation with d 1) for all exploited taxa in the database was determined. The taxon’s degree of association ab determined from the qualitative descriptions relating to it sity or co nness e part habitat (Table 2). As noted above, Striped bass prefers estua o oc the s’. the Striped bass received a score of 0.75 for estuaries and 0.5 for ‘other habitats’  into three categories (Figure 4), and it is ries with different degrees of membership (0 ility’ that the taxon is associated with that a pre-specified membership function for each le, the Striped bass (Morone saxatilis) has assumed that a taxon can associate with one or more categ to 1). A higher membership value means a higher ‘probab particularly category. The membership values are defined by of the length and versatility categories (Figure 4). For exam maximum length of 200 cm (TL). Thus, based on the defi Striped bass has a large body size with a membership of 1 maximum length estimates for all of the 1231 exploited t obtained from FishBase and other published literature for in In this paper, versatility refers to the taxon’s ability to inhabit ratio between the number of associated habitats to the to instance, based on descriptions in the SPECIES Table of Fish and ‘other habitats’. Given that the total number of defined seagrass, seamount, other habitats, while excluding sh versatility of Striped bass is estimated to be 0.4. Based on t right),  4A, left), ere ar xa in the Se tabas ertebrates.  different habitat types an d as th l number of defined habi  1). Fo ase, Striped bass is associated with estuaries  physical habitats is five (coral reef, estuary, f/slope/abyssal a e defined membership functions (Figure 4B, edium, with a membership of approximately , database ci ifferent habitats (Table  to each h itat is s den mmo in th icular ries and als curs in ‘o r habitat Thus, .  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 33 Maximum distance of habitat effect Maximum distance of habitat effect (maximum effective distance) refers to the maximum distance from the nearest perimeter of the habitat within which the ‘attraction’ effect to their associated taxa exists. This is defined as the maximum effective distance by the maximum length and habitat versatility of the taxa using a heuristic rule matrix (Table 3). For example: IF imum length is large (1) AND versatility is moderate (0.52), THEN maximum occurrence distance from the associated habitat is high (0.52). Here, the number in p represents the degree of membership to the cat e  fi equals the taxon’s average m  The maximum effective distance from the associated habitat can be estimated from the ‘centroid value’ h conclusion categories, weighted by a taxon’s degree of 0 km, respectively. Thus if, for example, a taxon has membership values of 0.2 and 0.5 to small and Estimating relative abundance in a spatial cell  the computations. Firstly, it is assumed that the habitat always outine provides an explicit and consistent way to incorporate habitat considerations into distribution Table 2 Common descriptions on taxa’s relative association to habitat and their assigned weighting factor. The weighting factor for ‘other habitats’ is assumed to be 0.1 when no information on habitat association is available. Description Weighting factor Absent/rare 0.00 Occasionally, sometimes 0.25 Often, regularly, seasonally* 0.50 max Usually, abundant in, prefer 0.75 Always, mostly, only occurs 1.00 * If a taxon occurs in a habitat, but no description on the strength of the association is found, we assume a default score of 0.5. Table 3 Heuristic rules that define the maximum effective distance from the associated habitat. The bolded columns and rules represent the predicates (categories of maximum body size and arentheses egories. In  minimum the same nal degree embership value.  this example, the degree of membership is th memberships of the two predicates. When conclusion is reached from different rules, the of membership  taxon’s versatility), while the italics represent the resulted categories of maximum effective distance.  Maximum body size    Versatility Small Medium Large Low Small Small Small Moderate Moderate Moderate Large High Moderate High High of eac Figure 5 Maximum effective distance for Striped bass (Morone saxatilis) estimated from the habitat versatility and maximum length of that species (see text). membership. The centroid values for small, medium and large maximum effective distances were defined as 1 km, 50 km and 10 medium maximum effective distance, respectively, then the estimated maximum effective distance is: (0.2*1 + 0.5*50 + 0*100)/(0.2 + 0.5 + 0) = 36.1 km (Figure 5). The maximum effective distance is calculated for all exploited taxa in the database. Several assumptions are made to simplify occurs in the centre of a cell and is circular in shape. Secondly, the density of a taxon (per unit area) is assumed to be the same across any habitat types. Also, it is assumed that a linear decline in density from the habitat perimeter to the taxon’s maximum effective distance occurs for each taxon. Given these assumptions, the total relative abundance of a taxon in a cell equals the sum of abundance on and around its associated habitat: B’T = (αj + αj+1 · (1 – αj)) · (1 – A) … 1) where B’T is the final abundances, αj is the density away from the habitat from cell j, and A is the habitat area of the cell. The relative abundance resulting from the different habitat types is the sum of relative abundance, and is weighted by their importance to the taxon. Although these assumptions on the relationship between maximum length, habitat versatility and maximum distance from the habitat may render predicted distributions at a fine spatial scale uncertain, this r ranges.  Distribution of commercial fishes and invertebrates, Close, C. et al. 34 Filter 6: Equatorial submergence The submergence phen Richardson, that Arctic for omenon was already known to Charles Darwin, who wrote that “we hear from Sir J. ms of fishes disappear in the seas of Japan & of northern China, are replaced by radiolarian, drew attention to it. In most cases, including those which interest us here, submergence As noted above, there is little information on the depth distribution of most commercial species. As a  lower latitude limit (Llow). These four data points are often available in FishBase for fishes, and can be readily inferred for commercial invertebrates, as noted above. If it is for the upper limits of the depth distribution (Phigh), and one for the lower limits (Plow), with the assumption that both Phigh and Plow are ymmetrical about the Equator. In addition, maximum depths are assumed not to change poleward of 600 even distribution of the temperature gradient can be mimicked by constraining Phigh to be less concave than Plow. This is achieved by setting Dgm, the geometric mean of Dhigh and Dlow, as the lowest depth that Phigh can attain. Furthermore, in the case of a distribution spanning both hemispheres, Plow will have its lowest point (Dlow) at the Equator. Finally, it is assumed that if a computed Phigh intercepts zero depth at lower latitudes than 600 N and S, then Phigh is recomputed using the three points D0N=0 at 600 N, D0S at 600 S, and Dhigh and its latitude, which jointly define a parabola. Figure 6 illustrates three cases of submergence based on different constraints. When this process is applied to a distribution range based on latitudinal range and depth that does not account for submergence, the plots in Figure 7 have the effect of ‘shaving off’ the shallow end depth values at low latitudes, and similarly, shaving off the deep end depth values at high latitudes. This will have the effect of narrowing the habitat temperature ranges of the corresponding species.   other assemblages in the warmer latitudes & reappear on the coast of Tasmania, southern New Zealand & the Antarctic islands” (Pauly 2004, p. 198). Eckman (1967) gives the current definition: “animals which in higher latitudes live in shallow water seek in more southern regions archibenthal or live in shallow water seek in more southern regions archibenthal or purely abyssal waters […]. This is a very common phenomenon and has been observed by several earlier investigators. We call it submergence after V. Haecker [1906-1908] who, in his studies on pelagic increases towards the lower latitudes and therefore may be called equatorial submergence. Submergence is simply a consequence of the animal’s reaction to temperature. Cold-water animals must seek colder, deeper water layers in regions with warm surface water if they are to inhabit such regions at all.” Modifying the distribution ranges to account for equatorial submergence requires accounting for two constraints: (1) data scarcity; and (2) uneven distribution of environmental variables (temperature, light, food, etc.) with depth. result, only the following four data points were available for each taxon, namely: the shallow or ‘high’ end of the depth range (Dhigh), its deep or ‘low’ end (Dlow) of the depth range, the poleward limit of the latitudinal range (Lhigh), and its assumed that equatorial submergence is to occur, then it is logical to also assume that Dhigh corresponds to Lhigh, and that Dlow corresponds to Llow. Data scarcity can be further mitigated by assuming the shape of the function linking latitude and equatorial submergence. In this context, two parabolas are used, one s N and S. The un  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 35 (a) (b)     (c) Figure 6 Illustrative representations of ‘equatorial submergence’, given different depth/latitude data: (a) Case 1: Barndoor skate (Dipturus laevis) – When the shallow end of the depth range (Dhigh) is at lower latitudes than 600 N and S, the upper limit of the depth distribution (Phigh) is assumed to intercept zero at 600 N and S; (b) Case 2: When distribution range is spanning the North and South hemispheres, as in the case of the Warsaw grouper, Epinephelus nigritus, the lowest point of the lower limit (Plow) is at the Equator; (c) Case 3: Silver hake (Merluccius bilinearis). The poleward limit of the latitudinal range (Lhigh) is at higher latitudes than 600 N and S.  Distribution of commercial fishes and invertebrates, Close, C. et al. 36 (b)(a)    from the distribution range of the Warsaw Figure 7 ‘Equatorial submergence’ has the effect of ‘shaving off’ areas grouper, Epinephelus nigritus: (a) Original Distribution; (b) Distribution adjusted for ‘equatorial submergence’.  RESULTS AND DISCUSSION The results consist of distribution ranges generated through the above methods, incorporated in the Sea Around Us database, and available online (see www.seaaroundus.org). They can also be accessed (for fish cies) via FishBase (click ‘Sea Around Us distributispe ons’ under the ‘Internet sources’ in the Species Most importa tion ranges will serve as basis for all spatial catch allocation done with the ents or Ch the Marine Conservation Biology Institute for the support of ranges. Ald e coast in the Sea Around Us Project. The Sea Around Us Newsletter. No. 15, 1-2. astern Pacific roese, R., Capuli, E., Garilao,C., Pauly, D., 2000. The SPECIES Table. In: R. Froese, Pauly, D. (eds.), FishBase 2000: Concept, Design and Data Sources. ICLARM, Los Baños, Laguna, Philippines, pp. 76-85. Haecker, V., 1906-1908. Tiefesee-Radiolarien. Wissenschaftliche Ergebnisse der Deutschen Tiefsee-Expedition auf dem Dampfer "Valdivia" 1898-1899 14. Jena, Fischer. Kitchingman, A., Lai, L., 2004. Inferences of potential seamount locations from mid-resolution bathymetric data. In: T. Morato and D. Pauly (eds.), Seamounts: Biodiversity and Fisheries. Fisheries Centre Research Reports 12(5), pp. 7-12. Fisheries Centre, University of British Columbia. Kramer D.L., Chapman, M.R., 1999. Implications of fish home range size and relocation for marine reserve function. Environ. Biol. Fishes 55, 65-79 MacCall, A., 1990. Dynamic Geography of Marine Fish Populations. University of Washington Press, Seattle. Table). ntly, these distribu Sea Around Us Project. Therefore, we would be very thankful for feedback, i.e., suggested comm corrections, which we will strive to implement as soon as possible. ACKNOWLEDGEMENT The work documented here is an output of the Sea Around Us Project, initiated and funded by the Pew aritable Trusts, Philadelphia. We thank Sally Hodgson who worked on the distribution ranges of deep sea fish. We thank, finally, Adrian Kitchingman and Ahmed Gelchu for their work on the first generation of Sea Around Us distribution  REFERENCES er, J., 2003. Putting th Alverson, D.L., Pruter, A.L., Ronholt, L.L., 1964. A Study of Demersal Fishes and Fisheries of the Northe Ocean. Institute of Fisheries, The University of British Columbia, Vancouver. Ekman, S., 1967. Zoogeography of the Sea. Sidgwick & Jackson, London. F  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 37 Millennium Ecosystem Assessment, 2005. Ecosystems and Human Well-being, Wetlands and Water Synthesis. World Resources Institute, Washington, DC. NOAA, 2004. ETOPO2 - 2-Minute Gridded Global Relief  http://www.ngdc.noaa.gov/mgg/fliers/01mgg04.html). Pauly, D., Chua Thia-E d in Southeast Asia. AMBIO J. Human Environ. 1 Pauly, D., 2004. Darwin’s Fishes: An Encyclopedia of Ichthyology, Ecology and Evolution. Cambridge University Press, Cambridge. Watson, R., Alder, J., Kitchingman, A., P ttention. Mar. Pol. 29(3), 281-284 Watson, R., Kit ocus. Fish and Fishe Watson, R., Pauly, D., 2001. Systematic distortions in world fisheries catch trends. Nature 414, 534-536. Zeller, D., Froese, R., Pauly, D., 2005. On losing a ies and marine science data. Mar. Pol. 29, 69-73. Zeller, D., Pauly, D. , 344-355.  Data ng., 1988. The overfishing of marine resources: socioeconomic backgroun 7(3), 200-206. auly, D., 2005. Catching some needed a chingman, A., Gelchu, A., Pauly, D. 2004. Mapping global fisheries: sharpening our f ries 5: 168-177. nd recovering fisher , 2001. Visualisation of standardized life history patterns. Fish and Fisheries 2(4)  Common names of fish families, Palomares, M.L.D. et al. 38 A PRELIMINARY LIST OF ENGLISH COMMON NAMES ity of British Columbia, Nicolas Bailly I Leibniz-Institut für Meereswissenschaften (IfM-GEOMAR) 2006) of J. Nelson’s Fishes of the World, 34 lack English common names, of which 20 have common names ending in ‘id’, i.e., no true common names. Similarly, of n. for many organisms, and their standardization and stabilization, e.g., for ice from a number of available names (see, e.g., Robins et al., 1991 FOR AS YET UNNAMED FISH FAMILIES1 Maria Lourdes D. Palomares The Sea Around Us Project, Fisheries Centre, Univers 2202 Main Mall, Vancouver, BC V6T 1Z4 Canada; Email: m.palomares@fisheries.ubc.ca WorldFish Center, Philippine Office, Natural Resources Management, RRI College, Khush Hall, Los Baños, Laguna 4031, Philippines; Email: n.bailly@cgiar.org Rainer Froese Düsternbrooker Weg 20, Kiel 24105 Germany; Email: rfroese@ifm-geomar.de Daniel Pauly The Sea Around Us Project, Fisheries Centre, University of British Columbia, 2202 Main Mall, Vancouver, BC V6T 1Z4 Canada; Email: d.pauly@fisheries.ubc.ca ABSTRACT Of the 515 families recognized in the last edition ( the 530 families recognized in W.N. Eschmeyer’s 2006 edition of the Catalog of Fishes, 122 lack common names, while 8 have names ending in ‘id’. Given the need for such names in FishBase and other applications, common names were coined for these families, mainly by translating and adapting the scientific names. The common names proposed in this preliminary work do not overlap with already used English common names and meet the criteria of the American Fisheries Society’s (AFS) Committee on Names of Fishes. They are presented here in two separate lists, i.e., those without common names and those ending in ‘id’, from Amarsipidae to Xenisthmidae, with the etymologies and, where required, reasons for the choices taken. We also include a list of common names included in the 2006 edition of J. Nelson’s Fishes of the World, but not in the 1994 edition. INTRODUCTION Common names of organisms serve a number of functions, notably as bridge between specialists and the lay public (Palomares and Pauly, 1998). As such, common names must be widely understandable, and preferably, describe some peculiar and memorable features of the organisms in questio Multiple common names exist legal purposes, merely involves a cho and Nelson et al., 2004).                                                  1 Cite as: Palomares, M.L.D., Bailly, N., Froese, R., Pauly, D., 2006. A preliminary list fish families. In: Palomares, M.L.D., Stergiou, K.I., Pauly, D. (eds.), Fishes in Databa  of English common names for as yet unnamed ses and Ecosystems. Fisheries Centre Research Reports 14(4), pp. 38-48. Fisheries Centre, University of British Columbia [ISSN 1198-6727].  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 39 However, new, small or rare species and higher taxa of plants, animals and other organisms are rarely assigned common names in the scientific literature. In such cases, new common names must be coined. Robins et al. (1991) proposed a series of criteria for new common names in fishes, notably: 1. “A single vernacular name shall be accepted for each species or taxonomic unit included and no two species on the list shall have the same proposed name”; 2. “Only clearly defined and well-marked taxonomic entities shall be assigned common names ately tied to the scientific name) and names intended to honor persons are discouraged in that they are without descriptive value”; 3. nciful, metaphorical, distinctive and original and describe structural attributes, color and color patterns, ecological characteristics and geographic distribution; 4. n as common names. However, commonly-employed names adopted from traditional English usage are given considerable latitude in taxonomic placement;  5. oided if possible, but names in general use need not be rejected on this basis alone”; and 6. ., adaptation of the scientific name to English, e.g., Adrianichthyidae to adrianichthyids, are not considered as true common names. The mo broken of these ategorized taxa. Similarly, the most recent edition (2006) of Eschmeyer’s Catalog of Fishes recognizes 530 families, 72 of which are bro  names ally, the August 2006 version of FishBase (www.fishbase.org (which shall not be intim “Names shall not violate the tenets of good taste” and are appropriate if they are colorful, romantic, fa Native names are welcome for adoptio “The duplication of common names of fishes and other organisms should be av Common names ending in ‘id’, i.e st recent edition (2006) of Nelson’s Fishes of the World recognizes 515 families, 100 of which are down to 206 non-nominal subfamilies. Of these families, 34 do not have true common names; and , 7 have no common names, 20 have names ending in ‘id’, 7 are rec ken down into 138 non-nominal subfamilies. Of these families, 122 lack common names, while 8 have ending in ‘id’. Fin ) recognizes 530 families, 79 of which are broken down into 170 non-nominal subfamilies. Of these families, 74 lack mes and 2 have names ending in ‘id’. nt taxon. On the other hand: when a family is split into subfamilies, the nominal subfamily sometimes takes the same common name as the whole family. When . e second desirable, e ombrida mposite c e of nas, bonitos ist name f one t be difficult to choose between ‘  ‘mac esent n d in kee p nges in taxonomic category. ison of t  l  fam ich no e r  T generat n that j h en shB S AN c na e tr ing G dicti  Perseus, e dictio olvefo common na Note that in this contribution, we only venture to propose common names for the currently-recognized fish families. Proposals for common names of currently-recognized subfamilies (and tribes) could be the subject of a future contribution. For instance, when a subfamily is raised to family level, a new common name is required (e.g., Dalatiidae: Etmopteriinae in Nelson, 1994, raised to Etmopteriidae in Nelson, 2006). The taxa of the family-group (super-, sub-, families and tribes) could carry over their common names irrespective of their rank and/or their pare families are lumped into one family and are taken down to the subfamily level, the new family takes either the common name of the nominal subfamily or a list of the common names of all or some subfamilies Note that th   case is not , which is the l .g, Sc e with the co ommon nam mackerels, tu would then of common tunas’ and s of 3 tribes o kerels’ to repr of the two subfamilies. I the family. The commo names ai ping trace of the taxa irres ective of the cha Compar he lists mentioned above eft us with 25 ilies for wh his list w s true common names ar ed with the in ntioavailable and fo ub  which we suggest English common names. a te it should be s ected to the scrutiny of fis experts, and th  encoded into Fi ase. MATERIAL D METHODS The scientifi mes of the families wer nary anslated us reek and Latin onaries (notably the onlin  at www.perseus.tufts.edu/cgi-bin/res rm), a ly gave the nd/or by f  name descri nt etymolog e ain talo base by W. Eschmeye shBase and also accessib gh the Internet (see www.calacademy.org  consulting the source o y of the type genus of th r as included  Fi the family s, which usually also bed and freque family. The m  source here was the Ca le throu g of Fishes data  in ).  Common names of fish families, Palomares, M.L.D. et al. 40 The translations were then compared with the common names of species in the family, and with the English common names for species and higher taxa already in FishBase. Non-English common names were translated literally using Babel Fish (babelfish.altavista.com), e.g., Chinese names in Chinese characters. Note that not all of the literal translations made sense. Thus, only the recurring words used in the common names of species within a family were noted. If a translation led to a unique common name, and was acceptable using the criteria of Robins et al. (1991), the name was used. Otherwise, a new name was coined, based on marked attributes of the species in the family in question. Also, the following were applied: t perpetuated as common names, e.g., Bathyclupeidae translates In the process of comparing the list of families currently included in FishBase with the lists of scientific h no common names were assigned are included in the lists of suggested common names presented in the following pages. T 1. If a family is monotypic, the common name of the species is used, in plural, to distinguish it from the species name, and to avoid the need for changing the name should more species be joined to the family; 2. Similar to rule 16 of Robins et al. (1991), the common name of the respective order is used in composite names where appropriate; 3. Variations of existing family common names were used only within the same order, e.g., wasp scorpionfishes (new) and scorpionfishes are both in the order Scorpaeniformes; 4. Similar to rule 4 of Robins et al. (1991), simple names were preferred, such as ‘Lutefishes’ instead of ‘Guitar characins’ for Citharinidae; 5. Misleading family names are no straightforwardly to ‘deep sea herrings’; however, species in this family belong to the Order Perciformes and to the Order Clupeiformes. Thus, the name ‘deep-sea scalyfins’ was coined instead from a prominent character of members of this family; and 6. As far as possible, names should be ‘telling’ and easy to remember, i.e., reflect obvious characters or relationships. RESULTS and common names of Nelson (1994; 2006) and Eschmeyer (2006), we found 9 families which Nelson (2006) has recategorized into subfamilies or lumped with other families (Table 1). We also found that some of the families for which Nelson (2006) used common names ending in ‘id’ have FishBase English names. The families for whic able 1. Families recognized in FishBase (www.fishbase.org), Eschmeyer (2006) and Nelson (1994), but reclassified in elson (2006). Included also are families recognized in FishBase but with no proper common name in Nelson (2006). Scientific name Reclassification Order Common nam  N (Nelson, 2006) e Source of common name Adrianichthyidae – Beloniformes Adrianichthyids; Ricefishes Nelson (2006); FishBase (18 August 2006 version) A B B rainbowfishes August 2006 version) Cottocomephoridae Lumped with Cottidae (p. 334) Scorpaeniformes – – Dentatherinidae Subfamily Dentatherininae (p. 273) Atheriniformes Tusked silversides Nelson (1994); FishBase (18 August 2006 version) Eschmeyeridae Lumped with Scorpaenidae (p. 320) Scorpaeniformes – – Lateolabracidae Lumped with Moronidae (p. 344) Perciformes Asian seaperches Eschmeyer (2006) Synaceiidae Tribe Synaceiini (p. 324) Scorpaeniformes Stonefishes Nelson (1994); FishBase (18 August 2006 version) Zanclorhynchidae Lumped with Congiopodidae (p. 327) Scorpaeniformes – – racanidae Subfamily Aracaninae (p. 455) Tetraodontiformes – – adidae Subfamily Badinae (p. 382) Perciformes – – edotiidae Subfamily Bedotiinae (p. 2710 Atheriniformes Madagascar Nelson (1994); FishBase (18   Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 41 The following presents the scientific names (in bold characters) of the 59 families considered here, the Apistidae (Scorpaeniformes, scorpionfishes and flatheads): Greek, ‘apistos’, i.e., ‘suspicious’ (Romero, aeniformes, scorpionfishes and flatheads): Greek, ‘kottos’, i.e., a river fish, derived from ‘kotta’, i.e., ‘head’; Greek, ‘komê’, ‘komes’, i.e., ‘hair’, ‘mane’; Greek, ‘pherein’, i.e., ‘to carry’ Neosebastidae (Scorpaeniformes, scorpionfishes and flatheads): Greek, ‘neos’, i.e., ‘new’; Greek, characterized with dentaries each with one huge fang (Nelson, 1994). Sole representative, Omosudis lowii , 1994). ily is not available in FishBase. Available common names include the words ‘scorpionfish’, ‘rockfish’ and ‘deep-sea’. Four of the 5 species in FishBase are all found in deep waters; the exception is a Order in which they belong (in parenthesis), their etymology, the proposed common name (in bold characters), and where required, the reason for the choice of the proposed name. Separate lists are presented for: (i) 7 families for which Nelson (2006) has not included an English common name; (ii) 20 families whose common names end in ‘id’; (iii) 7 families for which we would like to propose alternative common names to those used in Nelson (2006); and (iv) 25 families for which Nelson (1994) had no common names and the corresponding common names published in Nelson (2006), as well as the likely rationale for the choices that Nelson (2006) made in coining the common names.2 (i) Families without English common names 2002); Latin ‘apis’, i.e., ‘bee’ (Liddel and Scott, 1889). Wasp scorpionfishes. English names are available only for Apistus carinatus (Bloch and Schneider, 1801) and Apistops caloundra (De Vis, 1886) and these include the word ‘waspfish’. ‘Wasp fishes’ is preoccupied by Family Tetrarogidae (Scorpaeniformes). Apistus carinatus is distinguished by a brightly-colored and long pectoral fin which, when spread, deters predators (Kuiter and Tonozuka, 2001). Cottocomephoridae (Scorp (Romero, 2002). Bighead sculpins. ‘Sculpins’ is preoccupied by Family Cottidae (Scorpaeniformes) and ‘bullhead’ is in the composite common name for Family Heterodontidae (Heterodontiformes). Some available Russian common names include the word ‘bighead’, while some Chinese common names include the word ‘frog head’ and ‘shell-lake’. ‘sebastes’, i.e., ‘august’, ‘admirable’, epithet given to the Roman emperor Augustus (Romero, 2002). Gurnard scorpionfishes. Many of the available common names include the words ‘gurnard perch’. Omosudidae (Aulopiformes, grinners): Greek, ‘omo’, i.e., ‘shoulder’; Latin, ‘sudis’, ‘i.e., esox, fish of the Rhine, cited by Plinius 9.15; also ‘stake’ (Romero, 2002). Hammerjaws. Members of this species are Günther, 1887 assigned the FishBase English name of ‘hammerjaw’. Parabembridae (Scorpaeniformes, scorpionfishes and flatheads): Greek, ‘para’, i.e., ‘from the side of’, ‘from’, ‘beside’, ‘alongside’; Greek, ‘bembras’, ‘membras’, i.e., a kind of sprat or anchovy (Liddle and Scott, 1889). Sprat-like flatheads. Members are characterized with a depressed head and pelvic fins below the pectoral base (Nelson Plectrogenidae (Scorpaeniformes, scorpionfishes and flatheads): Greek, ‘plektron’, i.e., anything to strike with, e.g., ‘stick’; Greek, ‘genos’, ‘gene’, i.e, ‘race’ (Romero, 2002). Stinger flatheads. Members of this family have heads usually with spines and ridges, and venom gland in dorsal, anal and pelvic spines (Nelson, 1994). Some Chinese common names include the word ‘flathead’. Setarchidae (Scorpaeniformes, scorpionfishes and flatheads): Latin, ‘saeta’, i.e., a thick, stiff hair on an animal, ‘bristle’ (Liddel and Scott, 1889); and ‘arch’. Deep-sea bristly scorpionfishes. Description of the fam pelagic species.                                                    2 Proposed English common names for families included in an earlier version of this paper provided to J. Nelson before the publication of the 2006 edition of Fishes of the World are marked with asterisks.  Common names of fish families, Palomares, M.L.D. et al. 42 (ii) Families with common names ending in ‘id’ Acestrorhynchidae (Characiformes, characins): Greek, ‘akestra’, i.e., ‘needle’; Greek, ‘rhyngchos’, i.e., ‘jaw’ (Romero, 2002). Smallscale pike characins. Members of this family are characterized by very elongate (pike-like) bodies covered with small scales, conical teeth and strong canines are present on the premaxilla (Nelson, 1994). Available Chinese common names include the words ‘fat carp’, Spanish, Portuguese, German and French names include the word ‘dog’ and ‘tooth’. Note that ‘dogtooth’ is used in composite common names by some species of cardinalfishes, grenadiers, groupers, herrings, lampfishes, snappers and tunas and proposed as English name for Family Cynodontidae (see below). ‘Pike characin’ is names of species in the Family Ambassidae (Perciformes). ‘Pomfrets’ is preoccupied by the Family Aphyonidae (Ophidiiformes, cusk eels): Greek, ‘aphyo’, ‘aphye’, i.e., ‘sardine’, ‘anchovy’, in the sense of lupea’, i.e., ‘sardine’ lizard (Romero, 2002). Largescale deep-sea lizardfishes. ‘Deep-sea lizardfish’ is preoccupied by  Greek, ‘sauros’, i.e., ‘lizard’; k lizardfish’ (retained as the FishBase name) and ‘deep-water greeneye’. nce stringed s to used by two species of Oligosarcus (Characidae, Characiformes) and two species of Acestrorhynchus. ‘Pike characids’ is preoccupied by Family Ctenoluciidae (Characiformes). Amarsipidae (Perciformes, perch-likes): Greek, ‘a’, i.e., without; Greek, ‘marsipos’, i.e., ‘bag’ (Romero, 2002). Bagless glassfishes. These species have translucent bodies without coloration and do not have pharyngeal sacs (Nelson, 1994). Only one common name is available, i.e., for Amarsipus carlsbergi Haedrich, 1969, in Chinese, and which translates to ‘non-pouch pomfret’. ‘Glassfish’ is used in composite Bramidae (Perciformes). ‘Silver pomfret’ is preoccupied by Pampus argenteus (Euphrasen, 1788) (Stromateidae, Perciformes). whitish, with the color of sardine (Romero, 2002). Blind cusk eels. The only English name available, for Aphyonus gelatinosus Günther, 1878, is ‘gelatinous blindfish’. In addition, many of the available Chinese names in FishBase include the root word ‘blind’, which corroborates with the diagnosis that the eyes of members of this Family are rudimentary (Nelson, 1994). Bathyclupeidae (Perciformes, perch-likes): Greek, ‘bathys’, i.e., ‘deep’; Latin, ‘c (Romero, 2002). Deep-sea scalyfins. Members of this family have dorsal and anal fins covered with scales. Some of the available common names include the words ‘deep water’, ‘bottom’ and ‘herring’. ‘Herrings’ is preoccupied by Order Clupeiformes. ‘Scalyfin’ is used in composite common names as adjectives for some members of Pomacentridae, Sciaenidae, Haemulidae and Serranidae (all Perciformes). Bathysauroididae (Aulopiformes, grinners): Greek, ‘bathys’, i.e., ‘deep’; Greek, ‘saurodes’, similar to a Bathysaurus ferox Günther, 1878 (Synodontidae, Aulopiformes). ‘Largescale lizardfish’ is used by Saurida undosquamis (Richardson, 1848) and Saurida brasiliensis Norman, 1935. The available Chinese common name for the sole representative, Bathysauroides gigas (Kamohara, 1952), translates to ‘Large-scale deep sea nine spines fish’. Bathysauropsidae (Aulopiformes, grinners): Greek, ‘bathys’, i.e., ‘deep’; Greek, ‘opsis’, i.e., ‘appearance’ (Romero, 2002). Lizard greeneyes. New family in Nelson (2006; not in August 2006 version of FishBase). Members of this family are mesobenthic and widespread (Nelson, 2006). The available Chinese common name for one of the three species in this family, Bathysauropsis gracilis (Günther, 1878) translates to ‘filament body deep sea nine spines fish’, while the other available English names are ‘blac ‘Greeneyes’ is preoccupied by Family Chlorophthalmidae (Aulopiformes), which are also found in deep waters. Citharinidae (Characiformes, characins): Greek, ‘kitharia’, i.e., ‘cittern’, a Renaissa instrument like a guitar with a flat pear-shaped body, also ‘guitar’ and ‘lute’ (Merriam-Webster online dictionary). Lutefishes. ‘Guitarfishes’ is preoccupied by Family Rhinobatidae (Rajiformes). Cynodontidae (Characiformes, characins): Greek, ‘kyon’, ‘kyonos’, i.e., ‘dog’; Greek, ‘odous’, i.e., ‘teeth’ (Romero, 2002). Dogtooth characins. Members of this family are carnivorous with dentary canine stab prey (Nelson, 1994). Many of the available Brazilian common names include the word ‘dog’ while many of the Chinese common names include ‘carp’. ‘Daggertooths’ is preoccupied by Anotopteridae (Aulopiformes), ‘sabertooth fishes’ by Evermannellidae (Aulopiformes), ‘fangtooths’ by Anoplogasteridae (Beryciformes), and ‘sawtooths’ by Serrivomeridae (Anguilliformes).  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 43 Diplophidae (Stomiiformes, lightfishes and dragonfishes): Greek, ‘diploos’, i.e., ‘double’; Greek, ‘phos’, i.e., ‘light’ (Romero, 2002). Porthole lightfishes. New family in Nelson (2006; not in August 2006 version of FishBase). The available Chinese common names include the words ‘double’, ‘light’, and ‘lamp’. s) and Hemisorubim platyrhynchos (Valenciennes, 1840) (Pimelodidae, Siluriformes); and Poeciliopsis gracilis (Heckel, 1848) (Poeciliidae, iformes, catfishes): Greek, ‘erethismos’, ‘erethizein’, i.e., ‘irritate’ (Romero, 2002). South Asian river catfishes. Members of this family have four pairs of barbels and the adipose fin is lson, Hemiodontidae (Characiformes, characins): Greek, ‘hemi’, i.e., ‘half’; Greek, ‘odous’, i.e., ‘tooth’, ‘teeth’ rds in common. Lophichthyidae (Lophiidormes, anglerfishes): Greek, ‘lophos’, i.e., ‘crest’; Greek, ‘ichthys’, i.e., ‘fish’ Myrocongridae (Anguiliformes, eels and morays): Greek, ‘myros’, i.e., male of moray eel; Latin, ‘conger’, ): Greek, ‘para’, i.e., ‘the side of’; Greek, ‘odous’, i.e., ‘teeth’ The available English name includes ‘porthole’, which is used in composite names by: Gonostoma elongatum Günther, 1878 and Cyclothone microdon (Günther, 1878) (Gonostomatidae, Stomiiformes); Dianema longibarbis Cope, 1872 (Callichthyidae, Siluriforme Cyprinodontiformes). Erethistidae (Silur usually large (Nelson, 1994), found in rivers, streams and moving freshwaters of southern Asia (Ne 2006). Note that ‘river catfish’ is used in composite names of other siluriform catfishes, e.g., Clupisoma garua (Hamilton, 1822) (Schilbeidae), Corydoras metae Eigenmann, 1914 (Callichthyidae), Hemibagrus nemurus (Valenciennes, 1840) (Bagridae), Pangasius pangasius (Hamilton, 1822) (Pangasidae) and Porochilus meraukensis (Weber, 1913) (Plotosidae). (Romero, 2002). Halftooths. Members of this family have small and toothless lower jaws (Nelson, 1994). Available Chinese common names include the words ‘half tooth fat carp’. Heptapteridae (Siluriformes, catfishes): Greek, ‘hepta’, i.e., ‘seven’; Greek, ‘pteron’, i.e., ‘fin’ (Romero, 2002). Three-barbeled catfishes. A speciose family whose members are not externally differentiable with members of the Family Pimelodidae (long-whiskered catfishes; see Nelson, 2006). Members of this family usually have naked skin, three pairs of barbels, large adipose fin and a deeply forked caudal fin (Nelson, 2006). Available common names do not have any root wo Hispidoberycidae (Stephanoberyciformes, pricklefishes, bigscales and gibberfishes): Latin, ‘hispidus’, i.e., ‘rough’, ‘shaggy’, ‘hairy’, ‘bristly’ (Whitaker, 1998-2000). Spiny-scale pricklefishes. This family is monospecific. Hispidoberyx ambagiosus Kotlyar, 1981 is a deep sea fish with spinulose scales, operculum with a long, stout spine, the dorsal with 4-5 spines and the anal fin with 3 spines (Kotlyar, 2004). ‘Pricklefishes’ is preoccupied by Family Stephanoberycidae (Beryciformes), ‘spinyfins’ by Family Diretmidae (Beryciformes). (Romero, 2002). Crested frogfishes. Members of this family have no humped nape, the first dorsal spines are modified to an illicium (i.e., lure) and have palatine teeth (Nelson, 1994). The only representative is Lophichthys boschmai Boeseman, 1964 (FishBase common name: Boschma’s frogfish). The available Chinese common name includes the word ‘wave’, i.e., ‘crest’. i.e., ‘conger’ (Romero, 2002). Atlantic red eels. This family is monogeneric. ‘Moray’ is preoccupied by Family Muraenidae (Order Anguiliformes) and ‘conger’ is preoccupied by Family Congridae (Order Anguiliformes). Normanichthyidae (Scorpaeniformes, scorpionfishes and flatheads): Named after John Roxburgh Norman, British ichthyologist 1898-1944 (Romero, 2002). Barehead scorpionfishes. Members of this family have armorless heads, one spine on pelvic fin and lack ribs (Nelson, 1994). Ostracoberycidae (Perciformes, perch-likes): Greek, ‘ostrakon’, i.e., an earthern vessel, tile, also the hard shell of testaceous animals, as snails, muscles, tortoises (Liddle and Scott, 1889). Shellskin alfonsinos. Members of this family have a prominent spine extending backward from the lower limb of preopercle (Nelson, 1994). ‘Alfonsinos’ is preoccupied by Family Berycidae (Beryciformes, sawbellies). Parodontidae (Characiformes, characins (Romero, 2002). Scrapetooths. Members of this family have ventral mouths with teeth modified for scraping algae off rocks, highly mobile and enlarged premaxillaries, no adipose eyelid and with expanded and flattened pectoral fins (Nelson, 2006). Available Chinese common names include the words ‘cheek’,  Common names of fish families, Palomares, M.L.D. et al. 44 ‘tooth’, and ‘fat carp’. Some of the available Spanish common names include the word ‘mouse’. Some of the Brazilian Portuguese common names include the word ‘pen knife’. Tetrabrachiidae (Lophiiformes, anglerfishes): Greek, ‘tetra’, i.e., ‘four’; Greek, ‘brachion’, i.e., ‘arm’ (Romero, 2002). Four-armed frogfishes. FishBase English name of sole species in FishBase, ’ (Romero, 2002). Collared wrigglers. Available common names include the word ‘wriggler’. elson (2006) uses ‘picarel porgies’. The name ‘picarels’ is not preoccupied and is simpler than ‘picarel porgies’. ’ (Romero, 2002). Crocodile toothfishes. The only available English common name is h is used in Nelson (2006) is also preoccupied by Serranus cabrilla (Linnaeus, 1758) (Serranidae, Perciformes) and used in Chaudhuriidae (Synbranchiformes, spiny eels): From ‘chaudhuria’, i.e., a Burmese local name for a fish s’, ‘earthworm’ being used in the composite common name for Yirrkala lumbricoides (Bleeker, 1853) (Ophichthidae, Anguiliformes). We believe that h fishes. Some of the available English common names include the word ‘swallower’, the name used in Nelson (2006) but which is preoccupied by es, characins): From ‘Curimatá’, a locality in Piauí State, Brazil, and used in Creole French as a local name for a fish in French Guyana (Romero, 2002). Toothless characins. matter, microdetritus, microvegetation, and filamentous algae common in those habitats (Nelson, 1994). Ereuniidae (Scorpaeniformes, scorpionfishes and flatheads): Greek, ‘ereyn’, ‘aireoo’, i.e., ‘to catch’; also Greek, ‘ereyna’, ‘ereynes’, i.e., ‘inquiry’, ‘search’ (Romero, 2002). Deepwater bullhead sculpins. The available Chinese names include the word ‘bullhead’, which is appropriate as members of this family have Tetrabrachium ocellatum Günther, 1880, is ‘four-armed frogfish’. Xenisthmidae (Perciformes, perch-likes): Greek, ‘xenos’, i.e., ‘strange’, ‘rare’; Greek, ‘isthmia’, i.e., ‘neck’, ‘throat’, ‘narrow passage (iii) Alternative common names for families with common names in Nelson (2006) Centracanthidae (Perciformes, perch-likes): Greek, ‘kentron’, i.e., ‘thorn’, ‘sting’; Greek, ‘akantha’, i.e., ‘thorn’ (Romero, 2002). Picarels. Most of the species in this family have composite English names including the word ‘picarel’. N Champsodontidae (Perciformes, perch-likes): Greek from Egyptian, ‘champsai’, i.e., ‘crocodile’; Greek, ‘odous’, i.e., ‘teeth ‘gaper’ for Champsodon capensis Regan, 1908. Available Chinese common names include the word ‘toothfish’. ‘Crocodilefish’ is preoccupied by Cymbacephalus beauforti (Knapp, 1973) and ‘toothfish’ is preoccupied by several species of the Family Nototheniidae (Perciformes). ‘Gaper’ whic the composite name of Chaunax stigmaeus Fowler, 1946 (Chaunacidae, Lophiiformes). However, ‘crocodile toothfish’ is not preoccupied by any species. (Romero, 2002), named after B.L. Chaudhuri, an Indian Ichthyologist. Spineless eels. Members of this family have no dorsal or anal fin spines. Several of the available English composite common names include the words ‘spineless eel’. Nelson (2006) used ‘earthworm eel ‘spineless eel’ represents the most striking morphological character of this group without having to refer to another animal, e.g., ‘earthworm’. Chiasmodontidae (Perciformes, perch-likes): Greek, ‘chiasma’, i.e., ‘cross’, ‘chiasmos’, i.e., ‘diagonal’; Greek, ‘odous’, i.e., ‘tooth’, ‘teeth’ (Romero, 2002). Snaketoot Family Saccopharyngidae (Saccopharyngiformes). Some of the available Chinese common names include the words ‘snake-toothed’ and occasionally ‘fork-toothed’ and ‘long-toothed’. Curimatidae (Characiform Available Chinese common names include the words ‘toothless fat carp’ while Spanish common names include ‘smallmouth’. A distinct characteristic of the members of this family is the loss or the reduction of dentition on the fifth upper pharyngeal tooth plate (Nelson, 1994). The word ‘characin’ probably came from Latin, ‘characias’, ‘characiae’, i.e., ‘reed’ for making ‘stakes’, a kind of ‘spurge’ (Whitaker, 1998-2000; Pliny the Elder, 1906), derived from Greek, ‘charax’, i.e., ‘a pointed stake’, ‘a vine-prop or pole’ (Liddle and Scott, 1889). Romero (2002) provides the following: Greek, ‘charax’, i.e., a marine fish; Latin, ‘forma’, i.e., ‘shape’. Members of this family are usually found in riverine and lacustrine habitats and feed on organic ‘Reedfish’ is preoccupied by Erpetoichthys calabaricus Smith, 1865 (Polypteridae, Polypteriformes); ‘smallmouth’ is preoccupied by Haemulon chrysargyreum Günther, 1859 (Haemulidae, Perciformes). ‘Carps’ is preoccupied by Order Cypriniformes. Nelson (2006) used ‘toothless characiforms’; the word ‘characiforms’ deviates from ‘characins’.  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 45 large heads compared to their narrow and long tails (Nelson, 1994). ‘Bullhead’ is used in the composite name for ‘bullhead sharks’ (Heterodontidae, Heterodontiformes). Nelson (2006) used ‘deepwater sculpins’, which is used by Myoxocephalus thompsonii (Girard, 1851) (Cottidae, Scorpaeniformes); note that the word ‘sculpin’ is also preoccupied by Family Cottidae (Scorpaeniformes). Monognathidae (Saccopharyngiformes, swallowers and gulpers): Greek, ‘monos’, i.e., ‘only’; Greek, ‘gnathos’, i.e., ‘jaw’ (Romero, 2002). Onejaws. Members of this family lack maxilla and premaxilla n’, indicating that the family name might have been derived from the name of the city. Toothcarps. The two species represented have common names which include (Nelson, 1994). ‘Gulpers’ is preoccupied by Family Eurypharyngidae (Saccopharyngiformes). Nelson (2006) used ‘onejaw gulpers’. Valenciidae (Cyprinodontiformes, rivulines, killifishes and live bearers): Available common names include the words ‘Valencia’ and ‘Spai ‘toothcarp’. ‘Toothcarp’ also used in composite names for some species of Family Cyprinodontidae (pupfishes) and Poeciliidae (poeciliids) both belonging to the Order Cyprinodontiformes. Nelson (2006) used ‘Valencia toothcarps’. (iv) Families without common names in Nelson (1994) and which have new names in Nelson (2006)3 Banjosidae (Perciformes, perch-likes): English, ‘banjo’, i.e., musical instrument with a drumlike body, a fretted neck, and usually four or five strings (Merriam-Webster Online Dictionary; www.m-w.com/cgi- bin/dictionary). Banjofishes* from ‘Banjofish’ for Banjos banjos (Richardson, 1846). Bathylutichthyidae (Scorpaeniformes, scorpionfishes and flatheads): Greek, ‘bathys’, i.e., ‘deep’; Greek, ‘louso’, ‘louteon’, i.e., ‘bath’, ‘to inmerse’; Greek, ‘ichthys’, i.e., ‘fish’ (Romero, 2002). Antarctic sculpins. Centrophoridae (Squaliformes, bramble, sleeper and dogfish sharks): Greek, ‘kentron’, i.e., ‘thorn’, l’ (Romero, 2002). Short-tail eels*. ‘Short tail conger’ is preoccupied by Paraconger similis (Wade, 1946) ively broad gill openings and one straight strong spine on both opercle and subopercle; they are uncommon or relatively rare and hless tongue. The proposed English common name is the chosen FishBase common name for the sole Known only from South Georgia Island, Antarctica. Centrogeniidae (Perciformes, perch-likes): Greek, ‘kentron’, i.e., ‘thorn’, ‘sting’; Greek, ‘genos’, i.e., ‘race’ (Romero, 2002). False scorpionfishes*. Taken from the English name of Centrogenys vaigiensis (Quoy and Gaimard, 1824), sole representative of this family. ‘sting’; Greek, ‘pherein’, i.e., ‘to carry’ (Romero, 2002); Latin, ‘phor’, ‘phoreo’, with several meanings including ‘putting food into one’s mouth’. Gulper sharks*. Many of the available English names include the words ‘dogfish’ and ‘gulper’. However, ‘dogfishes’ (Squalidae, Squaliformes) and ‘gulpers’ (Eurypharyngidae, Saccopharyngiformes) are both preoccupied. Colocongridae (Anguilliformes, eels and morays): Greek, ‘kolos’, i.e., ‘tail’; Latin, ‘conger’, i.e., ‘sea ee (Congridae, Anguiliformes). The available Chinese and Czech common names include the words ‘short tail conger eel’ and ‘serpentine’. Draconettidae (Perciformes, perch-likes): Greek, ‘drakos’, ‘drakaina’, i.e., ‘dragon’; Greek, ‘nessa’, ‘netta’, i.e., ‘duck’ (Romero, 2002). Slope dragonets*. Members of this family have relatively big eyes compared to their small heads, two nostrils on each side of the head, with relat are found on the edge of the continental shelf or on seamounts (Nelson, 1994). The available Chinese names include the words ‘thick thorn lizard’. ‘Lizardfishes’ is preoccupied by the Family Synodontidae (Aulopiformes), ‘dragonets’ by the Family Callionymidae (Perciformes). Gymnarchidae (Osteoglossiformes, bony tongues): Greek, ‘gymnos’, i.e., ‘naked’; Greek, ‘archo’, i.e., the extreme of the anus (Romero, 2002). Abas*. This family is monospecific and the fish has a toot representative, Gymnarchus niloticus Cuvier, 1829 and is based on the Ijo language (Nigerian) common name.                                                  3 See footnote (2) on p. 41.  Common names of fish families, Palomares, M.L.D. et al. 46 Hepsetidae (Characiformes, characins): Greek, ‘epsetas’, i.e., ‘boiled’; also Greek, ‘oí epsetoi’, i.e., ‘certain fishes’; may be related to ‘psetta’, i.e., ‘grouper’ (Romero, 2002). African pikes. This family is monospecific and is found widespread in Africa from Senegal to Angola including Niger, Volta, Chad, e words ‘short’, ‘cheek’, and ‘python’. Nelson (2006) took the AFS official common name and accepted as the FishBase common name for Pythonichthys al fin-base. They also possess an electric organ which discharges discrete pulses (Albert, 2003). Nelson (2006) took the English common name used for m with spines and also has a series of 5 isolated dorsal spines (Nelson, 1994). ‘Pricklebacks’ is preoccupied by Family Stichaeidae (Perciformes) and ‘sticklebacks’  have eyes which can be minute or large or plate-like and without lenses; the pectoral, pelvic and caudal rays can be elongated on which they stand, and the jaw extends past the ’, i.e., a kind of fish (Romero, 2002). Pencilfishes*. Members of this family have elongate, cylindrically-shaped bodies with fairly large scales, Microstomatidae (Osmeriformes, smelts): Greek, ‘mikros’, i.e., ‘small’; Greek, ‘stoma’, i.e., ‘mouth’  the composite name for the Family Argentinidae (Salmoniformes), ‘pencil’ is used in composite names by the Family Trichomycteridae elt’ is preoccupied by Family Osmeridae (Salmoniformes).  in composite names of the Family Myctophidae. Nelson (2006) adapted the FAO English and accepted FishBase common name for Scopelengys tristis Alcock, 1890. roximal radial tact with the cleithrum, has six branchiostegal rays and lack of lateral line (Nelson, 1994). Some Ogowe, Democratic Republic of the Congo and upper Zambezi Rivers, as well as in the Cunene, Okavango, and Kafue systems; also widespread in central and West Africa but absent in the Nile River, Zambian Congo and the Great Lakes (Skelton, 1993). Heterenchelyidae (Anguilliformes, eels and morays): Greek, ‘heteros’, i.e., ‘other’; Greek, ‘enchelys’, i.e., ‘eel’ (Romero, 2002). Mud eels*. Members of this family have large mouths and are scaleless (Nelson, 1994). Available Chinese common names include th asodes Rosenblatt & Rubinoff, 1972. Hypopomidae (Gymnotiformes, knifefishes): Greek, ‘hypo’, i.e., ‘under’; Greek, ‘pomas’, ‘pomatos’, i.e., ‘cover’ (Romero, 2002). Bluntnose knifefishes. Members of this family have no teeth on both jaws, snout moderate to short length, small eyes. They resemble eels because they have no caudal or dorsal fin but have the anal-fin origin ventral or posterior to pector Brachyhypopomus brevirostris (Steindachner, 1868). Indostomidae (Gasterosteiformes, sticklebacks and seamoths): Latin, ‘induere’, i.e., ‘to cover’; Greek, ‘stoma’, i.e., ‘mouth’ (Romero, 2002). Armored sticklebacks. Members of this family have slender bodies with a covering of bony scutes; operculu by Family Gasterosteidae (Gasterosteiformes). Ipnopidae (Aulopiformes, grinners): Greek, ‘ipnos’, i.e., ‘oven’, ‘kiln’ (Romero, 2002). Deep-sea tripod fishes. Members of this family orbit of the eye (Nelson, 1994). Most of the common names available for this family include the words ‘deep-sea’, ‘tripodfish’, ‘spiderfish’ and ‘deep-pool fish’, while some include the words ‘stove eye’, ‘grid eye’, ‘net eye’. ‘Tripodfish’ used in composite common names in the Family Triacanthidae (Tetraodontiformes). Lebiasinidae (Characiformes, characins): Greek, ‘lebias lacking a frontal/parietal fontanel and the cheek well covered by orbital and opercular bones (Nelson 1994). Many of the available common names include ‘pencilfish’, ‘fat carp’ and ‘tetra’. ‘Pencil’ used in composite names by the Family Trichomycteridae (Siluriformes), ‘carp’ is preoccupied by Order Cypriniformes and ‘tetra’ is used in composite names for fishes of the Family Alestiidae (Characiformes). (Romero, 2002). Pencil smelts. Members of this family have large eyes (more than twice the length of snout), small mouths, have spineless fins (Nelson, 1994). Available common names include the words ‘south’, ‘argentine’, and ‘pencilsmelt’. ‘Argentine’ is used in (Siluriformes) and ‘sm Neoscopelidae (Myctophiformes, lanterfishes): Greek, ‘neos’, i.e., ‘new’; Greek, ‘skopelos’, i.e., the name of a fish cited by Cuvier, 1817; Greek, ‘skopelos’, i.e., ‘reef’, ‘rock’ (Romero, 2002). Blackchins. Some members of this family have photophores (Nelson, 1994). Many of the available common names include the word ‘lanterfish’, ‘glowfish’, ‘lampfish’. ‘Lanternfishes’ is preoccupied by Family Myctophidae (Myctophiformes) and ‘lampfish’ is used Odontobutidae (Perciformes, perch-likes): Greek, ‘odont’, i.e., tooth; Latin, ‘buttos’ from ‘butinê’, i.e., a flask covered with plaited osier (Liddell and Scott, 1889). Freshwater sleepers. Members of this family may be distinguished from other goboid families by the large scapula which excludes the p from con  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 47 of the available names include the word ‘sleeper’ and ‘pond’. ‘Sleeper’ is preoccupied by Family Eleotridae (Perciformes). Pseudotrichonotidae (Aulopiformes, grinners): Greek, ‘pseudes’, i.e., ‘false’, ‘falsely’; Greek, ‘thirx’, i.e., use barbells and are usually found in mountain streams. Ra dae (Lampriformes, velifers, tube-eyes and ribbonfishes): Latin, ‘radius’, i.e., ‘radius’; sca compressed and attenuated posteriorly to a thin caudal filament. The sole representative ‘tap posite English names for some members of Family Trachipteridae Samaridae hes): Latin, ‘samara’, i.e., seed of the elm (Romero, 2002). Could ddel and Scott, 1889). eye flounder’. ‘Right- es). Nelson (2006) (Es er, 2006). Infantfishes. Members of this family are small and neotenic, with transparent brus’, i.e., ‘lip’ ich have serrated tive,  Sternoptychidae (Stomiiformes, lightfishes and dragonfishes): Greek, ‘sternon’, i.e., ‘chest’, ‘breast’; Greek, ‘ptyx’, ‘ptychose’, i.e., ‘fold’, ‘crease’ (Romero, 2002). Marine hatchetfishes. Members of this family have branchiostegal photophores and pseudobranch (Nelson, 1994). Available common names include the words ‘hatchetfish’, ‘hatchet belly’, ‘axe’, ‘pearlside’, and ‘bristle mouth’. ‘Hatchetfish’ is also used by the Family Gasteropelecidae, i.e., freshwater hatchetfishes. Symphysanodontidae (Stomiiformes, lightfishes and dragonfishes): Greek, ‘symphysis’, i.e., ‘grown together’; Greek, ‘an’, i.e., ‘without’; Greek, ‘odous’, i.e., ‘teeth’ (Romero, 2002). Slopefishes*. Available common names include the words ‘slope’, ‘shelf’, ‘covered tooth’. DISCUSSION The names presented here are, as mentioned above, suggestions that can and will be discussed in the ichthyological community, then will be entered into FishBase. Following current practice, they also serve as basis for coining new common names for species that lack such names, e.g., by adding modifiers to the family common names. This may contribute to the fishes being the first very speciose group of organisms with all species having common names. ACKNOWLEDGEMENTS The authors wish to acknowledge Dr Konstantinos Stergiou for checking the Greek translations and Mr William Cheung for checking the translation and interpretation of Chinese common names.  ‘hair’; Greek, ‘noton’, i.e., ‘back’ (Romero, 2002). Sand diving lizardfishes. ‘Sanddivers’ is the name d for Family Trichonotidae (Perciformes). Psilorhynchidae (Cypriniformes, carps): Greek, ‘psilos’, i.e., ‘bald’, ‘hairless’; Greek, ‘rhynchos’, i.e., ‘jaw’ (Romero, 2002). Mountain carps. Members of this family have small inferior mouths, fleshy lips and no diicephali Greek, ‘kephale’, i.e., ‘head’ (Romero, 2002). Tapertails*. Members of this family have elongated leless bodies, of this family is Radiicephalus elongatus Osório, 1917, whose FishBase English common name is ertail’. This is used in com (Lampriformes) and Family Engraulidae (Clupeiformes).  (Pleuronectiformes, flatfis also be Greek, ‘sêma’, i.e., ‘sign’, ‘mark’, ‘token’ of the star on a horse’s head (Li Crested flounders. Most available English common names include the words ‘right eye flounder’ is used in composite name for Family Pleuronectidae (Pleuronectiform adapted a modified form of the scientific name. Schindleriidae (Perciformes, perch-likes): Named after Dr. D.W. Schindler, University of Alberta, Canda chmey bodies and many undeveloped cartilage and bones. Some of the available common names include the words ‘infant’ and ‘precocious’. Scombrolabracidae (Perciformes, perch-likes): Latin, ‘scomber’, i.e., ‘mackerel’; Latin, ‘la (Romero, 2002). Longfin escolars. Members of this family are deep water fishes wh operculum and preoperculum and a protusible maxilla (Nelson, 1994). The sole representa Scombrolabrax heterolepis Roule, 1921 has the FishBase common name ‘longfin escolar’. ‘Escolar’ is used in composite names in the Family Gempylidae (Perciformes).  Common names of fish families, Palomares, M.L.D. et al. 48  REFERENCES Albe ds.), C 6. Eschmeyer, W. (ed.), 1998. Catalog of Fishes. California Academy of Science. San Francisco, USA. Eschmeyer, W. (ed.), 2006. Catalog of Fishes. W onic publication. www.calacademy.org rt, J.S., 2003. Hypopomidae (bluntnose knifefishes), In: Reis, R.E., Kullander, S.O., Ferraris, C.J. Jr. (e hecklist of the Freshwater Fishes of South and Central America. EDIPUCRS, Porto Alegre, Brasil, pp. 494-49 orld Wide Web electr , version (08/2006). Froese, R., Pauly, D. (ed ww.fishbase.orgs.), 2006. FishBase. World Wide Web electronic publication. w , version (08/2006). Kotlyar, A.N., 2004. Family Hispidoberycidae Kotlyar 198. Hispidoberycids. Calif. Acad. Sci. Annotated Checklists of Fishes (26), 2. Kuiter, R. H. , Tonozuka, appers, Muraenidae - Lutjanidae. Zoonetics, Liddell and Scott, 1889. An intermediate Greek-English Lexicon. Clarendon Press Oxford. www.perseus.tufts.edu/cgi- T., 2001. Pictorial Guide to Indonesian Reef Fishes. Part 1. Eels-Sn  Australia. bin/resolveform, version (15/08/2006). Nelson, J.S., 1994. Fishe Nelson, J Palomares, M.L.D., Pauly, D., 2000. COMMON NAMES Table. In: Froese, R., Pauly, D. (eds.), FishBase 2000.  Concepts, Design and Data Sources. ICLARM, Los Baños, Laguna, Philippines, pp. 85-92. Pliny the Elder. 1669. Naturalis Historia. Translat arl Friedrich Theodor Mayhoff. Lipsiae. Teubner. 1906. Robins, C.R s Important to North Americans. American Fisheries Society Special Publication 21, AFS, Bethesda, Maryland, USA. 02. An Etymological Dictionary of Taxonomy. Madrid, unpublished. Cited in FishBase. s of the World. ., New York. .S., 2006. Fishes of the World. Fourth edition. John Wiley and Sons, Inc., New York. Third edition. John Wiley and Sons, Inc ed from Latin by K ., Bailey, R.M., Bond, C.E., Brooker, J.R., Lachner, E.A., Lea, R.N., Scott, W.B., 1991. World Fishe Romero, P., 20 Schmettkamp, W., 1985. Die Namen unserer Aquarienfische. Landbuch-Verlag GmbH, Hannover. Skelton, P.H., 1993. A Complete Guide to the Freshwater Fishes of Southern Africa. Southern Book Publishers. Whitaker, W., 1998-2000. Latin-to-English Dictionary. AbleMedia. www.blemedia.com, version (15/08/2006).   Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 49 GROWTH, REPRODUCTION AND FOOD OF THE MUDSKIPPER, PERIOPHTHALMUS BARBARUS ON MUDFLATS OF FREETOWN, SIERRA LEONE1 Ibrahim Turay Institute of Marine Biology and Oceanography (IMBO), Fourah Bay College, University of Sierra Leone, Freetown J. Michael Vakily2 Institute of Marine Biology and Oceanography (IMBO), Fourah Bay College, University of Sierra Leone, Freetown Maria Lourdes D. Palomares The Sea Around Us Project, Fisheries Centre, University of British Columbia, 2202 Main Mall, Vancouver, BC V6T 1Z4 Canada; Email: m.palomares@fisheries.ubc.ca Daniel Pauly The Sea Around Us Project, Fisheries Centre, University of British Columbia, 2202 Main Mall, Vancouver, BC V6T 1Z4 Canada; Email: d.pauly@fisheries.ubc.ca n 364 specimens of mudskipper, Periophthalmus barbarus (Linnaeus, 1766) (Family Gobiidae) sampled from July 1992 to April 1993 on a small mudflat near Freetown, deals with the uction and feeding habits of a species that is often studied in terms of its set er, Periophthalmus barbarus (Linnaeus, 1766) (Family Gobiidae, SubFamily Oxudercinae, Order Perciformes, Figure 1, insert) is the most abundant fish in the mangrove swamps  is driven into the mud. Then, leaves and mud are used to turn the opening into the mouth of a trap (Figure 1). known as ‘gballar’ is then used to smoothen the surface of the leaves, which may also be baited by a sprinkling of powder-dried crab, or additional ‘gballar’. In neighboring ece of broom stick are used, instead of bamboo cane, to construct a gear ABSTRACT This study, based o morphometrics, growth, reprod interesting amphibious behavior, but which, in Sierra Leone, represents a food resource for coastal dwellers. Results include length-weight relationships for females and males, a seasonally oscillating growth curve (TL∞ = 17 cm, K = 0.89 ± 0.30 year-1, C = 0.75 and WP = 0.95), with a strong end-of-year reduction of growth related to the cold ‘Harmattan’ wind, and a spawning peak in May-June, at the on of the rainy season. First maturity occurs at about 8 cm; the food consists of small crustaceans, polychaets, insects, mollusks and detritus. INTRODUCTION The Atlantic mudskipp along the coast of Sierra Leone. People, notably around the coastal villages of Bonthe, Wale, Yeloboya and Shenge, catch mudskipper, locally known as ‘jumbo fish’, using traps made of binodal segments of bamboo cane. One node is cut off, and the cane Powdered okra or a substance locally Guinea, a broad leaf and a pi similar in operation to the above-described trap.                                                  1 Cite as: Turay, I., Vakily, J.M., Palomares, M.L.D., Pauly, D., 2006. Growth, food and reproduction of the mudskipper, Periophthalmus barbarus on mudflats of Freetown, Sierra Leone. In: Palomares, M.L.D., Stergiou, K.I., Pauly, D. (eds.), Fishes in Databases and Ecosystems. Fisheries Centre Research Reports 14(4), pp. 49-54. Fisheries Centre, University of British Columbia ael.Vakily@googlemail.com [ISSN 1198-6727]. 2 Present address: CSRP, Dakar, Senegal; Email: Mich  Growth, reproduction and food of mudskipper, Turay, I. et al.  50  ), Figure 1 The mudskipper, Perioph with schematic representations of b Den-Khadhri, 1984). MATERIALS AND MET thalmus b us (Linnaeus, 1766) (Family Gobiidae, in t ada d fro FAO urrows and of the b oo-and-le se n Sier n  fro HODS Catch composition statistics relating to udsk per no pp  to ist, a  this study is thu caught e firs uth o uly  Apri  digg  int at of Co wn, a uartier’ of Freetown (8° 29’ 10’ N; 13° 15’ 30’ E). ere done 48 sa les  th earest mm sing n  ca r; th l (TL) n d le th ) a  some othe pres ted i ura esented  ref  TL  cm ndi ual h weights were determine th-frequency ana  all performed using the F AT twar ting  mat ty stage by mp ng g dos atic dice  weight) y esti tin ‘fec dity i.e., e nu er of eggs in the tudied gh ex ina n o tom h nd n s a of 15. m;  lar st female asur 0 . F re 2 ship of female and le mudskippers; the corresponding relation ip fo  = 0. 3.19 fo hi  =  w  log ea rsio as een standard length and total length is given by: SL (cm) = 0.86·TL (cm) -0.0914 (Figure 2B) arbar ser pte m amb aves trap u d i ra Leo e (adapted m the m ip  do t a ear  ex nd s entirely based on 364 specimens mudskipper burrows in the mudfl Morphometric measurements w measurements taken included tota (1993), but the measurements pr to the nearest 0.01 g. The leng (Gayanilo et al., 1996). Reproduction was studied by no (GSI = gonad weight·100/body by th t a or fr m J  1992 to l 1993 by ing o ngoto  ‘q  on 3 mp , to e n , u  a ver ier lipe e  and sta d er to ar ng (SL nd rs en n T y  below  in . I vid  fis d lyses were iS sof e  gonad uri s, co uti ona om  in s and b ma g un ’,  th mb gonads of mature female mudskipper. Food and feeding habits were s ‘occurrence’ methods (Hynes, 1950). RESULTS AND DISCUSSION The largest mudskipper observed wa shows the length-weight relation the population as a whole is W r  throu am tio f s ac contents, a the ‘point’ a d  male 4 c the ge  me ed 13.  cm igu A  ma sh r 0072·TL r w ch n  348 and hose  lin r ve n h an 2 = 0.961. The relationship betwFishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 51 0 5 10 15 20 25 30 35 40 45 50 W = 0.0072 * TL3.19 0 5 10 15 20 Tota length (cm) W ei gh t (g ) R =  0.980; n = 348 A 0 2 4 6 8 10 12 14 SL (cm) = 0.816 * TL (cm)-0.0914 B 0 2 S ta nd ar d le ng th  (c m ) R = 0.996; n = 150 4 6 8 10 12 14 16 Total length (cm)   Figure 2 Relationships of length to weight (A) and total to standard leng us barbarus (Linnaeus, 1766) (Family Gobiidae), removed from their burrow n, Sierra Leone from July 1992 to April 1993.  Table 1 summarizes the length-frequency (L/F) data used for the growth analyses, whose results are given in some details (including some intermediate outputs), as they nicely illustrate how gentle ‘massaging’ of an L/F data set can lead to improved estimates of growth parameters, here those of the von Bertalanffy model (von Bertalanffy, 1938) and one of its seasonally oscillating variants (Pauly, 1987). The method of Wetherall (1986) was applied to an accumulation of the data in Table 1; it led to estimates of L∞ = 17 cm was retained for all analyses using ELEFAN, a are scanned, in very small (Figure 3a). This was Table 1 Length-frequency data for 34 (Linnaeus, 1766) (Family Gobiidae), remo he mudflats of Congotown, Freetown, Sierra Leone. Midlength (TL, cm) Jul 1992 Aug Oct N ths (B) of male and female Periophthalm s in the mudflats of Congotown, Freetow 8 mudskippers, Periophthalmus barbarus ved from their burrows in t ov Dec Jan 1993 Feb Mar Apr 0.25 0 0 0 0 0 0 0 0 0 L∞ ranging from 16 to 18 cm, depending on which length range was included in the analysis. Hence a value of routine of the FiSAT software. A first pass of the ELEFAN subroutine in which K values steps, from K = 0.1 to K = 10 year-1, yielded an essentially flat structure 0.75 0 0 0 1.25 0 0 0 1.75 0 0 0 2.25 5 1 0 2.75 1 1 0 3.25 0 0 0 3.75 0 0 1 4.25 1 2 1 4.75 2 4 1 5.25 1 1 3 5.75 0 2 1 6.25 0 2 2 6.75 3 5 1 7.25 1 3 2 1 3 2 2 8 2 7.75 4 6 3 1 2 3 1 16 1 9.75 10 3 4 10.25 4 3 2 10.75 1 1 1 13.25 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 15.75 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 2 0 0 0 1 1 0 1 0 0 0 2 2 1 1 1 1 1 2 3 1 4 1 1 2 0 2 9 0 8.25 5 3 4 8.75 10 8 1 9.25 4 2 2 2 4 1 3 5 4 2 2 5 1 11 2 1 4 1 4 6 1 2 2 3 4 4 1 3 2 1 5 2 1 1 1 2 2 1 2 11.25 3 4 1 0 1 0 3 0 0 11.75 5 2 3 4 0 0 1 1 1 12.25 1 0 1 1 1 0 0 0 1 12.75 1 0 1 1 0 0 0 0 0 13.75 1 0 0 0 0 0 0 0 0 14.25 0 0 0 0 0 0 0 0 0 14.75 0 0 0 0 0 0 0 0 0 15.25 0 0 0 attributed to the class interval (0.5 cm) of the L/F data in Table 1, which was too small 16.25 0 0 0 0 0 0 0 0 0 Total 63 53 35 26 30 23 30 68 20  Growth, reproduction and food of mudskipper, Turay, I. et al. 52 to clearly indicate the progression of cohorts. Subsequent analysis of L/F files with gradually increasing intervals did not yield a marked improvement (Figures 3b-3g, easily done with the ‘slicing’ routine of FiSAT). The second pass, which assumed seasonal growth (C = 0.5, WP = 0.2, commonly occurring in fish) led to the identification of a peak, associated with K = 0.80 year-1 and a class interval of 1.25 cm (Figure 3j). This was used, along with L∞ = 17 cm, to (partly) correct the L/F file for the effect of incomplete selection by and incomplete recruitment to the gear, using the catch-curve approach of Pauly (1987). This led, in the third pass, to the emergence of a clear peak, centered around K = 0.81 year-1. The fourth and final pass consisted of optimizing the parameter related to seasonal growth oscillations, likely to occur in fish ex osed to ‘automatic search routine’ of ELEFAN, with K, C and WP as free parameters, then using the best ( t 17 cm as for all previous Figure per panel) s the growth curve obtained ult of th k. The asymp ength (L 7 cm, K = 0.89 rate at which L∞ is approached, C =  expresses the a tude of ther strong given that C can take values between 0 and e WP = 0.95 is the ‘winter point’, i.e., the time of the year (expressed as fraction of 1) when growth is slowest; WP corresponds here to late November 1992, which falls in the wet, ‘intermediate’ this period as the main spawning season. There is some indication of a second minor spawning at the end of d, minor cohort (lower panel of Figure 4).  p seasonal differences in (air) temperature of about 10°C (Findlay, 1978). This was done using the estimates of C and WP thus obtained, by running the K scanning routine with C and WP fixed at 0.75 and 0.95, respectively and L∞ a analyses). 4 (up  show as a res is wor totic l ∞) is 1  year-1 is the  0.75 mpli Figure 3 Response surface analysis of the of the von Bertalanffy growth function showing resolution of ELEFAN following modification of the lengt in Table 1, for TL  parameter K  the increasing h-frequency data ∞ =17 cm. the seasonal growth oscillations, which are here ra 1, and wher season (Vakily and Pauly, 1993). This growth curve, which also fits very small mudskippers, has its origin in mid June, suggesting the year, leading to a secon Table 2 presents the key results of the maturity study, from which a length at first maturity of about 9 cm can be inferred for the females and 7 cm for the males. The means of these two values yield Lm/L∞ = 0.65. as estimate of the ‘reproductive load’ (Cushing 1981) for mudskippers.  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 53   GSI ranged in females from 2.0 to 3.5, with a mean of 2.5 GSI and % of mature fish were plotted against months. However, low sample size and gaps in the series rendered these data inconclusive; all that emerged is that they did not contradict the timing of the spawning season(s) proposed above. Apparent fecundity was 11400 eggs g-1 of female gonads, in fish of 8 cm TL, i.e., about 2,100 eggs g-1 body weight. However, spawning probably occurs several times during (and outside) the main spawning season, and hence this estimate of fecundity represents only a minimum. The food items occurring most frequently in mudskipper stomachs were crustaceans (3 1%), polychaets (26%), insects (9%), and mollusks (3%) with detritus forming the rest. There was a slight tendency for the occurrence of crustaceans to increase and for polychaets to decrease with increasing mudskipper size. ACKNOWLEDGEMENTS The field work upon which this paper is based was performed by the first author for a BSc (Honours) thesis supervised by Professor D.E.B. Claytor, Fourah Bay College, University of Sierra Leone, Freetown. The first author would also like to thank Mr. Dean-Kadri, whose own thesis on mudskipper proved extremely helpful.  Figure 4 Von Bertalanffy Growth Function for Periophthalmus barbarus (Linnaeus, 1766) (Family Gobiidae), removed from their burrows in the mudflats of Congotown, Freetown, Sierra Leone from July 1992 to April 1993, with TL∞ = 17 cm, K = 0.89 year-1, C = 0.75 and WP = 0.95. Upper panel: main cohort, originating in June. Lower panel: second cohort, originating earlier in the year. Table 2 Relative frequency of immature and mature mudskipper, Periophthalmus barbarus (Linnaeus, 1766) (Family Gobiidae), removed from their burrows in the mudflats of Congotown, Freetown, Sierra Leone from July 1992 to April 1993. Midlength (TL, cm) Females Immature/mature Males Immature/mature 3.0 0/0 2/0 5.0 1/0 20/0 7.0 10/0 24/44 9.0 22/56 17/47 11.0 3/38 10/19 13.0 0/7 0/6 15.0 0/0 0/1  Growth, reproduction and food of mudskipper, Turay, I. et al. 54 REFERENCES Bertalanffy, L. von uman Biol. 10(2), 181-213. Deen-Khadri, A., 1984 y College, University of Freetown, Freetown, Sierra Leone. Findlay, I.W.O., 1978. Marine biology of the Sierra Leone River estuary. The physical environment. Bull. Inst. Biol. Ocean. 3(1), 48-64. Cushing, D.H., y of Wisconsin Press, M Pauly, D., 1987. A review of the ELEFAN system for analysis of length-frequency data in fish and aquatic invertebrates. In: Pauly, D., Morgan, G.R. (eds.), Length-based Methods in Fisheries Research. ICLARM Conf. rah Bay College, University of Freetown, Freetown, Sierra Leone. rdinella off Sierra Leone. In: Bard, F.X., Koranteng, K.A. les de l’Upwelling Côtier du Ghana et de la Côte d’Ivoire. a, Ghana. Editions ORSTOM, Paris, pp. 425-436. th and mortality parameters from length-frequency data. ., 1938. A quantitative theory of organic growth (inquiries on growth laws II). H . Some Aspects of the Biology of the Mudskipper. B. Sc. Thesis, Fourah Ba 1981. Fisheries Biology: A Study in Population Dynamics. Second Edition. Universit adison. Proc. 13, pp. 7-34. Manila, Philippines. Turay, I., 1993. Investigation of Various Aspects of the Biology of the Mudskipper, Periophtahlmus koelreuteri, B. Sc. Thesis, Fou Vakily, J.M., Pauly, D., 1993. Seasonal movements of Sa (eds.), Dynamique et Usage des Resources en Sardinel Actes de la recontre du DUSRU, 5-8 Octobre 1993, Accr Weatherall, J., 1986. A new method for estimating grow FishByte 4(1), 12-14.   Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 55 NO S, The Sea Around tish Columbia, 2202 Main M heries.ubc.ca In July 1976, I became by default the scie l survey conducted by R/V Mutiara 4 in the southern ted in Pauly et al. (1996). During the cruise, we captured a large sawfish (Family Pristidae), which we identifi smalltooth sawfish Pristis pectin (Figure 1). At the time, all lar were finned by the Indonesian 4, the fins hanged to dry, garland-like, from the ship’s rigging, and later sold on th market by the crew as bonus. The s separately fo This sawfish, of 270-300 cm (total length, i.e., including the saw-like snout), differed from others in that I weighted its different body sawfish is ‘Critically endangered’ on the IUCN Redlist (Baillie et al., 2004), both because it is caught as Pauly, D., Martosubroto, P., Saeger, J., 1996. The Mutiara 4 surveys in the Java and southern South China Sea, 976. In: D. Pauly and P. Martosubroto (eds.), Baseline Studies in Biodiversity: The Fish esia. ICLARM Studies and Reviews 23, pp. 47-54.                                                 TE ON THE WEIGHT OF BODY PARTS, INCLUDING FIN OF THE SMALLTOOTH SAWFISH PRISTIS PECTINATA1 Daniel Pauly  Us Project, Fisheries Centre, University of Bri all, Vancouver, BC V6T 1Z4 Canada; Email: d.pauly@fis ntific leader of a demersa Indonesian part of the South China Sea (south and east of Singapore), documen ed as the ata Latham 1794 ge sharks we caught  crew of the Mutiara e ‘soup fin’ aw was sold r the curios market. parts. The results of this rather bloody exercise are given in Table 1. Thirty years later, the smalltooth by-catch by trawlers fishing for shrimps and fish (as was the Mutiara 4), and targeted for its fins. Little did we know. REFERENCES Baillie, J.E.M., Hilton-Taylor, C., Stuart, S.N., 2004. 2004 IUCN Red List of Threatened Species. A global species assessment. IUCN, Gland, Switzerland and Cambridge, UK. Massey, L.L., Harper, D.E., 1993. Selected Computer Images of Southeastern U.S. Marine Fishes. NOAA Tech. Mem. NMFS-SEFSC-333. November 1974 to July 1 Resources of Western Indon   Figure 1 The smalltooth sawfish, Pris Latham 1794, Family Pristidae (drawing from tis pectinata  Massey and Harper, 1993). Table 1 Weight (in kg and as % of the total body weight, i.e., 201.7 kg) of different body parts of a mature female smalltooth sawfish, caught on June 18, 1976, at Station 154 (~0025’S; 103054’E, 15 m depth) in Pauly et al. (1996) Body Weight % Body parts (extremities) Weight % part Head 47.5 23.5 Saw 2.0 1.0 Trunk 95.0 47.1 Dorsal fins (2) 3.5 1.7 Liver 26.0 12.9 Pectoral fins (2) 10.0 5.0 Other inner organs 8.2 4.1 Ventral fins (2) 5.0 2.5 Blood (recovered) ~1.0 0.5 Caudal fin 3.5 1.7  1 Cite as: Pauly, D., 2006. Note on the weight of body parts, including fins, of the smalltooth sawfish Pristis pectinata. In: Palomares, M.L.D., Stergiou, K.I., Pauly, D. (eds.), Fishes in Databases and Ecosystems. Fisheries Centre Research Reports 14(4), p. 55. Fisheries Centre, University of British Columbia [ISSN 1198-6727].  Overview of biological data of Greek stocks of anchovies and sardines, Somarakis, S. et al. 56 AN OVERVIEW OF BIOLOGICA ANCHOVY AND SARDINE STOCK L DATA RELATED TO S IN GREEK WATERS1 Stylianos Somarakis University of Patras, Department of Biology, GR 26500 Patra, Greece; Email: somarak@upatras.gr sianis f Biology, Department of Zoology, ki, Greece; Email: dtsianis@in.gr achias ine Research, lion, Crete, Greece f i,  a n European anchovy (Engraulis encrasicolus) and sardine (Sardina small-sized pelagic species in Greek waters, making up 30% of th total purse seine landings (Stergiou et al., 1997a). Despite their im stocks have never been studied systematically, nor has their exploit with the exception of their landings (see e.g., Stergiou, 1989, 1990 general, most of their landings in Greek waters derive from the trawlers is prohibited, i.e., the percentage of small pelagic fish in th trawlers cannot exceed 5%. Management measures to protect the purse seiners from the 10th of December to the end of February. In this paper we attempt to present a brief overview of the av concerning anchovy and sardine in Greek waters. GENETIC STRUCTURE Various studies conducted in recent years indicate a significant eastern Mediterranean and Black Seas (Spanakis et al., 1989; M i a Dimitrios E. T Aristotle University of Thessaloniki, School o Laboratory of Ichthyology, 54124, Thessaloni Athanassios M Hellenic Centre for Mar P.O. Box 2214, 71003, Irak Konstantinos I. S Aristotle University of Thessaloniki, School o Laboratory of Ichthyology, 54124, Thessalonik ABSTRACT European anchovy (Engraulis encrasicolus) and sardine small pelagic species in Greek waters. We present here studies concerning their biology and ecology. Some new a first maturity. INTRODUCTION pilchardus) are the two most important e total Greek landings and 59% of the portance for the Greek fisheries, their ation been monitored on a yearly basis, , 1991, 1992; Stergiou et al., 1997). In purse seine fleet. Fishing with pelagic e total marketable fraction of demersal se species include a closed season for ailable biological and ecological data genetic structuring for anchovy in the agoulas et al., 1996; Machias et al., ctic population (Machias et al., 2001b). l., 1996; Machias et al., 2000a) provide tergiou  Biology, Department of Zoology, Greece; Email: kstergio@bio.auth.gr (Sardina pilchardus) are the most important  brief overview of available data sources and alyses are also presented concerning lengths at 2000a). This is not the case for sardine which conforms to a panm Surveys of genetic variation in mitochondrial DNA (Magoulas et                                                  1 Cite as: Somarakis, S., Tsianis, D.E., Machias, A., Stergiou, K.I., 2006. An overvi stocks in Greek waters. In: Palomares, M.L.D., Stergiou, K.I., Pa ew ical data related to anchovy and sardine uly, D. (eds.), Fishes in Databases and Ecosystems. Fisheries Centre Research Reports 14(4), pp. 56-64. Fisheries Centre, University of British Columbia [ISSN 1198-6727]. of biolog  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 57 consistent evidence that the anchovy stocks do not form a genetically homogeneous population. Barriers for gene flow have been suggested between the northern and southern Aegean, and between the northern Adriatic and Ionian Seas. Two different genetic stocks of anchovy are currently recognized in Greek waters: the eastern stock (Aegean type) and the western stock (Ionian type). REPRODUCTION Information on spawning time and grounds of anchovy and sardine has been obtained from 1992 to 2001 within the framework of three EU and one national research projects. These projects aimed, among others, at delimiting the spawning grounds and times of these species in Greek waters, and the development and application of ichthyoplankton-based methods, specifically the daily egg production method (DEPM), in the north Aegean (Somarakis, 1993; Somarakis et al., 2000a; Ramfos et al., 2000) (see also Stergiou and Georgopoulos, 1993, for the relationship between distribution of phytoplankton pigments and landings of small pelagics in the Greek Seas). Consequently, the highest egg densities have been typically observed over the northern Aegean Sea continental shelf (Figure 2a). A potential spawning location for anchovy around the island of Lemn s  0 for estimating the spawning bi mass of the stocks (Tsimenides et al., 1995a; 1998; Machias et al., 2000a; 001b). o 1 2 3 4 5 6 oct nov dec jan feb mar apr may jun jul aug sep Month G SI (% ) 0 1 2 3 4 oct nov dec jan feb mar apr may jun jul aug sep G SI (% ) 50 150 250 350 450 eg gs /m 2 (a) (b)  Data relevant to anchovy and sardine  periods in the Greek Seas: (a) north n Sea. Monthly evolution of the gonosomatic  (GSI=100*ovary weight/ovary-free weight) ardine (□) and anchovy (■) and mean ance of anchovy eggs in the plankton (●). omarakis (1993), Voulgaridou and iou (2003) and Tsianis (2003); (b) central  and Ionian Seas. Monthly evolution of the matic index (GSI=100*ovary t/eviscerated weight) for sardine in the n (○) and the Ionian (□) as well for the vy in the Aegean (●) and the Ionian Sea ta from Machias et al. (2000a; 2001b). Bars: 2 The spawning period of anchovy in the Greek Seas extends from May to September (Somarakis, 1993; 1999; Machias et al., 2000a; Tsianis et al., 2003), but some spawning activity can be observed up to December in the central Aegean Sea (S. Somarakis, unpublished data). Spawning peaks at around June in all areas studied so far (Figure 1): 1992, north Aegean Sea, (Somarakis, 1993); 1998, central Aegean Sea (Machias et al., 2000a); 1998, central Ionian Sea (Machias et al., 2000a); 2002, NW Aegean Sea (Tsianis, 2003). The major spawning grounds of anchovy in the Aegean Sea are located in areas characterized by wide continental shelf and enrichment processes associated with the outflow from large rivers or the Black Sea Water (BSW) Figure 1 spawning Aegea index for s abund Data from S Sterg Aegean gonoso weigh Aegea ancho (■).Da ±SE.  22 23 24 25 26 40 41    1000    100    10    0.1    0 Lemnos island Dardanelles NORTH AEGEAN SEA (a) 20 21 22 23 24 25 38 39    1000    100    10    0.1    0 CENTRAL AEGEAN IONIAN SEA (b) 20 21 22 23 24 25 38 39    1000    100    10    0.1    0 CENTRAL AEGEAN IONIAN SEA (c)  Figure 2 Distribution and abundance of anchovy and sardine eggs from ichthyoplankton surveys. (a) Anchovy, north Aegean Sea, June 1996. (b) Anchovy, central Aegean and Ionian Seas, June 1999. (c) Sardine, central Aegean and Ionian Seas, winter 2000-2001. Data from Tsimenides et al. (1998), Somarakis et al. (2001, 2002). o has never been surveyed. Nevertheless, it is likely to be an important spawning ground, since this area is under the direct influence of BSW and the associated enhancement of biological production (Somarakis, 1999).  Overview of biological data of Greek stocks of anchovies and sardines, Somarakis, S. et al. 58       Figure 3 Biweekly mean total length (in cm) of sard Stergiou 2003; Tsianis, 2003). ine in NW Aegean, 07/1996-05/2003 (from Voulgaridou and  e coastal e 0 s ce earlier in the a (January- x data for suggest the er-December second peak 003). This is th with time eral, smaller 03). During rveys in the 1; Somarakis ere found  m depth e tactics of ers differ substantially to rely mainly scera during haracterized tch fecundity anias, 2003, r hand, the  seems to be zo ri n NE Aegean Sea continental shelf, Somarakis (19   Larval abundance 60 120 180 240 300 360 93 94 95 96 N o/ m 2 Temperature 16 17 18 19 o C 20 Salinity 34.5 35 35.5 36 36.5 37 37.5 ps u Zooplankton 15 20 25 30 35 40 m l/m 2 Egg size 0.19 0.20 0.20 0.21 0.21 93 94 95 96 m m 3 Relative fecundity 250 650 Spawning frequency 0.10 0.20 0.30 0.40 0.50 S Adult condition 0.52 0.53 0.54 0.55 0.56 K Egg abundance 0 100 200 300 400 500 N o/ m 2 Year  Figure 4 Means (±95% confidence intervals) for temperature (0-40 m); salinity (0-40 m); zooplankton omarakis (1999). displacement volumes, anchovy egg and larval abundance; and egg size from ichthyoplankton surveys and fecundity, spawning frequency and somatic condition of anchovy females from concurrent adult surveys in the Thracian Sea (NE Aegean Sea). Redrawn from S 350 450 550 eg gs /g Data on the spawning period of sardine exist for th areas of the central Aegean and Ionian Seas as w north Aegean Sea (Machias et al., 2001b; Ganias, 2 et al., 2003b; Voulgaridou and Stergiou, 2003; T (Figure 1). In the coastal waters of central Gree takes place from October to May but it peaks Aegean (November-January) than in the Ionian Se February) (Figure 1b). The gonadosomatic inde sardine in the north Aegean Sea (Figure 1a) existence of two spawning peaks: one in Novemb and another in March. The importance of the increased during the period 1996-2000 (Tsianis, 2 attributed to the decrease in the mean total leng (Figure 3; Tsianis, 2003) given that, in gen sardines spawn later than larger ones (Ganias, 20 the 1999 and 2000-2001 ichthyoplankton su central Aegean and Ionian Seas (Ganias et al., 200 et al., 2001), the highest egg abundance values w inshore, mainly in waters shallower than 100 (Figure 2c). Available studies suggest that the reproductiv anchovy and sardine in Greek wat ll as for the 03; Ganias ianis 2003) , spawning (Somarakis, 1999; Ganias, 2003). Sardines seem on fat reserves stored in the muscles and vi summer to reproduce during the winter. They are c by low seasonal and interannual variability in ba and spawning frequency (Somarakis et al., 2002, G Ganias et al., 2003a; 2004). On the othe reproductive effort of summer spawning anchovy closely associated with adult prey fields (meso- i.e., the energy allocated to reproduction derives p the food intake. In comparing batch fecundity a frequency estimates between June 1993 and June 1 oplankton), marily from d spawning 995 over the 99) showed  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 59 that adult food availability was higher in 1993,   when waters were significantly cooler and of Concurrently, female anchovies w e g nu us larg d eg at a hi  frequency (short inter- spawnin ) (Figure 4 The observations were cons t with a ration- related e ic in ncho (Somarakis et al., 2000b raveya et al., 2001; Somarakis, 2005). Lengths at first maturity vy and sardine have be for entral n a Ionian Seas from samples collected experimentally during DEPM surveys (Figure 5 e  m d). An  rea maturity at approximat 5 mm total length and sa nd mm, at t completion of the first year of life. AGE AN H There are  a th s anchovy  l readings ha ly validated (Table 1). Published growth curve comparisons ears have not shown es for either anchovy or 991; ffy from tal hing es ge, fish ovy  the ring n n clearly indicate that the mean length of sardine years (e.g., Figure 3). The potential effect of lower salinity. ere in b tter condition, producin gher mero e-size gs spawning g interval ). se isten reproductiv tact ; Ma  a vy of ancho en estimated  the c  Aegea nd and se  Appendix for etho ely 10 chovy ch rdine at arou  115 i.e., he D ROWTG very few age nd grow Greek waters and oto tudies for ith  and sardine in ve not been general among stocks, areas, and y any significant differenc sardine (Tserpes and Tsimenides, 1 Nikoloudakis et al., 2000). Von Bertalan growth parameters (Table 1) estimated samples collected from landings, experimen trawling, and onboard the commercial fis fleet clearly indicate that commercial catch might be biased with respect to lengths-at-a at least for the one and two year old (Figure 6). Indeed, selection of larger anch schools by the Greek purse seine fishery in north Aegean Sea was evident when compa length frequencies obtained from the commercial landings with those of experime trawling (with ‘a proportional to abunda allocation of pelagic trawl samples) durin DEPM survey in June 1995 (Figure 7). Since 1996, the Laboratory of Ichthyology of the Aristotle University of Thessaloniki has been collecting biweekly data on length, weight, sex ratio, gonadosomatic index and condition of both species in the NW Aegean Sea. The results tal ce’ g a (Voulgaridou and Stergiou, 2003; Tsianis 2003) and anchovy (Loukmidou and Stergiou, 2000) in the NW Aegean Sea has declined in recent 0 10 20 30 40 70 100 0 140 Total Length (mm) % rdin50 60 80 90 80 90 100 110 12 130 150 160 170 Sa e 0 10 60 80 110 120 0 Total Len ovy  F  ri %) o sardine. Pe males per leng in the central Aeg n (●) Seas. Fitted logistic curves are also shown. 100 20 30 40 50% 70 80 90 90 100 130 140 150 160 17 gth (mm) Anch igure 5 Length at first matu ty (at 50 f anchovy and th class rcent of mature fe ean (□) and Ionia  Anchovy 0 50 100 150 200 0 1 2 3 To ta lL en gt h (m m ) 4 Sardine 150 200 h (m m ) 0 50 0 1 2 To ta l 100 3 4 Age (years) Le ng t  experimental sampling (Machias et al., 2000a, 2001b). Broken and dotted lines: estimation based on sampling the landings (Tsimenides et al., 1995a, Tserpes and Tsimenides, 1991; Kallianiotis et al., 2003 respectively).  Figure 6 Von Bertalanffy growth curves for anchovy and sardine in Greek waters. Parameters of the models are given in Table 1. Solid lines: estimation based on  Overview of biological data of Greek stocks of anchovies and sardines, Somarakis, S. et al. 60 environmental variability on such a decline is under investigation. Table 1 Estimates of the von Bertalanffy growth parameters d sard r Species Area Year (year-1) L∞ (mm) pling Method Reference for anchovy an ine in G eek waters. K t0 Sam A  N ege Sea  0.280 191 seine s  (1995a) nchovy orth A an 1993  -2.480 Purse landing Otoliths Tsimenides et al.  N gean 1998 00 seine landings u   N gean 75  Purse seine landings  T n Se 76 seine landings l.  Central Aegean 1998- 0.509 175 -0.888 Experimental pelagic Otoliths Machias et al. Sardin  NW Aegean S 6- 9 208 andings Length- frequency Voulgaridou and Stergiou (2003)   and Ionian Seas - 4 181  seine l   and es (1991)  Central Aegean and Ionian Seas 1999- 2001 0.314 191 -1.839 Experimental pelagic trawl samples and onboard sampling Otoliths Machias et al., (2001b) W Ae  Sea 1997- 0.75 2  Purse Length- Loukmido frequency (1998) analysis W Ae  Sea 2000- 2003 0.77 1 Length- Tsianis et frequency  al. (2003) analysis Otoliths Kallianiotis et ahracia a 2000- 2001 0.49 1 -1.276 Purse (2003) and Ionian Seas 1999 trawl samples and onboard sampling (2000a) e NW Aegean Sea 1996- 0.80 219  Purse seine landings Length- Tsianis (2003) 2003 frequency analysis ea 199 199 0.86  Purse seine l analysis Aegean  1983 198 0.300 -3.210 Purse andings Scales Tserpes Tsimenid BIOMASS ESTIMATES Direct biomass estimates for anchovy and sardine through acoustic and/or egg surveys have been obtained in recent years in the framework of various EU and national projects (Tsimenides et al.,   199 2001b). First 5a; 1998; Machias et al., 2000a; acoustic surveys were the Ts the es coa ias et al., 1996; 1997; 2000a, b; 2001a, b; me  DEPM for the estimation of biomass of anchovy in the north be or spawning 0 to 32000 t.  conducted in north Aegean in 1987-1988, in the central Aegean in 1989-1990, and in south Aegean in 1991, all aiming to study  echo-distribution of small pelagic fish assemblages (Tsimenides, 1989; imenides et al., 1995b). Later surveys in 1995-2001 were focused on the study of vertical and horizontal distribution and timation of biomass of anchovy and sardine in the north Aegean and the stal areas of the central Aegean and Ionian Seas (Mach Giannoulaki et al., 1999; 2001; 2003; Tsimenides et al., 1998; Maravelias et al., 1997). The acoustic thod has been applied concurrently with Aegean during June 1995, and in the central Aegean and Ionian Seas in June 1999, as well as for the estimation of the biomass of sardine in the latter area in winter 2000-01 (Tables 2 and 3). The abundance estimates of anchovy ranged between 40000 and 45000 t in the north Aegean and tween 11000 and 15000 t in the central Aegean and Ionian Seas. The estimates of sardine biomass (total biomass) in the central Aegean and Ionian Seas ranged from 2000 0 95 100 105 110 115 120 125 130 135 140 145 150 155 160 165 170 17 5 10 15 5 % 20 25 30 Total length (mm) Pelagic trawling Purse seine landings Fr eq ue nc y (% ) % Fr eq ue nc ncy an acoustic survey and from purse seine landings. Data from y (% )  Figure 7 North Aegean Sea, June 1995. Length freque distributions of anchovy from experimental pelagic trawling during Tsimenides et al. (1998).  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 61 Table 2 Biomass estimates of anchovy stocks in the Greek Seas (in tonnes, with CVs in parentheses). DEPM: Daily  production method. LVPA: Length-based virtual population analysis. r Month Region egg Yea  Acoustic DEPM survey LVPA References of direct surveys survey (spawning biomass) 199  3 June North Aegean  40643 (0.276) 40236 Tsimenides et al. (1995a, 1998) 1995 June North Aegean 44601 (0.120) 42708 (0.181)  Tsimenides et al. (1998) Machias et al. (1997) Somarakis et al. (1997) 1996 June North Aegean 39475 (0.132)   Tsimenides et al. (1998) 8 July Central Aegean 14261 (0.298)  199 13446 Machias et al. (2000a, 2001a) and Ionian Seas  1999 June Central Aegean and Ionian Seas 14511 (0.280) 11861 (0.278) 13044 Machias et al. (2000a, 2001a) Somarakis et al. (2002)  nian Sea; in tonnes Year Month Acoustic DEPM survey LVPA References Table 3 Biomass estimates for sardine stocks in the Greek Seas (central Aegean and Io with CVs in parentheses). DEPM: Daily egg production method. LVPA: Length-based virtual population analysis. of direct surveys survey (Spawning Biomass) 1999 December  32594 (0.301) 27086 Machias et al. (2001b) 2000-2001 December- February 19826 (0.429) 24207 (0.225) 30317 Machias et al. (2001b) Somarakis et al. (2001)  SU bio experimental sampling and surveys have been generally highly discontinuous in time and space, mainly Ga sity of Thessaly [in Greek with English Abstract]. ember 1999. Rapp. Comm. int. Mer Medit. 36, 268. Ga  S., Koutsikpoulos, C., Machias, A., Theodorou, A., 2003b. Ovarian atresia in the Mediterranean Gan . Pattern of oocyte development and batch fecundity in the Gia 75-288. ic Conference of Ichthyologists, pp. 61-64. fish. Mar. Ecol. Prog. Ser. 265, 243- 253. MMARY Based on the above-mentioned studies, it is clear that there is a lack of regular, long-time series of data on mass and biological parameters of anchovy and sardine throughout the Greek waters. Indeed, linked to the existence of on-going projects and ending with the specific project that initiated them. This represents an impediment to the study of the effect of fishing and environmental variability on small pelagic fishes (see, e.g., Stergiou and Lascaratos, 1992; Stergiou et al., 1997), and thus, to their management, which is based exclusively on technical measures. REFERENCES Fryer, R.I., 1991. A model of between-haul variation in selectivity. ICES J. Mar. Sci. 48, 281–290. nias, K., 2003. Oceanographic and Biological Study of Sardine Egg Production, Sardina pilchardus (Walbaum, 1792) in Coastal Waters of Central Greece. PhD Thesis, Univer Ganias, K., Somarakis, S., Caragitsou, E., Koutsikopoulos, C., Machias, A., Theodorou, A., 2001. Differential egg production of sardine off the central Hellenic coasts in Dec Ganias, K., Somarakis, S., Machias, A., Theodorou, A., 2003a. Evaluation of spawning frequency in a Mediterranean sardine population. Mar. Biol. 142, 1169-1179. nias, K., Somarakis, sardine, Sardina pilchardus sardina. Mar. Biol. 83, 1327-1332. ias, K., Somarakis, S., Machias, A., Theodorou, A., 2004 Mediterranean sardine. Fish. Res. 67, 13-23. nnoulaki, M., Machias, A., Tsimenides, N., 1999. Ambient luminance and vertical migration of sardine (Sardina pilchardus). Mar. Ecol. Progr. Ser. 173, 2 Giannoulaki, M, Machias, A., Somarakis, S., Koutsikopoulos, C., Manousakis, L., Tsimenides, N., 2001. Study of sardine (Sardina pilchardus Walb.) distribution in the central Aegean and Ionian Seas by hydroacoustics. In: Proceedings of the 10th PanHellen Giannoulaki, M., Machias, A., Koutsikopoulos, C., Haralabous, J., Somarakis, S., Tsimenides, N., 2003. Effect of coastal topography on the spatial structure of the populations of small pelagic  Overview of biological data of Greek stocks of anchovies and sardines, Somarakis, S. et al. 62 Kallianiotis, A., Papantoniou, V., Euthimiadis, K., Panora, D., Argyri, A., 2003. Age and growth of anchovy Engraulis encrasicolus (Linnaeus, 1758) in Thracian Sea. In: Proceedings of the 11th PanHellenic Conference of Ichthyologists, pp. 43-46. Loukmidou, S.P., 1998. Biology and Dynamics of Anchovy Engraulis encrasicolus (Linnaeus, 1758) (Pisces: Engraulidae) in Thermaikos Gulf. M.Sc. Thesis, Aristotle University of Thessaloniki [in Greek with English Lou ulis Ma tribution of anchovy and sardine in north Aegean th Ma ousakis, L., Kapantagakis, A., Tsimenides, N., 1997. Estimation of s. In: Proceedings of the 10th PanHellenic Conference of Ichthyologists, pp. 57-60. o 98/030, Final Report. ulis encrasicolus). Mol. Biol. Evol. 13(1), Ma atial distribution of pelagic species in the Nikoloudakis, G., Machias, A., Somarakis, S., Koutsikopoulos, C., Tsimenides, N., 2000. Comparison of growth in two . 6  PanHellenic Symp. Oceanogr. Fish. 6(2), pp. 88-93.  [in Greek with English abstract]. rakis, S., 2005. Marked inter-annual differences in reproductive parameters and daily egg production of anchovy in the northern Aegean Sea. Belg. J. Zool. 135, 247-252. Somarakis, S., Machias, A., Kapantagakis, A., Tsimenides, N., 1997. Application of the Daily Egg Production Method (DEPM) for the estimation of the northernorth Aegean Sea anchovy stock in June 1995. In: Proc. 5th Panhellenic Symp. Oceanogr. Fish. 5(2), pp. 43-46. Somarakis, S., Machias, A., Koutsikopoulos, C., Maraveya, E., Giannoulaki, M., Tsimenides, N., 2000a. Distribution of anchovy and its spawning grounds off the central Aegean and Ionian Seas. In: Proc. 6th Panhellenic Symp. Oceanogr. Fish. 6(2), pp. 94-98. Somarakis, S., Maraveya, E., Tsimenides, N., 2000b. Multispecies ichthyoplankton associations in epipelagic species: Is there any intrinsic adaptive function? Belg. J. Zool., 130 (Suppl 1), 125-129. abstract]. kmidou, S.P., Stergiou, K.I., 2000. Length-weight relationships and length frequencies of anchovy, Engra encrasicolus (Linnaeus, 1758), in the Thermaikos Gulf. Proc. 6th PanHellenic Symp. Oceanogr. Fish. 6(2), 109-113. chias, A., Somarakis, S., Tsimenides, N., 1996. Comparative dis basin: June 1995. Proc. 17  Scient. Symp. Greek Soc. Biol. Sci. 17, 135-137. chias, A., Somarakis, S., Giannoulaki, M., Man the northern Aegean anchovy stock in June 1995 by means of hydroacoustics. In: Proc. 5th PanHellenic Symp. Oceanogr. Fish. 5(2), 47-50. Machias, A., Somarakis, S., Drakopoulos, P., Magoulas, A., Koutsikopoulos, C., 2000a. Evaluation of the Southern Greek Anchovy Stocks. Institute of Marine Biology, Crete. Project 97-0048, Final Report. Machias, A., Somarakis, S., Koutsikopoulos, C., 2000b. The anchovy stocks in the central Aegean and Ionian Seas. Hellenic Fish. News 233, 89-95 [in Greek]. Machias, A., Giannoulaki, M., Somarakis, S., Koutsikopolos, C., Manousakis, L., Kapantagakis, A., Tsimenides, N., 2001a. Estimation of the anchovy (Engraulis encrasicolus) stocks in the central Aegean and Ionian Seas by means of hydroacoustic Machias, A., Somarakis, S., Magoulas, A., Koutsikopoulos, C., 2001b. Evaluation of the Southern Greek Sardine Stocks. Institute of Marine Biology, Crete, Contract N Magoulas, A., Tsimenides, N., Zouros, E., 1996. Mitochondrial DNA phylogeny and the reconstruction of the population history of a species: The case of the European anchovy (Engra 178-190. ravelias, C.D., Machias, A., Somarakis, S., Tsimenides, N., 1997. Sp northern Aegean Sea: use of geostatistics and environmental variables. In: Proc. 5th PanHellenic Symp. Oceanogr. Fish. 5(2), pp. 119-122. Maraveya, E., Somarakis, S., Machias, A., 2001. Batch fecundity of anchovy, Engraulis encrasicolus, in the NE Aegean Sea. Rapp. Comm. int. Mer Medit. 36, 299. anchovy stocks. In: Proc. 6th PanHellenic Symp. Oceanogr. Fish. 6(2), pp. 104-108. Petrakis, G., Stergiou, K.I., 1997. Size selectivity of diamond and square mesh codends for four commercial Mediterranean fish species. ICES J. Mar. Sci. 54, 13–23. Ramfos, A., Koutsikopoulos, C., Fragopoulou, N., Machias, A., Somarakis, S., Pyronovaki, E., Lykakis, I., 2000. Hydrology and biological features in the coastal areas of Central Greece during the anchovy spawning period. In: Proc th Somarakis, S., 1993. Contribution to the Study of the Planktonic Stages of Anchovy in the Aegean Sea. MSc Thesis, University of Crete [in Greek with English abstract]. Somarakis, S., 1999. Ichthyoplankton of the Northeastern Aegean Sea with Emphasis on Anchovy Engraulis encrasicolus (Linnaeus, 1758) (June 1993, 1994, 1995, 1996). PhD Thesis, University of Crete Soma  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 63 Somarakis, S., Ganias, K., Koutsikopoulos, C., Machias, A., Papaco Production Method (DEPM) to small sardine stocks in the easter nstantinou, C., 2001. Applying the Daily Egg n Mediterranean (central Greece). ICES Study 6, 16-23. y in th e eastern Mediterranean using 27 thly anc lis  in Sea. Fish. Bull. he distribution of phytoplankton pigments and ishery of y aga, The I  Quarte ), 34-37. atic v nd t hovy/sar tio in Hellenic waters. Geojournal  1996. Sex ratio, as at red  iol. 49, 5 572. os, D. netos, A., mezoglo 997. Th nic Seas cs, n. Mar iol. Ann. R 5-538. tion  growth rate differences in populations of  ia eas. Cybiu -22. n of Biological Parameters of Anchovy and Sardine in NW Aegean. M.Sc. essaloniki [in Greek with English abstract]. nis, D.E., Moutopoulos, D.K., Stergiou, K.I., 2003. Elements of biology and reproduction of Engraulis encrasicolus (Linnaeus, 1758) in NW Aegean. In: Proceedings of the 1st PanHellenic Conference of Ecologists, Thessaloniki. Tsimenides, N., 1989. The acoustic survey program in Greek waters. FAO Fish. Rep. 412, 127-134. Tsimenides, N., Somarakis, S., Magoulas, A., Tserpes, G., 1995a. Evaluation of the Anchovy Stocks in the Aegean Sea. Hellenic Center for Marine Research, Final report. Project no XIV-1/MED/91/011. Tsimenides, N., Bazigos, G., Georgakarakos, E., Kapadagakis, A., 1995b. Distribution of acoustic pelagic fish populations in the northernorth Aegean Sea. In: Proceedings of the 1st World Fisheries Congress, Athens, pp. 59- 65. Tsimenides, Ν., Papaconstantinou, K., Kapandagakis, A., Machias, A., Somarakis, S., Petrakis, G., Karagitsou, I., 1998. Development of Greek Fishery. EPET 125, Final Report. Voulgaridou, P., Stergiou, K.I., 2001. Trends in various biological parameters of the European sardine, Sardina pilchardus (Walbaum, 1792), in the Eastern Mediterranean Sea. Sci. Mar. 67 (Suppl. 1), 269-280.  Group on the Estimation of Spawning Biomass for Sardine and Anchovy (SGSBSA), Lisbon, 22-25/10/2001. Somarakis, S, Koutsikopoulos, C., Machias, A., Tsimenides, N., 2002. Applying the daily egg production method to small stocks in highly heterogeneous seas. Fish. Res. 55, 193-204. Spanakis, E., Tsimenides, N., Zouros, E., 1989. Genetic differences between populations of sardine (Sardina pilchardus) and anchovy (Engraulis encrasicolus) from the Aegean and Ionian Seas. J. Fish Biol. 35, 417-437. Stergiou, K.I., 1989. Modelling and forecasting the fishery of pilchard (Sardina pilchardus) in Greek waters using ARIMA time series models. J. Cons. Int. Explor. Mer 4 Stergiou, K.I., 1990. A seasonal autoregressive model of the anchovy Mediterranean. Fish. Bull. (U.S.) 88, 4 Stergiou, K.I., 1991. Describing and for vector autoregressions. Fish. Res. 11, 1 Stergiou, K.I., 1992. Variability of mon (U.S.) 90, 211-215. Stergiou, K.I., Georgopoulos, D., 1993. T (Engraulis encrasicolus) in the Hellenic Se Stergiou, K.I., Laskaratos, A., 1997. Clim 41(3), 245-254. Stergiou, K.I., Economidis, P., Sinis, A., western north Aegean Sea. J. Fish B Stergiou, K.I., Christou, E., Georgopoul chemistry, biology and fisheries. Ocea Tserpes, G., Tsimenides, N., 1991. Evalua (Walbaum, 1792) from the Aegean and Ion Tsianis, D.E., 2003. Temporal Variatio Thesis, Aristotle University of Th Engraulis encrasicolus fisher e eastern 11-414. casting the sardine-anchovy complex in the -141. catches of hovy Engrau  encrasicolus  the Aegean  the f  anchov as. N CLARM rly 16(2  ariability a he anc dine ra  spawning se on and size maturity of  bandfish in 61– , Ze  Souver u, C., 1 e Helle : physi . B ev. 35, 41   of n s Sardina pilchardus m 15, 15 Tsia  Overview of biological data of Greek stocks of anchovies and sardines, Somarakis, S. et al. 64 APPENDIX: ESTIMATION OF LENGTH AT FIRST MATURITY Lengt k of the spawning season ( Fish were collected with a pelagi gically in female specimens (Somarakis et al., 2001; 2002). The size at which 50% of females were mature (size at maturity, L50) was estimated from the relationship between percentage P of mature fish at length class L. This relationship, widely used for maturity studies (e.g., Stergiou ), is described by the logistic function: P = e(v1+v2 L) / (1+e(v1 + v2 · L)),and the value an be estimated from the expressions: L50 = -v1/v2, L25 = [-Ln(3)-v1]/v2, L75 = [L The proportion of mature fish for each 10 mm length class was calculated by sex and v1, v2 were calculated us algorithm likelihood, ln(L) (Petrakis and Stergiou, 1997). A test for over– was obtained by e deviance statistic ∆ Table h at first maturity of anchovy and sardine was estimated from samples collected during the pea during DEPM surveys) in summer 1999 and winter 2000-2001, respectively. c trawl and maturity was assessed histolo  et al., 1996  of L50, L25, L75 c n(3)-v1]/v2.  A1 Estimated parameters for the size at maturity, L50. SE: Standard error; CI: 95% confidence intervals; ∆: deviance statistic; df: degrees of freedom tal lengths in mm. dine . All lengths are to Anchovy SarParameter estimates Aegean Sea Ionian Sea Aegean Sea Ionian Sea ing Fryer’s (1991) by maximising the log– v1 -28.175 -100.739 -24.992 -15.554 v2 0.270 0.963 0.211 0.137 SΕ v1 2.957 8.536 2.505 1.147 dispersion estimating th and comparing it to a χ2 distribution on N-2 degrees of freedom (Petrakis and Stergiou, 1997). The data are over– dispersed if ∆ > χ2. The standard errors and the 95% confidence intervals of the estimated value of L50 were calculated following procedures described in Petrakis and Stergiou (1997). Results of the analysis are given in Table A1.  SΕ v2 0.028 0.082 0.021 0.010 L50 104.41 104.62 118.39 113.83 CI L50 102.9–105.6 104.3-104.9 116.5-120.3 111.9–115.7 L25 100.34 103.48 113.18 105.79 L75 108.48 105.77 123.59 121.87 ∆ 12.610 7.941 22.708 22.637 Df 13 13 14 14 χ2 22.362 22.362 23.685 23.684  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 65 A COMPARISON OF GROWTH PARAMETERS OF AUSTRALIAN MARINE FISHES NORTH AND SOUTH OF 28° SOUTH1 Claire Andersen2 University of Queensland Brisbane QLD 4072 Australia Daniel Pauly The Sea Around Us Project, Fisheries Centre, University of British Columbia, 2202 Main Mall, Vancouver, BC V6T 1Z4 Canada; Email: d.pauly@fisheries.ubc.ca ABSTRACT Estimated growth parameters in fish contribute to knowledge of fish biology, and assist in the appropriate management of commercial fisheries resources. Parameters can also be used to test theories of organic growth. Here, one such test is conducted, of the theory that the availability of oxygen to the internal tissu  bly unpublished documents from Australian government agencies. All rowth parameter sets, corresponding to fish populations north or south of 28°S, were assigned a mean water temperature, and analyzed using multiple log-linear regression of K vs. temperature and L∞. The n in which both L∞ and temperature had a significant effect on K. Thus, nmental temperature, which varies strongly along the east and west coast of Australia, had an effect n on the biology of fish, and for management of fisheries. Few studies have looked simultaneously at the growth of a large number of e valuable insights this can provide (Cury and Pauly, 2000). d hence can devote more of it to growth, with the result that they reach larger size (Figure 1). In the von Bertalanffy equation (see below), asymptotic length, L∞, is positively and closely  other sources, notably unpublished documents from Australia. All growth    es of fish limits their rate of growth, as proposed earlier by the second author. More precisely, what is being tested is a follow-up of that theory, i.e., that, other things being equal, fishes in the cold, high-latitude part of their overall distribution range will have von Bertalanffy growth parameters (L∞, K) different (higher L∞, lower K) from those in the warm, low-latitude part of that distribution range. To test this, growth parameters of Australian fishes were assembled from FishBase, and complemented with growth parameters from other sources, nota g result was a multiple regressio enviro on the growth parameters of Australian fishes, as predicted by the theory tested here. INTRODUCTION Growth parameters of fish are an important source of informatio species, in spite of th The purpose of this study was to compare the growth parameters of Australian fishes along a latitude (i.e., temperature) gradient as a way of indirectly testing the theory formulated by Pauly (1981, 1984, and see Pauly, 2006, this volume), which states that the growth of fish is linked with their oxygen supply. Other things being equal, fish at low temperature should use less of the oxygen available to them for maintenance, an correlated with maximum size, and inversely correlated with the parameter K. Hence, at low temperatures, L∞ should be high, and K low, and the converse should apply at high temperatures (Longhurst and Pauly, 1987). To test this, growth parameters of Australian fishes were assembled from FishBase, and complemented with growth parameters from                                               th of 28° South. In: Palomares, M.L.D., Stergiou, K.I., Pauly, D. (eds.), Fishes in Databases and Ecosystems. Fisheries Centre Research Reports claire.andersen@dpi.qld.gov.au 1 Cite as: Andersen, C., Pauly, D., 2006. A comparison of growth parameters of Australian marine fishes north and sou 14(4), pp. 65-68. Fisheries Centre, University of British Columbia [ISSN 1198-6727]. 2 Present address: Department of Primary Industries and Fisheries, GPO Box 46, Brisbane, Queensland 4001, Australia; Email: .  Growth of Australian marine fishes, Andersen, C., Pauly, D. 66 parameter sets, corresponding to fish populations north or the south of 280 S, were assigned a mean water temperature.  Figure 1 Schematic representation and body weight in fishes. Main relative gill area (and hence oxyge temperature will have a low main species  of the relationship between relative gill area (and hence relative oxygen supply) tenance metabolism determines the maximum size that can be reached because n supply) must decline with body weight. A: A fish exposed to low environmental tenance metabolism and reaches a larger size. B: Exposure of a fish of the same to high temperatures causes rapid denaturation of body protein, requiring more O2 to be diverted to protein synthesis, and hence to maintenance of metabolism. Other things being equal, this fish will remain smaller than that e). in A (see also Pauly 2006, this volum  MATERIALS AND METHODS Growth parameters were obtained from the POPGROWTH Table of FishBase (www.fishbase.org). These wth parameters from other sources, notably various publishedwere complemented with gro unpublished documents from Australia (see References). The von Bertalanffy growth function (VBGF) for length has the form: L  and … 1) ∞ ere to grow a very long time (indefinitely, actually), K is a coefficient of dimension e-1, expressing the rate at which L∞ is approached, and to is the (usually negative) age the fish would gth zero if they had always grown in the manner predicted by the equation (which they data. When this was not available from FishBase or th  growth parameter sets, corresponding to the mean water temperature at the sampling using the sampling Ce t = L∞ (1 – e-K(t-to)) where Lt is the predicted mean length at time t, L  is the asymptotic length, or the mean length the fish would reach if they w tim have had at len usually do not). The length-at-age data from which these growth curves paramaters were usually obtained from reading otoliths or vertebrae. Growth parameters were also inferred from length-frequency and tagging-recapture e references below, a temperature was assigned to all location and depth (actual or inferred), based on data available from the Australian Oceanographic Data ntre (www.aodc.gov.au/). RE logK = 0.1652 + 0.0245·Temp - 0.681·logL∞ … 2) SULTS AND DISCUSSION The following multiple regression was obtained from the 190 sets of growth parameters obtained:  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 67 where K is in year-1, Temp in 0C, L∞ in cm, and log refers to base 10. The r2 value was 0.544 and the effect temperature was significant (P<0.01), in addition to Lof  was solved for logK using 10 Australian waters north of 28  parameters in the opposite quadrant were com however, is that high inc rate, and hence the amount requirements to be met. This dif tha Th offer them food of different suitability and in different no coa ther, food can be lues We an RE ished report held in NIWA library, Wellington. ock structure, Mar. Freshwat. Res. 47(2), 109- Bro ge, growth and mortality of red throat emperor, Lethrinus miniatus, from the Cam gy and Interaction in the Northern Australian Small Mackerel Fishery. Cu  reproduction and growth of fishes. Ecol. Res. 15, 101-106.                                                 ∞, in predicting K. This different temperatures, as shown in Figure 2, i.e., a plot of logK vs. logL∞, with isotherms at 30, 20, and 0C superimposed. As might be seen, the fish in the high K – low L∞ quadrant of Figure 2 stem overwhelmingly from 0 S, while the fish with growth sampled south of 280S. These findings are patible with a number of biological mechanisms. The most likely mechanism, temperature reduces the scope for growth of fishes, by reasing their metabolic of water having to be passed across the gills in order for their respiratory  task is made even more ficult by the fact that less oxygen is dissolved in warm n in cold water (Pauly 1981). e different habitats of fishes, whether estuarine, coastal, neritic or oceanic, Figure 2 Plot of the von Bertalanffy parameter (log) K vs. the parameter L∞ in 190 population of Australian marine fishes north and south of 280 South. The warm-water populations from the North tend to have lower L∞ and higher K than those in the South, and vice versa for those in the colder South. The isotherms for 10, 20 and 30 C were drawn using Equation 2 (see text). 0densities. However, there is  reason to assume that the suitability or density of food available to fish of different species should change, along the Australian sts, in a close relationship to temperature or latitude (Longhurst and Pauly, 1987). Ra assumed to vary randomly, and to represent one of the causes for the variability around the va predicted by the multiple regression (Equation 1) and consequently, in Figure 2.  thus feel that the evidence we presented support the existence of a strong linkage between fish growth d respiration. FERENCES3 Annala, J.H., Sullivan, K.J., O'Brien, C.J. (compilers), 1999. Report from the Fishery Assessment Plenary, April, 1999: Stock Assessments and yield estimates. Unpubl Begg, G., Sellin, M., 1998. Age and growth of school mackerel (Scomberomous queenslandicus) and spotted mackerel (S. munroi) in Queensland east-coast waters with implications for st 120. wn, I., Sumpton, W., 1998. A southern Great Barrier Reef: Queensland, Australia. Bull. Mar. Sci. 62(3), 905-917. eron, D.S., Begg, G.A., Fisheries Biolo Fisheries Research and Development Corporation [Internal report]. ry, P., Pauly, D., 2000. Patterns and propensities in  3 Some of these references document sources of growth parameters (or growth data from which such parameters could be computed), and have not been cited in the text; the growth parameters were all subsequently entered in FishBase.  Growth of Australian marine fishes, Andersen, C., Pauly, D. 68 Fairclough, D.V., Dimmlich, W.F., Potter, I.C., 2000. Length and age compositions and growth rates of the Australian herring, Arripis Georgiana, in different regions. Mar. Freshwat. Res. 51(6), 631-640. Francis, R., Pa  annual ring counts: validation by tagging an at. Res. 43, 1069-1089. Gillanders, B.M., Ferrell, D.J., Andrew, N.L., 1999. Ageing methods for yellowtail kingfish, Seriola lalandi, and results from age- and size-based growth models. Fish. Bull. 97, 812-827. Hoyle, S, Brown, I., Dichm d Fish Stock Assessment and Monitoring Progra /161. Horn, P.L ccius australis) Hyndes, G.A., Platell, M.E., Potter, I.C., Lenanton, R.C.J., 1998, Age composition, growth, reproductive biology and t of King George whiting, Sillaginodes punctata, in coastal waters of southwestern Australia. Fish. Bull. 1. e , J.M., Rawlinson, N.J.F., 1993. Age and growth of three species of Clupeidae from Kiribati, tropical central Pacific. J. Fish Biol. 43, 89-108. Morison, A.K., Coutin, P.C., Robertson, S.G., 1998. Age determination of black bream, Acanthopagrus butcheri  Australian waters. Aust. J. Mar. Freshwat. Res. 43, 1241-1267. ment of the Burnett River, Maroochy River and Pumicestone Passage. Queensland . (eds.), Fishes in Databases and Ecosystems. Fisheries Centre Research Reports 14(4), pp. 89-95. Fisheries Centre, University of British Columbia. ys auratus) in south-east Australian waters using capture- recapture data. New Zeal. J. Mar. Freshwat. Res. 13(2), 279-284. kano, H., Shimizu, M., 1998, Age, growth and reproduction of the Oceanic whitet k from the Pacific Ocean. Fish. Sci. 64(1), 14-20. Tilzey, R.R.J., 1994. The South East Fishery: A Scientific Review with Particular Reference to Quota Management. Bureau of Resource Sciences, Canberra. Wise, B.S., Potter, I.C., Wallace, J.H., 1994. Growth, movements and diet of the terapontid, Amniataba caudavittata, in an Australian estuary. J. Fish Biol. 45, 917-931.  ul, L.J., Mulligan, K.P., 1992. Ageing of adult snapper (Pagrus auratus) from otolith d oxytetracycline injection. Aust. J. Mar. Freshw ont, C., Sellin, M., Cosgrove, M., McLennan, M., 2000. Integrate m. Fisheries Research and Development Corporation Project no 94 ., 1997. An ageing methodology, growth parameters and estimates of mortality for hake (Merlu from around South Island, New Zealand. Mar. Freshwat. Res. 48, 201-209. recruitmen 96, 258-27 Hyndes, G.A., Potter, I.C., Hesp, S.A., 1996. Relationships between the movements, growth, age structures and reproductive biology of the teleosts Sillago burrus and S. vittata in temperate marine waters. Mar. Biol. 126, 549- 558. Longhurst, A., Pauly, D., 1987. Ecology of Tropical Oceans. Acad mic Press, San Diego. Milton, D.A., Blaber Morrison, A.K., Robertson, S.G., 1995. Growth, Age Composition and Mortality of Blue-eye Trevalla (Hyperoglyphe antarctica). Victorian Fisheries Research Institute Internal Report No. 220. Morrison, A.K., 1996, Age and Growth of Major Species in the South East Fishery. Central Ageing Facility, Marine and Freshwater Resources Institute, Department of Natural Resources and Environment, Queenscliff, Victoria. (Sparidae), from the Gippsland Lakes of south-eastern Australia indicates glow growth and episodic recruitment. Mar. Freshwat. Res. 49(6), 491-498. Moulton, P.L., Walker, T.I., Saddlier, S.R., 1992. Age and growth studies of gummy shark, Mustelus antarcticus (Günther) and school shark, Galeorhinus galeus (Linnaeus) from south O'Neill, M.F., 2000. Fishery Assess Department of Primary Industries Project Report Q099012. Pauly, D., 1981. The relationships between gill surface area and growth performance in fish: a generalization of von Bertalanffy's theory of growth. Ber. Deutsch. Wissenschaf. Komm. Meeresfors. 28(4), 251-282. Pauly, D., 1984. A mechanism for the juvenile-to-adult transition in fishes. J. Cons. int. Explor. Mer 41, 280-284. Pauly, D., 2006. Effects of lake and pond aeration on fish growth and related processes. In: Palomares, M.L.D, Stergiou, K.I., Pauly, D Sanders, M.J., Powell, D.G.M., 1979. Comparison of the growth rates of two stocks of snapper (Chrysophr Seki, T., Taniuchi, T., Na ip shar Simpfendorfer, C.A., Chidlow, J., McAuley, R., Unsworth, P., 2000. Age and growth of the whiskery shark, Furgaleus macki, from south-west Australia. Env. Biol. Fishes 58, 335-343. Tziournis,V., Kingsford, M.J., 1999. Reproductive biology and growth of the temperate damselfish Parma microlepis. Copeia 1999(2), 384-361.  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 69 GROWTH PARAMETERS AND LENGTH-LENGTH RELATIONSHIPS OF GREEK FRESHWATER FISHES1 Platonas K. Kleanthidis, Konstantinos I. Stergiou Aristotle University of Thessaloniki, School of Biology, Department of Zoology, Laboratory of Ichthyology, P.O. BOX 134, Thessaloniki 54124, Greece; Email: kstergio@bio.auth.gr ABSTRACT In this paper, we gathered from the literature: (a) a total of 80 length-length relationships for 20 Greek freshwater fish species and one hybrid; and (b) von Bertalanffy growth parameters for 54 freshwater fish stocks, belonging to 22 species and one hybrid. The relationship between log10L∞ and log10K for all stocks, excluding one outlier, was: log10K = -1.07·log10L∞ + 0.77 (r2 = -0.66, n = 53, P < 0.05). INTRODUCTION Relationships between different types of lengths (length-length relationships) are very important for comparative growth studies (Froese and Pauly, 2000; www.fishbase.org). In addition, growth parameters and maximum observed length, Lmax, and age, tmax, are also important for management, comparative growth studies, and testing life-history theories, with Lmax being used for the estimation of plethora biological parameters using existing empirical equations (e.g., Pauly, 1998; Froese and Binohlan, 2000). In this paper, ng to 20 Greek eshwater fish species and one hybrid (from nine lakes, three rivers and two lagoons) and Lmax and tmax values and the von Bertalanffy growth parameters for 54 Greek freshwater fish stocks, belonging to 22 Growth in length has been described using the von Bertalanffy (1938) growth function (VBGF): we gathered from the literature 80 length-length relationships referri fr species and one hybrid, from twelve lakes and two rivers in Greece. MATERIALS AND METHODS We gathered articles with biological data pertinent to Greek freshwater fish using the Aquatic Sciences and Fisheries Abstracts (ASFA), which cover peer-reviewed as well as grey literature articles. We also used any available unpublished theses and other technical reports. The following type of information was collected: (a) length-length relationships, expressed in cm; (b) maximum observed length and age, Lmax and tmax, in cm and year respectively; and (c) the von Bertalanffy growth parameters K, L∞ and to, in year-1, cm and year, respectively. The word ‘stock’ is used here to indicate sets of parameters corresponding to different sexes, years, and areas. Lt = L∞(1-e –K(t-t0)) … 1) where L∞ is the asymptotic length, i.e., the length a fish would reach if it were to grow indefinitely; K is the rate at which L∞ is approached (in year-1); and t0 is the theoretical origin of the curve, i.e., the age of the fish at zero length (in year). When the authors did not estimate VBGF growth parameters, we estimated them from the back-calculated length-at-ages provided by the authors, using the non-linear least-squares method (Gayanilo et al., 1994).                                                   1 Cite as: Kleanthidis, P.K., Stergiou, K.I., 2006. Growth parameters and length-length relationships of Greek freshwater Palomares, M.L.D., Stergiou, K.I., Pauly, D. (eds.), Fishes in Databases and Ecosyste isheries Centre Research Repo fishes. In: ms. F rts 14(4), pp. 69-77. Fisheries Centre, University of British Columbia [ISSN 1198-6727].  Growth parameters of Greek freshwater fishes, Kleanthidis, P.K., Stergiou, K.I.  70 ere collected from the e sample size ranged from 9 individuals, for Knipowitschia caucasica in the Evros River, to 2575 individuals, for Atherina boyeri in Trichonis Lake (Table 1). Sample size was not reported for 15 cases (Table 1). In 24 cases, the sample size was less than 200 individuals (Figure 1). For 49 cases, length-length relationships referred to both sexes combined, for 30 cases they were sex-specific (15 for males and 15 for females) and for 1 case it referred to immature fish (Table 1). 0 2 4 6 8 10 12 14 16 18 0 400 800 1200 1600 2000 2400 Sample size (individuals) N um be r o f s to ck s  Figure 1 wn in Table 1.  Table 2 summarizes the  fish stocks, belonging to 22 species and one hybrid, from twelve lak t to sampling frequency, samples were collected mainly on a m n  s  a lesser extent, on a seasonal or other ba  a  scale readings (in  other hard skeletal elements (i.e., opercular bones: 4 cases; fin spines: 3 cases; combination of skeletal elements: 4 cases) (Table 2). The von Bertalanffy growth parameters were provided in the original studies for 11 cases only, estimated using the non-linear method (in 7 cases), the Ford-Walford plot (in 3 cases) and Rafail’s (1973) method (in 1 case) using back-calculated length-at-ages (Table 2). In the remaining 43 cases we estimated growth parameters from the back-calculated (in 41 cases) or the observed (in 2 cases) length-at-ages provided in the original studies, using the non-linear regression method. Growth parameters referred to combined sexes for 22 cases (Table 2). Lmax was provided in 44 cases and ranged from 9.5 cm, for Pseudorasbova parva in Lake Mikri Prespa, to 76 cm, for Cyprinus carpio in Lake Vistonis (Table 2). It exhibited a primary mode at 20-25 cm and a secondary one at 35-40 cm (Figure 2a). A value of tmax was provided in 50 cases and ranged from 3 year, for Pseudorasbova parva in Lake Mikri Prespa, to 14 year for Barbus albanicus in Lake Kremasta (Table 2). For 34 cases, tmax ranged between 6 and 9 year (Figure 2b). The von Bertalanffy L∞ values ranged from 11.8 to 88 cm (median = 28.8 cm), with a primary mode between 25-30 cm and a secondary one at 40-50 cm (Figure 2c). The K values ranged from 0.081 to 0.577 year-1 (median = 0.139 year-1), with a mode at 0.10-0.15 year-1 (Figure 2d). Finally, t0 values ranged from -3.218 to 0.236 year (median = -0.798 year), with a mode at -1.0 to -0.5; 41 cases had values between -1 and 0 (Figure 2e). The Lmax/L∞ ratio ranged between 0.60 and 1.76 (mean = 0.84; median = 0.85), with a mode between 0.8-0.9. The relationship between Lmax and L∞ was logL∞ = 0.972·logLmax + 0.123 (r2 = 0.83, n = 44, P < 0.05). The relationship between logK and log < 0.05  tionship between logK vs. logL∞, excluding one stock, which seems to be an outlier (i.e., Cyprinus carpio in Lake Vistonis; Figure 3), was: logK = -1.07·logL∞ + 0.77 (r2 = -0.66, n = 53, P < 0.05). The rela RESULTS AND DISCUSSION Overall, 80 length-length relationships between total, fork and standard length w literature (Table 1), corresponding to 20 fish species and 1 hybrid from 9 lakes, 3 rivers, and 2 lagoons. Th  Sample size (i.e., number of individuals) for the cases sho biological parameters collected for the 54 es and two rivers in Greece. With respec onthly basis (in 22 cases) or based o a single ampling event (in 17 cases), and, to sis (in 7 and 4 cases, respectively) (Table 2). Sampling frequency cases (T ble 2). Age and growth were derived mainly fromwas not reported in 4 43 cases) and to a lesser extent from tmax was logK = -0.939·logtmax + 0.038 (r2 = 0.39, n = 50, P ).Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D.  71 The above mentioned relationships can be used for estimating one variable from the other for various ter fish species in Greece for which data are lacking. For this to yield reliable estimates, it will, be nec y to expand the data set by including more stocks and species. fre however, shwa essar a 0 4 8 12 16 0 20 M 4 engt 0 h (Lma 60 aximum l x, cm) N um be r of  s to ck s c 0 12 0 20 40 60 80 st oc ks 4 8 16 von Bertalanffy Loo (cm) N um be r o f b 0 4 8 12 61 N um be r o f st oc ks 1 3 5 Ma 7 um age ( 9 11 ) 13 15 xim tmax, yr d 0 20 0.00 0.10 0.20 0.30 0.40 0.50 0.60 st oc ks 10 30 von Bertalanffy K (1/yr) N um be r o f f 0 4 8 16 von Bertalanff o yr) N to c 12 -4 -3 -2 -1 0 y t  ( um be r of  s ks e 0 4 20 ax oo N um to c  Figure 2 Distribution of the values of (a) Lmax, (b) tmax, (c) Loo, (d) K, (e) Lmax/Loo, and (f) t0 ish stocks an  this study. 8 12 16 0.4 0.6 0.8 (Lm 1 L ) 1.2 1.4 )/( be r of  s ks  for the Greek freshwater f alysed in  -1.2 -1 -0.4 -0.2 0 0.9 1.1 1.3 1.5 1.  (l -1 ) -0 - .8 0.6 7  fish 1.9 s. Loo og; eek  cm  fr ) esh K  (l og ; y r  Figure 3 Plot of logK vs. logL∞ for 54 Gr water  stock   G ro w th  p ar am et er s of  G re ek  fr es hw at er  fi sh es , K le an th id is , P .K ., St er gi ou , K .I . 72 T a  t he  le n gt h- le n gt h re la ti on sh ip s (Y  =  a + bX , i n  c m ; T L  =  to ta l l en gt h;  F L  =  fo rk  le n gt h;  S L  =  s ta n da rd  le n gt h)  o f 20  G re ek  f re sh w at er  f is h sp ec ie s an d on e h hr ee  r iv er s an d tw o la go on s.  S ex  ( M  =  m al es ; in e I =  i at ur e  s am pl e si ze .   Ye ar X b ce  b le  1  P ar am et er s of yb ri d in  n in e la ke s,  t  F = females; C = sexes com b d;  m m  fi sh );  N , Sp ec ie s Ar ea  Se x N  Y   a Re fe re n Ab r La ke  V ol vi  19 89 -9 C  L FL 1.  a nd  E co no m id is  ( 19 96 ) am is  b ra m a 1 63 1 T  10 2 0. 84 59  Va lo uk as Ab r La ke  V ol vi  19 89 -9 C   FL 0.  a nd  E co no m id is  ( 19 96 ) Ab r La ke  V ol vi  19 89 -9 C   TL  0.  a nd  E co no m id is  ( 19 96 ) Al La ke  K or on i 19 86 -8 C  L FL  99 3)  Al La ke  K or on i 19 86 - C  L SL  1. 20 2  Al La ke  M ik ri P 19 8 C   SL  0. po nt  ( 19 87 ) Al La ke  M ik ri P 19 8 C 60   SL  0. po nt  ( 19 87 ) Al o La ke  V is to ni - C   TL  0. d Si ni s (1 98 6)  Al o La ke  V is to ni - C   TL  0.  a nd  S in is  ( 19 86 ) Al o La ke  V ol vi  19 78  M    SL 81 ) Al o La ke  V ol vi  19 78  F 1  SL  Al o La ke  V ol vi  19 78  C   SL  Al o La ke  V ol vi  19 78  M    FL  Al o La ke  V ol vi  19 7 F   FL  Al o La ke  V ol vi  19 7 C   FL 1)  M es so lo ng hi 19 89 - M  3 SL 1. 99 6)  M es so lo ng hi 19 89 -9 F 6 SL 1. 99 6)  M es so lo ng hi 19 89 -9 C 9 SL 1. 99 6)  Et ol ik on  la g. 19 89 -9 M  9 SL 1. 99 6)  Et ol ik on  la g. 19 89 -9 F 5 SL 1. 99 6)  Et ol ik on  la g. 19 89 -9 C 4 SL 1. 99 6)  M es so lo ng hi 19 89 -9 M  SL 1. 99 6)  Ap M es so lo ng hi 19 89 -9 F 7 SL 1.  Ap s M es so lo ng hi 19 89 -9 C 0 SL 1.  At h  T ric ho n 19 89 -9 C 5 TL  ( 19 97 ) Ba rb  K re m a 19 82  M  L FL 1. 87 ) Ba rb La ke  K re m a 19 82  F L FL 0. 98 7)  Ba rb La ke  K re m a 19 82  C L FL 1. l. (1 98 7)  Ba rb La ke  K re m a 19 82  M  L SL  1. l. (1 98 7)  Ba rb La ke  K re m a 19 82  F  SL  1. 98 7)  Ba rb La ke  K re m a 19 8 C L SL  1. 98 7)  Ch a La ke  M ik ri P 19 90 - C   FL 1. tr id is  ( 19 95 ) Ch a La ke  M ik ri P 19 90 - C   TL  0. tr id is  ( 19 95 ) Ch La ke  M ik ri P 19 90 - C 9  FL 0. 19 95 ) Ch de s La ke  V ol vi  19 8 M  90   TL  0.  Ch de s La ke  V ol vi  19 8 F 80   TL  0.  La ke  V ol vi  19 83  C   TL  0. La ke  V ol vi  19 83  M    SL  1. Ch vi  19 83  F L SL  Ch Ch al ca st 19 83  L n am is  b ra m a 1 74 SL  89 7 -0 .2 17 5 Va lo uk as am is  b ra m a 1 68 SL 80 5 -0 .8 82 2 Va lo uk as bu rn us  a lb ur nu s us b r a 8 44 9 T  1. 09 5 0. 16 06  Po lit ou (1 bu rn  a l u nu s a 88  44 9 T 0. 50 05  Po lit ou  ( 19 93 ) bu rn us  a lb ur nu s re sp a 5 60 FL 95 6 9 0. 02 96  Cr iv el li an d D u bu rn us  a lb ur nu s re sp a 5 TL 44  0. 04 58  Cr iv el li an d D u sa  c as pi a vi st on ic a p s 38 FL 86 6 0. 52 17  Ec on om id is  a n di s sa  c as ia  v is to ni ca  s 38 SL 78 3 0. 55 05  Ec on om i sa  m ac ed on ic a 12 3 TL  1. 28 2  1. 27 6 -0 .3 98 0 Si ni s (1 9 sa  m ac ed on ic a 21 TL -0 .3 71 0 Si ni s (1 98 1) sa  m ac ed on ic a 33 4 TL  1. 27 2 -0 .3 12 0 Si ni s (1 98 1) sa  m ac ed on ic a 12 3 TL  1. 16 5  1. 15 9 -0 .5 67 0 Si ni s (1 98 1) 1) sa  m ac ed on ic a 8 21 1 TL -0 .4 96 0 Si ni s (1 98 sa  m ac ed on ic a ia u 8 9 33 4 4 TL  1. 16 4 -0 .5 64 0 Si ni s (1 98  Ap ha ni us  fa sc t s  la g.  ( Re ba ki a)  1 3  TL   18 0 0. 09 00 Le on ar do s (1 Ap ha ni us  fa sc ia tu s  la g.  ( Re ba ki a)  1 45  TL   17 0 0. 11 30  Le on ar do s (1 Ap ha ni us  fa sc ia tu s  la g.  ( Re ba ki a)  1 79  TL   17 0 0. 12 00  Le on ar do s (1 Ap ha ni us  fa sc ia tu s  ( As tr ov its a)  1 21  TL   19 0 0. 08 00  Le on ar do s (1 Ap ha ni us  fa sc ia tu s  ( As tr ov its a)  1 23  TL   17 0 0. 12 00  Le on ar do s (1 Ap ha ni us  fa sc ia tu s  ( As tr ov its a)  1 45  TL   17 0 0. 14 00  Le on ar do s (1 Ap ha ni us  fa sc ia tu s  f at us   la g.  ( Al yk es ) 1 28 3 TL   18 0 0. 12 00  Le on ar do s (1 ha ni us as ci  la g.  ( Al yk es ) 1 29  TL   19 0 0. 09 00  Le on ar do s (1 99 6) ha ni us  fa sc ia tu  la g.  ( Al yk es ) 1 58  TL   18 0 0. 11 00  Le on ar do s (1 99 6) er in a bo ye ri La ke is  0 25 7 FL   0. 91 5 0. 08 23  St ou m bo ud i e t a l. us  a lb an ic us  La ke us  a lb s st a - T  12 0 0. 13 80  D ao ul as  e t a l. (1 9 l. (1 an ic u  st a - T  98 0 0 2. 66 50  D ao ul as  e t a us  a lb an ic us  st a - T  00  2. 12 00  D ao ul as  e t a us  a lb an ic us  al c  st a - T 19 0 0. 34 40  D ao ul as  e t a us  ba ni us ni s st a - TL 16 0 0. 77 00  D ao ul as  e t a l. (1  a l. (1 us  a lb a cu  st a 2 - T 17 0 1 0. 64 00  D ao ul as  e t al ca lb ur nu s be lv ic re sp a 91  36 9 TL  20  -0 .1 59  Si ni s an d Pe al ca lb ur nu s be lv ic lburnus be re sp a 91  36 9 6 SL 81 6 -0 .1 64  Si ni s an d Pe  nd al ca lvica  re sp a 91  3 SL  91 7 -0 .3 38  Si ni s a  P et rid is  ( al ca lb ur nu s ch al co i 3 FL 89 9 -0 .0 53 0 Ko kk in ak is  ( 19 92 ) al ca lb ur nu s ch al co i 3 2 FL 91 3 -0 .3 31 0 Ko kk in ak is  ( 19 92 ) Ch al ca lb ur nu s ch al co id es  37 0 FL 90 7 -0 .2 08 0 Ko kk in ak is  ( 19 92 ) Ch al ca lb ur nu s ch al co id es  90 28 0 FL F 07 0 0. 18 50  Ko kk in ak is  ( 19 92 ) 1. 05 0 0. 47 90  Ko kk in ak is  ( 19 92 ) al ca lb ur nu s ch al co id es  La ke  V ol al ca lb ur nu s ch al co id es  La ke  V ol vi  19 83  C 37 0 FL  SL  1. 05 0 0. 39 80  Ko kk in ak is  ( 19 92 ) lb ur nu s ch al co id es  La ke  V i on is  M  26 0 F  TL  0. 93 3 -0 .3 29 0 Ko kk i ak is  ( 19 92 )    F is he s in  D at ab as es  a nd  E co sy st em s,  P al om ar es , M .L .D ., St er gi ou , K .I ., P au ly , D . 73 T a b le  1  c on ti n ue d.  Sp ec ie s Ar ea  Ye ar  Se x N  Y X b a Re fe re nc e Ch al ca lb ur nu s ch al co id es  La ke  V is to ni s 19 83  F 25 3 FL  TL  0. 92 1 -0 .1 08 0 Ko kk in ak is  ( 19 92 ) Ch al ca lb ur nu s ch al co id es  La ke  V is to ni s 19 83  C 51 3 FL  TL  0. 92 6 -0 .1 97 0 Ko kk in ak is  ( 19 92 ) Ch al ca lb ur nu s ch al co id es  La ke  V is to ni s 19 83  M  26 0 FL  SL  1. 03 0 0. 82 60  Ko kk in ak is  ( 19 92 ) Ch al ca lb ur nu s ch al co id es  La ke  V is to ni s 19 83  F 25 3 FL  SL  1. 06 0 0. 29 40  Ko kk in ak is  ( 19 92 ) Ch al ca lb ur nu s ch al co id es  La ke  V is to ni s 19 83  C 51 3 FL  SL  1. 05 0 0. 51 50  Ko kk in ak is  ( 19 92 ) Kn ip ow its ch ia  c au ca si ca  Ev ro s Ri ve r 19 83 -8 4 M  23 0 SL  TL  0. 83 7 0. 02 24  Ke vr ek id is  e t a l. (1 99 0)  Kn ip ow its ch ia  c au ca si ca  Ev ro s Ri ve r 19 83 -8 4 F 15 8 SL  TL  0. 85 3 -0 .0 19 8 Ke vr ek id is  e t a l. (1 99 0)  Kn hi a ca uc as ic a Ev ro s Ri ve r 98 4 63  ev re k l. (1 99 0)  ts ch ia  c au ca si ca  Ev ro s Ri 19 T .  03 2 re ki al . ( s ce ph al us  Ri hi os  S S . 51 7 op hi t 88 ) ep ha lu s Ri hi os  S F .9 82 8 ph it 88 ) til is  La ke  K or T 24 5 ag e  ( 19 til is  La ke  K or T 30 3 ag e  ( 19 til is  La ke  K or 87 0 ag eo ou  ( 19 til is  La ke  D oi 19 TL  71 0 ph ito 19 93 a til is  La ke  D oi 19 T . 30 0 ph it 93 a  p le ur ob ip un ct La ke  K r  F .7  11 0 ou la  ( 19 s pl eu ro bi pu nc t La ke  K r  F .7 0  0 ou la  ( 19 s pl eu ro bi pu nc t La ke  K r   F .  0 ou la  ( 19  p le ur ob ip un ct La ke  K re S .1 50  ou la  ( 19 eu ro bi pu nc t La ke  K re S .0 3 05 0 ou la  ( 19 eu ro bi pu nc t La ke  K re S .1 5 08 0 ou la  ( 19  p ar va  La ke  M i F .9 0  54 0 ec c . ( 1 La ke  M i pa  S 0. 97 20 1 ve lli  po n La ke  M i pa  S 0. 96 34 5 ve lli  po n bu rn us La ke  M i pa  S 0. 97 24 2 ve lli  po n bu rn us La ke  M i pa  S 0. 98 29 0 ve lli  po n La ke  V ol T 38 6 ag e  ( 19 La ke  V ol T 64 7 ag e  ( 19 ut ilu s  La ke  V ol T 57 ag e  ( 19 ie ns is  La ke  L ys a  F .0 8  88 0 na r al . ( La ke  T ri 19 SL  1. 20 41 0 ar ra s  ( 19 La ke  T ri 19 SL  1. 17 20 0 ar ra s  ( 19 ia til is La ke  T ri 19 S .1 8 83 0 ar ra s  ( 19 ta  fa rio  As pr op ot  s tr ea m  19 47 0 ag e  e t a Sa lm o tr ut ta  fa rio  As pr op ot am os  s tr ea m  19 81 -8 2 C8 - FL  SL  1. 18 0 -1 .1 19 0 Pa pa ge or gi ou  e t a l. (1 98 3)  Sa lm o tr ut ta  fa rio  As pr op ot am os  s tr ea m  19 81 -8 2 C9 - FL  SL  1. 19 0 -1 .2 36 0 Pa pa ge or gi ou  e t a l. (1 98 3)  Sc ar di ni us  e ry th ro pt ha lm us  La ke  K as to ria  19 80  C1 0 34 2 SL  TL  0. 78 0 -0 .0 11 2 Pa pa ge or gi ou  a nd  N eo ph ito u (1 98 2)  Sc ar di ni us  e ry th ro pt ha lm us  La ke  K as to ria  19 80  C1 1 34 2 SL  TL  0. 87 0 -0 .1 00 7 Pa pa ge or gi ou  a nd  N eo ph ito u (1 98 2)  *  r ef er re d as  R ut ilu s pl eu ro bi pu n ct at us . 1  T L  =  8 0 -1 10  m m . 2  T L  =  1 10 -1 50  m m . 3  T L  =  1 50 -2 30  m m . 4  T L < 12 0  m m . 5 T L  =  1 21 -1 8 0  m m . 6  T L > 18 1 m m . 7  S L < 10 0  m m . 8 SL  =  1 0 1- 17 5 m m . 9  S L > 17 5 m m . 1 0 T L < 10 0  m m .11  T L > 10 1 m m . ip ow its c 1 3- 8 I 9 SL  TL  0. 68 6 0. 20 K id is  e t a Kn ip ow i ve r 83 -8 4 C 39 7 SL  L 0 84 7 -0 .0  Ke v di s et  19 90 ) Le uc is cu s tr ea m  19 84  C 48 9 FL  L 1 20 9 5 -0 .2  N e ou  ( 19  Le uc i cu s c tr ea m  19 84  C 48 9 TL  L 0 6 -0 .3  N eo ou  ( 19  Pe rc a flu vi a on ia  19 75  C1 24  SL  L 0. 76 6 0. 0  Pa p or gi ou 77 ) Pe rc a flu vi a u ia on ia  19 75  C2 12 0 SL  L TL  0. 85 3 0. 88 6 -0 .0  Pa p or gi ou r 77 ) Pe rc a fl v on ia  19 75  C3 23 0 SL  -0 .0  Pa p gi 77 ) Pe rc a flu vi a ra ni  89 -9 2 C 31 7 SL  0. 82 1 0. 5 N eo u ( ) Pe rc a flu vi a lu s ra ni  89 -9 2 C 31 7 FL  L 0 97 7 8 0. 6 N eo ou  ( 19 ) Ph ox in el at us * em as ta   19 82  M - TL  L 0 0 4. 6  D a s et  a l. 87 ) Ph ox in el lu at us * em as ta 19 82  F - TL  L 0 0 6. 39 2  D a s et  a l. 87 ) Ph ox in el lu ho i at us * em as ta 19 82  C - TL  L 1 08 0 6 0. 15 3 4  D a s et  a l. 87 ) P x ne llu s at us * m as ta  19 82  M  - TL  L 1 0 0. 4 D a s et  a l. 87 ) Ph ox in el lu s pl at us * m as ta  19 82  F - TL  L 1 0 2. 3 D a s et  a l. 87 ) Ph ox in el lu s pl at us * m as ta  19 82  C - TL  L 1 0 0. 4 D a s et  a l. 87 ) Ps eu do ra sb or a kr i P re sp a 19 90  C 24 5 SL  L 0 1 0. 0  Ro s hi  e t a l 99 3)  Ru til us  r ub ili o kr i P re s 19 85  C 60  FL  L 8 0. 0 Cr i an d D u t (1 98 7)  Ru til us  r ub ili o kr i P re s 19 85  C 60  TL  L 8 0. 0 Cr i an d D u t (1 98 7)  Ru til us  r ub ili o X Al  a lb ur nu s kr i P re s 19 85  C 60  FL  L 0 0. 0 Cr i an d D u t (1 98 7)  Ru til us  r ub ili o X Al  a lb ur nu s kr i P re s 19 85  C 60  TL  L 1 0. 0 Cr i an d D u t (1 98 7)  Ru til us  r ut ilu s vi  19 78  C4 10 2 SL  L 0. 77 0 -0 .1  Pa p or gi ou 79 ) Ru til us  r ut ilu s vi  19 78  C5 11 0 SL  L 0. 78 0 0. 2  Pa p or gi ou 79 ) Ru til us  r vi  19 78  C6 21  SL  L 0. 81 0 0. 01  Pa p or gi ou 79 ) Ru til us  y lik im ac hi - C 31 4 TL  L 1 0 0. 0  Le o do s et  20 00 b)  Sa la ria  fl uv ia til is ch on is  88 -9 1 M  32  TL  0 -0 .0  Ps et  a l. 97 ) Sa la ria  fl uv ia til is ch on is  88 -9 1 F 15 9 TL  0 0. 1 Ps et  a l. 97 ) Sa la ria  fl uv ch on is  88 -9 1 C 19 1 TL  L 1 0 0. 0 Ps  e t a l. 97 ) Sa lm o tr ut am os 81 -8 2 C7 - FL  SL  1. 05 0 0. 6  Pa p or gi ou l. (1 98 3)    G ro w th  p ar am et er s of  G re ek  fr es hw at er  fi sh es , K le an th id is , P .K ., St er gi ou , K .I .  74 ic al  p ar am et er s fo r 22  G re ek  f re sh w at er  f is h sp ec ie s an d on e hy br id , f ro m  t w el ve  la ke s an d tw o ri ve rs . K  in  y ea r- 1 , L ∞ in  c m , a n d t 0  in  y ea r,  a re  t he  v on  B er ta la n ff y er s.  F , fr eq ue n cy  o f  ( M  =  m on th ly ; S =  al ; B  = on e s gl e sa  t w o sa m pl i s;  U  es ; F  =  fe m al es ; se xe s co m bi n ed );  M E P ,  B e la =  pl ot : Sp a  e 98 n on -l in ea r st er is k de n ot es  t ha ff y pa ra m et e  c al t  i n m ax m ax dy  l en gt h in  c m  a n d ag e  e le m en t n g (S  =  s ca le s;  O B   b on es ; V  =  v er e;  F S n pi n  l  u st i ti on  th ra m e (B  =  b ac k- ca le n gt h;  gt h;  S L  =  s ta n da rd  le   Ye ar  F x K t 0 m ax SA  L m ax /L  T a b le  2  B io lo g pa ra m et sa m pl in g se as on =  b im on th ly ; O   in m pl in g;  T  = n g =  u n de fi n ed ) 9 ; Se x (M  =  m al C  =  m et ho d us ed  f or  t he  e st im at io n  o f th e re  v on rt a n ff y pa ra m et er s (F W  F or d- W al fo rd  , m ax im um  b rr e t al . 1 ; R  =  e in  yR af ai l, 19 73 ; N L  =  re gr es si on ; a t vo n  B er ta la n rs  w e cu la ed  t hi s st ud y) ; L  a n d t o ar  r es pe ct iv el y;  S A , us ed  f or  a ge i =  o pe rc ul ar te br a =  fi  s es );  L , en gt h se d fo r th e e m a of  e pa te rs  lc ul at ed ; T L  =  t ot al  F L  =  fo rk  le n n gt h) . Sp ec ie s Ar ea Se  L ∞  M EP  L t m ax L ∞ Re fe re nc e Ab ra m is  b ra m a La 19 89 - M  0. 10 2 2 .3 14  S B.  F Va l  ( 19 96 ) ke  V ol vi  91  M  45 .  -0 51  FW  40 .5  L 0. 90  ou ka s an d Ec on om id is Ab ra m is  b ra m a La 19 89 -  0. 09 4 7 .4 42 .2  14  S B.  F Va l  ( 19 96 ) Al bu rn us  a lb ur nu s La ke  K or on ia  19 86 -  M  0. 57 7 12 .4  -0 .1 53  R 11 .6  4 S B.  T L 0. 94  Po lit ou  ( 19 93 ) rn us  La 19 86 -  .3 6 3 L 14 .9  7 S B.  T L 0. 97  Po li rn us  La 19 86 -  .3 5 5 L 14 .9  7 S B.  T L 0. 96  Po li on ic a La 19 78  0. 40 2 3 N L*   S B.  T L - Si ni Al os a m ac ed on ic a La ke  V ol vi  19 78  O  F 0. 40 3 18 .5  -1 .4 94  N L*  - - S B.  T L - Si ni  a 19 78  O  .3 7 8 L 33 .1  S B.  T L 1. 76  Si ni er i La 19 92 - U  0. 37 5 3 W 4 S B. F Le o s La 19 82  S 0. 14 4 8 N L*  20 .1  9 O B.  S  B.  F 0. 78  D ao ul 9)   La 19 82  S 0. 08 3 9 N L*  28 .8  O B.  S  B.  F 0. 76  D ao ul Ri 19 84   .1 6 1 L 26 .0  6 S B.  F 0. 70  La  0. 34 0 3 N L*  13 .6  4 S B.  F 0. 95  Si ni is  ( 19 95 ) La  0. 11 0 5 N L*  22 .0  8 S B.  F 0. 66  Si ni 99 5)  La 19 90 -9 1 M  0. 10 7 9 N L*  22 .0  8 S B.  F 0. 63  Si ni de s La 19 83  O  0. 12 8 1 L* 19 .4  S B.  F  0. 69  Ko kk i de s La 19 83  O  0. 13 7 6 L* 22 .5  S B.  F  0. 88  Ko kk i de s La 19 83  O  0. 13 7 2 L* 22 .5  S B.  F  0. 86  Ko kk i La 19 83  O   .1 0 6 L* 20 .0  6 S B.  F 0. 60  Ko kk i Ch al ca lb ur nu s ch al co id es  La ke  V is to ni s 19 83  O  F 0. 10 9 32 .6  -1 .6 54  N L*  23 .7  8 S B.  F L 0. 73  Ko kk i 19 92 ) La 19 83  O  .1 0 0 L* 23 .7  8 S B.  F 0. 72  Ko kk i La 19 73  O  .2 2 0 L* 76 .0  7 O B B. 0. 86  Ts i lu s Ri 19 84  U  .1 4 8 L 37 .0  8 S B.  F 0. 81  s ce ph al us  a lb us  La 19 87 - B 0. 18 2 9 L* S B.  F - Ec o ce ph al us  a lb us  La 19 87 B .1 9 7 L* 5 S  F - Ec o )  0. 90  Ec on om ou  e t a l. (1 99 1)  Le uc cu s ‘sv al liz e’  La ke  K re m as ta  19 82  S F 0. 31 7 22 .9  0. 11 2 N L*  20 .8  7 S B.  F L 0. 91  Ec on om ou  e t a l. (1 99 1)  ke  V ol vi  91  M F 50 .  -0 06  FW  L 0. 83  ou ka s an d Ec on om id is 88  M Al bu rn us  a lb u ke  K or on ia  88  M F 0 9 15 .  -0 .2 32  N to u (1 99 3)  Al bu rn us  a lb u ke  K or on ia  88  M C 0 2 15 .  -0 .2 84  N to u (1 99 3)  Al os a m ac ed ke  V ol vi  O  M  18 .  -1 .2 70  - - s (1 98 1)  s (1 98 1) Al os a m ac ed on ic a L ke  V ol vi  C 0 4 18 .  -1 .5 05  N * 10  s (1 98 1)  At he rin a bo y ke  T ric ho ni s 93  C 12 .  0. 01 8 F  11 .0    L 0. 89  na rd os  e t a l. (1 99 3)  Ba rb us  a lb an ic u ke  K re m as ta  M  25 .  -0 .8 05  L as  a nd  E co no m id is  ( 19 8 Ba rb us  a lb an ic us ke  K re m as ta  F 37 .  -0 .9 03  14  L as  a nd  E co no m id is  ( 19 89 ) Ba rb us  c yc lo le pi s1 hi os  s tr ea m  B C 0 8 37 .  -0 .1 79  N * L N eo ph ito u (1 98 7)  Chalc alburnus be lvica  ke  M ik ri Pr es pa  19 90 -9 1 M M  14 .  -1 .7 45  L s an d Pe tr id Chalc alburnus be lvica  ke  M ik ri Pr es pa  19 90 -9 1 M F 33 .  -1 .4 05  L s an d Pe tr id is  ( 1 Chalc alburnus be lvica  ke  M ik ri Pr es pa  C 34 .  -1 .2 57  L s an d Pe tr id is  ( 19 95 ) Ch al ca lb ur nu s ch al co i ke  V ol vi  M  28 .  -2 .1 33  N  7 L na ki s (1 99 2)  Ch al ca lb ur nu s ch al co i ke  V ol vi  F 25 .  -3 .2 18  N  8 L na ki s (1 99 2)  Ch al ca lb ur nu s ch al co i ke  V ol vi  C 26 .  -2 .6 28  N  8 L na ki s (1 99 2)  Ch al ca lb ur nu s ch al co id es  ke  V is to ni s M 0 5 33 .  -1 .8 21  N  L na ki s (1 99 2)  na ki s ( Ch al ca lb ur nu s ch al co id es  ke  V is to ni s C 0 7 33 .  -1 .7 29  N  L na ki s (1 99 2)  Cy pr in us  c ar pi o ke  V is to ni s C 0 6 88 .  -0 .7 69  N   T L m en id is  ( 19 76 ) Le uc is cu s ce ph a hi os  s tr ea m  C 0 1 45 .  -0 .5 86  N * L N eo ph ito u (1 98 8)  Le uc is cu ke  M or no s 88  M  27 .  -0 .8 32  N  - 5 L  no m ou  e t a l. (1 99 7)  Le uc is cu s ke  M or no s -8 8 F 0 2 28 .  -0 .5 69  N  -   B. L  no m ou  e t a l. (1 99 7 Le uc is cu s ‘sv al liz e’  La ke  K re m as ta  19 82  S M  0. 41 2 20 .4  0. 23 6 N L*  18 .4  6 S B.  F L is  F is he s in  D at ab as es  a nd  E co sy st em s,  P al om ar es , M .L .D ., St er gi ou , K .I ., P au ly , D .  75  T a b le  2  c on ti n ue d.  Sp ec ie s Ar ea  L P L t SA  L L m ax ∞ re nc Ye ar  F Se x K t ∞ 0 M E m ax m ax / L Re fe e Pe rc a flu vi a r til is  La ke  D oi an i 19 89 -9 2 U  C 0. 31 7 23 .9  -1 .2 60  N L*  23 .0  5 O B.  S  B.  T L 0. 96  N eo ph ito u (1 99 3a ) Pe rc a flu vi a Pe rc a flu vi a Pe rc a flu vi a 9 u Ph ox in el lu s bi tu s l. (1 Ph ox in el lu s bi tu s B 0. 70  D ao ul l. (1 Ps eu do ra sb or a p 1 0. 81  cc Ru til us  r ub i  B Ru til us  r ub i  B Ru til us  r ub i B eo ch ar i Ru til us  r ub i al bu .9 02  B  ( 19 87 ) Ru til us  r ut i v B 0. 70  P ou  ( 19 79 ) Ru til us  y lik i . B et  a l. (2 00 0b ) Sa lm o tr ut . B ou  e t . ( 19 83 ) Sc ar di ni us  ch i B (1 98 1)  Sc ar di ni us  ch i B (1 98 1)  Sc ar di ni us  ch i B (1 98 1)  Sc ar di ni us   . B et  a l Sc ar di ni us   . B et  a l Sc ar di ni us  i  - B et  a l Sc ar di ni us  i  0. 13 4 41 .4  - B et  a l Sc ar di ni us op ht ha s or 0. 17 8 23 .9  - B to u Si lu ru s ar is ho ni s ch i  0. 09 9 44 .5  - B (1 98 1)  Si lu ru s ar is ho ni s ch i  0. 10 0 46 .4  - B (1 98 1)  Si lu ru s ar is ho ni s ch i 0. 10 0 45 .5  - B (1 98 1)  Ti nc a tin ca  m vo ti 0. 20 5 32 .4  - * B u (1 99 3b ) Ti nc a tin ca  go ri O B   ( 19 99 ) 1 re fe rr ed  a s B a s io n R ut i ro bi pu n s. ar t lm us rr ra s ar i te lis . til is  til is  til is   p le ur o  p le ur o a lio  lio  lio  lio  X  A l lu s ie ns is ta  fa rio  ac ar na ni cu s ac ar na ni cu s ac ar na ni cu s ac ar na ni cu s ac ar na ni cu s ac ar na ni cu s ac ar na ni cu s  e ry th r to te lis La ke  K or La ke  K or La ke  K or La ke  K re m a La ke  K re m a La ke  M ik r La ke  T ric La ke  T ric La ke  P a La ke  M ik r La ke  V ol La ke  L ys As pr op ot La ke  T ric La ke  T ric La ke  T ric La ke  T ric La ke  T ric La ke  L ys La ke  L ys La ke  K as t La ke  T ric La ke  T ric La ke  T ric La ke  P a La ke  V e fe rr ed  a s on ia  on ia  on ia  st a st a i P re sp a ho ni s ho ni s m vo tis  i P re sp a i m ac hi a am os  s tr ea m  ho ni s- Ly si m a ho ni s- Ly si m a ho ni s- Ly si m a ho ni s ho ni s m ac hi a m ac hi a ia  - Ly si m a -L ys im a -L ys im a s tis  lu s pl eu 19 76  19 76  19 76  19 82  19 82  19 84 -8 5 19 91 -9 2 19 78 -7 9 19 78 -7 9 19 83 -8 4 19 85  19 78  - 19 81  19 77 -7 9 19 77 -7 9 19 77 -7 9 - - - - 19 80  19 77 -7 9 19 77 -7 9 19 77 -7 9 19 88  19 88   3 O  M  O  F O  C S M  S F U  C M  M M  F M  C O  C O  C S C O  C M  M  M  F M  C M  M M  F M  M M  F O  C M  M M  F M  C M  C T C ed  a s Sc 0. 16 4 0. 15 0 0. 15 9 0. 13 6 0. 11 9 0. 15 7 0. 08 1 0. 15 9 0. 13 6 0. 13 4 0. 11 2 0. 12 3 di ni us  e ry 23 . 24 . 23 . 26 .2  28 .9  0. 24 5 11 .8  0. 12 4 26 .2  - 0. 11 3 30 .0  - 0. 27 5 20 .6  - 25 .7  33 .3  0. 08 4 32 .3  - 0. 21 8 26 .7  - 34 .5  42 .6  42 .6  0. 12 6 42 .8  - 0. 13 4 39 .7  -  47 .1 54 .0  hr op7  - 5 - 9 - - - - 1 1 1 -0 -1 0 0 -0 0 1  -0 0 0 0 0 0 0 -0 ht ha 0. 95 2 1. 21 2 1. 08 6 0. 64 0 0. 74 8 .3 73  N L*  .2 19  .0 40  .2 16  .2 95  N L*  12 6 N L 41 7 N L*  .1 55  N L*  0. 09 4 N L*  0. 05 5 N L*  83 0 N L 18 0 N L .7 90  .7 00  N L .5 21  .3 05  N L*  .3 37  N L*  .3 21  N L*  .4 01  N L .7 90  N L*  . N L*  - N L*  - N L*  23 .0  N L*  N L*  9. 5 N L*  18 .8  N L*  25 .8  N L*  20 .5  N L*  18 .5  23 .0  28 .5  24 .0  27 .0  33 .0  33 .0  - - N L - - N L*  18 .0  39 .0  40 .0  40 .0  29 .0  36 .6  ed  a s P a 9  9 17 .6  7 20 .1  9 7 9 8 6 12  10  8 7 7 7 - 10   11  7 10  10  10  10  7 si lu ru S S S O B O B 3 S S S O B S S S S S S S S S S S S FS  FS  FS  S .S .V st o B.  T L B.  T L B.  T L B.  F L . F L O . F L . F L . F L . F L . F L . T L . F L . S L . . T L . T L . F L . F L . F L . F L . T L . . . . F L O . T L - -  0. 96  0. 72  0. 86  1. 00  0. 72  0. 88  0. 90  TL  0. 78   0. 77   0. 77  - - - - 0. 75  TL  0. 88  TL  0. 86  TL  0. 88  0. 90  0. 68   Pa  Pa Pa 0. 67  D ao ul Ro se D ao ul D ao ul N eo ph it (1 98 9)  Cr iv el a Le on ar do s Pa Ili a Ili a Ili a Le on ar do s Le on ar do s Le on ar do s Le on ar do s Pa (1 98 2)  Ili a Ili a Ili a N eo ph it Si ni spa ge or gi pa ge or gi pa ge or gi o as  e t a as  e t a hi  e t a l as  ( 19 81 ) as  ( 19 81 ) ou  a nd  T li an d D up on t pa ge or gi  pa ge or gi do u do u do u     pa ge or gi do u do u do u o  e t al ou  ( 1 ou  ( 1  ( 1 9 87 ) 98 7)  . ( 19 93 ) h  a l . ( 20 00 a)  . ( 20 00 a)  . ( 20 00 a)  . ( 20 00 a)  ou  a nd  N eo ph i  97 7)  97 7)  97 7)  pu nc ta pu nc ta rv a bu rn us  2 2 rn us  al is . 3 a a a a a a ct at u 3 3 lm u  m er id 4 to te lis 4 to te lis 4 rb u 2 re re fe rr 4 r ef e   Growth parameters of Greek freshwater fishes, Kleanthidis, P.K., Stergiou, K.I. 76 REFERENCES Bertalanffy, L. von., 1938. A quantitative theory of organic growth (Inquiries on growth laws II). Human Biol. 10(2), 181-213. Crivelli, A.J., Dupont, F., 1987. Biometrical and biological features of Alburnus alburnus X Rutilus rubilio natural hybrids from Lake Mikri Prespa, northern Greece. J. Fish Biol. 31, 721-733. Daoulas, C., 1981. Contribution to the Biology of Rutilus rubilio (Bonaparte, 1837), (Pisces, Cyprinidae), in Lake Trichonis (Greece). University of Thessaloniki, Thessaloniki, Greece, Doctorate Thesis [in Greek, with English abstract]. Daoulas, C., Economidis, P., 1989. Age, growth and feeding of Barbus albanicus Steindachner in the Kremasta reservoir, Greece. Arch. für Hydrobiol. 114(4), 591-601.  Center of Marine Research. (Greece), 12 [in Greek, with English Eco es (Pisces, Clupeidae) provenant ia vistonica. J. Nat. Hist. 20, Eco s ‘svallize’ in the Kremasta reservoir (Greece). Hydrobiologia 213, 99-111. Kavala, Greece, pp. 265-268 [in Greek, with English abstract]. Froese, R., Binohlan, C., 2000. Empirical relationships to estimate asymptotic length, length at first maturity, and Froese, R., Pauly, D., 2000. Fishbase 2000: Concepts, Design and Data Sources. ICLARM, Los Baños, Philippines. Ga T): User’s Guide. FAO Computerized Information Series, Fisheries, Rome.  abstract]. Ke the biology and ecology of Knipowitschia  Helgol. Meeresunters. 44, 173-187. macedonicus Stephanidis, 1971 (Pisces: Cyprinidae) of the Systems Volvi and Vistonis. University of Thessaloniki, Leo ) in the Mesolongi and eek, with English Leonardos, I., Kokkinidou, A., Agiannitopoulos, A., Giris, S., 2000a. Population structure and reproductive strategy of ngress, 20-23 January 2000, Messolonghi, Greece, pp. 137-140 [in Greek, with English abstract]. Leonardos, I., Kokkinidou, A., Mourloukou, D., Karakyriakos, D., 2000b. Age growth and mortality of Rutilus ylikiensis (Stephanidis, 1939) (Pisces: Cyprinidae) in the Lysimachia lake. In: Proceedings of the Ninth Hellenic Ichthyological Congress, 20-23 January, Messolonghi 2000, Greece, pp. 53-56 [in Greek, with English abstract]. Leonardos, I., Petridis, D., Kokkinidou, A., 1993. Dynamic and exploitation of sand smelt (Atherina boyeri) in Lake Trichonis. In: Proceedings of the Sixth Hellenic Ichthyological Congress, 4-6 June 1993, Xanthi, Greece, pp. 11 [in Greek]. Neophitou, C., 1987. A study of some autoecological parameters of southern barbel (Barbus meridionalis R.) in the Rentina stream, Greece. J. Applied Ichthyol. 3, 24-29. Neophitou, C., 1988. Autecology of chub, Leuciscus cephalus (L.), in a Greek stream, and the use of the pharyngeal bone in fish predator-prey studies. Aquacul. Fish. Manage. 19, 179-190. Daoulas, C., Koussouris. T., Psarras, T., 1987. Ecology and Possibilities of Fisheries Management of the Artificial Lake of Kremasta. Special Publication National abstract]. nomidis, P.S., Sinis, A.I., 1986. Situation taxinomique et comparaisons des Alos des lacs Volvi et Vistonis (Grèce). Description d’une nouvelle sous-espèce: Alosa casp 723-734. nomou, A.N., Daoulas, C., Economidis, P., 1991. Observations on the biology of Leuciscu Economou, A.N., Daoulas, C., Stoumboudi, M.Th., 1997. On the biology of Leuciscus cephalus albus from Mornos Dam (Greece). In: Proceedings of the Fifth National Symposium on Oceanography and Fisheries, 15-18 April 1997, length at maximum yield per recruit in fishes, with a simple method to evaluate length frequency data. J. Fish Biol. 56, 758-773. yanilo Jr, F.C., Sparre, P., Pauly, D., 1994. The FAO-ICLARM Stock Assessment Tools (FISA Iliadou, K., 1981. The Biology of the Fish Species Scardinius erythrophthalmus and Parasilurus aristotelis of Lakes Lysimachia and Trichonis, Greece. University of Patras, Patras, Greece, Doctorate Thesis [in Greek, with English vrekidis, T., Kokkinakis, A.K., Koukouras, A., 1990. Some aspects of caucasica (Teleostei: Gobiidae) in the Evros Delta (north Aegean Sea). Kokkinakis, A., 1992. Comparative Study of the Biology and Dynamics of the Fish Chalcalburnus chalcoides Thessaloniki, Greece, Doctorate Thesis [in Greek, with English abstract]. nardos, I., 1996. Population Dynamics of Toothcarp (Aphanius fasciatus Nardo, 1827 Etolikon Lagoons. University of Thessaloniki, Thessaloniki, Greece, Doctorate Thesis [in Gr abstract]. an endemic species Scardinius acarnanicus (Stephanidis, 1939) in two W. Greece Lakes (Lysimachia and Trichonis). In: Proceedings of the Ninth Hellenic Ichthyological Co  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 77 Neophitou, C., 1993a. Ecological study of perch (Perca fluviatilis L.) in Lake Doirani. Geotech. Scient. Issues 4(3), 38- 47. Neophitou, C. a Hydrobiol. 35(4), 367 Neophitou, C., Theochari, V., 1989. Biology of the south European roach (Rutilus rubilio) in Lake Pamvotida. Scient. Annals Depart. Forestry Nat. Environ. 13, 50-110 [in Greek, with English abstract]. Papageorgiou, N., 1977. Age and growth of perch, L.) in the lake of Ag. Vasileios. Thalassographica 1(3), 245-265 [in Greek]. Papageorgiou, N ilus rutilus (L.) in Lake Volvi. J. Fish Biol. Papageorgiou, N., Neophitou, C., 1982. Age, growth and fecundity of the rudd (Scardinius erythrophthalmus L.) in Lake Kastoria. Thalassographica 2(5), 5-15. Papageorgiou, N., Neophitou, ction of brown trout (Salmo trutta fario) in the Asprop Pauly, D., 1998. Tropical fishes: patterns and propensities. J. Fish Biol. 53 (Suppl.), 1-17. Politou, C-Y., 1993. Biology and Dynamics o us (L., 1758) in Lake Koronia. University of Thessaloniki, Thessaloniki, Greece, act]. Psarras, T uction of Salaria fluviat 5-18 April 1997, Kavala, Greece, pp. 261-264 [in Greek, with English abstract]. Rafail, S.Z., 1973. A simple and precise method th curve. Mar. Biol. 19, 354-358. Rosecchi, E., Crivelli, A.J., Catsadorakis, G., 1993. T ent and impact of Pseudorasbora parva, an exotic fis  223- 231. Sinis, A., 1981. L’Autoecologie de l’Espèce Endémique Alosa (Caspialosa) macedonica (Vinciguerra) (Pisces: ), du Lac Volvi. Aristotle University of Thessaloniki, Doctorate Thesis (in Greek). isso,  pril 1997, Kavala, Greece, pp. 257-260 [in Greek, with English abstract].  N., 19 6. The relationship between fish length and the length of the operculum for the carp in Lake , 1993b. Some biological data on tench (Tinca tinca (L.)) in Lake Pamvotida (Greece). Act -379. Perca fluviatilis ( ., 1979. The length-weight relationship, age, growth and reproduction of the roach Rut 14, 529-538. C.N., Vlachos, C.G., 1983. The age, growth, and reprodu otamos stream. Acta Hydrobiol. 25/26(3/4), 451-467. f the Fish Alburnus alburn ate Thesis [in Greek, with E Doctor nglish abstr h., Barbieri-Tseliki, R., Economou, A.N., 1997. First data on the feeding and biology of reprod ilis. In: Proceedings of the Fifth National Symposium on Oceanography and Fisheries, 1  for fitting a von Bertalanffy grow he establishm h species introduced into Lake Mikri Prespa (north-western Greece). Aquatic Cons. Mar. Freshw. Ecosys. 3, Osmeridae Sinis, A.I., Meunier, F.J., Francilon-Vieillot, H., 1999. Comparison of scales, opercular bones, and vertebrae to determine age and population structure in tench, Tinca tinca (L. 1758) (Pisces, Teleostei). Israel J. Zool. 45, 453- 465. Sinis, A.I., Petridis, D., 1995. Age structure and reproductive pattern of Chalcalburnus belvica (Karaman, 1924) in Lake Mikri Prespa (Northwestern Greece). Israel J. Zool. 41, 569-580. Sparre, P., Ursin, E., Venema, S.H., 1989. Introduction to tropical fish stock assessment. Part 1: Manual FAO Fish. Tech. Pap. 306 (1). Stoumboudi, M.Th., Psarras, Th., Barbieri-Tseliki, R., 1997. Reproductive cycles of atherina (Atherina boyeri R 1810) from Trichonis Lake (Greece). In: Proceedings of the Fifth National Symposium on Oceanography and Fisheries, 15-18 A Tsimenidis, 7 Vistonis. Thalassographica 1(1), 53-63 [in Greek]. Valoukas, V.A., Economidis, P.S., 1996. Growth, population composition and reproduction of Bream Abramis brama (L.) in Lake Volvi, Macedonia, Greece. Ecol. Freshw. Fish 5, 108-115.   Growth and population trends in Grand Canyon native fishes, Walters, C. et al. 78 ASSESSMENT OF GROWTH AND APPARENT POPULATION TRENDS IN GRAND CANYON NATIVE FISHES FROM TAG-RECAPTURE DATA1 Carl Walters Fisheries Centre, University of British Columbia, 2202 Main Mall, Vancouver, BC V6T 1Z4; Email: c.walters@fisheries.ubc.ca SWCA Inc., t 400 South St., Suite 201, Salt Lake City, Utah 84101, USA; Email: rvaldez@swca.com ses from Glen Canyon Dam (GCD) may have caused declines in native s). Presumption of continued decline has prompted expensive proposals to restore more favorable physical habitat conditions for these fishes, by altering operation of Glen Canyon releases, restoration of seasonality in flows, even restoration of turbidity by transport of materials past Lake Powell). Such proposals would be not only directly costly to implement, but also While abundances may have dropped initially after GCD was filled, there is little evidence to support ution monitoring, and         Michael Douglas Department of Biology, Arizona State University, Tempe, AZ 85287, USA; Email: m.douglas@asu.edu William R. Persons Arizona Fish and Game Department, 2221 West Greenway Road, Phoenix, AZ 85023, USA; Email: bpersons@gf.state.az.us Richard A. Valdez 56 Wes ABSTRACT Mean growth curves and individual variation in asymptotic body length are estimated for the humpback chub (Gila cypha), flannelmouth sucker (Catostomus latippinus), and bluehead sucker (C. discobolus), from growth of fish tagged in the Colorado River. Age distributions for 1991-1994 are back-calculated from the individual growth curves, to provide assessments of apparent natural mortality rates and/or recruitment trends. Declines in relative abundance with age are consistent with natural mortality rates predicted from the growth parameters for populations with stable recruitment, but there are relatively more old chubs and bluehead suckers than would be expected from natural mortality rate estimates based on tag-recapture models. Either the tag-recapture methods have overestimated natural mortality rate, or chubs and bluehead suckers have had considerable decline in recruitment rates since the mid-1980s. INTRODUCTION Beginning in 1963, cold water relea warm-water fish species of the Colorado River, especially the humpback chub (Gila cypha) (listed as endangered under the US Endangered Species Act), flannelmouth sucker (Catostomus latippinus), and bluehead sucker (C. discobolu Dam (warm water destructive to some ecological values that have developed in conjunction with clear, cold water releases from the dam (rainbow trout fishery, riparian bird community including peregrine falcons). claims of continued decline toward extinction. Intensive fisheries monitoring programs have been carried out by various agencies since the late 1970s, including tagging studies, size distrib                                          1 Cite as: Walters, C., Douglas, M., Persons, W.R., Valdez, R.A., 2006. Assessment of growth and apparent population trends in Grand Canyon native fishes from tag-recapture data. In: Palomares, M.L.D., Stergiou, K.I., Pauly, D. (eds.), Fishes in Databases and Ecosystems. Fisheries Centre Research Reports 14(4), pp. 78-88. Fisheries Centre, University of British Columbia [ISSN 1198-6727].  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 79 index sampling for juvenile and adult densities using size structure data show evidence of continued recrui (older fish) as would be expected from non-recruiting adult density sampling data, though changes in methods an recapture estimates have been obtained since the la spawning in the Little Colorado River; these estimate Low ratios of juveniles to adults in size-frequency sampl (Valdez and Ryel, 1995), but this ratio comparison is different methods, in different habitats, with unknown population collected by each method). This paper shows that tag recovery data from the 19 flannelmouth, and bluehead sucker populations are assumption of continued rapid population decline. T that were tagged and subsequently (0.5-6.8 years la From changes in size of these fish and the assumption we can back-calculate an apparent age at tagging for ea each tag sample. Assuming the tagged fish were a repres least for older fish), the age distributions can then b various hypotheses about recruitment success and rate of populatio distribution data give estimates of apparent annual m close to mortality rate predictions from growth pa estimated from tag-recapture models. Were populati considerably higher apparent survival rates (relatively older fish and fewer you various fishing gears. For most sampling sites, the tment, and no clear trend toward larger body sizes  populations. There are no clear trends in the d sites make long-term comparisons suspect. Mark- te 1980s for the humpback chub ‘subpopulation’ s are highly variable and show no consistent trend. es have been cited as evidence of low recruitment not valid since juveniles and adults are collected by differences in sampling rates (proportions of total 90s are consistent with the hypothesis that chub, relatively stable, and are not consistent with an he analysis is based on substantial samples of fish ter) recovered for growth and survival estimation.  that fish grow according to von Bertalanffy curves, ch individual, and construct an age distribution for entative sample of the population age structure (at e compared to expected age distributions under n decline. Specifically, the age ortality rate, and these estimates are surprisingly rameters (Pauly, 1980) though lower than rates ons declining during the 1980s, we would expect ng ones). m e f z t ata set. Though fish were tagged and recovered o Lake Mead, most of the tagging and recovery was  (at least 6 months of growth) and 148 of 184 bluehead sucker records (again at least 6 months growth). The very . (Hilborn and Walters, 1992): La = L∞·(1-e ) … 1) fish tagged at lengths LS are recovered after time periods T at lengths LR, Fabens (1965) showed that t e von Bertalanffy model can be written as: METHODS The tag-recovery data used in this analysis are fro contractors to the Grand Canyon Monitoring and R Game, Arizona State University, U.S. Fish and Wildli database at Arizona State University. We also analy Department of Fish and Game during the 1980s (ADFG because sample sizes were only large enough for grow very similar to those from the larger, more recent d throughout the Grand Canyon from Glen Canyon Dam t from in or near the mouth of the Little Colorado River (LCR). Of 9191 total chub tag recoveries in the data base, we considered 1676 records usable for the analysis (no obvious recording errors, at least one year of growth from time of tagging to recovery), along with 386 of 1127 flannelmouth sucker records  a variety of tagging programs carried out by search Center (Arizona Department of Fish and e Service) mainly during 1991-94, archived on a ed much smaller data sets collected by Arizona , 1987), but do not report those results separately h analysis and growth parameter estimates were large number of ‘immediate’ (within a few days of tagging) chub and flannelmouth recoveries were used to assess length measurement error patterns, and this assessment indicated the average measurement error for typical-sized fish (200-400 mm) was roughly 4 mm (8 mm standard deviation of differences between immediate length measurements) Growth in length of fishes is generally very well described by a von Bertalanffy growth function of the form -K(a-ao) where La = length at age a; a = age (years), relative to the apparent age at zero body length; ao = age correction for non-zero length at age 0 (positive if growth is less than von Bertalanffy prediction for young fish); L∞ = asymptotic body length (at infinite age); K = ‘growth’ (actually metabolism) parameter. For analysis of tag recovery data, where h LR = LS + (L∞-LS)·(1 - e-KT) … 2)  Growth and population trends in Grand Canyon native fishes, Walters, C. et al. 80 This equation provides a nonlinear regression model for estimation of L∞ and K given a set of LR, LS  this method have been suggested to account for individual variation in the L∞, K t notably an extension of the Fabens model that accounts at least for variation in L∞ individual fish. In this method, the growth  (2) prediction, but with a mean-zero normally-distributed deviation Ei in the asymptotic m the population asymptotic average size L∞, i.e., L(i)∞ = L ∞ + Ei. age-1 length L  based on scale analysis and analysis n seasonal juvenile length frequency samples (chub estimate from Valdez and ly ages 2-3). Here, ‘constrained’ means varying only L∞ in the nonlinear growth regression, while calculating K from the relation K = -log (1 - L1 / L∞). uite small compared to variation in L∞. If so, the apparent L(i) ∞ and age a(i) of each individual fish i can be calculated from its LS, using the following relationships derived from Equations (1) and (2), provid e population K and L  are known: en only LS ‘shrinks’ when additional information (Gi, T) is provided about the individual’s L(i)∞. The ‘age’ distribution sample resulting from application of Equations (3) and (4) can be expressed as estimated age proportions pa of the sample (pa = na/n, na = estimated number of age a fish and n = sample size). For further analysis, we need to make some assumptions or alternative hypotheses about the population proportions Pa from which the pa were sampled. We note that these population proportions are observations. A key reason for the common use of Equation (2), besides allowing for variable times T to recovery, is that it expresses the growth curve in a form that does not depend explicitly on current age a, which is most often unknown. Various improvements in 0 100 parameters, mos (Wang et al., 1995; Wang, 1998). We tried the Wang (1998) method, but found it gave poor estimates for the nuisance parameter representing variance in L∞. We then decided to use a modification of the Wang et al. (1995) maximum likelihood method, based on assuming a normal distribution of L∞ values among observation Gi = LR - LS for each individual i is assumed to be Gi = (L∞ + Ei - LS) · (1 - e-KT), which is the Equation size for individual i fro Additionally, for chub we constrained the estimates of L∞, K to pass through independent estimates of 1 of modal progression i Ryel, 1995). For flannelmouth suckers, we were unable to use such a constraint to improve the estimation, since we found the L1 from modal progression analysis (around 80-90 mm) to be considerably lower than the age 0-1 growth that would be predicted from observed annual growth of tagged fish in the 200-300 mm size range (most like Wang et al. (1995) suggest that variation among individuals in K should generally be q LR data and K ed th ∞ L(i)∞ = L∞+Ei = LS+(LR - LS)/(1 - e-KT) … 3) a(i) = -(1/K)log(1 - LS/L(i) ∞) … 4) Note here that errors in estimation of K will tend to cause ages for all individuals to change in the same way (K too low causes all ages to be too high, K too high makes all fish look too young). Equations (3) and (4) essentially provide an individual-based ‘age-length key’ for converting length to age, given information from growth after initial tagging about individual variation in L∞. This age calculation is illustrated in Figure 1, which shows how large age uncertainty giv 200 300 400 500 600 0 5 10 15 20 Age Le ng th Age information from length at tagging  0 100 300 0 5 10 15 20 400 500 600 ng t 200 Age Le h Additional age information from length at recovery and time to recovery  Figure 1 Illustration of how tag-recapture data can provide improved estimates of fish age. Individual variation in growth curves implies that age is highly uncertain given only length at first capture (top  But given length at second capture and time until that  panel). capture, the individual’s growth curve and hence age can be determined much more accurately, at least for younger fish.  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 81 Pa = vaNa/(ΣvaNa) where v … 5)  Pa is the vulnerable number vaNa of age a fish ion size summed over ages, ΣvaNa. Based on yon scientists about how fish shift their ow, we assume that vulnerability va increases d function. For convenience, we assume the … 6)  the power parameter v represents steep ss e-edged’ vulnerability curve around the age a ). Luckily, we found the results presented below to be largely insensitive to choices ah and v. Thus,  va = 0 for ages 0 and 1, and va = 1 for ages 2 cation can cause modest underestimation of n size has been declining (see below). ival rate S (= e-M, where M is the annual e Na of Equation (5). These should be related N  = N Sa-1 = N e-M(a - 1) … 7) 1 without loss of generality since absolute contrast, for a population that is growing or hould be related by … 8) oth survival rate and population growth. The mposition sample will be higher than S if the ly recruited’ ages (va = 1, a≥amin), eqs. (7) and -(M+r)(a - amin)(1 - e-M - r) if age is treated as a d as continuous (here the terms 1 - e-M - r  a a so  7 he  assumption of ‘knife-edge’ selection (va = 1, a≥amin) i mates much higher than obtained by other method declining population (or some unknown problem a an a is an age-specific vulnerability to sampling. That is, in the population, divided by the total vulnerable populat inspection of the data and discussions with experienced Grand Can distributions and become vulnerable to sampling as they gr asymptotically from 0 to 1.0 with age, according to a sigmoi following form: va = av/(ahv+av) where ah is age at 50 % vulnerability (age where v = 0.5) and of the sigmoid function (high v values imply a steeper, more ‘knif ne h and for maximum likelihood analysis, we elected to assume and older (4 and older for bluehead sucker); this simplifi mortality rates and hence favor the hypothesis that populatio In a stable population with annual age-independent surv instantaneous natural mortality rate), it is easy to predict th according to a 1 1 where N1 is average age 1 recruitment. We can take N1 = numbers cancel in the calculation of P in Equation (5). In declining exponentially at annual rate r (Nt+1 = Nter), the Na s Na = Na-1Se-r = Na-1e-M - r i.e., abundances at successive ages should contain effects of b apparent survival rate (Na/Na-1, Pa for high a) from an age co population is declining (r<0). If we examine only data for ‘ful (8) imply that the stable age proportions should vary as Pa = e discrete variable, or Pa = e -(M+r) (a - amin) (M + r) if age is treate M+r represent the sum and integral of N  and a over ages). For initial analyses of age vulnerability patterns, we defined goodness of fit of alternative hypotheses about P  simple sum of squares criterion to compare ted age proportions pa to the estim SS = Σ a: … 9) ft Excel) to search for estimates of annual would minimize SS, with P a(pa - Pa)2 We then used a simple search procedure (Solver in Micro survival rate S and the vulnerability parameters ah and v that the stable population ‘null’ predictions of N a calculated using ). This approach allows definition of at least re near the estimated ones, and it indicated s reasonable for the chub and flannelmouth s would imply r<<0, i.e., the sample p a from Equation ( ranges of S that would predict sample age proportions anyw that the data. S esti come from a a had  with those independent estimates of S). S te an increasing population, but might also imals to tagging or recapture (failure in estimates much lower than from other methods might indic indicate some decrease in vulnerability of larger, older monotonic vulnerability assumption, Equation (6)).  Growth and population trends in Grand Canyon native fishes, Walters, C. et al. 82 Using the above relationships and an approach suggested by Wang et al. (1995), we then developed likelihood functions for the (LS, G) data in relation to the parameters L∞, K and Z = M + r (Appendix 1). These functions involve transformation from assumptions about the random variables Ei and age at capture, for which we can make reasonable statistical assumptions (Ei normal, age at capture sampled from exponential proportions at age), to No measurement errors, no K variation 10 15 20 25 30 35 40 st im at ed  a ge  (y ea rs ) 0 5 0 10 20 30 40 True age (years) E  Measurement errors, no K variation 0 5 10 15 20 25 30 35 40 0 10 20 30 True age (years) E st im at ed  a ge  (y ea rs ) 40 the (LS, G, T) data. Various simplified sum of squares and reduced likelihood functions for fitting the data were also tested for accuracy and bias using Monte Carlo procedures (Appendix 2), and these generally gave similar results for the relatively large sample sizes available for analysis. Monte Carlo tests of the likelihood estimation procedure indicate that it is quite robust to errors in measurement of individual fish lengths and to variation among individuals in the growth K parameter. These sources of variation apparently do not cause bias in estimates of population mean growth parameters K and L∞ , or apparent total mortality rate Z, though they do cause considerable variation in estimated ages for older fish (Figure 2). A key advantage of the likelihood approach is that it use of Bayesian assessment methods to provide prob distributions for the parameters, integrating uncer over sample variation in individual growth rates and range of possible estimates for the auxiliary data L1 method gives a more conservative assessment of stat uncertainty (wider statistical limits) than would si approaches like bootstrapping, since there is no simple way to include uncertainty about L1 in such pro edures. tandard Bayes techniques (Gelman et al., 1995) and derivations in Walters and Ludwig (1994) imply the marginal likelihood integrated over the ‘nuisance’ variance of Ei, should be proportional to the log(l) derived in Appendix 1. Summing such likelihood values over a numerical grid of (L∞, Z, K) values gives an approximate marginal likelihood of the data given each parameter; this ‘likelihood profile’ can be treated as a posterior probability distribution for the parameter, in essence assuming a flat (uninformative) prior for it. In intuitive terms, the marginal distribution for Z is wider than it wo be if we estimated Z only from apparent age composition estimated using only the most likely K, L∞, since the marginal distribution accounts for uncertainty in these par as well. RESULTS Results from fitting the growth data are presented in able 1 and Figures 3-4. Despite much obvious variation in measured growth rates, the maximum likelihood method  allows ability tainty over a . This istical mpler c Measurement errors, high K variation 20 25 30 35 40 E st im at ed  a ge  (y ea rs ) 0 5 10 15 0 10 20 30 40 True age (years)  Figure 2 Simulated effect of estimation and measurement errors on true vs. calculated ages of individual fish, assuming sample size of 100 fish used in estimation of population mean K, L∞. Simulated data generated using standard deviation of 60 mm among individual L∞ values. First panel shows that even this high growth variation does not cause bias or appreciable error in age estimation (variation due solely to variation in estimates of population K, L∞). Second panel shows spread in estimated ages, iation at each capture (observed standard deviation in measurements of individual fish recaptured within a few days after tagging). Third panel shows additional spread caused by having both measurement errors and variation in individual K values; individual K values S uld ameters especially for older fish, caused by having length measurement errors with 4 mm standard dev T gives a reasonably good fit to the overall growth rate vs. length at tagging relationship (Figure 3). There is not much indication of nonlinearity in this relationship (which would invalidate the von Bertalanffy model), though there is apparently high variation in individual asymptotic lengths. Reconstructed age-size observations (Equation (4) ages) also show reasonable fits to the von Bertalanffy model (Figure 4), with the caveats that assumed to have multiplicative normal variation with a coefficient of variation of 0.5 (somewhat higher than estimated for individual fish that were captured more than once).  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 83 (1) samples for younger ages are missing (except for L1 constraint mentioned above), and (2) the flannelmouth data suggest lower growth rate for at least the first year of life than predicted by the von Bertalanffy model for older fish. Marginal probability distributions for the parameters (Figure 5) indicate that L∞ and K are fairly well determined for all three species, though there is more uncertainty for bluehead sucker due to smaller sample size and lack of smaller fish in the tag sample. Estimated K parameters (Table 1) and natural mortality rates calculated from K, L∞, and average water temperature (10oC, mean mainstem Colorado temperature below Glen Canyon Dam) using the Pauly (1980) equation are reasonable for fish that have been called relatively long-lived and slow-growing.    Figure 3 Apparent growth rates (annual change in length per year, measured as (LR - LS)/T) vs. length at tagging for Grand Canyon fishes. Lines show growth rate trajectories for individual fish that were recovered more than once with at least six months growth (or one year in case of chub) after each recovery. Thick line shown is simple linear regression fitted to the data.  Humpback chub 0 100 200 300 400 500 600 0 5 10 15 20 25 30 35 Apparent age (years) Le ng th  ( m m ) Predicted Observed   Flannelmouth sucker 0 100 200 300 400 500 600 700 800 0 2 Le ng th  (m m ) 4 6 8 10 12 14 Apparent age (years) Predicted Observed  Figure 4 Fits of the von Bertalanffy growth curve to length at first capture vs. apparent age at that capture, for Grand Canyon fishes. Growth rate vs. length trajectories for individual fish that were recaptured more than once (Figure 3) indicate some violation of the assumption that all fish have the same growth K parameter. There are not enough multiple recaptures for detailed analysis of variation in K among individuals; for chub and flannelmouth, individual K estimates from growth rate vs. length regressions have a coefficient of variation of about 0.3, which is enough to cause considerable random error in age estimation for older fish (Figure 2) but not enough to cause bias in Monte Carlo tests of mortality rate estimation. For chub and Bluehead sucker 0 50 100 150 200 250 300 350 400 0 5 10 15 20 25 30 Apparent age (years) L en g th  ( m m ) Predicted Observed   Growth and population trends in Grand Canyon native fishes, Walters, C. et al. 84 bluehead sucker, there is a worrisome tendency for fish recaptured a second time less than 1 year after initial capture to show considerably lower growth rate after the second capture (apparent high K values), indicating a possible short-term effect of handling on growth rate. Table 1 Estimates of growth parameters and natural mortality rate. G-L regression estimates of K and L∞ are from simple linear regression of annual growth rate ((LR - LS)/T) on length at tagging; growth likelihood estimates of these parameters are from likelihood function for growth (LR - LS) only. Total likelihood estimates are with the likelihood function derived in Appendix 1. Regression estimates of apparent total mortality rate (Z) are slopes of fitting numbers at age to the exponential decay model Na = Noe-Za. Parameter Estimation procedure Humpback chub Flannelmouth sucker Bluehead sucker K G-L regression 0.15 0.38 0.089  Growth likelihood 0.16 0.41 0.32  Total likelihood 0.22 0.46 0.34 L∞ G-L regression 381 512 355  Growth Likelihood 387 522 280  Total Likelihood 376 538 262 Z (S = e-Z) Age regression 0.11 0.22 0.40  Total Likelihood 0.12 0.32 0.37  Pauly (1980) 0.16-0.18 0.27-0.31 0.14-0.30  Apparent age distributions are presented in Figure 6, with ‘best fit’ predicted population proportions Pa based on maximum likelihood estimates of apparent total mortality rate Z = M+r. A good visual fit is not expected for such distributions, since the sample sizes for pa are small. The estimates of Z are relatively insensitive to uncertainty about the K, L1, L∞ estimates used to reconstruct apparent individual ages (see marginal probability distributions, Figure 5). The Z estimates can also be made to vary by about 0.02-0.05 by changing the minimum age included in mortality rate estimates (younger ages apparently less vulnerable at least for chub and bluehead). These sensitivity tests all indicate that M, or more precisely the sum M+r, is fairly well determined by the data. Note, in Figure 6, that the humpback chub age distribution appears to have systematic underrepresentation of younger ages and overrepresentation of older ages, hinting at possible recruitment decline over time. Chub age proportions by year of tagging also suggest Table 1 also reports independent estimates of natural mortality rate M from tagging studies during the 1990s. Note that for chub and bluehead sucker, the best fitting M is considerably lower than these independent estimates. Taken at face value, comparing these M estimates to the best fit Z values would indicate stability for flannelmouth sucker (or perhaps a slight annual decline), but possibly rapid decline for humpback chub and bluehead sucker. However, the Z es of M for all s  possible decline in relative abundance of young (< 7 year old) fish from 1991 to 1994 (Figure 7), especially when sample sizes are inflated by including all fish with time to recapture > 0.5 year. Apparent mortality rate 0.0 0.3 0 0.2 0.4 0.6 Instantaneous rate (Z) P os te ri or  p ro ba bi lit y HBC FMS BHS  Growth K parameter 0.0 0.3 0.6 0.1 0.3 0.5 K P os te ri or  p ro ba bi lit y HBC FMS BHS  Asymptotic body length timates are close to estimates  from the Pauly (1980) equation pecies. 0.0 200 300 400 500 60 Asymptotic length (mm) P os t 0.3 0.6 er io r pr ob ab ili ty HBC FMS BHS  Figure 5 Bayes posterior probability distribution for population mean growth (K, L∞) and apparent total mortality rate Z parameters, calculated from the likelihood function defined in  = bluehead sucker. Appendix 1. HBC = humpback chub, FMS = flannelmouth sucker, BHS  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 85 Humpback chub 0.1 0.12 0.14 0 0.02 0.04 0.06 0 5 10 15 20 25 3 0.08 0 Age (yr.) P ro po rt io n of  fi sh Observed Predicted Bluehead sucker 0.2 0 0.05 0.1 0 5 10 15 20 25 30 3 0.15 5 Age (yrs.) P ro po rt io n of  s am pl e Observed Predicted   Flannelmouth sucker 0.1 0 0.02 0.04 0.06 0.08 0 5 10 15 20 25 Age (yrs.) P ro po rt io n of  s am pl e Observed Predicted  instantaneous mortality rate Z equal to the maximum likelihood estimate; essentially the same age composition curve is predicted for each species by assuming Z equal to the natural mortality rate M predicted using the Pauly (1980) empirical relationship between growth parameters and M.  Figure 6 Apparent age distributions at age of first tagging for Grand Canyon fishes. Curves show age distribution expected for stable population with annual Humpback chub age proportions by year of tagging 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0 5 10 15 20 25 30 Apparent age (years) P ro po rt io n of  fi sh 1991, n=380 1992, n=667 1993, n=539 1994, n=59  Humpback chub age proportions by year of tagging, including all fish with >0.5 yr time to recapture 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0 5 10 15 20 25 30 Apparent age (years) P ro po rt io n of  fi sh 1991, n=912 1992, n=1227 1993, n=998 1994, n=203  Figure 7 Apparent age distributions at age of first tagging for humpback chub, separated by year of tagging. Note apparent scarcity of younger fish after 1991, especially for larger sample created by including all fish with >0.5 year time to recapture (rather than 1 year to recapture).  DISCUSSION Observed patterns of growth rate vs. length at tagging (Figure 3) leave little doubt that relatively large numbers of old fish were sampled for tagging. Far more old fish were tagged than would be expected from the low annual survival rates estimated by mark-recapture models (see Table 1), if the populations were stable. This means that the overall results of our analysis can be interpreted in at least three ways: 1. Stable population hypothesis: tagged fish were representative of population age composition, mortality rates are low as predicted from the Pauly (1980) relationship between mortality rate and K, L∞, and the mark-recapture survival estimates have provided underestimates of survival for some reason; 2. Nonrepresentative sampling hypothesis: populations are stable, with higher mortality rates than predicted by the relationship of Pauly (1980), but younger fish were underrepresented in collecting fish for tagging due to ontogenetic differences in habitat preference; 3. Declining populations hypothesis: age composition sampling was representative and survival rates are low, but there are relatively few young fish in the populations due to declining recruitment. We know of no other instance where the Pauly relationship has so grossly underestimated natural mortality rate for a species with low K as to lend credibility to hypotheses (2)-(3), and there is no evidence from analysis of length-frequency samples of smaller fish that growth rates (and hence K) are much higher  Growth and population trends in Grand Canyon native fishes, Walters, C. et al. 86 than we have estimated. In fact the opposite appears to be the case for flannelmouth suckers; juveniles It is o ve statement about population trends, but uncertainties about natural survival rates and age-selectivity in sampling make this impossible. However, east some recruitment well after construction of GCD; most of the fish tagged were almost certainly less than  of tagged fish fro s apparently still exist in a few locations. of h. There will likely be continued reliance on analysis methods based on tag recovery and (<2  causes a relatively high mortality risk; and rel  and recruitment rates. Age estimates vary widely among individuals t length interval, i.e., growth rates vary widely among individuals so that length alone is a very oor predictor of age. This means that length frequency distributions can appear stable on time scales of a decade or more, even if there were nearly complete recruitment failure. It is essential to continue gathering age composition information either by direct age sampling, or by methods like the tagging-based procedure described here. If tagging is continued, considerably more care should be taken to avoid the large measurement errors evident from comparing recorded lengths of fish recovered immediately after tagging (see Figure 2, top two panels). A very interesting feature of the flannelmouth data is the discrepancy between first year growth rate measured from length-frequency sampling of shoreline habitats vs. the substantially higher predicted age 0-1 growth given growth, performance of older tagged fish. It appears that warmed shoreline-backwater areas are either not adequate to insure ‘normal’ first year growth in the face of the unnaturally cold waters that first year juveniles must often face (while dispersing to warm areas, and when such areas are flushed by flow changes), or that first year juveniles are driven to use relatively poor shoreline growth habitats by some other factor such as predation risk. Interestingly, rainbow trout (Oncorhynchus mykiss) just below GCD have similar, strong ontogenetic habitat shifts and also display lower first year growth than would be expected from growth rates of older fish. It is difficult to see why juvenile rainbows would concentrate in shoreline areas except to avoid predation risk, since there are large offshore areas of relatively slow water and shoreline areas have relatively low food concentrations due to diurnal water level fluctuation associated with operation of GCD. It would not be necessary to seek indirect evidence of population trend from age composition data if reliable methods could be developed for direct assessment of changes in relative abundance over time. Unfortunately, past sampling and abundance index programs have specifically targeted sites where biologists expected to catch fish, and this could easily lead to ‘hyperstability’ (Hilborn and Walters, 1992) in catch rate indices (catch rates remaining high in preferred habitats despite population declines). It may be impractical with existing sampling methods to regularly visit enough additional sampling sites to avoid this problem in the future. If adaptive, experimental management programs are developed for GCD that involve comparing relative abundance trends under alternative water management regimes, a critical research investment prior to implementing such programs should be in development of better abundance indexing procedures for large, turbid rivers. For example, side-scan sonar now being used in the Canyon for geomorphology monitoring (sand accumulation on the river bottom) might be adapted to ‘count’ fish over long river reaches. rearing in shoreline/backwater refuge areas appear to have lower first year growth rates than predicted by the von Bertalanffy model. bviously frustrating to be unable to make a more definiti it is clear from the distributions of apparent ages that all three populations were achieving at l 20 years. old, implying they were recruited well after construction of GCD. That is, growth implies that adult size distributions cannot be interpreted simply as closed cohorts of old fish left over m healthy populations prior to GCD, though such closed cohort It is understandable that Grand Canyon policy has been to avoid killing any native fishes for direct analysis age and growt size distribution data. But these methods cannot be tested and improved much without eventually: (1) killing a few fish for validation of age estimates based on eq. (3)-(4); (2) including more small fish 00 mm) in the tagging programs even if tagging these fish (3) evaluation of possible reasons for underestimation of survival rate from mark-recapture methods. The estimates of apparent age indicate that analysis of the adult size distribution alone will not provide a iable index of changes in mortality of any adul p  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 87 REFERENCES Arizona Department of Fish and Game (ADFG), 1987. Evaluation of varied flow regimes on aquatic resources of Glen and Grand Canyons. Arizona Fish and Game Department, Phoenix, Az. [Available from National Technical Information Service, Springfield VA 22161, report PB88-183439.] Fabens, A.J., 1965. Properties and fitting of the von Bertalanffy growth curve. Growth 29, 265-289. Gelman, A., Carlin, J.B., Stern, H.S., Rubin, D.B., 1995. Bayesian Data Analysis. Chapman & Hall, London. Hilborn, R., Walters, C., 1992. Quantitative Fisheries Stock Assessment. Chapman & Hall, New York. Pauly, D., 1980. On the interrelationships between natural mortality, growth parameters, and mean environmental temperature in 175 fish stocks. J. Cons. int. Explor. Mer. 39, 175-192. Valdez, R.A., Ryel, R.J., 1995. Life History and Ecology of the Humpback Chub (Gila cypha) in the Colorado River, Grand Canyon, Arizona. Contract Report to Bureau of Reclamation, Salt Lake City, Utah. Contract report No. 0- CS-40-09110. Wang, You-Gan, 1998. An improved Fabens method for estimation of growth parameters in the von Bertalanffy model with individual asymptotes. Can. J. Fish. Aquat. Sci. 55, 397-400. Wang, You-Gan, Thomas, M.R., Sommers, I.F., 1995. A maximum likelihood approach for estimating growth from tag- recapture data. Can. J. Fish. Aquat. Sci. 52, 252-259. Walters, C.J, Ludwig, D., 1994. Calculation of Bayes posterior probability distributions for key population parameters. Can. J. Fish. Aquat. Sci. 51, 713-722.  APPENDIX 1. COMBINED LIKELIHOOD FUNCTIONS FOR GROWTH AND SURVIVAL PARAMETERS Suppose a sample i = 1…n fish of ages ai≥amin has been gathered from a population with a stable age distribution under population growth rate r and age-independent natural mortality rate M, hence displaying the apparent mortality rate Z = M + r. This population should have age proportions P(a) = e-Z(a-amin) (1 - e-Z) if ages are discrete, or P(a) = e-Z(a – amin) Z if ages are continuous (recruitment is a continuous process). These expected values for P(a) hold whether or not there has been historical random variation in recruitment. Assume each of these fish grows according to a von Bertalanffy growth curve with common K but individual asymptotic length L(i)∞ = L∞+Ei, where the Ei are normally distributed with mean 0 and variance σ2: p(E) = (2Πσ2)-1/2exp(-E2/2σ2). For each fish, we observe the length at first capture LS and growth G = LR - LS over an arbitrary period T until it is next recaptured. To construct a likelihood function for these (LS,G) observations, we need to express them in terms of the probability statements P(a) about age and p(E) about E. That is, we need to transform the probability distributions from (a, E) to (LS, G). Assuming LS and G are independent, this transformation gives the likelihood of a given observed combination. The transformation can be expressed as l(LS,G| L∞,K,M) = ΠiP(ai)p(Ei)||Ji||-1 … A1) In this likelihood function, each ai is calculated as ai = -(1/K)log(1 - LS/L(i)∞), i.e., by eqs. (3)-(4), and each Ei is calculated as Ei = Gi/(1 - e-KT)+LS - L∞. The term) ||Ji||-1 is the absolute value of the Jacobian of the transformation from (a,E) to (Ls,G), and it is given for each observation by ||Ji||-1 = KL(i)∞ e -K(ai - ao)Hi where Hi = 1 - e-KT. Note that l(LS,G| L∞, K, Z) depends on the parameters L∞, K, and M in a complex, nonlinear way through the effect of the parameters on ai and Ei, so we must expect to use numerical search procedures to find maximum likelihood estimators and/or posterior probability distributions for the parameters. For maximum likelihood estimation it is generally easier to work with log(l). Taking the logarithm of l(LS,G| L∞, K, M) above, discarding constant terms that do not affect the maximization (or Bayesian analysis), and evaluating the nuisance parameter σ2 at its maximum likelihood estimate s2 = ΣiE2i/n conditional on the other parameters, we obtain the reduced log-likelihood function: log(l) = -(n/2)log(s2) - Σiln(Hi)+nlog(Z/K) - ZΣi(ai - amin) - Σilog(L(i) ∞)+KΣi(ai - ao) … A2)  Growth and population trends in Grand Canyon native fishes, Walters, C. et al. 88 For analysis of growth data only (estimation of K and L∞ without assuming that a random sample of the population age composition was obtained), the appropriate reduced log-likelihood function is just the first two terms of log(l); log(l) is actually quite simple to calculate in a spreadsheet format, and to maximize using spreadsheet functions like Excel’s Solver: (1) enter the LS, G, T data in columns and define cells for the parameters L∞, K, Z; (2) calculate a column of Ei values (from Ei = Gi/(1 - e-KT)+LS - L∞) and a column of L(i) ∞ = L∞+Ei values from the parameter values; (3) calculate the sum terms (s2, etc.) of log(l) using the spreadsheet SUMPRODUCT function; and (4) assemble these terms into the log(l) formula. K can be elimi 1 is measured exactly so orporated in log(l), if it is assumed that the estimate has a normally distributed error with known standard deviation σ1. Simply subtract the term (L*1 - L1)2/(2σ12) from log(l), where L*1 = L∞(1 - e-K). Then assuming a very nown perfectly) becomes equivalent to removing K from the estimation by calculating it as - /L ). PPENDIX 2. M CARLO TESTS OF ESTIMATION METHODS We conducted a variety of Monte Carlo simulation tests to evaluate possible biases in the estimation procedures. Each test consisted of generating 1000 samples of 300 fish each, doing the estimation procedures on each sample, and tabulating mean and variance of parameter estimates over the samples. Each fish was assigned an age at capture from a stable (exponential) age distribution with minimum capture age 4 years, and a normally distributed L(i)∞ (most tests used a standard deviation of 60 mm, roughly what we estimated for the actual data). For some tests, each individual was also assigned a unique K(i), normally distributed around the population K for that test. Times to recapture were simulated as tion tests revealed a number of potential biases. First, the maximum likelihood estimates of L∞ ward, and K given L1 is biased upward by the same relative amounts when overestimated (see Appendix 1), and L∞ underestimated, when independent information about L1 is ignored, unless age at first capture is reduced to 1 year. This effect was also seen in the original data analysis, and reflects lack of information about K when only older fish are included in the sample. Third, estimates of Z are biased slightly upward (5%) so natural survival rate S is biased slightly downward (e.g., 0.76 when should be 0.78, 0.90 when should be 0.91). Since our main concern here is about apparent survival rates, this bias could tend to mask effects of negative population r values (i.e., make population appear stable when in fact it is actually declining slowly). Fourth, individual K(i) can vary with a standard deviation of up to 0.1 around a mean of 0.2 without causing bias in the growth and survival parameter estimates, though higher K variation of course causes increased variation in the parameter estimates (higher apparent variation in L(i)∞ , K, and S). Fifth, including Lee’s phenomenon effects in the fake data (decreasing mean L(i)∞ with increasing age) causes downward bias in L(i)∞ and in estimated survival rate. ixth, measurement errors of the magnitude apparent in the data can badly bias the estimates, unless the ensored’ to eliminate individuals with low T (<0.5 year.) and negative observed growth G. Censoring the data by eliminating such individuals does not cause any obvious bias in the estimated growth and mortality parameters, which is a bit surprising considering that older individuals are more likely to display negative apparent G and hence to be omitted from the analysis. Finally, distributions of L∞ estimates have shapes and variances quite close to the posterior distributions calculated using Bayesian methods, though with slightly smaller variances due to not explicitly considering uncertainty about L1.  nated from the likelihood function above by assuming mean length at age 1, L K = -log(1 - L1/L∞). But uncertainty in the independent estimate of L1 can be easily inc small σ1 (L1 k log(1 - L1 ∞ A ‘pathology’ can arise in maximization of log(l) when the number of young fish (just above age amin) in the sample differs considerably by chance or age sampling bias from the number expected under exponential decline. In such cases, log(l) can be made larger just by increasing K so as to drive apparent ages ai of such young fish below the cutoff age amin for inclusion in the mortality components of the likelihood function. A ONTE uniform over the interval 0-3 years. Simulated ‘true’ individual sizes at marking and recapture were calculated with the von Bertalanffy model from the assigned ages, L(i)∞, and K(i), without measurement errors. Normally distributed random measurement errors were added to the true LS and LR values for each fish to generate ‘observed’ sizes. The simula are biased slightly (5%) down age at first capture is high (amin = 4). These biases can only be corrected by reducing the age at first capture and/or providing very precise values of L1. Second, K is much more likely to be pathologically S data are ‘c  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 89 EFFECTS OF LAKE AND POND AERATION ON FISH GROWTH AND RELATED PROCESSES20 Daniel Pauly The Sea Around Us Project, Fisheries Centre, University of British Columbia, 2202 Main Mall, Vancouver, BC V6T 1Z4 Canada; Email: d.pauly@fisheries.ubc.ca ABSTRACT The basic principles of the growth of fish (and aquatic invertebrates) are recalled, with emphasis on the fact that oxygen, while continuously required for maintenance, cannot be stored for later use. Hence the n is extracted via the gills and transported into the body of fish limits, at any tim ir scope for activity, growth and food conversion efficiency. This is shown to be consistent with the increased growth and food conversion efficiency observed by aquaculturists who aerate their ponds. Some implications for aeration of larger water bodies, such as lakes and reservoirs, are presented. INTRODUCTION The following is a brief presentation of a theory, elaborated in more detail in Pauly (1979, 1981, 1984, 1986, 1998) and in Longhurst and Pauly (1987) of how f  g e he  t explains some of the observed direct and indirect effects  t n h h elated processes (Loyacana, 1974; Hollerman and Boyd, 1980), and that it can be used to predict some of the effects of aerating larger water bodies such as lakes and reservoirs. This theory applies to any submerged animal breathing through gill; the example presented below refers to fish in the narrow sense (i.e., to teleosts), although the principles it illustrates also apply to other fishes and aquatic invertebrates (see Pauly, 1998). We shall assume for simplicity’s sake that fish consist of, and feed, only on proteins. The theory presented below can accommodate more realistic body composition and diets (van Dam and Pauly, 1995), but dealing with this does not change its main points. THEORY OF FISH GROWTH Fish are aerobic heterotrophs – with some exceptions, such as common carp (Cyprinus carpio), which may operate anaerobically at very low temperatures, a feature which is not a concern here. As fish feed, their food is assimilated i.e., broken down into amino acids; part of the amino acid pool is oxidized, and the energy thus bound used to form ATP, used for activity (i.e., muscle contraction), and, along with building blocks drawn from the amino acid pool, used for synthesis of native protein. This synthesis is required for net growth, but also, even more importantly, for replacing proteins that have spontaneously denatured (i.e., lost their quaternary and tertiary structures). Such spontaneous denaturation – a mildly exergonic reaction requiring neither O2 nor ATP – is a characteristic of live proteins. Indeed, it expresses a basic feature of life itself: that living organisms will spontaneously decay, i.e., fail to maintain their structure integrity unless entropy is ‘pumped out’ (Schrödinger, 1944).                                                   rate at which oxyge e, the ish row. Emphasis is giv n to t  fact hat it  of pond aera ion o  fis  growt and r   20 Cite as: Pauly, D., 2006. Effects of lake and pond aeration on fish growth and related processes. In: Palomares, M.L.D., Stergiou, K.I., Pauly, D. (eds.), Fishes in Databases and Ecosystems. Fisheries Centre Research Reports 14(4), pp. 89-95. Fisheries Centre, University of British Columbia [ISSN 1198-6727].  Lake and pond aeration on fish growth, Pauly, D. 90 Important here are: • That the rate of spontaneous denaturation of proteins can be assumed proportional to protein mass (i.e., roughly proportional to body weight); and • That this rate of spontaneous denaturation, being due to thermally-induced vibrations of protein molecules, increases with temperature, with a Q10 usually ranging from 2 to 4 (Winberg, 1971; Regier et al., 1990). • That the relation between the metabolic rate of fish and water temperature is reasonably well described by Krogh’s ‘normal curve’ (Table 1). Protein synthesis as mentioned above requires O2 to be where needed (in cells’ mitochondria); for this to be the case O2 must have been brought in via the circulatory system, through the gills from the water surrounding a fish. Transfer of O2 through the gills of fish follows Fick's Law: Q = dP·G·U/WBD ... 1) where Q is the O2 uptake (e.g., ml·hour-1), dP is the O2 pressure difference on either side of the gill membrane (in atm), G is the respiratory area of the gills (total area of respiratory lamellae), U is Krogh's diffusion constant, i.e., the quantity of O2 (in ml) which diffuses through an area of 1 mm2 in one minute for a given type of tissue when the pressure gradient is one atm O2 µ-1, and WBD is the water-blood distance, i.e., the thickness of the membrane separating water and blood, in µ (Hughes and Morgan, 1973; Hughes, 1984). Of the four parameters which influence Q, only G varies with body weight (W), i.e., G = a·Wd ... 2) where ‘a’ is a multiplicative factor used here as ‘gill area index’, and ‘d’ is an exponent ranging in fish between 0.50 (in cyprinodonts, Winberg, 1961) and 0.95 (in tuna, Muir and Hughes, 1969), but never reaching unity, at least not in well-studied cases covering a wide range of body weight. Thus, gill surface area can be expected to be a key variable when attempts are made to explain the wide difference of growth performance occurring among species of fishes. This can be shown by using the parameters W∞ and K of the von Bertalanffy growth function (VBGF), whose simplest version has, for weight, the form Wt = W∞ (1 - e- (K(t-to))3 ... 3) where Wt is the weight at age t, W∞ is the mean weight the fish would reach if they were to live indefinitely, K expresses the rate at which W∞ is approached, and to is the theoretical ‘age’ the fish would have at W = 0. From W∞ and K, a growth performance index Φ can be derived, i.e., Φ = log10(K) + 2/3 log10(W∞) ... 4) which takes similar values among different populations of the same species, and hence can be used to compare the growth performance of different fishes (Pauly, 1979; 1994). Figure 1 shows that the gill area index of 37 species of teleosts, ranging from guppies to tunas, i.e., selected to cover a wide range of asymptotic sizes and ecologies, significantly and positively correlates with their Table 1 Values of the temperature (t) correction factor (q) for converting respiratory rates to 20°C, according to the ‘normal curve’ of Krogh (1914) (from Winberg, 1971). t q t q t q t q 5 5.19 12 2.16 19 1.09 26 0.609 6 4.55 13 1.94 20 1.00 27 0.563 7 3.98 14 1.74 21 0.920 28 0.520 8 3.48 15 1.57 22 0.847 29 0.481 9 3.05 16 1.43 23 0.779 30 0.444 10 2.67 17 1.31 24 0.717 - - 11 2.40 18 1.20 25 0.659 - -  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 91 growth performance index. It might be argued at this point that if gill size is limiting, then fish should, over evolutionary time, have developed larger gills. The answer to this is that they have: their gills are suitable for rapid growth up to the size at (first) reproduction, i.e., to the size which is crucial to their evolutionary fitness. It is only from that size that the limiting effect of low relative gill area manifests itself (Pauly, 1984; 1994). Moreover, a growth limitation would occur at some stages: whatever the initial endowment, gill area, being a surface cannot, for geometrical reasons, keep up with the growing volume it is supposed to supply with oxygen. Given (2), we also have: Q = a’·Wd in which Q is defined as in (1), W and d as in (2), and a’ is a proportionality constant. Equation (5) implies that relative gill area, and hence the O2 available for growth and routine metabolism in fish, decline as size increases. This decline occurs in proportion to a power of weight equal to 1-d, down to a level where Q is, at W∞, just enough for maintenance, i.e., that level of activity and of protein synthesis that is sufficient to compensate for spontaneous protein denaturation (Figure 2A). Thus the level of metabolism corresponding to W∞ is, by definition, an estimate of maintenance metabolism, and any factor that increases maintenance metabolism (e.g., elevated temperature, or reduced food density, by increasing the level of activity required to secure the required food) will have the effect of reducing W∞ (Figure 2B). This explains why, e.g., the fish of North American freshwaters tend to reach larger maximum sizes at their cold northern end than at the warm, southern ends of their range (see data in Carlander, 1969; 1977), or why Australian fishes have asymptotic sizes that are higher in the (cold) south than in the north of that country (Andersen and Pauly, 2006, this volume). An important variable for managers of aquaculture ponds is food conversion ratio (FCR), defined as the given amount of fish flesh elated to a concept commonly eries scie  food conversion efficiency (K1) of Iv ough l/FCR = K1, the latter being defined, for any time interval by: K  = growth increment/food consump ... 6) te K1 wth (e.g., Paloheimo and Dickie, 1966), or m  (6), r pr in eq ation (3). One of these w s presented  A·(1 -W 1 is t fficie eight , W∞ as defined in equation (3), d is implied in the version of the VBGF presented here [there are other version of the VBGF, incorporating other, more l f d, see Pauly, 1981; Temming, 1994b; and Essington et al., 2001], and A is a ro and one, and expressing the fraction of the ingested food that is available r protein synthesis. It can be expected that A will be related to the nitrogen content of the food (Pandian and Marian, 1985). ... 5)  Figure 1 Relationshi in Equation 2) and the g ts and two sharks as exp  4; adapted from data in Pauly 1979). ps between an index of the gill area (a’ rowth performance of 35 species of teleos ressed by their value of Φ (see Equation amount of food required to produce a u  fish . This is r lev (1966) thrsed in n ece, th 1 tion Various approaches exis equation t to rela  with the VBGF as e  and gro ore precisely by Temming esented u a (1994a), i.e., K  = /W )l-d ... 7) 1 ∞ where K he food conversion e ncy at w  W  is set at 2/3, as rea istic values o factor constrained between ze fo  Lake and pond aeration on fish growth, Pauly, D. 92 APPLICATION OF THE THEORY TO AQUACULTURE POND AERATION There is an extensive literature on pond aeration, which tends to emphasize its technological aspects (see e.g., Boyd et al., 1988), and only one of its numerous biological effects: the reduction of mortality due to (early-morning) oxygen deficiency. However, aeration has numerous other biological effects, notably, beneficial effects on food conversion and growth (see Table 2). Strangely enough, these effects of aeration appear to date not to have been related to any theory of growth. The point here is that the observations in Table 2 are fully consistent with the theory presented above stating that fish growth is generally oxygen- limited. On the other hand, they flatly contradict conventional theories of fish growth, which tend to concentrate exclusively on ad hoc postulates of local food scarcity (see, e.g., Weatherley and Gill, 1987). This suggests that quantitative predictions (i.e., hypotheses) concerning the response of fish to pond aeration made on the basis of that theory, represent ‘strong inferences’ sensu Platt (1964), the testing of which is likely to advance a field still dominated by empirical approaches. APPLICATION OF THE THEORY TO LAKE AND RESERVOIR AERATION One of the corollaries of the above theory is that destratifying a lake such that its overall oxygen content is increased (Fast and Hulquist, 1989) should result, other things being equal, in improved growth of the fish therein, both by directly facilitating respiration, and by increasing the size of those water layers that have both suitable temperature and food. Table 2 Response to aeration of some cultivated fish species (√: increase explicitly noted; -: item not mentioned). Common name Scientific name Location Increase of:     Cona Growth Survival Harvest Profits Source Common carp Cyprinus carpio Szarvas, Hungary - - - √ √ Abdul Amir (1988) Silv tichthys nobilis Szarvas, - - - - √ Abdul Amir (1988) Big  Jap Taiwan - - √ Anon. (1988a) Tila Cha Hyb X Aristichthys nobilis USA  - Shireman et al. (1983) a) food consumption and/or conversion  Figure 2 Illustrating how, given a certain G-line (determined by a’ and d in Equation 5), maintenance metabolism determines asymptotic weight (W∞), because relative gill area (and hence oxygen supply) must decline with body weight. A. Fish exposed to a low level of stress (e.g., environmental temperature, abundant food). B. Fish exposed to a higher level of stress (high temperature, causing rapid denaturation of body protein, and/or low food density, requiring O2 to be diverted to foraging, rather than protein synthesis). Note that ‘scope for growth’ and food conversion efficiency can both be directly related to the difference, in these graphs, between the G-line and the level of routine metabolism. er carp Artis Hungary head carp Hypophthalmichthys molitrix Szarvas, Hungary - - - - √ Abdul Amir (1988) anese eel Anguilla japonica Lukang, √ √ pia Cichlidae Singapore - √ √ - √ Anon. (1988b) nnel catfish Ictalurus punctatus Alabama, USA √ √ √ √ √ Hollerman and Boyd (1980) rid carp Ctenopharyngodon idella Florida, - √ √ -    Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 93 This is illustrated here by a scheme in which stratification reduces the habitat of a fish population. Let us ification such that the O2-free hypolimnion moves up, reducing the amount of benthos accessible in 10°C water. In such a case, the fish will have to undertake nto the warm epilimnion, and thus expose themselves to higher temperatures. Let us further assume that integrating time/temperature profiles of these forays suggests the fish to live, on the average, in a temperature of 12°C. Other things being equal, and given Krogh's normal curve (Table 1), this will raise O2 consumption by about 25%. Thus we have: 1.25 = (W∞(10°C)/W∞(12°C))1-d ... 6) from which W∞(12°C) = 328g. Given equation (4) and the initial value of ∞(12°C) i.e., K(12°C) = 1.05 year-1. Thus, given the above theory and ancillary field information, we can predict qualitatively and quantitatively how food conversion efficiency and growth (and hence also natural mortality, see Pauly 1980) will change, given changes in the thermal stratification of a lake (Figure 3). Similar procedures can be applied to assess the impact of the distribution of O2, wo  documents a habitat – Chesapeake Bay – then in need of destratification, i.e., where areas of their original habitat. Th ove, based on first principles and easily verifiable assumptions, can be easily eriments and for Without such development of the above theory, or of a modification thereof, the observed impacts of ments of a well-   assume a lake with a warm epilimnion, a cool (10°), well-oxygenated mesolimnion and a small, oxygen- free hypolimnion (Figure 3). Let us further assume a population of cold-water fish, limited to the mesolimnion, feeding at 10°C, and having, for a value of d = 0.8, the VBGF parameters W∞(10°C) = 1,000g (Figure 3), and K(10°C) = 0.5 year-1. Let us now imagine a change in the lake's strat feeding forays i  K(10°C) = 0.5 year-1, one can also estimate the value of K corresponding to W , and thus to reexamine as an example, the rk of Coutant (1985, 1987, 1990), which Figure 3 Schematic representation of a stratified lake, with each layer offering a different O2/temperature combination to resident fishes. The insert in the lower right corner shows the growth striped bass – especially the large ones, for which oxygen supply was a problem – had curves resulting from the two scenarios in the text. become unable to grow and feed in certain CONCLUSION e theory presented ab developed to provide a comprehensive framework both for interpreting aeration exp predicting potential effects of aeration in various water bodies. aeration on ponds and lakes will continue to be perceived as isolated facts, and not as ele articulated system of principles allowing strong inferences and rapid advances.  Lake and pond aeration on fish growth, Pauly, D. 94 ACKNOWLEDGEMENTS Pond ed and included here, only tably Ab ., 1988. Aeration proves its worth with carp. Fish Farmer-Internat. File 2(1), 4-5.  65-68. Fisheries Centre, University of British Columbia. s 1(1), 4. Boyd, C.E., Tucker, C.S., 1979. Emergency aeration of fish ponds. Trans. Amer. Fish. Soc. 108(3), 299-306. Boyd, C.E., Taufik Ahmad, Zhang La-fa, 1988. Evaluation of plastic pipes, paddle wheel aerations. Aquacul. Eng. 7, 63- Carlander, K.D., 1969. Handbook of Freshwater Fishery Biology. Vol. 1. The Iowa State University Press, Ames, Iowa. Ca y. Vol. 2. The Iowa State University Press, Ames, Iowa. Co .C., 1985. Striped bass, temperature, and dissolved oxygen: a speculative hypothesis for environmental risk. Co lity? Environ. Biol. Fishes 18(3), 161-172.  changing climate. Trans. Ess ., 2001. The von Bertalanffy growth function, bioenergetics, and the ast, A.W., Hulquist, R.G., 1989. Oxygen and temperature relationship in nine artificially aerated California reservoirs. Calif. Fish Game 75(4), 213-217. Hollerman, W.D., Boyd, C.E., 1980. Nightly aeration to increase production of channel catfish. Trans. Amer. Fish. Soc. 109, 446-452. Hughes, G.M., 1984. Measurement of gill area in fishes: practices and problems. J. Mar. Biol. Ass. (U.K.) 64, 637-655. Hughes, G.M., Morgan, M., 1973. The structure of fish gills in relation to their respiratory function. Biol. Rev. 48, 419- 475. Iv1ev, V.S., 1966. The biological productivity of waters. J. Fish. Res. Board Can. 23(11), 1727-1759. Krogh, A., 1914. The quantitative relation between temperature and standard metabolism in animals. Int. Z. Phys.- Chem. Biol. 1, 491-508. Longhurst, A., Pauly, D., 1987. Ecology of Tropical Oceans. Academic Press, San Diego, California. Loyacano, H.A., 1974. Effect of aeration in earthern ponds on water quality and production of white catfish. Aquaculture 3, 261-271. Muir, B.S., Hughes, G.H., 1969. Gill dimensions for three species of tunny. J. Exp. Biol. 51, 271-285. Paloheimo, J.E., Dickie, L.M., 1966. Food and growth of fishes III. Relations among food, body size, and growth efficiency. J. Fish. Res. Board Can. 23, 1209-1248. Pandian, T.J., Marian, M.P., 1985. Nitrogen content of food as an index of absorption efficiency in fishes. Mar. Biol. 85, 301-311. Pauly, D., 1979. Gill size and temperature as governing factors in fish growth: a generalization of von Bertalanffy's growth formula. Ber Inst. Meereskd. Univ. KieI. No. 63. I thank Dr. R.S.V. Pullin for his most useful comments on the draft of this contribution, originally prepared in 1996, and which was intended as chapter in a book, to be edited by colleagues, on ‘Lake and Aeration’. The book project fell through, but this contribution was rescu minimally updated (3 new references), as I feel it has not, 10 years later, lost any of its pertinence, no for the interpretation of growth data in FishBase. REFERENCES dul Amir, A Andersen, C., Pauly, D., 2006. A comparison of growth parameters of Australian marine fishes north and south of 280 South. In: Palomares, M.L.D., Stergiou, K.I., Pauly, D. (eds.), Fishes in Databases and Ecosystems. Fisheries Centre Research Reports 14(4), pp. Anon., 1988a. The miracle of Taiwan's eel culture. Aqua-O2 New Anon., 1988b. Veteran nets greater tilapia yields. Aqua-O2 News 1(2), 6. 72. rlander, K.D., 1977. Handbook of Freshwater Fishery Biolog Costa-Pierce, B.A., Pullin, R.S.V., 1989. Stirring ponds as a possible mean of increasing aquaculture production. Aquabyte (ICLARM) 2(3), 5-7. utant, C Trans. Amer. Fish. Soc. 114(1), 31-61. utant, C.C., 1987. Thermal preference: when does an asset become a liabi Coutant, C.C., 1990. Temperature-oxygen habitat for freshwater and coastal striped bass in a Amer. Fish. Soc. 119, 240-253. ington, T.E., Kitchell, J.F., Walters, C.J consumption rates of fish. Can. J. Fish. Aquat. Sci. 58, 2129-2138. F  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 95 Pauly, D., 1980. On the interrelationships between natural mortality, growth parameters and mean environmental temperature in 175 fish stocks. J. Cons. int. Explor. Mer 39(3), 75-192. Pauly, D., 1981. The relationships between gill surface area and growth performance in fish: a generalization of von Bertalanffy's theory of growth. Meeresforschung/Rep Mar. Res. 28(4), 251-282. Pauly,D., 1984.A mechanism for the juvenile-to-adult transition in fishes. J. Cons. int. Explor. Mer 41, 280-284. Pauly, D., 1986. A simple method for estimating the food consumption of fish populations from growth data and food conversion experiments. U.S. Fish.Bull. 84(4), 827-842. Pauly, D.,1994. On the Sex of Fish and the Gender of Scientists: Essays in Fisheries Science. Chapman and Hall, London. Pauly, D. 1998. Why squids, though not fish, may be better understood by pretending they are. In: Payne, A.I.L., Lipinski, M.R., Clarke, M.R., Roeleveld, M.A.C. (eds.), Cephalopod Biodiversity, Ecology and Evolution. South African J. Mar. Sci. 20, pp. 47-58. Platt, J.R., 1964. Strong inference. Science 146(3642), 347-353. Regier, H.A., Holmes, I.A., Pauly, D., 1990. Influence of temperature changes on aquatic ecosystems: an interpretation of empirical data. Trans. Amer. Fish. Soc. 119, 374-389. Schrödinger, E., 1944.What is Life? Cambridge University Press, Cambridge. Shireman, J.V., Aldridge, F.J., Rottmann, R.W., 1983. Growth and food habits of hybrid carp, Ctenopharyngodon idella X Aristichthys nobilis, from aerated and non-aerated ponds. J. Fish Biol. 23, 595-604. Temming, A., 1994a. Food conversion efficiency and the von Bertalanffy growth function. Part I: A modification of Pauly's model. Naga, the ICLARM Quarterly 17(1), 38-39. Temming, A., 1994b. Food conversion efficiency and the von Bertalanffy growth function. Part II and conclusion: extension of the new model to the generalized von Bertalanffy growth function. Naga, the ICLARM Quarterly 17(4), 41-45. van Dam, A.A., Pauly, D., 1995. Simulation of the effects of oxygen on food consumption and growth of Nile tilapia, Oreochromis niloticus (L.). Aquacul. Res. (26), 427-440. Weatherley, A.H. and H.S. Gill. 1987. The Biology of Fish Growth. Academic Press, London. Winberg, G.G., 1961. New information in the metabolic rate in fishes. Voprosy lkhtiologii 1(1), 157-165 [Fish. Res. Board Can. Transl. Ser. (362), 11 p.] Winberg, G.G., 1971. Method for the Estimation of Production of Aquatic Animals. Academic Press, London.       FISHERIES CENTRE RESEARCH REPORTS       ISSN 1198-6727  1993 Volume 1(1) – 2006 Volume 14(4)  Pitcher, T., Chuenpagdee, R. (eds) (1993) Commercial whaling - the issues reconsidered. Fisheries Centre Research Reports 1(1):36pp. Pitcher, T., Chuenpagdee, R. (eds) (1993) Decision making by commercial fishermen: a missing component in fisheries management? Fisheries Centre Research Reports 1(2):75pp. Pitcher, T., Chuenpagdee, R. (eds) (1994) Bycatch in fisheries and their impact on the ecosystem. Fisheries Centre Research Reports 2(1):86pp. Bundy, A., Babcock, E. (eds) (1995) Graduate student symposium on fish population dynamics and management. Fisheries Centre Research Reports 3(2):33pp. Pitcher, T., Chuenpagdee, R. (eds) (1995) Harvesting krill: ecological impact, assessment, products and markets. Fisheries Centre Research Reports 3(3):82pp. Pauly, D., Christensen, V., Haggan, N. (eds) (1996) Mass- balance models of north-eastern Pacific ecosystems. Fisheries Centre Research Reports 4(1):131pp.   Pitcher, T. (ed) (1996) Reinventing fisheries management. Fisheries Centre Research Reports 4(2):84pp.  Pitcher, T. (ed) (1997) The design & monitoring of marine reserves. Fisheries Centre Research Reports 5(1), 47pp. Dalsgaard, J., Pauly, D. (1997) Preliminary mass-balance models of north-eastern Pacific ecosystems. Fisheries Centre Research Reports 5(2), 33pp. Pitcher, T., Watson, R., Courtney, A., Pauly, D. (1998) Assessment of Hong Kong’s inshore fishery resources. Fisheries Centre Research Reports 6(1):148pp. Pauly, D., Weingartner, G. (eds) (1998) Use of Ecopath with Ecosim to evaluate strategies for sustainable exploitation of multi-species resources. Fisheries Centre Research Reports 6(2):49pp. Vasconcellos, M., Preikshot, D. (eds) (1998) Graduate student symposium on fish population dynamics and management. Fisheries Centre Research Reports 6(3):40pp. Okey, T., Pauly, D. (eds) (1998) Trophic mass-balance model of Alaska’s Prince William Sound ecosystem for the post-spill period 1994-1996. Fisheries Centre Research Reports 6(4):143pp. [n/a; superceded by 7(4), 1999] Pauly, D., Pitcher, T., Preikshot, D., Hearne, J. (eds) (1998) Back to the future: reconstructing the Strait of Georgia ecosystem. Fisheries Centre Research Reports 6(5):99pp. Bonfil, R., Munro, G., Sumaila, U., Valtysson, H., Wright, M., Pitcher, T., Preikshot, D., Haggan, N., Pauly, D. (eds) (1998) Distant water fleets: an ecological, economic and social assessment. Fisheries Centre Research Reports 6(6):111pp. Trites, A., Livingston, P., Mackinson, S., Vasconcellos, M., Springer, A., Pauly, D. (1999) Ecosystem change and the decline of marine mammals in the eastern Bering Sea. Fisheries Centre Research Reports 7(1):106pp. Pitcher, T. (ed.) (1999) Evaluating the benefits of recreational Fisheries. Fisheries Centre Research Reports 7(2):169pp. Haggan, N., Beattie, A., Pauly, D. (eds) (1999) Back to the future: reconstructing the Hecate Strait ecosystem. Fisheries Centre Research Reports 7(3):65pp. Okey, T., Pauly, D. (eds) (1999) A trophic mass-balance model of Alaska's Prince William Sound ecosystem for the post-spill period 1994-1996, 2nd Edition. Fisheries Centre Research Reports 7(4):137pp.  Guénette, S., Chuenpagdee, R., Jones, R. (2000) Marine protected areas with an emphasis on local communities and indigenous peoples: a review. Fisheries Centre Research Reports 8(1):56pp.  Pauly, D., Pitcher, T. (eds) (2000) Methods for evaluating the impacts of fisheries on North Atlantic ecosystems. Fisheries Centre Research Reports 8(2):195pp. Hunter, A., Trites, A. (2001) An annotated bibliography of scientific literature (1751-2000) pertaining to Steller sea lions (Eumetopias jubatus) in Alaska. Fisheries Centre Research Reports 9(1):45pp.  Watson, R., Pang, L., Pauly, D. (2001) The marine fisheries of China: development and reported catches. Fisheries Centre Research Reports 9(2):50pp. Zeller, D., Watson, R., Pauly, D. (eds) (2001) Fisheries impacts on North Atlantic ecosystems: catch, effort and national/regional data sets. Fisheries Centre Research Reports 9(3):254pp. Guénette, S., Christensen, V., Pauly, D. (eds) (2001) Fisheries Impacts on North Atlantic ecosystems: models and analyses. Fisheries Centre Research Reports 9(4):344pp. Pitcher, T., Sumaila, R., Pauly, D. (eds) (2001) Fisheries impacts on North Atlantic ecosystems: evaluations and policy exploration. (2001) Fisheries Centre Research Reports 9(5):94pp. Preikshot, D., Beattie, A. (eds) (2001) Fishing for answers: analysis of ecosystem dynamics, trophic shifts and salmonid population changes in Puget Sound, 1970-1999. Fisheries Centre Research Reports 9(6):35pp.  Tyedmers, P., Ward, B. (2001) A review of the impacts of climate change on BC's freshwater fish resources and possible management responses. Fisheries Centre Research Reports 9(7):12pp.   Sumaila, U., Alder, J. (eds) (2001) Economics of marine protected areas. Fisheries Centre Research Reports 9(8):250pp.  Pitcher, T., Vasconcellos, M., Heymans, S., Brignall, C., Haggan, N. (eds) (2002) Information supporting past and present ecosystem models of northern British Columbia and the Newfoundland shelf. Fisheries Centre Research Reports 10(1):116pp. Pitcher, T., Cochrane, K. (2002) The use of ecosystem models to investigate multispecies management strategies for capture fisheries. Fisheries Centre Research Reports 10(2):156pp.  Pitcher, T., Buchary, E., Trujillo, P. (eds) (2002) Spatial simulations of Hong Kong's marine ecosystem: ecological and economic forecasting of marine protected areas with human- made reefs. Fisheries Centre Research Reports 10(3):169pp. Ainsworth, C., Heymans, J., Pitcher, T., Vasconcellos, M. (2002) Ecosystem models of northern British Columbia for the time periods 2000, 1950, 1900 & 1750. Fisheries Centre Research Reports 10(4):41pp.  Pitcher, T., Heymans, J., Vasconcellos, J. (eds) (2002) Ecosystem models of Newfoundland for the time periods 1995, 1985, 1900 and 1450. Fisheries Centre Research Reports 10(5):74pp.  Haggan, N., Brignall, C., Peacock, B., Daniel, R. (eds) (2002) Education for aboriginal fisheries science and ecosystem management. Fisheries Centre Research Reports 10(6):49pp.  Pitcher, T., Power, M., Wood, L. (eds) (2002) Restoring the past to salvage the future: report on a community participation workshop in Prince Rupert, BC.  Fisheries Centre Research Reports 10(7): 55pp. Pauly, D., Palomares, M.L.D. (eds) (2002) Production systems in fishery management. Fisheries Centre Research Reports 10(8):28pp. Haggan, N., Brignall, C., Wood, L. (eds) (2003) Putting fishers’ knowledge to work conference proceedings.  Fisheries Centre Research Reports 11(1):504pp. Seaman, W., Smiley, B., Pitcher, T., Wood, L. (eds) (2003) Research and monitoring of marine reefs using volunteer divers. Fisheries Centre Research Reports 11(2):107pp. Sumaila, U.R. (ed) (2003) Three essays on the economics of fishing. Fisheries Centre Research reports 11(3):33pp. Burek, K.A., Gulland, F.M.D., Sheffield, G., Keyes, E., Spraker, T.R., Smith, A.W., Skilling, D.E., Evermann, J., Stott, J.L., Trites, A.W. (2003) Disease agents in Steller sea lions in Alaska: a review and analysis of serology data from 1975-2000. Fisheries Centre Research Reports 11(4):26pp. Heymans, J.J. (ed) (2003) Ecosystem models of Newfoundland and southeastern Labrador: additional information and analyses for ‘Back to the Future’. Fisheries Centre Research Reports 11(5):79pp. Zeller, D., Booth, S., Mohammed, E., Pauly D. (eds) (2003) From Mexico to Brazil: central Atlantic fisheries catch trends and ecosystem models. Fisheries Centre Research Reports 11(6):264pp.  Pitcher, T. (ed) (2004) Back to the future: advances in methodology for modelling and evaluating past ecosystems as future policy goals.  Fisheries Centre Research Reports 12(1):158pp. Kavanagh, P., Pitcher, T. (2004) Implementing Microsoft Excel software for RAPFISH: a technique for the rapid appraisal of fisheries status.  Fisheries Centre Research Reports 12(2):75pp. Stanford, R., Pitcher, T. (2004) Ecosystem simulations of the English Channel: climate and trade-offs.  Fisheries Centre Research Reports 12(3):103pp.  Morato, T., Pauly, D. (2004) Seamounts: biodiversity and fisheries.  Fisheries Centre Research Reports 12(5):78pp.  Watson, R., Hoshino, E., Beblow, J., Revenga, C., Kura, Y., Kitchingman, A. (2004) Fishing gear associated with global marine catches. Fisheries Centre Research Reports 12(6):32pp. Palomares, M.L.D., Pauly, D. (2004) West African marine ecosystems: models and fisheries impacts.  Fisheries Centre Research Reports 12(7):209pp. Guénette, S., Christensen, V.  (eds) (2005) Food web models and data for studying fisheries and environmental impacts on Eastern Pacific ecosystems.  Fisheries Centre Research Reports 13(1):237pp. Chuenpagdee, R., Bundy, A. (eds) (2005)  Innovation and outlook in fisheries: an assessment of research presented at the 4th World Fisheries Congress.  Fisheries Centre Research Reports 13(2):113pp. Sumaila, U.R., Volpe, J., Liu, Y. (2005) Ecological and economic impact assessment of sablefish aquaculture in British Columbia.  Fisheries Centre Research Reports 13(3):33pp. Haggan, N., Pitcher, Tony J. (eds) (2005) Ecosystem simulation models of Scotland’s west coast and sea lochs. Fisheries Centre Research Reports, Vol. 13(4):67pp. Bhathal, B. (2005) Historical reconstruction of Indian marine fisheries catches, 1950-2000, as a basis for testing the ‘Marine Trophic Index’.  Fisheries Centre Research Reports 13(5):122pp. Miller, E.H., Trites, A.W., Wiig, Ø. (2005) International survey of scientific collections of Steller sea lions. Fisheries Centre Research Reports 13(6):69pp. Palomares, M.L.D., Pruvost, P., Pitcher, T.J., Pauly, D. (eds) (2005) Modeling Antarctic marine ecosystems. Fisheries Centre Research Reports 13(7):98pp. Sumaila, U. R., Marsden, A.D. (eds) (2006) Proceedings of the 2005 North American Association of Fisheries Economists Forum. Fisheries Centre Research Reports 14(1): 220pp.  Pitcher , T.J. , Kalikoski, D. (2006) Evaluations of compliance with the UN Code of Conduct for Responsible Fisheries for the top 32 fishing countries: Volume 1. Fisheries Centre Research Reports 14(2a): (in press).  Pitcher , T.J., Kalikoski, D. (2006) Evaluations of compliance with the UN Code of Conduct for Responsible Fisheries for the top 32 fishing countries: Volume 2. Fisheries Centre Research Reports 14(2b): (in press).  Alder, J., Pauly, D. (eds) (2006) On the multiple uses of forage fish: from ecosystems to markets. Fisheries Centre Research Reports 14(3): 120 pp.  Palomares, M.L.D., Stergiou, K.I., Pauly, D. (eds) (2006) Fishes in databases and ecosystems. Proceedings of the 4th annual FishBase mini-Symposium, Fisheries Centre, UBC, Vancouver, 6 September 2006. Fisheries Centre Research Reports 14(4): 95 pp.       Fisheries Centre Research Reports publish results of research work carried out, or workshops held, at the UBC Fisheries Centre. The series focuses on multidisciplinary problems in fisheries management and aims to provide a synoptic overview of the foundations, themes and prospects of current research. Fisheries Centre Research Reports are distributed to appropriate workshop participants or project partners and are recorded in Aquatic Sciences and Fisheries Abstracts.  All reports are available for free downloading from the Fisheries Centre website.  Where available, back issues of research reports may be obtained on request for the cost of shipping.  For further information, please contact: Fisheries Centre, University of British Columbia Phone:   604 822-2731; Fax:   604 822-8934 2202 Main Mall, Vancouver, BC  V6T 1Z4  Canada E-mail:    office@fisheries.ubc.ca  Website:  www.fisheries.ubc.ca

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.37052.1-0074760/manifest

Comment

Related Items