UBC Faculty Research and Publications

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

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-FCRR_2006_14-4.pdf [ 5.35MB ]
Metadata
JSON: 52383-1.0074760.json
JSON-LD: 52383-1.0074760-ld.json
RDF/XML (Pretty): 52383-1.0074760-rdf.xml
RDF/JSON: 52383-1.0074760-rdf.json
Turtle: 52383-1.0074760-turtle.txt
N-Triples: 52383-1.0074760-rdf-ntriples.txt
Original Record: 52383-1.0074760-source.json
Full Text
52383-1.0074760-fulltext.txt
Citation
52383-1.0074760.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  2  Scientific impact of FishBase, Stergiou, K.I., Tsikliras, A.C.  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 19952006, 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  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 citations and implies an annual mean rate of 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).  160 y = 1.2e^0.47x 140  2  n=11, r = 0.96, p<0.01  120 Number of citations  FishBase was not subjected to formal peerreview 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).  100 80 60 40 20 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).  The number of Scopus citations per year increased 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  4  Scientific impact of FishBase, Stergiou, K.I., Tsikliras, A.C.  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). Number of citations 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 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).  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 journals (there were also 5 citations in conference proceedings). Thus, more than half of the citations to FishBase and of the sources of such citations 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. FishBase was also cited in 48 aquatic and 18 general books as well as in three general (e.g., Lévêque and Mounolou, 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; 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, 18, 25%  aquatic books, 48, 66%  Figure 3 Number of citations (left) and percentages (right) of citations in books (source: Google Books, accessed 7 July 2006).  To sum up, the analysis presented here shows that FishBase is also very successful in terms of scientific impact, when the latter is evaluated based on ‘traditional’ bibliometric indices. Naturally, the number of citations to FishBase is much larger than the one presented here, since it is most probably cited in many other scholarly publication types not covered by Scopus, some of which are covered by Google Scholar (e.g., theses, technical reports, and technical papers, if online). In addition, the analyses of citations 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 part of the public knowledge infrastructure that people from all walks of life refer to it (Cornelia Nauen, European Commission, Brussels, pers. comm.).  ACKNOWLEDGEMENT The authors would like to thank Dr Rainer Froese (University of Kiel, Germany) for his comments and suggestions.  REFERENCES Bauer, H.H., 1992. Scientific Literacy and the Myth of the Scientific Method. University of Illinois Press. 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. Christensen, V., Guénette, S., Heymans, J.J., Walters, C., Watson, R., Zeller, D., Pauly, D., 2003. Hundred-year 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. CIESM, 2003. Mediterranean Biological Time Series. CIESM Workshop Monograph Series 22. CIESM publications, 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 yield per recruit in fishes, with a simple method to evaluate length frequency data. J. Fish Biol. 56, 758-773. Froese, R., Binohlan, C., 2003. Simple methods to obtain preliminary growth estimates for fishes. J. Appl. Ichthyol. 19, 376-379.  6  Scientific impact of FishBase, Stergiou, K.I., Tsikliras, A.C.  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 as compared with other ecosystems. Belg. J. Zool. 135 (2), 139-143. Garfield, 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.  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), FUEMNUPÉLIA (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).  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].  1  8  Common names of Brazilian freshwater fishes, Freire, K.  RESULTS AND DISCUSSION 400 300 Number of species  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.  200 100  Trachelyopterus galeatus  0 0  10  20  30  40  Number of common names per species  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)  Hypostomus derbyi (0)  Hypostomus vaillanti (0)  Ancistrus formoso (0)  Hypostomus fluviatilis (0)  Hypostomus variipictus (0)  Ancistrus stigmaticus (0)  Hypostomus garmani (0)  Hypostomus wuchereri (0)  Harttia rhombocephala (0)  Hypostomus gomesi (0)  Lithodoras dorsalis (6)  Hemiancistrus chlorostictus (1)  Hypostomus jaguribensis (0)  Loricaria parnahybae (0)  Hemiodontichthys acipenserinus (0)  Hypostomus lima (0)  Megalechis thoracata (9)  Hemipsilichthys cameroni (0)  Hypostomus luetkeni (0)  Hemipsilichthys garbei (0)  Hypostomus macrops (0)  Hemipsilichthys gobio (8)  Hypostomus nudiventris (0)  Hemipsilichthys mutuca (0)  Hypostomus obtusirostris (0)  Pterygoplichthys anisiti (0)  Hoplosternum littorale (6)  Hypostomus papariae (0)  Pterygoplichthys etentaculatus (10)  Hypostomus alatus (2)  Hypostomus plecostomus (12)  Rineloricaria cubataonis (0)  Hypostomus albopunctatus (1)  Hypostomus pusarum (0)  Rineloricaria kronei (0)  Hypostomus auroguttatus (0)  Hypostomus regani (1)  Rineloricaria lanceolata (1)  Hypostomus carvalhoi (0)  Hypostomus scabriceps (13)  Rineloricaria latirostris (1)  Hypostomus commersonii (18)  Hypostomus unae (0)  Rineloricaria lima (10)  Pogonopomoides parahybae (0) Pseudancistrus luderwaldti (0)  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).  Cascudo Ancistrus  Cascudo de espinhos  Bacu (4)  Acarí lima (0)  Bacu de pedra  Acary lima (0)  Hemipsilichthys  Anhã (3)  Aperta galha (0)  Cascudinho (7)  Barbado (2)  Cascudo espinho (1)  Cascudo barbudo (0)  Cascudo barbado (0) Cascudo pardo (0) Cascudo piririca (0)  Cascudo lima (0) Rineloricaria lima Megalechis thoracata  Mãe de anhã (1)  Cascudo asa branca  Camboatá (4) Curite (0) Hoplosternum  Cascudo espada (5)  Tamboatá (1) Tamoatá (3) Tamuatá (1) Acari (4)  Hypostomus alatus Cascudo chitão (0)  Cascudo cinzento (0) Cascudo trepa pau (0) Jaruitaquara (0) Vieja (10 Acarí (4)  Lithodoras  Daqueiro (1) Pacu (9) Vacu (3)  Cascudo chinelo (2)  Couraçado (0)  Atipa (0)  Bacu pedra (0)  Viola (4)  Rineloricaria lanceolata  Acarí juba (0)  Acari  Acarí roncador (1)  Acarí amarelo (0)  Acarijuba (0)  amarelo  Pterygoplichthys  Carinhanha (0)  Coroncho (0)  Cascudo  Panaque (0)  amarelo  Acarí pedral (0)  Vacarí (0)  Armadilho (0)  Yau urá (0)  Bode (1)  Hypostomus plecostomus  Pirá tatu (0)  Bode de Igarapé (0)  Bacu puá (0)  Bodó (2)  Cascudo cachichô (0)  Guacari (0) Hypostomus  Cascudo comum (1) Cascudo prêto (0)  Jaru itaquara (0)  Guacarí (1)  Jaruitacoara (0)  Iarú urá (0) Hypostomus  Cascudo preto (3)  Hypostomus scabriceps  Uacarí (0) Uará urá (0) Yaru itacura (0)  Cascudo barata (0)  Hypostomus regani  Cascudo viola (6)  Rineloricaria  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 Santo 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 (Brachyplatystoma rousseauxii and Pellona castelnaeana) according to the National Plan for the Development of Recreational Fisheries in Brazil (PNDPA, 2006). If one is to compare catches originating from commercial and recreational fisheries, this difference should be considered. Piramutaba represents Brachyplatystoma vaillantii according to IBAMA (2005) and is considered an overexploited resource (Anonymous, 1999; Ruffino, 2005). Even though this is not a unique association (as this common name was also related to Platysilurus mucosus in Acre state; Begossi et al.,  10  Common names of Brazilian freshwater fishes, Freire, K.  1999), it does not cause any problem to catch statistics due to the inexistence of catch records for 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 Portuguese (40%) and less influence from Ameridian languages (24%). This may be due to Amerindian influence being stronger in the interior of the country.  Table 1 Descriptors used in the core, and in the first and second modifiers of the common names of Brazilian freshwater and marine fishes. Descriptor Freshwater Marine Primary lexeme 857 1793 Morphology 461 557 Color pattern 408 444 Inanimate object 297 314 Non-fish animal 206 419 Behaviour 202 158 Modification for size 199 232 Plant 124 162 Other 106 228 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  Most of the descriptors used in the names of Brazilian freshwater and marine fishes represent primary lexemes or are associated with the morphology or color pattern of the species (Table 1). A test of independence between the descriptor and the environment (fresh and marine water) was performed and the null hypothesis of independence was rejected (χ2 = 212.8; df = 14). Descriptors related to 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.  Curimatã  30  3  Cat ch (tonnes; 10 )  35  25  Piramutaba  20 15  Dourada  10 5 0 2000  2001  2002  2003  2004  Figure 4 Catches from Brazilian continental waters for the three main species (2000-2004).  The spatial coverage of the name database for 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. For the remaining 19 states, there are less than 50 names and none were recorded for Amapá, Roraima, Rondônia, Mato Grosso do Sul, and Espírito Santo. This is the first database of names ever compiled for the analysis of Brazilian freshwater species. Ten other sources have been selected to be included in the database in the next future.  ACKNOWLEDGEMENTS  I would like to thank Juarez Rodrigues for encoding part of the common names compiled here; Carolina Minte-Vera, Fernando Becker, Juarez Rodrigues, Maurício Cetra, Paulo Chaves, and Projeto PróVárzea/IBAMA for providing lists of names and books, including unpublished materials.  REFERENCES Anonymous, 1999. V reunião do grupo permanente de estudos sobre a piramutaba. Série Estudos Pesca, n. 26. Becker, F. G., Grosser, K.M., Milani, P.C.C., Braun, A.S., 2006. Peixes. Chapter 19. In: Becker, F.G.; Ramos, R. A. and 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. Begossi, A., Garavello, J.C., 1990. Notes on the ethnoichthyology of fishermen from the Tocantins River (Brazil). Acta Amazonica 20(único), 341-351.  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. CBPDS, 2004. Confederação Brasileira de Pesca e Desportos Aquáticos [www.antares.com.br/cbpds/html/nova.htm], accessed in August 20, 2006. Ferreira, A.B.H., 999. Dicionário Aurélio Eletrônico - Século XXI. Rio de Janeiro, Brasil, Lexicon Informática Ltda. Freire, K.M.F., Pauly, D., 2003. What’s in there? Common names of Brazilian marine fishes. In: Haggan, N., Brignall, 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 marine fishes and its effect on catch statistics. J. Ethnobiol. 25(2), 279-296. FUEM-NUPÉLIA/SUREHMA/ITAIPU BINACIONAL, 1987. Relatório anual do projeto ‘Ictiofauna e Biologia Pesqueira’. Nupélia, Universidade Estadual de Maringá. Maringá, Universidade Estadual de Maringá. Godinho, H.P., Godinho, A. L., 2003. Águas, Peixes e Pescadores do São Francisco das Minas Gerais. Belo Horizonte, PUC Minas. Godoy, M.P. , 1987. Peixes do Estado de Santa Catarina. Florianópolis, Brazil, Editora da UFSC. 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. IBAMA, 2003. Estatística da Pesca do Brasil. 2001. Grandes regiões e unidades da federação. Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis/CEPENE, Tamandaré-PE. IBAMA, 2004a. Instrução normativa no. 5 de 21 de maio de 2004. Anexo 1. 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, desenhos e aquarelas, 160 p.; V3 Florence, desenhos e aquarelas). Mutti Pedreira, J. M. , 1971, Dicionário de Peixes de Couro do Brasil. Belém, SUDAM, Assessoria de Programação e Coordenação Divisão de Documentação. Nelson, J.S., Crossman, E.J., Espinosa-Pérez, H., Findley, L.T., Gilbert, C.R., Lea, R.N., Williams, J.D., 2004. Common and Scientific Names of Fishes from the United States, Canada, and Mexico. 6th ed., American Fisheries Society, Special Publication 29, Bethesda, Maryland. 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. Palomares, 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.  12  Prominence trend in maximum lengths, Bailly, N.  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) Nominal subspecies Subspecies  29,023 193 281  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 Table 2 Statistics on maximum lengths (Lmax) in FishBase (as of 18 each centimetre rounded to the nearest Aug. 2006). Taxa with Lmax Number All taxa With Lmax 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.  For one of or both sexes For both sexes For one of sexes For male/unsexed only For female only Missing for both sexes Missing for male/unsexed Missing for female Where Female > Male/unsexed >= 100 cm < 100 cm >= 80 cm < 80 cm  24,804 1,082 23,722 23,395 327 4,693 5,020 28,088 544  n = 29,497 % 84.1 3.7 80.4 79.3 1.1 15.9 17.0 95.2 1.8  n = 24,804 % 100.0 4.4 95.6 94.3 1.3 18.9 20.2 113.2 2.2  1,136 23,668 1,568 23,236  3.9 80.2 5.3 78.8  4.6 95.4 6.3 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).  14  Prominence trend in maximum lengths, Bailly, N.  Number of taxa  1500  1000  500  100  Figure 1 Number of taxa (species and subspecies) for each centimetre of maximum length from 0 to 100 cm.  300  Figure 2 Number of taxa (species and subspecies) per each centimetre of maximum length from 80 to 300 cm.  550  Figure 3 Number of taxa (species and subspecies) per each centimetre of maximum length from 300 to 550 cm.  0 0  10  20  30  40  50  60  70  80  90  Maximal lengths (cm) 100 cm: 217 taxa  Number of taxa  100  50  0 80  100  120  140  160  180  200  220  240  260  280  Maximal lengths (cm) 7  300 cm: 27  6  Number of taxa  5 4 3 2 1 0 300  350  400  450  Maximal lengths (cm)  500  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D.  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 are dominant followed by digits 5, 1 and 2, and digit 9 is definitely the least-represented digit. The signal fades progressively when moving towards the upper tens but the same tendency remains strong for tens, digits 5 and 9, and to a lesser degree for digits 1 and 2; digits 6, 3, 4, 8 and 7 are more variable.  15  Table 3 Mean rank of digits 0-9 among tens 10-70 when ordered down by number of species with the same unit 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 9.7 0.4 9 10  DISCUSSION Fitting laws Albers and Albers (1983): Prominence in the decimal system The results shown in 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 {s*10i | s∈{1, 1.5, 2, 3, 5, 7} | i∈ℵ} [the notation is ours] (Hertwig et al., 1999), for numbers preferentially chosen by persons for estimations (of prices in Albers’ work). Indeed 5, 10, 15, 20, 30, 50, 100, 150, 200, 300, 500 are prominent in our results (Figure 4). Nevertheless, 1 and 2, and 7 and 70, are not prominent. Two complementary explanations can be suggested: the theory is about estimations without measurements, and does not completely apply to the latter case (for 7 and 70); and the maximum length distribution among fishes (approximately a Poisson distribution) strongly constrains the results (in particular 1 and 2). 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 a strong tendency of authors to round to the highest ten or hundred, which confirms the strong constraint of the data distribution. For the second digit, the results are similar: digits 0, 5, 1, and 2 between 11 and 99; and 120, 250, 310, 320, 350, 430, 450 beyond (not all shown on 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.  Table 4 Taxa numbers by ranges of maximum lengths in tens from 0 to 99 cm, and by hundreds from 100 to 999 cm. Range Number Range Number by tens of taxa by hundreds of taxa 0-9 8,305 0-99 23,668 10-19 6,637 100-199 868 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 60-69 604 600-699 6 70-79 435 700-799 4 80-89 231 800-899 1 90-99 198 900-999 1  Benford (1938): first digit law The distributions of the first digit between 0 and 99, and 100 and 999, follow Benford’s law (1938), which predicts a decreasing number of species grouped by tens, and by hundreds (Table 4), as a logarithmic distribution p(i) = log10(1+1/i) for i = 1 to 9 (our number fits the logarithmic tendency, but not exactly the values computed from the equation). The Poisson distribution of our data explains why the distribution of the first digit fits with Benford’s law. A more rigorous demonstration can be found in Hill (1996).  16  Prominence trend in maximum lengths, Bailly, N.  5 10 15 20  3  30  Number of taxa [log(n +1)]  40  50 60  100  2  120  150  200  250  300  1  500  400 430  0  100  200  300  400  500  Maximal lengths (cm) 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*10i | 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 encoding in FishBase. Data are encoded as published. However, values with tenth of millimetres for lengths beyond 50 cm are rounded to the nearest millimetre. As mentioned above, authors tend to round at the highest tens or hundreds. In which publications? In general, in publications that first describe a species, the length is not rounded because only few specimens are usually measured, and the maximum length is extracted by the FishBase encoders and 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 maximum length reported in FishBase). 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 starts to appear for species described from 1990 onwards: 25, 30, 40, 50, 70, 80, and 120, although 6 is still the highest value). For species described before 2000, the FishBase Team used mainly large syntheses, such as the FAO catalogues, regional accounts, identification sheets, regional faunas and checklists and revisions to a lesser 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 involved unit conversion (inches in cm, for instance). Conclusions We demonstrated that, overall, FishBase may over-estimate maximum lengths. In synthetic works, it is usually not possible to track back the origin of the information used by authors. Unfortunately, with overexploitation of natural resources, the lengths recorded in nature tend to decrease, and the reported maximum lengths cannot be validated in the field. Species, museum collection, and document databases, 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 number of taxa: n-2, n-1, n, n+1 for a given maximum length. This however generates other biases. Another more complex solution is to fit a distribution analytically, either a Poisson distribution, or by ignoring lengths below 5 cm or by summing 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.  18  Age and growth of Mediterranean fishes, Stergiou, K.I. et al.  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  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].  1  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).  France  Italy  Croatia Yugoslavia Greece  Spain Turkey  Algeria  Syria Lebanon  Tunisia  Marocco  Egypt  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).  Number of stocks 0  5  10  15  20  Pagellus erythrinus Engraulis encrasicolus  Species  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).  Merluccius merluccius Mullus barbatus Spicara smaris Diplodus annularis Boops boops  Figure 2 The most intensively-studied species in the Mediterranean Sea.  20  Age and growth of Mediterranean fishes, Stergiou, K.I. et al.  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 %).  3.0  Asymptotic length (L∞, log, cm)  2.5  2.0  1.5  1.0  0.5  y = 0.99x + 0.075 2 r = 0.93 n=265  0.0 Maximum length ranged between 4.2 cm 0.0 0.5 1.0 1.5 2.0 2.5 (Aphia minuta, Balearic Islands, Spain) and Maximum recorded length (Lmax, log, cm) 215 cm (Xiphias gladius, southern Aegean Sea, Greece) and had a mean value of Figure 3 The relationship between asymptotic length (L∞, log, 37.6 cm, while L∞ ranged between 4.7 and cm) and maximum recorded length (Lmax, log, cm) for 265 220.1 cm for the same species Mediterranean marine fish stocks. (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.  Growth coefficient (K, log, 1/y)  1.0 0.5  Coryphaena hippurus  0.0  Lepidopus caudatus  -0.5 -1.0  Macrouridae  -1.5 -2.0 0.0  0.5  1.0  1.5  2.0  2.5  Asymptotic length (L∞, log, cm)  Figure 4 The relationship between the growth coefficient (K, log, year-1) and asymptotic length (L∞, log, cm) for 340 Mediterranean marine fish stocks. The outliers are circled.  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 Lepidopus caudatus. Finally, the theoretical age at zero length, t0, ranged 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.  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.  22  Trophic levels of North Aegean sea fishes, Karachle, P.K., Stergiou, K.I.  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)  24  Trophic levels of North Aegean sea fishes, Karachle, P.K., Stergiou, K.I.  Table 1 (continued) Species  Monochirus hispidus 1,2,3 Mullus surmuletus Oblada melanura 1 Pagellus acarne Pagellus bogaraveo Pagellus erythrinus Pagrus pagrus Parablennius gattorugine Phycis blennoides 1 Pomatomus saltatrix 1,2 Raja clavata 1,2 Raja miraletus 1 Raja radula Sardina pilchardus Sardinella aurita Sarpa salpa 1 Sciaena umbra 1 Scomber japonicus 1,2 Scomber scombrus Scorpaena notata1 Scorpaena porcus Scyliorhinus canicula 1 Serranus cabrilla Serranus hepatus Serranus scriba Sphyraena sphyraena 1,2 Spicara maena Spicara smaris Spondyliosoma cantharus 1 Symphodus tinca Symphurus nigrescens 1 Torpedo marmorata 1 Trachinus draco 1 Trachurus mediterraneus Trachurus trachurus Trisopterus minutes Uranoscopus scaber Xyrichtys novacula 1  N 24 55 56 63 72 59 10 4 20 6 7 3 3 752 230 25 11 371 204 42 96 34 34 99 81 104 282 118 82 221 10 118 25 627 133 167 70 12  Length range 9.2-12.8 9.1-23.1 12.6-22.7 10.5-19.2 9.3-23.1 8.4-16.4 10.2-15.5 13.4-17.9 8.1-37.4 13.1-18.5 25.6-46.5 22.6-33.9 21.8-32.0 7.6-16.7 8.4-23.9 11.7-19.5 12.2-16.0 8.8-26.8 13.3-27.4 8.3-17.8 8.2-26.4 24.1-45.1 9.5-23.1 5.7-13.1 10.6-23.6 21.6-45.1 9.0-20.2 7.0-18.5 9.7-14.0 11.1-22.0 6.4-11.9 8.8-37.3 15.0-30.5 7.0-25.8 6.3-3.9 5.7-24.5 8.7-26.9 12.3-17.1  TROPH 3.19±0.32 3.19±0.37 3.11±0.42 3.84±0.55 4.43±0.76 3.30±0.39 3.36±0.34 2.11±0.09 3.55±0.59 4.50±0.80 3.90±0.67 3.82±0.54 3.97±0.69 3.14±0.29 3.20±0.32 2.00±0.00 3.53±0.54 3.99±0.57 4.37±0.54 3.60±0.62 3.90±0.69 4.41±0.58 3.90±0.67 3.77±0.63 3.94±0.66 4.30±0.46 3.24±0.34 3.49±0.46 3.41±0.46 2.95±0.25 3.35±0.51 4.39±0.67 4.19±0.66 4.01±0.64 3.58±0.50 4.13±0.64 4.43±0.75 3.37±0.51  TROPHFB 3.40±0.51 3.00±0.12 3.50±0.45 3.50±0.46 3.40±0.47 3.70±0.61 2.90±0.29 3.70±0.58 4.50±0.55 3.80±0.59 3.80±0.74 3.70±0.54 2.80±0.23 3.00±0.00 2.00±0.00 3.70±0.65 3.10±0.43 3.70±0.56 3.50±0.50 3.90±0.65 3.70±0.55 3.40±0.47 3.50±0.56 3.80±0. 62 4.00±0.51 4.20±0.70 3.00±0.04 3.30±0.43 3.10±0.45 3.30±0.43 4.50±0.80 4.20±0.71 3.60±0.58 3.60±0.58 3.80±0.53 4.40±0.70 3.10±0.32  no TROPH estimates available from the Aegean Sea; no TROPH estimates available from the Mediterranean; 3 no TROPH estimates available in FishBase (date: 4/7/2005). 1  2  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.  (b)  (a)  42.7 (32)  4.50  TROPH TROPH  number of %% number ofspecies species  50.0  25  25.0  8.0 (6)  9.3 (7)  3.50  2.50  6.7 (5)  TROPH = 1.2392+0.6459TROPHFB N=75, R2 = 0.54, SEb=0.09, p<0.01  0.0  1.50 0-0.2  0.2-0.4  0.4-0.6  0.6-0.8  1.50  0.8-1.0  2.50  3.50  4.50  TROPHFB  TROPHFB |TROPH-TROPH TROPH-TROPH FB FB| 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 relationship between TROPH and TROPHFB for species for which regional TROPH values are available could be used to refine the general TROPHFB values for species for which no regional estimates are available (e.g., rare species). To generate more accurate predictions, the number of species for which regional TROPHs are available should be increased.  1.20  |TROPH-TROPHFB|  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 reported diet in FishBase, from which TROPH is estimated, is composed mainly of zooplankton (Pinnegar and Polunin, 2000) and bony fish (Khoury, 1984).  N=75, R2 = 0. 03, p p=0.16  0.80  0.40  0.00 0  1  2  3  Number o off stomachs (log) 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).  26  Trophic levels of North Aegean sea fishes, Karachle, P.K., Stergiou, K.I.  REFERENCES Bell, J.D., Harmelin-Vivien, M.L., 1983. Fish fauna of French Mediterranean Posidonia oceanica seagrass meadows. 2. Feeding habits. Tethys 11, 1-14. CIESM, 2000. Fishing down the Mediterranean (www.ciesm.org/online/monographs/Corfou.html).  food  webs?  CIESM  Workshop  Series,  12.  Froese, R., Pauly, D. (eds.), 2005. FishBase. World Wide Web electronic publication. URL: www.fishbase.org. Froese, R., Garthe, S., Piatkowski, U., Pauly, D., 2005. Trophic signatures of marine organisms in the Mediterranean as compared with other ecosystems. Belg. J. Zool. 135 (2), 139-143. 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, 411429. Khoury, C., 1984. Ethologies alimentaires de quelques espèces de poisons de l'herbier de Posidonies du Parc National de Port-Cros. In: Boudouresque C.F., Jeudy de Grissac A. and Olivier J. (eds.), International Workshop Posidonia oceanica Beds, GIS Posidonie Publications, France 1, pp. 335-347 Lindeman, R.E., 1942. Trophic-dynamic aspect of ecology. Ecology 23, 399-418. Pauly, D., Christensen, V., 1995. Primary production required to sustain global fisheries. Nature 374, 255–257. Pauly, D., Watson, R., 2005. Background and interpretation of the ‘Marine Trophic Index’ as a measure of biodiversity. Phil. Trans. R. Soc. (Ser. B) 360, 415-423. 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, 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.  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. 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]. 1  28  Distribution of commercial fishes and invertebrates, Close, C. et al.  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  18  18 27 21  67 61  37  61  31 34 77  71  71  51  87  41  57  47  81  81 58  FAO #  FAO Name  88  48  88  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  34  Atlantic, East Central  58  Antarctic (South Indian)  71  Pacific, West Central  37  Mediterranean & Black Sea  67  Pacific, North East  81  Pacific, South West  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 that the United Nations Food and Agriculture Organization (FAO) uses for statistical reporting purposes.  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 semienclosed 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 notably 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.  30  Distribution of commercial fishes and invertebrates, Close, C. et al.  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 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.  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.  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 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, defined as “[t]he depth range (in 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 identification sheets, and guides to the commercial species of various countries or regions), or online sources. One of these sources was OBIS where, in some cases, the deepest record was taken to estimate a taxon’s maximum depth. Where no information was available, the depth range of a similar species was applied.  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.  A further refinement of a taxon’s depth range is its relative abundance within the water 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. In this context, it is assumed that the relative abundance of a taxon in a spatial cell is, in part, determined by the fraction derived from the number of habitats that a taxon associates with in that same cell, and by how 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 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).  32  Distribution of commercial fishes and invertebrates, Close, C. et al.  B Degree of membership  Degree of membership  A 1.0 0.8 0.6 0.4  Medium Medium  Small Small  Large Large  0.2  1.0 0.8 0.6  Low Low  0.4  Moderate Moderate  High High  0.2 0.0  0.0 0  50  100  150  Maximum length (cm)  200  250  0  0.2  0.4  0.6  0.8  1  Versatility  Figure 4 Fuzzy membership functions for the three categories of (A) maximum length and (B) taxon’s versatility. Habitat versatility is defined as ratio of number of habitat types in which a taxon 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 Other habitats – Muddy/sandy/rocky bottom 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 into three categories (Figure 4), and it is assumed that a taxon can associate with one or more categories with different degrees of membership (0 to 1). A higher membership value means a higher ‘probability’ that the taxon is associated with that particularly category. The membership values are defined by a pre-specified membership function for each of the length and versatility categories (Figure 4). For example, the Striped bass (Morone saxatilis) has a maximum length of 200 cm (TL). Thus, based on the defined membership functions (Figure 4A, left), Striped bass has a large body size with a membership of 1 (membership ranges from 0 to 1). There are maximum length estimates for all of the 1231 exploited taxa in the Sea Around Us Project database, obtained from FishBase and other published literature for invertebrates. In this paper, versatility refers to the taxon’s ability to inhabit different habitat types and is defined as the ratio between the number of associated habitats to the total number of defined habitats (Table 1). For instance, based on descriptions in the SPECIES Table of FishBase, Striped bass is associated with estuaries and ‘other habitats’. Given that the total number of defined physical habitats is five (coral reef, estuary, seagrass, seamount, other habitats, while excluding shelf/slope/abyssal and inshore/offshore), the versatility of Striped bass is estimated to be 0.4. Based on the defined membership functions (Figure 4B, right), the versatility of Striped bass is classified as low to medium, with a membership of approximately 0.48 and 0.52 respectively. Determining habitat association Based on qualitative descriptions from the published sources, databases such as FishBase and/or through personal communications from experts, each taxon’s degree of association with different habitats (Table 1) for all exploited taxa in the database was determined. The taxon’s degree of association to each habitat is determined from the qualitative descriptions relating to its density or commonness in the particular habitat (Table 2). As noted above, Striped bass prefers estuaries and also occurs in ‘other habitats’. Thus, the Striped bass received a score of 0.75 for estuaries and 0.5 for ‘other habitats’.  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D.  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 maximum 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 parentheses represents the degree of membership to the categories. In this example, the degree of membership is the minimum memberships of the two predicates. When the same conclusion is reached from different rules, the final degree of membership equals the taxon’s average membership value.  33  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 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 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  The maximum effective distance from the associated habitat can be estimated from the ‘centroid value’ of each conclusion categories, weighted by a taxon’s degree of Figure 5 Maximum effective distance for Striped membership. The centroid values for small, medium and bass (Morone saxatilis) estimated from the large maximum effective distances were defined as 1 km, habitat versatility and maximum length of that 50 km and 100 km, respectively. Thus if, for example, a species (see text). taxon has membership values of 0.2 and 0.5 to small and 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. Estimating relative abundance in a spatial cell Several assumptions are made to simplify the computations. Firstly, it is assumed that the habitat always 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 routine provides an explicit and consistent way to incorporate habitat considerations into distribution ranges.  34  Distribution of commercial fishes and invertebrates, Close, C. et al.  Filter 6: Equatorial submergence The submergence phenomenon was already known to Charles Darwin, who wrote that “we hear from Sir J. Richardson, that Arctic forms of fishes disappear in the seas of Japan & of northern China, are replaced by 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 radiolarian, drew attention to it. In most cases, including those which interest us here, submergence 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. As noted above, there is little information on the depth distribution of most commercial species. As a 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 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 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 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 symmetrical about the Equator. In addition, maximum depths are assumed not to change poleward of 600 N and S. The uneven 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.  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.  36  (a)  Distribution of commercial fishes and invertebrates, Close, C. et al.  (b)  Figure 7 ‘Equatorial submergence’ has the effect of ‘shaving off’ areas from the distribution range of the Warsaw 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 species) via FishBase (click ‘Sea Around Us distributions’ under the ‘Internet sources’ in the Species Table). Most importantly, these distribution ranges will serve as basis for all spatial catch allocation done with the Sea Around Us Project. Therefore, we would be very thankful for feedback, i.e., suggested comments or 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 Charitable Trusts, Philadelphia. We thank the Marine Conservation Biology Institute for the support of 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 ranges.  REFERENCES Alder, J., 2003. Putting the coast in the Sea Around Us Project. The Sea Around Us Newsletter. No. 15, 1-2. Alverson, D.L., Pruter, A.L., Ronholt, L.L., 1964. A Study of Demersal Fishes and Fisheries of the Northeastern Pacific Ocean. Institute of Fisheries, The University of British Columbia, Vancouver. Ekman, S., 1967. Zoogeography of the Sea. Sidgwick & Jackson, London. Froese, 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.  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 Data http://www.ngdc.noaa.gov/mgg/fliers/01mgg04.html). Pauly, D., Chua Thia-Eng., 1988. The overfishing of marine resources: socioeconomic background in Southeast Asia. AMBIO J. Human Environ. 17(3), 200-206. Pauly, D., 2004. Darwin’s Fishes: An Encyclopedia of Ichthyology, Ecology and Evolution. Cambridge University Press, Cambridge. Watson, R., Alder, J., Kitchingman, A., Pauly, D., 2005. Catching some needed attention. Mar. Pol. 29(3), 281-284 Watson, R., Kitchingman, A., Gelchu, A., Pauly, D. 2004. Mapping global fisheries: sharpening our focus. Fish and Fisheries 5: 168-177. 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 and recovering fisheries and marine science data. Mar. Pol. 29, 69-73. Zeller, D., Pauly, D., 2001. Visualisation of standardized life history patterns. Fish and Fisheries 2(4), 344-355.  38  Common names of fish families, Palomares, M.L.D. et al.  A PRELIMINARY LIST OF ENGLISH COMMON NAMES FOR AS YET UNNAMED FISH FAMILIES1  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  Nicolas Bailly  WorldFish Center, Philippine Office, Natural Resources Management, IRRI College, Khush Hall, Los Baños, Laguna 4031, Philippines; Email: n.bailly@cgiar.org  Rainer Froese  Leibniz-Institut für Meereswissenschaften (IfM-GEOMAR) 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 (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 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 question. Multiple common names exist for many organisms, and their standardization and stabilization, e.g., for legal purposes, merely involves a choice from a number of available names (see, e.g., Robins et al., 1991 and Nelson et al., 2004).  Cite as: Palomares, M.L.D., Bailly, N., Froese, R., Pauly, D., 2006. A preliminary list of English common names for as yet unnamed fish families. In: Palomares, M.L.D., Stergiou, K.I., Pauly, D. (eds.), Fishes in Databases and Ecosystems. Fisheries Centre Research Reports 14(4), pp. 38-48. Fisheries Centre, University of British Columbia [ISSN 1198-6727].  1  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 (which shall not be intimately tied to the scientific name) and names intended to honor persons are discouraged in that they are without descriptive value”; 3. “Names shall not violate the tenets of good taste” and are appropriate if they are colorful, romantic, fanciful, metaphorical, distinctive and original and describe structural attributes, color and color patterns, ecological characteristics and geographic distribution; 4. Native names are welcome for adoption as common names. However, commonly-employed names adopted from traditional English usage are given considerable latitude in taxonomic placement; 5. “The duplication of common names of fishes and other organisms should be avoided if possible, but names in general use need not be rejected on this basis alone”; and 6. Common names ending in ‘id’, i.e., adaptation of the scientific name to English, e.g., Adrianichthyidae to adrianichthyids, are not considered as true common names. The most recent edition (2006) of Nelson’s Fishes of the World recognizes 515 families, 100 of which are broken down to 206 non-nominal subfamilies. Of these families, 34 do not have true common names; and of these, 7 have no common names, 20 have names ending in ‘id’, 7 are recategorized taxa. Similarly, the most recent edition (2006) of Eschmeyer’s Catalog of Fishes recognizes 530 families, 72 of which are broken down into 138 non-nominal subfamilies. Of these families, 122 lack common names, while 8 have names ending in ‘id’. Finally, the August 2006 version of FishBase (www.fishbase.org) recognizes 530 families, 79 of which are broken down into 170 non-nominal subfamilies. Of these families, 74 lack common names and 2 have names ending in ‘id’. 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 parent 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 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 the second case is not desirable, e.g, Scombridae with the composite common name of mackerels, tunas, bonitos, which is the list of common names of 3 tribes of one of the two subfamilies. It would then be difficult to choose between ‘tunas’ and ‘mackerels’ to represent the family. The common names aid in keeping trace of the taxa irrespective of the changes in taxonomic category. Comparison of the lists mentioned above left us with 25 families for which no true common names are available and for which we suggest English common names. This list was generated with the intention that it should be subjected to the scrutiny of fish experts, and then encoded into FishBase.  MATERIALS AND METHODS The scientific names of the families were translated using Greek and Latin dictionaries (notably Perseus, the online dictionary at www.perseus.tufts.edu/cgi-bin/resolveform), and/or by consulting the source of the family names, which usually also described and frequently gave the etymology of the type genus of the family. The main source here was the Catalog of Fishes database by W. Eschmeyer as included in FishBase and also accessible through the Internet (see www.calacademy.org).  40  Common names of fish families, Palomares, M.L.D. et al.  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: 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 not perpetuated as common names, e.g., Bathyclupeidae translates 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 In the process of comparing the list of families currently included in FishBase with the lists of scientific 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 which no common names were assigned are included in the lists of suggested common names presented in the following pages. Table 1. Families recognized in FishBase (www.fishbase.org), Eschmeyer (2006) and Nelson (1994), but reclassified in Nelson (2006). Included also are families recognized in FishBase but with no proper common name in Nelson (2006). Scientific name Adrianichthyidae  Reclassification (Nelson, 2006) –  Order Beloniformes  Aracanidae Badidae Bedotiidae  Subfamily Aracaninae (p. 455) Subfamily Badinae (p. 382) Subfamily Bedotiinae (p. 2710  Tetraodontiformes Perciformes Atheriniformes  Cottocomephoridae Dentatherinidae  Lumped with Cottidae (p. 334) Subfamily Dentatherininae (p. 273) Lumped with Scorpaenidae (p. 320) Lumped with Moronidae (p. 344) Tribe Synaceiini (p. 324)  Scorpaeniformes Atheriniformes  Lumped with Congiopodidae (p. 327)  Eschmeyeridae Lateolabracidae Synaceiidae Zanclorhynchidae  Scorpaeniformes  Common name Adrianichthyids; Ricefishes – – Madagascar rainbowfishes – Tusked silversides –  Source of common name Nelson (2006); FishBase (18 August 2006 version) – – Nelson (1994); FishBase (18 August 2006 version) – Nelson (1994); FishBase (18 August 2006 version) –  Perciformes  Asian seaperches  Eschmeyer (2006)  Scorpaeniformes  Stonefishes  Scorpaeniformes  –  Nelson (1994); FishBase (18 August 2006 version) –  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 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 Apistidae (Scorpaeniformes, scorpionfishes and flatheads): Greek, ‘apistos’, i.e., ‘suspicious’ (Romero, 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 (Scorpaeniformes, 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’ (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’. Neosebastidae (Scorpaeniformes, scorpionfishes and flatheads): Greek, ‘neos’, i.e., ‘new’; Greek, ‘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 characterized with dentaries each with one huge fang (Nelson, 1994). Sole representative, Omosudis lowii 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, 1994). 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 family 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 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.  42  Common names of fish families, Palomares, M.L.D. et al.  (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 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 names of species in the Family Ambassidae (Perciformes). ‘Pomfrets’ is preoccupied by the Family Bramidae (Perciformes). ‘Silver pomfret’ is preoccupied by Pampus argenteus (Euphrasen, 1788) (Stromateidae, Perciformes). Aphyonidae (Ophidiiformes, cusk eels): Greek, ‘aphyo’, ‘aphye’, i.e., ‘sardine’, ‘anchovy’, in the sense of 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, ‘clupea’, i.e., ‘sardine’ (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 lizard (Romero, 2002). Largescale deep-sea lizardfishes. ‘Deep-sea lizardfish’ is preoccupied by 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, ‘sauros’, i.e., ‘lizard’; 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 ‘black lizardfish’ (retained as the FishBase name) and ‘deep-water greeneye’. ‘Greeneyes’ is preoccupied by Family Chlorophthalmidae (Aulopiformes), which are also found in deep waters. Citharinidae (Characiformes, characins): Greek, ‘kitharia’, i.e., ‘cittern’, a Renaissance stringed 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 canines to 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’. 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, Siluriformes) and Hemisorubim platyrhynchos (Valenciennes, 1840) (Pimelodidae, Siluriformes); and Poeciliopsis gracilis (Heckel, 1848) (Poeciliidae, Cyprinodontiformes). Erethistidae (Siluriformes, 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 usually large (Nelson, 1994), found in rivers, streams and moving freshwaters of southern Asia (Nelson, 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). Hemiodontidae (Characiformes, characins): Greek, ‘hemi’, i.e., ‘half’; Greek, ‘odous’, i.e., ‘tooth’, ‘teeth’ (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 words in common. 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). Lophichthyidae (Lophiidormes, anglerfishes): Greek, ‘lophos’, i.e., ‘crest’; Greek, ‘ichthys’, i.e., ‘fish’ (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’. Myrocongridae (Anguiliformes, eels and morays): Greek, ‘myros’, i.e., male of moray eel; Latin, ‘conger’, 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): Greek, ‘para’, i.e., ‘the side of’; Greek, ‘odous’, i.e., ‘teeth’ (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’,  44  Common names of fish families, Palomares, M.L.D. et al.  ‘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, 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’ (Romero, 2002). Collared wrigglers. Available common names include the word ‘wriggler’.  (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’. Nelson (2006) uses ‘picarel porgies’. The name ‘picarels’ is not preoccupied and is simpler than ‘picarel porgies’. Champsodontidae (Perciformes, perch-likes): Greek from Egyptian, ‘champsai’, i.e., ‘crocodile’; Greek, ‘odous’, i.e., ‘teeth’ (Romero, 2002). Crocodile toothfishes. The only available English common name is ‘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’ which is used in Nelson (2006) is also preoccupied by Serranus cabrilla (Linnaeus, 1758) (Serranidae, Perciformes) and used in the composite name of Chaunax stigmaeus Fowler, 1946 (Chaunacidae, Lophiiformes). However, ‘crocodile toothfish’ is not preoccupied by any species. Chaudhuriidae (Synbranchiformes, spiny eels): From ‘chaudhuria’, i.e., a Burmese local name for a fish (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 eels’, ‘earthworm’ being used in the composite common name for Yirrkala lumbricoides (Bleeker, 1853) (Ophichthidae, Anguiliformes). We believe that ‘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). Snaketooth fishes. Some of the available English common names include the word ‘swallower’, the name used in Nelson (2006) but which is preoccupied by Family Saccopharyngidae (Saccopharyngiformes). Some of the available Chinese common names include the words ‘snake-toothed’ and occasionally ‘fork-toothed’ and ‘long-toothed’. Curimatidae (Characiformes, 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. 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 matter, microdetritus, microvegetation, and filamentous algae common in those habitats (Nelson, 1994). ‘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’. 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  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 (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 ‘Spain’, 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 ‘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/cgibin/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. 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. Centrophoridae (Squaliformes, bramble, sleeper and dogfish sharks): Greek, ‘kentron’, i.e., ‘thorn’, ‘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 eel’ (Romero, 2002). Short-tail eels*. ‘Short tail conger’ is preoccupied by Paraconger similis (Wade, 1946) (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 relatively broad gill openings and one straight strong spine on both opercle and subopercle; they are uncommon or relatively rare and 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 toothless tongue. The proposed English common name is the chosen FishBase common name for the sole representative, Gymnarchus niloticus Cuvier, 1829 and is based on the Ijo language (Nigerian) common name.  3  See footnote (2) on p. 41.  46  Common names of fish families, Palomares, M.L.D. et al.  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, 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 the words ‘short’, ‘cheek’, and ‘python’. Nelson (2006) took the AFS official common name and accepted as the FishBase common name for Pythonichthys 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 pectoral fin-base. They also possess an electric organ which discharges discrete pulses (Albert, 2003). Nelson (2006) took the English common name used for 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; operculum with spines and also has a series of 5 isolated dorsal spines (Nelson, 1994). ‘Pricklebacks’ is preoccupied by Family Stichaeidae (Perciformes) and ‘sticklebacks’ by Family Gasterosteidae (Gasterosteiformes). Ipnopidae (Aulopiformes, grinners): Greek, ‘ipnos’, i.e., ‘oven’, ‘kiln’ (Romero, 2002). Deep-sea tripod fishes. Members of this family 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 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’, i.e., a kind of fish (Romero, 2002). Pencilfishes*. Members of this family have elongate, cylindrically-shaped bodies with fairly large scales, 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). Microstomatidae (Osmeriformes, smelts): Greek, ‘mikros’, i.e., ‘small’; Greek, ‘stoma’, i.e., ‘mouth’ (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 the composite name for the Family Argentinidae (Salmoniformes), ‘pencil’ is used in composite names by the Family Trichomycteridae (Siluriformes) and ‘smelt’ is preoccupied by Family Osmeridae (Salmoniformes). 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 in composite names of the Family Myctophidae. Nelson (2006) adapted the FAO English and accepted FishBase common name for Scopelengys tristis Alcock, 1890. 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 proximal radial from contact with the cleithrum, has six branchiostegal rays and lack of lateral line (Nelson, 1994). Some  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., ‘hair’; Greek, ‘noton’, i.e., ‘back’ (Romero, 2002). Sand diving lizardfishes. ‘Sanddivers’ is the name used 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 barbells and are usually found in mountain streams. Radiicephalidae (Lampriformes, velifers, tube-eyes and ribbonfishes): Latin, ‘radius’, i.e., ‘radius’; Greek, ‘kephale’, i.e., ‘head’ (Romero, 2002). Tapertails*. Members of this family have elongated scaleless bodies, compressed and attenuated posteriorly to a thin caudal filament. The sole representative of this family is Radiicephalus elongatus Osório, 1917, whose FishBase English common name is ‘tapertail’. This is used in composite English names for some members of Family Trachipteridae (Lampriformes) and Family Engraulidae (Clupeiformes). Samaridae (Pleuronectiformes, flatfishes): Latin, ‘samara’, i.e., seed of the elm (Romero, 2002). Could also be Greek, ‘sêma’, i.e., ‘sign’, ‘mark’, ‘token’ of the star on a horse’s head (Liddel and Scott, 1889). Crested flounders. Most available English common names include the words ‘righteye flounder’. ‘Righteye flounder’ is used in composite name for Family Pleuronectidae (Pleuronectiformes). Nelson (2006) adapted a modified form of the scientific name. Schindleriidae (Perciformes, perch-likes): Named after Dr. D.W. Schindler, University of Alberta, Canda (Eschmeyer, 2006). Infantfishes. Members of this family are small and neotenic, with transparent 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, ‘labrus’, i.e., ‘lip’ (Romero, 2002). Longfin escolars. Members of this family are deep water fishes which have serrated operculum and preoperculum and a protusible maxilla (Nelson, 1994). The sole representative, Scombrolabrax heterolepis Roule, 1921 has the FishBase common name ‘longfin escolar’. ‘Escolar’ is used in composite names in the Family Gempylidae (Perciformes). 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.  48  Common names of fish families, Palomares, M.L.D. et al.  REFERENCES Albert, J.S., 2003. Hypopomidae (bluntnose knifefishes), In: Reis, R.E., Kullander, S.O., Ferraris, C.J. Jr. (eds.), Checklist of the Freshwater Fishes of South and Central America. EDIPUCRS, Porto Alegre, Brasil, pp. 494-496. Eschmeyer, W. (ed.), 1998. Catalog of Fishes. California Academy of Science. San Francisco, USA. Eschmeyer, W. (ed.), 2006. Catalog of Fishes. World Wide Web electronic publication. www.calacademy.org, version (08/2006). Froese, R., Pauly, D. (eds.), 2006. FishBase. World Wide Web electronic publication. www.fishbase.org, 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, T., 2001. Pictorial Guide to Indonesian Reef Fishes. Part 1. Eels-Snappers, Muraenidae Lutjanidae. Zoonetics, Australia. Liddell and Scott, 1889. An intermediate Greek-English Lexicon. Clarendon Press Oxford. www.perseus.tufts.edu/cgibin/resolveform, version (15/08/2006). Nelson, J.S., 1994. Fishes of the World. Third edition. John Wiley and Sons, Inc., New York. Nelson, J.S., 2006. Fishes of the World. Fourth edition. John Wiley and Sons, Inc., New York. 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. Translated from Latin by Karl Friedrich Theodor Mayhoff. Lipsiae. Teubner. 1906. Robins, C.R., Bailey, R.M., Bond, C.E., Brooker, J.R., Lachner, E.A., Lea, R.N., Scott, W.B., 1991. World Fishes Important to North Americans. American Fisheries Society Special Publication 21, AFS, Bethesda, Maryland, USA. Romero, P., 2002. An Etymological Dictionary of Taxonomy. Madrid, unpublished. Cited in FishBase. 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  ABSTRACT This study, based on 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 morphometrics, growth, reproduction and feeding habits of a species that is often studied in terms of its 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 onset 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 mudskipper, Periophthalmus barbarus (Linnaeus, 1766) (Family Gobiidae, SubFamily Oxudercinae, Order Perciformes, Figure 1, insert) is the most abundant fish in the mangrove swamps 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 is driven into the mud. Then, leaves and mud are used to turn the opening into the mouth of a trap (Figure 1). Powdered okra or a substance locally 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 Guinea, a broad leaf and a piece of broom stick are used, instead of bamboo cane, to construct a gear similar in operation to the above-described trap. 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 [ISSN 1198-6727]. 2 Present address: CSRP, Dakar, Senegal; Email: Michael.Vakily@googlemail.com 1  50  Growth, reproduction and food of mudskipper, Turay, I. et al.  Figure 1 The mudskipper, Periophthalmus barbarus (Linnaeus, 1766) (Family Gobiidae, insert adapted from FAO), with schematic representations of burrows and of the bamboo-and-leaves trap used in Sierra Leone (adapted from Den-Khadhri, 1984).  MATERIALS AND METHODS Catch composition statistics relating to the mudskipper do not appear to exist, and this study is thus entirely based on 364 specimens caught by the first author from July 1992 to April 1993 by digging into mudskipper burrows in the mudflat of Congotown, a ‘quartier’ of Freetown (8° 29’ 10’ N; 13° 15’ 30’ E). Morphometric measurements were done on 348 samples, to the nearest mm, using a vernier caliper; the measurements taken included total (TL) and standard length (SL) and some others presented in Turay (1993), but the measurements presented below refer to TL in cm. Individual fish weights were determined to the nearest 0.01 g. The length-frequency analyses were all performed using the FiSAT software (Gayanilo et al., 1996). Reproduction was studied by noting gonad maturity stages, by computing gonadosomatic indices (GSI = gonad weight·100/body weight) and by estimating ‘fecundity’, i.e., the number of eggs in the gonads of mature female mudskipper. Food and feeding habits were studied through examination of stomach contents, and the ‘point’ and ‘occurrence’ methods (Hynes, 1950).  RESULTS AND DISCUSSION The largest mudskipper observed was a male of 15.4 cm; the largest female measured 13.0 cm. Figure 2A shows the length-weight relationship of female and male mudskippers; the corresponding relationship for the population as a whole is W = 0.0072·TL3.19 for which n = 348 and whose log linear version has an r2 = 0.961. The relationship between standard length and total length is given by: SL (cm) = 0.86·TL (cm) -0.0914 (Figure 2B)  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D. 50  14  A  45  B -0.0914  SL (cm) = 0.816 * TL (cm) R = 0.996; n = 150  3.19  W = 0.0072 * TL R = 0.980; n = 348  40  12  Standard length (cm)  35 Weight (g)  51  30 25 20 15  10 8 6 4  10 2  5 0  0 0  5  10  15  Tota length (cm)  20  0  2  4  6  8  10  12  14  16  Total length (cm)  Figure 2 Relationships of length to weight (A) and total to standard lengths (B) of male and female Periophthalmus barbarus (Linnaeus, 1766) (Family Gobiidae), removed from their burrows in the mudflats of Congotown, Freetown, 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∞ ranging from 16 to 18 cm, depending on which length range was included in the analysis. Hence a value of L∞ = 17 cm was retained for all analyses using ELEFAN, a routine of the FiSAT software. A first pass of the ELEFAN subroutine in which K values are scanned, in very small steps, from K = 0.1 to K = 10 year-1, yielded an essentially flat structure (Figure 3a). This was attributed to the class interval (0.5 cm) of the L/F data in Table 1, which was too small  Table 1 Length-frequency data for 348 mudskippers, Periophthalmus barbarus (Linnaeus, 1766) (Family Gobiidae), removed from their burrows in the mudflats of Congotown, Freetown, Sierra Leone. Midlength Jul Aug Oct Nov Dec Jan Feb Mar Apr (TL, cm) 1992 1993 0.25 0 0 0 0 0 0 0 0 0 0.75 0 0 0 0 0 0 0 0 0 1.25 0 0 0 0 0 0 0 0 0 1.75 0 0 0 0 0 0 0 0 0 2.25 5 1 0 0 0 0 0 0 0 2.75 1 1 0 0 0 0 0 0 0 3.25 0 0 0 0 0 0 0 0 0 3.75 0 0 1 0 0 0 0 0 0 4.25 1 2 1 0 0 0 0 0 0 4.75 2 4 1 2 2 0 0 0 1 5.25 1 1 3 1 0 1 0 0 0 5.75 0 2 1 2 2 1 1 1 1 6.25 0 2 2 1 2 3 1 4 1 6.75 3 5 1 1 2 0 2 9 0 7.25 1 3 2 1 3 2 2 8 2 7.75 4 6 3 1 2 3 1 16 1 8.25 5 3 4 2 4 1 3 5 4 8.75 10 8 1 2 2 5 1 11 2 9.25 4 2 2 1 4 1 4 6 1 9.75 10 3 4 2 2 3 4 4 1 10.25 4 3 2 3 2 1 5 2 1 10.75 1 1 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 13.25 0 0 0 0 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 0 0 0 0 0 1 15.75 0 0 0 0 0 0 0 0 0 16.25 0 0 0 0 0 0 0 0 0 Total 63 53 35 26 30 23 30 68 20  52  Growth, reproduction and food of mudskipper, Turay, I. et al.  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 exposed to seasonal differences in (air) temperature of about 10°C (Findlay, 1978). This was done using the ‘automatic search routine’ of ELEFAN, with K, C and WP as free parameters, then using the best 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∞ at 17 cm as for all previous analyses). Figure 4 (upper panel) shows the growth curve obtained as a result of this work. The asymptotic length (L∞) is 17 cm, K = 0.89 year-1 is the rate at which L∞ is approached, C = 0.75 expresses the amplitude of the seasonal growth oscillations, which are here rather strong given that C can take values between 0 and 1, and where 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’ season (Vakily and Pauly, 1993). Figure 3 Response surface analysis of the parameter K of the von Bertalanffy growth function showing the increasing resolution of ELEFAN following modification of the length-frequency data in Table 1, for TL∞ =17 cm.  This growth curve, which also fits very small mudskippers, has its origin in mid June, suggesting this period as the main spawning season. There is some indication of a second minor spawning at the end of the year, leading to a second, minor cohort (lower panel of Figure 4). 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  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.  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.  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 Females Males (TL, cm) Immature/mature 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  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.  54  Growth, reproduction and food of mudskipper, Turay, I. et al.  REFERENCES Bertalanffy, L. von., 1938. A quantitative theory of organic growth (inquiries on growth laws II). Human Biol. 10(2), 181-213. Deen-Khadri, A., 1984. Some Aspects of the Biology of the Mudskipper. B. Sc. Thesis, Fourah Bay 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., 1981. Fisheries Biology: A Study in Population Dynamics. Second Edition. University of Wisconsin Press, Madison. 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. 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, Fourah Bay College, University of Freetown, Freetown, Sierra Leone. Vakily, J.M., Pauly, D., 1993. Seasonal movements of Sardinella off Sierra Leone. In: Bard, F.X., Koranteng, K.A. (eds.), Dynamique et Usage des Resources en Sardinelles de l’Upwelling Côtier du Ghana et de la Côte d’Ivoire. Actes de la recontre du DUSRU, 5-8 Octobre 1993, Accra, Ghana. Editions ORSTOM, Paris, pp. 425-436. Weatherall, J., 1986. A new method for estimating growth and mortality parameters from length-frequency data. FishByte 4(1), 12-14.  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D.  55  NOTE ON THE WEIGHT OF BODY PARTS, INCLUDING FINS, OF THE SMALLTOOTH SAWFISH PRISTIS PECTINATA1 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 In July 1976, I became by default the scientific leader of a demersal survey conducted by R/V Mutiara 4 in the southern Indonesian part of the South China Sea (south and east of Singapore), documented in Pauly et al. (1996). During the cruise, we captured a large sawfish (Family Pristidae), which we identified as the smalltooth sawfish Pristis pectinata Latham 1794 (Figure 1). At the time, all large sharks we caught were finned by the Indonesian crew of the Mutiara 4, the fins hanged to dry, garland-like, from the ship’s rigging, and later sold on the ‘soup fin’ market by the crew as bonus. The saw was sold separately for the curios market. This sawfish, of 270-300 cm (total length, i.e., including the saw-like snout), differed from others  Figure 1 The smalltooth sawfish, Pristis pectinata Latham 1794, Family Pristidae (drawing from 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 Weight % part (extremities) 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  in that I weighted its different body parts. The results of this rather bloody exercise are given in Table 1. Thirty years later, the smalltooth sawfish is ‘Critically endangered’ on the IUCN Redlist (Baillie et al., 2004), both because it is caught as 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. Pauly, D., Martosubroto, P., Saeger, J., 1996. The Mutiara 4 surveys in the Java and southern South China Sea, November 1974 to July 1976. In: D. Pauly and P. Martosubroto (eds.), Baseline Studies in Biodiversity: The Fish Resources of Western Indonesia. ICLARM Studies and Reviews 23, pp. 47-54.  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].  56  Overview of biological data of Greek stocks of anchovies and sardines, Somarakis, S. et al.  AN OVERVIEW OF BIOLOGICAL DATA RELATED TO ANCHOVY AND SARDINE STOCKS IN GREEK WATERS1 Stylianos Somarakis  University of Patras, Department of Biology, GR 26500 Patra, Greece; Email: somarak@upatras.gr  Dimitrios E. Tsianis  Aristotle University of Thessaloniki, School of Biology, Department of Zoology, Laboratory of Ichthyology, 54124, Thessaloniki, Greece; Email: dtsianis@in.gr  Athanassios Machias  Hellenic Centre for Marine Research, P.O. Box 2214, 71003, Iraklion, Crete, Greece  Konstantinos I. Stergiou  Aristotle University of Thessaloniki, School of Biology, Department of Zoology, Laboratory of Ichthyology, 54124, Thessaloniki, Greece; Email: kstergio@bio.auth.gr  ABSTRACT European anchovy (Engraulis encrasicolus) and sardine (Sardina pilchardus) are the most important small pelagic species in Greek waters. We present here a brief overview of available data sources and studies concerning their biology and ecology. Some new analyses are also presented concerning lengths at first maturity.  INTRODUCTION European anchovy (Engraulis encrasicolus) and sardine (Sardina pilchardus) are the two most important small-sized pelagic species in Greek waters, making up 30% of the total Greek landings and 59% of the total purse seine landings (Stergiou et al., 1997a). Despite their importance for the Greek fisheries, their stocks have never been studied systematically, nor has their exploitation been monitored on a yearly basis, with the exception of their landings (see e.g., Stergiou, 1989, 1990, 1991, 1992; Stergiou et al., 1997). In general, most of their landings in Greek waters derive from the purse seine fleet. Fishing with pelagic trawlers is prohibited, i.e., the percentage of small pelagic fish in the total marketable fraction of demersal trawlers cannot exceed 5%. Management measures to protect these species include a closed season for purse seiners from the 10th of December to the end of February. In this paper we attempt to present a brief overview of the available biological and ecological data concerning anchovy and sardine in Greek waters.  GENETIC STRUCTURE Various studies conducted in recent years indicate a significant genetic structuring for anchovy in the eastern Mediterranean and Black Seas (Spanakis et al., 1989; Magoulas et al., 1996; Machias et al., 2000a). This is not the case for sardine which conforms to a panmictic population (Machias et al., 2001b). Surveys of genetic variation in mitochondrial DNA (Magoulas et al., 1996; Machias et al., 2000a) provide 1 Cite as: Somarakis, S., Tsianis, D.E., Machias, A., Stergiou, K.I., 2006. An overview of biological data related to anchovy and sardine stocks in Greek waters. In: Palomares, M.L.D., Stergiou, K.I., Pauly, 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].  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D.  GSI (%)  4  REPRODUCTION  450  (a)  3  350  2  250  1  150  0  50 oct  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), for estimating the spawning biomass of the stocks (Tsimenides et al., 1995a; 1998; Machias et al., 2000a; 2001b).  6  eggs/m2  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).  57  nov dec  jan  feb  mar  apr  may jun  jul  aug sep  nov dec  jan  feb  mar apr  may jun  jul  aug sep  (b)  5  GSI (%)  4 3 2 1 0 oct  Month  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) 41  (a)  Dardanelles  40 1000 100 10 0.1 0 22  Lemnos island NORTH AEGEAN SEA 23  24  25  26  (b)  1000 100 10 0.1 0  39  Figure 1 Data relevant to anchovy and sardine spawning periods in the Greek Seas: (a) north Aegean Sea. Monthly evolution of the gonosomatic index (GSI=100*ovary weight/ovary-free weight) for sardine (□) and anchovy (■) and mean abundance of anchovy eggs in the plankton (●). Data from Somarakis (1993), Voulgaridou and Stergiou (2003) and Tsianis (2003); (b) central Aegean and Ionian Seas. Monthly evolution of the gonosomatic index (GSI=100*ovary weight/eviscerated weight) for sardine in the Aegean (○) and the Ionian (□) as well for the anchovy in the Aegean (●) and the Ionian Sea (■).Data from Machias et al. (2000a; 2001b). Bars: ±SE.  CENTRAL AEGEAN  38 IONIAN SEA 20  21  22  23  24  25  (c)  1000 100 10 0.1 0  39  CENTRAL AEGEAN  38 IONIAN SEA 20  21  22  23  24  25  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).  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 Lemnos 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).  58  Overview of biological data of Greek stocks of anchovies and sardines, Somarakis, S. et al.  Figure 3 Biweekly mean total length (in cm) of sardine in NW Aegean, 07/1996-05/2003 (from Voulgaridou and Stergiou 2003; Tsianis, 2003).  Temperature  20  650  eggs/g  C o  18  16  250  Salinity  37.5  0.50  37  psu  Spawning frequency  0.40  36.5  S  36 35.5  0.30 0.20  35 34.5  0.10  Zooplankton  40  0.56  35  Adult condition  0.55  30  K  ml/m2  450 350  17  25  0.54 0.53  20 15  0.52  Egg abundance  Egg size  500  0.21  400  0.21  mm3  300 200  360  0.20 0.20  100 0  0.19 93  Larval abundance  94  95  96  300  No/m2  Available studies suggest that the reproductive tactics of anchovy and sardine in Greek waters differ substantially (Somarakis, 1999; Ganias, 2003). Sardines seem to rely mainly on fat reserves stored in the muscles and viscera during summer to reproduce during the winter. They are characterized by low seasonal and interannual variability in batch fecundity and spawning frequency (Somarakis et al., 2002, Ganias, 2003, Ganias et al., 2003a; 2004). On the other hand, the reproductive effort of summer spawning anchovy seems to be closely associated with adult prey fields (meso-zooplankton), i.e., the energy allocated to reproduction derives primarily from the food intake. In comparing batch fecundity and spawning frequency estimates between June 1993 and June 1995 over the NE Aegean Sea continental shelf, Somarakis (1999) showed  Relative fecundity  550  19  No/m2  Data on the spawning period of sardine exist for the coastal areas of the central Aegean and Ionian Seas as well as for the north Aegean Sea (Machias et al., 2001b; Ganias, 2003; Ganias et al., 2003b; Voulgaridou and Stergiou, 2003; Tsianis 2003) (Figure 1). In the coastal waters of central Greece, spawning takes place from October to May but it peaks earlier in the Aegean (November-January) than in the Ionian Sea (JanuaryFebruary) (Figure 1b). The gonadosomatic index data for sardine in the north Aegean Sea (Figure 1a) suggest the existence of two spawning peaks: one in November-December and another in March. The importance of the second peak increased during the period 1996-2000 (Tsianis, 2003). This is attributed to the decrease in the mean total length with time (Figure 3; Tsianis, 2003) given that, in general, smaller sardines spawn later than larger ones (Ganias, 2003). During the 1999 and 2000-2001 ichthyoplankton surveys in the central Aegean and Ionian Seas (Ganias et al., 2001; Somarakis et al., 2001), the highest egg abundance values were found inshore, mainly in waters shallower than 100 m depth (Figure 2c).  240 180 120 60 93  94  95  96  Year  Figure 4 Means (±95% confidence intervals) for temperature (0-40 m); salinity (0-40 m); zooplankton 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 Somarakis (1999).  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D.  There are very few age and growth studies for anchovy and sardine in Greek waters and otolith readings have not been generally validated (Table 1). Published growth curve comparisons among stocks, areas, and years have not shown any significant differences for either anchovy or sardine (Tserpes and Tsimenides, 1991; Nikoloudakis et al., 2000). Von Bertalanffy growth parameters (Table 1) estimated from samples collected from landings, experimental trawling, and onboard the commercial fishing fleet clearly indicate that commercial catches might be biased with respect to lengths-at-age, at least for the one and two year old fish (Figure 6). Indeed, selection of larger anchovy schools by the Greek purse seine fishery in the north Aegean Sea was evident when comparing length frequencies obtained from the commercial landings with those of experimental trawling (with ‘a proportional to abundance’ allocation of pelagic trawl samples) during a 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 clearly indicate that the mean length of sardine (Voulgaridou and Stergiou, 2003; Tsianis 2003) and anchovy (Loukmidou and Stergiou, 2000) in the NW Aegean Sea has declined in recent years (e.g., Figure 3). The potential effect of  90  100  110  120  130  140  150  160  170  160  170  Total Length (mm)  100 90 80 70 60 50 40 30 20 10 0  Sardine  80  90  100  110  120  130  140  150  Total Length (mm)  Figure 5 Length at first maturity (at 50%) of anchovy and sardine. Percent of mature females per length class in the central Aegean (□) and Ionian (●) Seas. Fitted logistic curves are also shown.  200 Total Length (mm)  AGE AND GROWTH  Anchovy  Anchovy  150 100 50 0 0  1  200 Total Lengt h (mm)  Lengths at first maturity of anchovy and sardine have been estimated for the central Aegean and Ionian Seas from samples collected experimentally during DEPM surveys (Figure 5 and see Appendix for method). Anchovy reach maturity at approximately 105 mm total length and sardine at around 115 mm, i.e., at the completion of the first year of life.  100 90 80 70 60 50 40 30 20 10 0 80  %  Concurrently, female anchovies were in better condition, producing numerous large-sized eggs at a higher spawning frequency (short interspawning interval) (Figure 4). These observations were consistent with a rationrelated reproductive tactic in anchovy (Somarakis et al., 2000b; Maraveya et al., 2001; Somarakis, 2005).  %  that adult food availability was higher in 1993, when waters were significantly cooler and of lower salinity.  59  2  3  4  3  4  Sardine  150 100 50 0 0  1  2 Age (years)  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 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).  60  Overview of biological data of Greek stocks of anchovies and sardines, Somarakis, S. et al.  environmental variability on such a decline is under investigation. Table 1 Estimates of the von Bertalanffy growth parameters for anchovy and sardine in Greek waters. t0 L∞ Species Area Year K Sampling Method (year-1) (mm) Anchovy  Sardine  North Aegean Sea NW Aegean Sea  1993  0.280  191  -2.480  Purse seine landings Purse seine landings  19971998  0.75  200  NW Aegean Sea  20002003  0.77  175  Thracian Sea  20002001  0.49  176  -1.276  Purse seine landings  Central Aegean and Ionian Seas  19981999  0.509  175  -0.888  NW Aegean Sea  19962003  0.80  219  Experimental pelagic trawl samples and onboard sampling Purse seine landings  NW Aegean Sea  19961999  0.86  208  Purse seine landings  Aegean and Ionian Seas  19831984  0.300  181  -3.210  Purse seine landings  Central Aegean and Ionian Seas  19992001  0.314  191  -1.839  Experimental pelagic trawl samples and onboard sampling  Purse seine landings  Otoliths Lengthfrequency analysis Lengthfrequency analysis Otoliths  Reference Tsimenides et al. (1995a) Loukmidou (1998) Tsianis et al. (2003) Kallianiotis et al. (2003)  Otoliths  Machias et al. (2000a)  Lengthfrequency analysis Lengthfrequency analysis Scales  Tsianis (2003)  Otoliths  Voulgaridou and Stergiou (2003) Tserpes and Tsimenides (1991) Machias et al., (2001b)  BIOMASS ESTIMATES  Frequency Frequenc % y (%)  Direct biomass estimates for anchovy and 30 sardine through acoustic and/or egg 25 surveys have been obtained in recent years Pelagic trawling in the framework of various EU and Purse seine landings 20 national projects (Tsimenides et al., 15 1995a; 1998; Machias et al., 2000a; 2001b). First acoustic surveys were 10 conducted in north Aegean in 1987-1988, 5 in the central Aegean in 1989-1990, and in 0 south Aegean in 1991, all aiming to study 95 100 105 110 115 120 125 130 135 140 145 150 155 160 165 170 175 the echo-distribution of small pelagic fish Total length (mm) assemblages (Tsimenides, 1989; Tsimenides et al., 1995b). Later surveys in Figure 7 North Aegean Sea, June 1995. Length frequency 1995-2001 were focused on the study of distributions of anchovy from experimental pelagic trawling during vertical and horizontal distribution and an acoustic survey and from purse seine landings. Data from the estimation of biomass of anchovy and Tsimenides et al. (1998). sardine in the north Aegean and the coastal areas of the central Aegean and Ionian Seas (Machias et al., 1996; 1997; 2000a, b; 2001a, b; Giannoulaki et al., 1999; 2001; 2003; Tsimenides et al., 1998; Maravelias et al., 1997). The acoustic method has been applied concurrently with DEPM for the estimation of biomass of anchovy in the north 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 between 11000 and 15000 t in the central Aegean and Ionian Seas. The estimates of sardine biomass (total or spawning biomass) in the central Aegean and Ionian Seas ranged from 20000 to 32000 t.  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 egg production method. LVPA: Length-based virtual population analysis. Year Month Region Acoustic DEPM survey LVPA References of direct survey (spawning biomass) surveys 1993  June  North Aegean  40643 (0.276)  1995  June  North Aegean  44601 (0.120)  1996 1998  June July  39475 (0.132) 14261 (0.298)  1999  June  North Aegean Central Aegean and Ionian Seas Central Aegean and Ionian Seas  14511 (0.280)  40236  Tsimenides et al. (1995a, 1998)  13446  Tsimenides et al. (1998) Machias et al. (1997) Somarakis et al. (1997) Tsimenides et al. (1998) Machias et al. (2000a, 2001a)  42708 (0.181)  11861 (0.278)  13044  Machias et al. (2000a, 2001a) Somarakis et al. (2002)  Table 3 Biomass estimates for sardine stocks in the Greek Seas (central Aegean and Ionian Sea; in tonnes with CVs in parentheses). DEPM: Daily egg production method. LVPA: Length-based virtual population analysis. Year Month Acoustic DEPM survey LVPA References of direct survey (Spawning Biomass) surveys 1999 2000-2001  December DecemberFebruary  19826 (0.429)  32594 (0.301) 24207 (0.225)  27086 30317  Machias et al. (2001b) Machias et al. (2001b) Somarakis et al. (2001)  SUMMARY Based on the above-mentioned studies, it is clear that there is a lack of regular, long-time series of data on biomass and biological parameters of anchovy and sardine throughout the Greek waters. Indeed, experimental sampling and surveys have been generally highly discontinuous in time and space, mainly 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. Ganias, K., 2003. Oceanographic and Biological Study of Sardine Egg Production, Sardina pilchardus (Walbaum, 1792) in Coastal Waters of Central Greece. PhD Thesis, University of Thessaly [in Greek with English Abstract]. Ganias, K., Somarakis, S., Caragitsou, E., Koutsikopoulos, C., Machias, A., Theodorou, A., 2001. Differential egg production of sardine off the central Hellenic coasts in December 1999. Rapp. Comm. int. Mer Medit. 36, 268. Ganias, K., Somarakis, S., Machias, A., Theodorou, A., 2003a. Evaluation of spawning frequency in a Mediterranean sardine population. Mar. Biol. 142, 1169-1179. Ganias, K., Somarakis, S., Koutsikpoulos, C., Machias, A., Theodorou, A., 2003b. Ovarian atresia in the Mediterranean sardine, Sardina pilchardus sardina. Mar. Biol. 83, 1327-1332. Ganias, K., Somarakis, S., Machias, A., Theodorou, A., 2004. Pattern of oocyte development and batch fecundity in the Mediterranean sardine. Fish. Res. 67, 13-23. Giannoulaki, M., Machias, A., Tsimenides, N., 1999. Ambient luminance and vertical migration of sardine (Sardina pilchardus). Mar. Ecol. Progr. Ser. 173, 275-288. 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 PanHellenic Conference of Ichthyologists, pp. 61-64. 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 fish. Mar. Ecol. Prog. Ser. 265, 243253.  62  Overview of biological data of Greek stocks of anchovies and sardines, Somarakis, S. et al.  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 abstract]. Loukmidou, S.P., Stergiou, K.I., 2000. Length-weight relationships and length frequencies of anchovy, Engraulis encrasicolus (Linnaeus, 1758), in the Thermaikos Gulf. Proc. 6th PanHellenic Symp. Oceanogr. Fish. 6(2), 109-113. Machias, A., Somarakis, S., Tsimenides, N., 1996. Comparative distribution of anchovy and sardine in north Aegean basin: June 1995. Proc. 17th Scient. Symp. Greek Soc. Biol. Sci. 17, 135-137. Machias, A., Somarakis, S., Giannoulaki, M., Manousakis, L., Kapantagakis, A., Tsimenides, N., 1997. Estimation of 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 hydroacoustics. In: Proceedings of the 10th PanHellenic Conference of Ichthyologists, pp. 57-60. Machias, A., Somarakis, S., Magoulas, A., Koutsikopoulos, C., 2001b. Evaluation of the Southern Greek Sardine Stocks. Institute of Marine Biology, Crete, Contract No 98/030, Final Report. 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 (Engraulis encrasicolus). Mol. Biol. Evol. 13(1), 178-190. Maravelias, C.D., Machias, A., Somarakis, S., Tsimenides, N., 1997. Spatial distribution of pelagic species in the 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. Nikoloudakis, G., Machias, A., Somarakis, S., Koutsikopoulos, C., Tsimenides, N., 2000. Comparison of growth in two 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. 6th PanHellenic Symp. Oceanogr. Fish. 6(2), pp. 88-93. 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 [in Greek with English abstract]. Somarakis, 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.  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D.  63  Somarakis, S., Ganias, K., Koutsikopoulos, C., Machias, A., Papaconstantinou, C., 2001. Applying the Daily Egg Production Method (DEPM) to small sardine stocks in the eastern Mediterranean (central Greece). ICES Study 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 46, 16-23. Stergiou, K.I., 1990. A seasonal autoregressive model of the anchovy Engraulis encrasicolus fishery in the eastern Mediterranean. Fish. Bull. (U.S.) 88, 411-414. Stergiou, K.I., 1991. Describing and forecasting the sardine-anchovy complex in the eastern Mediterranean using vector autoregressions. Fish. Res. 11, 127-141. Stergiou, K.I., 1992. Variability of monthly catches of anchovy Engraulis encrasicolus in the Aegean Sea. Fish. Bull. (U.S.) 90, 211-215. Stergiou, K.I., Georgopoulos, D., 1993. The distribution of phytoplankton pigments and the fishery of anchovy (Engraulis encrasicolus) in the Hellenic Seas. Naga, The ICLARM Quarterly 16(2), 34-37. Stergiou, K.I., Laskaratos, A., 1997. Climatic variability and the anchovy/sardine ratio in Hellenic waters. Geojournal 41(3), 245-254. Stergiou, K.I., Economidis, P., Sinis, A., 1996. Sex ratio, spawning season and size at maturity of red bandfish in western north Aegean Sea. J. Fish Biol. 49, 561–572. Stergiou, K.I., Christou, E., Georgopoulos, D., Zenetos, A., Souvermezoglou, C., 1997. The Hellenic Seas: physics, chemistry, biology and fisheries. Ocean. Mar. Biol. Ann. Rev. 35, 415-538. Tserpes, G., Tsimenides, N., 1991. Evaluation of growth rate differences in populations of Sardina pilchardus (Walbaum, 1792) from the Aegean and Ionian seas. Cybium 15, 15-22. Tsianis, D.E., 2003. Temporal Variation of Biological Parameters of Anchovy and Sardine in NW Aegean. M.Sc. Thesis, Aristotle University of Thessaloniki [in Greek with English abstract]. Tsianis, 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. 5965. 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.  64  Overview of biological data of Greek stocks of anchovies and sardines, Somarakis, S. et al.  APPENDIX: ESTIMATION OF LENGTH AT FIRST MATURITY Length at first maturity of anchovy and sardine was estimated from samples collected during the peak of the spawning season (during DEPM surveys) in summer 1999 and winter 2000-2001, respectively. Fish were collected with a pelagic trawl and maturity was assessed histologically 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 et al., 1996), is described by the logistic function: P = e(v1+v2 L) / (1+e(v1 + v2 · L)),and the value of L50, L25, L75 can be estimated from the expressions: L50 = -v1/v2, L25 = [-Ln(3)-v1]/v2, L75 = [Ln(3)-v1]/v2. The proportion of mature fish for each 10 mm length class was calculated by sex and v1, v2 were calculated using Fryer’s (1991) algorithm by maximising the log– likelihood, ln(L) (Petrakis and Stergiou, 1997). A test for over– dispersion was obtained by estimating the deviance statistic ∆ 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.  Table A1 Estimated parameters for the size at maturity, L50. SE: Standard error; CI: 95% confidence intervals; ∆: deviance statistic; df: degrees of freedom. All lengths are total lengths in mm. Anchovy Sardine Parameter estimates Aegean Sea Ionian Sea Aegean Sea Ionian Sea 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  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 tissues 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, notably unpublished documents from Australian government agencies. All growth 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 result was a multiple regression in which both L∞ and temperature had a significant effect on K. Thus, environmental temperature, which varies strongly along the east and west coast of Australia, had an effect on the growth parameters of Australian fishes, as predicted by the theory tested here.  INTRODUCTION Growth parameters of fish are an important source of information on the biology of fish, and for management of fisheries. Few studies have looked simultaneously at the growth of a large number of species, in spite of the valuable insights this can provide (Cury and Pauly, 2000). 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, and 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 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 other sources, notably unpublished documents from Australia. All growth  1 Cite as: Andersen, C., Pauly, D., 2006. A comparison of growth parameters of Australian marine fishes north and south of 28° South. In: Palomares, M.L.D., Stergiou, K.I., Pauly, D. (eds.), Fishes in Databases and Ecosystems. Fisheries Centre Research Reports 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: claire.andersen@dpi.qld.gov.au.  66  Growth of Australian marine fishes, Andersen, C., Pauly, D.  parameter sets, corresponding to fish populations north or the south of 280 S, were assigned a mean water temperature.  Figure 1 Schematic representation of the relationship between relative gill area (and hence relative oxygen supply) and body weight in fishes. Maintenance metabolism determines the maximum size that can be reached because relative gill area (and hence oxygen supply) must decline with body weight. A: A fish exposed to low environmental temperature will have a low maintenance metabolism and reaches a larger size. B: Exposure of a fish of the same species 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 in A (see also Pauly 2006, this volume).  MATERIALS AND METHODS Growth parameters were obtained from the POPGROWTH Table of FishBase (www.fishbase.org). These were complemented with growth parameters from other sources, notably various published and unpublished documents from Australia (see References). The von Bertalanffy growth function (VBGF) for length has the form: Lt = L∞ (1 – e-K(t-to))  … 1)  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 were to grow a very long time (indefinitely, actually), K is a coefficient of dimension time-1, expressing the rate at which L∞ is approached, and to is the (usually negative) age the fish would have had at length zero if they had always grown in the manner predicted by the equation (which they 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 data. When this was not available from FishBase or the references below, a temperature was assigned to all growth parameter sets, corresponding to the mean water temperature at the sampling using the sampling location and depth (actual or inferred), based on data available from the Australian Oceanographic Data Centre (www.aodc.gov.au/).  RESULTS AND DISCUSSION The following multiple regression was obtained from the 190 sets of growth parameters obtained: logK = 0.1652 + 0.0245·Temp - 0.681·logL∞  … 2)  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 of temperature was significant (P<0.01), in addition to L∞, in predicting K. This was solved for logK using different temperatures, as shown in Figure 2, i.e., a plot of logK vs. logL∞, with isotherms at 30, 20, and 100C superimposed. As might be seen, the fish in the high K – low L∞ quadrant of Figure 2 stem overwhelmingly from Australian waters north of 280 S, while the fish with growth parameters in the opposite quadrant were sampled south of 280S. These findings are compatible with a number of biological mechanisms. The most likely mechanism, however, is that high temperature reduces the scope for growth of fishes, by increasing their metabolic rate, and hence the amount of water having to be passed across the gills in order for their respiratory requirements to be met. This task is made even more difficult by the fact that less oxygen is dissolved in warm than in cold water (Pauly 1981). The different habitats of fishes, whether estuarine, Figure 2 Plot of the von Bertalanffy parameter (log) K vs. the parameter L∞ coastal, neritic or oceanic, in 190 population of Australian marine fishes north and south of 280 South. offer them food of different The warm-water populations from the North tend to have lower L∞ suitability and in different and higher K than those in the South, and vice versa for those in the colder South. densities. However, there is The isotherms for 10, 20 and 300C were drawn using Equation 2 (see text). no reason to assume that the suitability or density of food available to fish of different species should change, along the Australian coasts, in a close relationship to temperature or latitude (Longhurst and Pauly, 1987). Rather, food can be assumed to vary randomly, and to represent one of the causes for the variability around the values predicted by the multiple regression (Equation 1) and consequently, in Figure 2. We thus feel that the evidence we presented support the existence of a strong linkage between fish growth and respiration.  REFERENCES3 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. Unpublished report held in NIWA library, Wellington. 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 stock structure, Mar. Freshwat. Res. 47(2), 109120. Brown, I., Sumpton, W., 1998. Age, growth and mortality of red throat emperor, Lethrinus miniatus, from the southern Great Barrier Reef: Queensland, Australia. Bull. Mar. Sci. 62(3), 905-917. Cameron, D.S., Begg, G.A., Fisheries Biology and Interaction in the Northern Australian Small Mackerel Fishery. Fisheries Research and Development Corporation [Internal report]. Cury, P., Pauly, D., 2000. Patterns and propensities in reproduction and growth of fishes. Ecol. Res. 15, 101-106.  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.  68  Growth of Australian marine fishes, Andersen, C., Pauly, D.  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., Paul, L.J., Mulligan, K.P., 1992. Ageing of adult snapper (Pagrus auratus) from otolith annual ring counts: validation by tagging and oxytetracycline injection. Aust. J. Mar. Freshwat. 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., Dichmont, C., Sellin, M., Cosgrove, M., McLennan, M., 2000. Integrated Fish Stock Assessment and Monitoring Program. Fisheries Research and Development Corporation Project no 94/161. Horn, P.L., 1997. An ageing methodology, growth parameters and estimates of mortality for hake (Merluccius australis) from around South Island, New Zealand. Mar. Freshwat. Res. 48, 201-209. Hyndes, G.A., Platell, M.E., Potter, I.C., Lenanton, R.C.J., 1998, Age composition, growth, reproductive biology and recruitment of King George whiting, Sillaginodes punctata, in coastal waters of southwestern Australia. Fish. Bull. 96, 258-271. 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, 549558. Longhurst, A., Pauly, D., 1987. Ecology of Tropical Oceans. Academic Press, San Diego. Milton, D.A., Blaber, 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. 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. Morison, A.K., Coutin, P.C., Robertson, S.G., 1998. Age determination of black bream, Acanthopagrus butcheri (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 Australian waters. Aust. J. Mar. Freshwat. Res. 43, 1241-1267. O'Neill, M.F., 2000. Fishery Assessment of the Burnett River, Maroochy River and Pumicestone Passage. Queensland 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. (eds.), Fishes in Databases and Ecosystems. Fisheries Centre Research Reports 14(4), pp. 89-95. Fisheries Centre, University of British Columbia. Sanders, M.J., Powell, D.G.M., 1979. Comparison of the growth rates of two stocks of snapper (Chrysophrys auratus) in south-east Australian waters using capture- recapture data. New Zeal. J. Mar. Freshwat. Res. 13(2), 279-284. Seki, T., Taniuchi, T., Nakano, H., Shimizu, M., 1998, Age, growth and reproduction of the Oceanic whitetip shark from the Pacific Ocean. Fish. Sci. 64(1), 14-20. 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. 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.  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, we gathered from the literature 80 length-length relationships referring to 20 Greek freshwater 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 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. Growth in length has been described using the von Bertalanffy (1938) growth function (VBGF): 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).  Cite as: Kleanthidis, P.K., Stergiou, K.I., 2006. Growth parameters and length-length relationships of Greek 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. 69-77. Fisheries Centre, University of British Columbia [ISSN 1198-6727].  1  70  Growth parameters of Greek freshwater fishes, Kleanthidis, P.K., Stergiou, K.I.  RESULTS AND DISCUSSION Overall, 80 length-length relationships between total, fork and standard length were collected from the literature (Table 1), corresponding to 20 fish species and 1 hybrid from 9 lakes, 3 rivers, and 2 lagoons. The 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). 18  Number of stocks  16 14 12 10 8 6 4 2 0 0  400  800  1200  1600  2000  2400  Sample size (individuals) Figure 1 Sample size (i.e., number of individuals) for the cases shown in Table 1.  Table 2 summarizes the biological parameters collected for the 54 fish stocks, belonging to 22 species and one hybrid, from twelve lakes and two rivers in Greece. With respect to sampling frequency, samples were collected mainly on a monthly basis (in 22 cases) or based on a single sampling event (in 17 cases), and, to a lesser extent, on a seasonal or other basis (in 7 and 4 cases, respectively) (Table 2). Sampling frequency was not reported in 4 cases (Table 2). Age and growth were derived mainly from scale readings (in 43 cases) and to a lesser extent from 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 logtmax was logK = -0.939·logtmax + 0.038 (r2 = 0.39, n = 50, P < 0.05). The relationship 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).  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 freshwater fish species in Greece for which data are lacking. For this to yield reliable estimates, it will, however, be necessary to expand the data set by including more stocks and species. 16  a  12  Number of stocks  Number of stocks  16  8  4  0  8  4  20 40 60 Maximum lengt h (Lmax , cm)  1  3  5  30  c  12 8 4  0  11  13  15  0.50  0.60  d  20  10  20  40  60  0.00  80  0.10  0.20  0.30  0.40  von Bertalanffy K (1/yr)  von Bertalanffy Loo (cm) 16  e  16  Number of s toc ks  Number of s toc ks  9  0  0  20  7  Maximum age (t max , yr)  Number of stocks  Number of stocks  12  0 0  16  b  12 8 4 0  f  12  8  4  0 0.4  0.6  0.8  1  1.2  1.4  -4  (Lm ax )/(Loo)  -3  -2  -1  0  von Bertalanffy t o (yr)  Figure 2 Distribution of the values of (a) Lmax, (b) tmax, (c) Loo, (d) K, (e) Lmax/Loo, and (f) t0 for the Greek freshwater fish stocks analysed in this study.  0 -0.2  K (log; yr-1)  -0.4 -0.6 -0.8 -1 -1.2 0.9  1.1  1.3  1.5  1.7  1.9  Loo (log; cm )  Figure 3 Plot of logK vs. logL∞ for 54 Greek freshwater fish stocks.  Growth parameters of Greek freshwater fishes, Kleanthidis, P.K., Stergiou, K.I.  Abramis brama Abramis brama Abramis brama Alburnus alburnus Alburnus alburnus Alburnus alburnus Alburnus alburnus Alosa caspia vistonica Alosa caspia vistonica Alosa macedonica Alosa macedonica Alosa macedonica Alosa macedonica Alosa macedonica Alosa macedonica Aphanius fasciatus Aphanius fasciatus Aphanius fasciatus Aphanius fasciatus Aphanius fasciatus Aphanius fasciatus Aphanius fasciatus Aphanius fasciatus Aphanius fasciatus Atherina boyeri Barbus albanicus Barbus albanicus Barbus albanicus Barbus albanicus Barbus albanicus Barbus albanicus Chalcalburnus belvica Chalcalburnus belvica Chalcalburnus belvica Chalcalburnus chalcoides Chalcalburnus chalcoides Chalcalburnus chalcoides Chalcalburnus chalcoides Chalcalburnus chalcoides Chalcalburnus chalcoides Chalcalburnus chalcoides  Species  Area Lake Volvi Lake Volvi Lake Volvi Lake Koronia Lake Koronia Lake Mikri Prespa Lake Mikri Prespa Lake Vistonis Lake Vistonis Lake Volvi Lake Volvi Lake Volvi Lake Volvi Lake Volvi Lake Volvi Messolonghi lag. (Rebakia) Messolonghi lag. (Rebakia) Messolonghi lag. (Rebakia) Etolikon lag. (Astrovitsa) Etolikon lag. (Astrovitsa) Etolikon lag. (Astrovitsa) Messolonghi lag. (Alykes) Messolonghi lag. (Alykes) Messolonghi lag. (Alykes) Lake Trichonis Lake Kremasta Lake Kremasta Lake Kremasta Lake Kremasta Lake Kremasta Lake Kremasta Lake Mikri Prespa Lake Mikri Prespa Lake Mikri Prespa Lake Volvi Lake Volvi Lake Volvi Lake Volvi Lake Volvi Lake Volvi Lake Vistonis  Year 1989-91 1989-91 1989-91 1986-88 1986-88 1985 1985 1978 1978 1978 1978 1978 1978 1989-91 1989-91 1989-91 1989-91 1989-91 1989-91 1989-91 1989-91 1989-91 1989-90 1982 1982 1982 1982 1982 1982 1990-91 1990-91 1990-91 1983 1983 1983 1983 1983 1983 1983  Sex C C C C C C C C C M F C M F C M F C M F C M F C C M F C M F C C C C M F C M F C M 631 74 68 449 449 60 60 38 38 123 211 334 123 211 334 343 456 799 219 235 454 283 297 580 2575 369 369 369 90 280 370 90 280 370 260  N  Y TL SL SL TL TL FL TL FL SL TL TL TL TL TL TL TL TL TL TL TL TL TL TL TL FL TL TL TL TL TL TL TL SL SL FL FL FL FL FL FL FL  X FL FL TL FL SL SL SL TL TL SL SL SL FL FL FL SL SL SL SL SL SL SL SL SL TL FL FL FL SL SL SL FL TL FL TL TL TL SL SL SL TL  b 1.102 0.897 0.805 1.095 1.202 0.956 0.944 0.866 0.783 1.282 1.276 1.272 1.165 1.159 1.164 1.180 1.170 1.170 1.190 1.170 1.170 1.180 1.190 1.180 0.915 1.120 0.980 1.000 1.190 1.160 1.170 1.120 0.816 0.917 0.899 0.913 0.907 1.070 1.050 1.050 0.933  a 0.8459 -0.2175 -0.8822 0.1606 0.5005 0.0296 0.0458 0.5217 0.5505 -0.3980 -0.3710 -0.3120 -0.5670 -0.4960 -0.5640 0.0900 0.1130 0.1200 0.0800 0.1200 0.1400 0.1200 0.0900 0.1100 0.0823 0.1380 2.6650 2.1200 0.3440 0.7700 0.6400 -0.159 -0.164 -0.338 -0.0530 -0.3310 -0.2080 0.1850 0.4790 0.3980 -0.3290  Reference Valoukas and Economidis (1996) Valoukas and Economidis (1996) Valoukas and Economidis (1996) Politou (1993) Politou (1993) Crivelli and Dupont (1987) Crivelli and Dupont (1987) Economidis and Sinis (1986) Economidis and Sinis (1986) Sinis (1981) Sinis (1981) Sinis (1981) Sinis (1981) Sinis (1981) Sinis (1981) Leonardos (1996) Leonardos (1996) Leonardos (1996) Leonardos (1996) Leonardos (1996) Leonardos (1996) Leonardos (1996) Leonardos (1996) Leonardos (1996) Stoumboudi et al. (1997) Daoulas et al. (1987) Daoulas et al. (1987) Daoulas et al. (1987) Daoulas et al. (1987) Daoulas et al. (1987) Daoulas et al. (1987) Sinis and Petridis (1995) Sinis and Petridis (1995) Sinis and Petridis (1995) Kokkinakis (1992) Kokkinakis (1992) Kokkinakis (1992) Kokkinakis (1992) Kokkinakis (1992) Kokkinakis (1992) Kokkinakis (1992)  Table 1 Parameters of the length-length relationships (Y = a+bX, in cm; TL = total length; FL = fork length; SL = standard length) of 20 Greek freshwater fish species and one hybrid in nine lakes, three rivers and two lagoons. Sex (M = males; F = females; C = sexes combined; I = immature fish); N, sample size.  72  1  *  Area Lake Vistonis Lake Vistonis Lake Vistonis Lake Vistonis Lake Vistonis Evros River Evros River Evros River Evros River Rihios Stream Rihios Stream Lake Koronia Lake Koronia Lake Koronia Lake Doirani Lake Doirani Lake Kremasta Lake Kremasta Lake Kremasta Lake Kremasta Lake Kremasta Lake Kremasta Lake Mikri Prespa Lake Mikri Prespa Lake Mikri Prespa Lake Mikri Prespa Lake Mikri Prespa Lake Volvi Lake Volvi Lake Volvi Lake Lysimachia Lake Trichonis Lake Trichonis Lake Trichonis Aspropotamos stream Aspropotamos stream Aspropotamos stream Lake Kastoria Lake Kastoria  Year 1983 1983 1983 1983 1983 1983-84 1983-84 1983-84 1983-84 1984 1984 1975 1975 1975 1989-92 1989-92 1982 1982 1982 1982 1982 1982 1990 1985 1985 1985 1985 1978 1978 1978 1988-91 1988-91 1988-91 1981-82 1981-82 1981-82 1980 1980  Sex F C M F C M F I C C C C1 C2 C3 C C M F C M F C C C C C C C4 C5 C6 C M F C C7 C8 C9 C10 C11  N 253 513 260 253 513 230 158 9 397 489 489 24 120 230 317 317 245 60 60 60 60 102 110 21 314 32 159 191 342 342  Y FL FL FL FL FL SL SL SL SL FL TL SL SL SL SL FL TL TL TL TL TL TL SL FL TL FL TL SL SL SL TL TL TL TL FL FL FL SL SL  X TL TL SL SL SL TL TL TL TL SL FL TL TL TL TL TL FL FL FL SL SL SL FL SL SL SL SL TL TL TL FL SL SL SL SL SL SL TL TL  b 0.921 0.926 1.030 1.060 1.050 0.837 0.853 0.686 0.847 1.209 0.956 0.766 0.853 0.886 0.821 0.977 0.780 0.700 1.080 1.160 1.030 1.150 0.901 0.978 0.968 0.970 0.981 0.770 0.780 0.810 1.080 1.200 1.170 1.180 1.050 1.180 1.190 0.780 0.870  a -0.1080 -0.1970 0.8260 0.2940 0.5150 0.0224 -0.0198 0.2063 -0.0032 -0.2517 -0.3828 0.0245 -0.0303 -0.0870 0.5710 0.6300 4.6110 6.3920 0.1530 0.4450 2.3050 0.4080 0.0540 0.0201 0.0345 0.0242 0.0290 -0.1386 0.2647 0.0157 0.0880 -0.0410 0.1200 0.0830 0.6470 -1.1190 -1.2360 -0.0112 -0.1007  Reference Kokkinakis (1992) Kokkinakis (1992) Kokkinakis (1992) Kokkinakis (1992) Kokkinakis (1992) Kevrekidis et al. (1990) Kevrekidis et al. (1990) Kevrekidis et al. (1990) Kevrekidis et al. (1990) Neophitou (1988) Neophitou (1988) Papageorgiou (1977) Papageorgiou (1977) Papageorgiou (1977) Neophitou (1993a) Neophitou (1993a) Daoulas et al. (1987) Daoulas et al. (1987) Daoulas et al. (1987) Daoulas et al. (1987) Daoulas et al. (1987) Daoulas et al. (1987) Rosecchi et al. (1993) Crivelli and Dupont (1987) Crivelli and Dupont (1987) Crivelli and Dupont (1987) Crivelli and Dupont (1987) Papageorgiou (1979) Papageorgiou (1979) Papageorgiou (1979) Leonardos et al. (2000b) Psarras et al. (1997) Psarras et al. (1997) Psarras et al. (1997) Papageorgiou et al. (1983) Papageorgiou et al. (1983) Papageorgiou et al. (1983) Papageorgiou and Neophitou (1982) Papageorgiou and Neophitou (1982)  referred as Rutilus pleurobipunctatus. TL = 80-110 mm. 2 TL = 110-150 mm. 3 TL = 150-230 mm. 4 TL<120 mm. 5TL = 121-180 mm. 6 TL>181 mm. 7 SL<100 mm. 8SL = 101-175 mm. 9 SL>175 mm. 10TL<100 mm.11 TL>101 mm.  Chalcalburnus chalcoides Chalcalburnus chalcoides Chalcalburnus chalcoides Chalcalburnus chalcoides Chalcalburnus chalcoides Knipowitschia caucasica Knipowitschia caucasica Knipowitschia caucasica Knipowitschia caucasica Leuciscus cephalus Leuciscus cephalus Perca fluviatilis Perca fluviatilis Perca fluviatilis Perca fluviatilis Perca fluviatilis Phoxinellus pleurobipunctatus* Phoxinellus pleurobipunctatus* Phoxinellus pleurobipunctatus* Phoxinellus pleurobipunctatus* Phoxinellus pleurobipunctatus* Phoxinellus pleurobipunctatus* Pseudorasbora parva Rutilus rubilio Rutilus rubilio Rutilus rubilio X Alburnus alburnus Rutilus rubilio X Alburnus alburnus Rutilus rutilus Rutilus rutilus Rutilus rutilus Rutilus ylikiensis Salaria fluviatilis Salaria fluviatilis Salaria fluviatilis Salmo trutta fario Salmo trutta fario Salmo trutta fario Scardinius erythropthalmus Scardinius erythropthalmus  Species  Table 1 continued.  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D.  73  Growth parameters of Greek freshwater fishes, Kleanthidis, P.K., Stergiou, K.I.  Abramis brama Abramis brama Alburnus alburnus Alburnus alburnus Alburnus alburnus Alosa macedonica Alosa macedonica Alosa macedonica Atherina boyeri Barbus albanicus Barbus albanicus Barbus cyclolepis1 Chalcalburnus belvica Chalcalburnus belvica Chalcalburnus belvica Chalcalburnus chalcoides Chalcalburnus chalcoides Chalcalburnus chalcoides Chalcalburnus chalcoides Chalcalburnus chalcoides Chalcalburnus chalcoides Cyprinus carpio Leuciscus cephalus Leuciscus cephalus albus Leuciscus cephalus albus Leuciscus ‘svallize’ Leuciscus ‘svallize’  Species  Area Lake Volvi Lake Volvi Lake Koronia Lake Koronia Lake Koronia Lake Volvi Lake Volvi Lake Volvi Lake Trichonis Lake Kremasta Lake Kremasta Rihios stream Lake Mikri Prespa Lake Mikri Prespa Lake Mikri Prespa Lake Volvi Lake Volvi Lake Volvi Lake Vistonis Lake Vistonis Lake Vistonis Lake Vistonis Rihios stream Lake Mornos Lake Mornos Lake Kremasta Lake Kremasta  Year 1989-91 1989-91 1986-88 1986-88 1986-88 1978 1978 1978 1992-93 1982 1982 1984 1990-91 1990-91 1990-91 1983 1983 1983 1983 1983 1983 1973 1984 1987-88 1987-88 1982 1982  F M M M M M O O O U S S B M M M O O O O O O O U B B S S  Sex M F M F C M F C C M F C M F C M F C M F C C C M F M F  K 0.102 0.094 0.577 0.369 0.352 0.402 0.403 0.374 0.375 0.144 0.083 0.168 0.340 0.110 0.107 0.128 0.137 0.137 0.105 0.109 0.107 0.226 0.141 0.182 0.192 0.412 0.317 L∞ 45.2 50.7 12.4 15.3 15.5 18.3 18.5 18.8 12.3 25.8 37.9 37.1 14.3 33.5 34.9 28.1 25.6 26.2 33.6 32.6 33.0 88.0 45.8 27.9 28.7 20.4 22.9  t0 -0.351 -0.406 -0.153 -0.232 -0.284 -1.270 -1.494 -1.505 0.018 -0.805 -0.903 -0.179 -1.745 -1.405 -1.257 -2.133 -3.218 -2.628 -1.821 -1.654 -1.729 -0.769 -0.586 -0.832 -0.569 0.236 0.112  MEP FW FW R NL NL NL* NL* NL* FW NL* NL* NL* NL* NL* NL* NL* NL* NL* NL* NL* NL* NL* NL* NL* NL* NL* NL*  Lmax 40.5 42.2 11.6 14.9 14.9 33.1 11.0 20.1 28.8 26.0 13.6 22.0 22.0 19.4 22.5 22.5 20.0 23.7 23.7 76.0 37.0 18.4 20.8  tmax 14 14 4 7 7 10 4 9 14 6 4 8 8 7 8 8 6 8 8 7 8 5 5 6 7  SA S S S S S S S S S OB. S OB. S S S S S S S S S S S OB S S S S S  L B. FL B. FL B. TL B. TL B. TL B. TL B. TL B. TL B. FL B. FL B. FL B. FL B. FL B. FL B. FL B. FL B. FL B. FL B. FL B. FL B. FL B. TL B. FL B. FL B. FL B. FL B. FL Lmax/L∞ 0.90 0.83 0.94 0.97 0.96 1.76 0.89 0.78 0.76 0.70 0.95 0.66 0.63 0.69 0.88 0.86 0.60 0.73 0.72 0.86 0.81 0.90 0.91  Reference Valoukas and Economidis (1996) Valoukas and Economidis (1996) Politou (1993) Politou (1993) Politou (1993) Sinis (1981) Sinis (1981) Sinis (1981) Leonardos et al. (1993) Daoulas and Economidis (1989) Daoulas and Economidis (1989) Neophitou (1987) Sinis and Petridis (1995) Sinis and Petridis (1995) Sinis and Petridis (1995) Kokkinakis (1992) Kokkinakis (1992) Kokkinakis (1992) Kokkinakis (1992) Kokkinakis (1992) Kokkinakis (1992) Tsimenidis (1976) Neophitou (1988) Economou et al. (1997) Economou et al. (1997) Economou et al. (1991) Economou et al. (1991)  Table 2 Biological parameters for 22 Greek freshwater fish species and one hybrid, from twelve lakes and two rivers. K in year-1, L∞ in cm, and t0 in year, are the von Bertalanffy parameters. F, frequency of sampling (M = monthly; S = seasonal; B = bimonthly; O = one single sampling; T = two samplings; U = undefined); Sex (M = males; F = females; C = sexes combined); MEP, method used for the estimation of the von Bertalanffy parameters (FW = Ford-Walford plot: Sparre et al. 1989; R = Rafail, 1973; NL = non-linear regression; asterisk denotes that von Bertalanffy parameters were calculated in this study); Lmax and tmax, maximum body length in cm and age in year respectively; SA, element used for ageing (S = scales; OB = opercular bones; V = vertebrae; FS = fin spines); L, length used for the estimation of the parameters (B = back-calculated; TL = total length; FL = fork length; SL = standard length).  74  Lake Mikri Prespa Lake Volvi Lake Lysimachia Aspropotamos stream Lake Trichonis-Lysimachia Lake Trichonis-Lysimachia Lake Trichonis-Lysimachia Lake Trichonis Lake Trichonis Lake Lysimachia Lake Lysimachia Lake Kastoria  Rutilus rubilio X Alburnus alburnus Rutilus rutilus Rutilus ylikiensis Salmo trutta fario Scardinius acarnanicus3 Scardinius acarnanicus3 Scardinius acarnanicus3 Scardinius acarnanicus Scardinius acarnanicus Scardinius acarnanicus Scardinius acarnanicus Scardinius erythrophthalmus 1985 1978 1981 1977-79 1977-79 1977-79 1980  Year 1989-92 1976 1976 1976 1982 1982 1984-85 1991-92 1978-79 1978-79 1983-84 O O S O M M M M M M M O  M M M  F U O O O S S U  C C C C M F C M F M F C  M F C  Sex C M F C M F C  0.157 0.081 0.084 0.218 0.159 0.136 0.134 0.126 0.134 0.112 0.134 0.178  0.124 0.113 0.275  K 0.317 0.164 0.150 0.159 0.136 0.119 0.245  25.7 33.3 32.3 26.7 34.5 42.6 42.6 42.8 39.7 47.1 41.4 23.9  26.2 30.0 20.6  L∞ 23.9 23.7 24.5 23.9 26.2 28.9 11.8  -0.902 -1.295 -0.126 -0.417 -0.155 0.094 0.055 -0.830 -1.180 -0.790 -0.700 -0.521  -1.219 -1.040 -1.216  t0 -1.260 -0.952 -1.212 -1.086 -0.640 -0.748 -1.373  NL* NL* NL NL* NL* NL* NL* NL NL NL NL NL*  NL* NL* NL*  MEP NL* NL* NL* NL* NL* NL* NL*  18.5 23.0 28.5 24.0 27.0 33.0 33.0 18.0  18.8 25.8 20.5  Lmax 23.0 23.0 17.6 20.1 9.5  6 12 10 8 7 7 7 10 11 7  7 9 8  tmax 5 9 9 9 7 9 3  S S S S S S S S S S S S  S S OB  SA OB. S S S S OB OB S  B. FL B. TL B. FL B. SL B. TL B. TL B. TL B. FL B. FL B. FL B. FL B. TL  B. FL B. FL B. FL  L B. TL B. TL B. TL B. TL B. FL B. FL O. FL  0.099 44.5 -0.305 B. TL NL* 1977-79 Lake Trichonis-Lysimachia M M 39.0 10 FS 0.100 46.4 -0.337 B. TL NL* 1977-79 Lake Trichonis-Lysimachia M F 40.0 10 FS 0.100 45.5 -0.321 B. TL NL* 1977-79 Lake Trichonis-Lysimachia M C 40.0 10 FS 0.205 32.4 -0.401 B. FL NL* 1988 Lake Pamvotis M C 29.0 10 S OB.S.V 0.123 54.0 -0.790 NL* 1988 Lake Vegoritis T C 36.6 7 O. TL 1 referred as Barbus meridionalis. 2 referred as Rutilus pleurobipunctatus. 3referred as Scardinius erythrophthalmus. 4referred as Parasilurus aristotelis.  Silurus aristotelis4 Silurus aristotelis4 Silurus aristotelis4 Tinca tinca Tinca tinca  Lake Trichonis Lake Trichonis Lake Pamvotis  Rutilus rubilio Rutilus rubilio Rutilus rubilio  Doirani Koronia Koronia Koronia Kremasta Kremasta Mikri Prespa  Area Lake Lake Lake Lake Lake Lake Lake  Perca fluviatilis Perca fluviatilis Perca fluviatilis Perca fluviatilis Phoxinellus pleurobipunctatus2 Phoxinellus pleurobipunctatus2 Pseudorasbora parva  Table 2 continued. Species  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D.  0.88 0.86 0.88 0.90 0.68  0.72 0.70 0.88 0.90 0.78 0.77 0.77 0.75  0.72 0.86 1.00  Lmax/ L∞ 0.96 0.96 0.67 0.70 0.81  Daoulas (1981) Daoulas (1981) Neophitou and Theochari (1989) Crivelli and Dupont (1987) Papageorgiou (1979) Leonardos et al. (2000b) Papageorgiou et al. (1983) Iliadou (1981) Iliadou (1981) Iliadou (1981) Leonardos et al. (2000a) Leonardos et al. (2000a) Leonardos et al. (2000a) Leonardos et al. (2000a) Papageorgiou and Neophitou (1982) Iliadou (1981) Iliadou (1981) Iliadou (1981) Neophitou (1993b) Sinis et al (1999)  Reference Neophitou (1993a) Papageorgiou (1977) Papageorgiou (1977) Papageorgiou (1977) Daoulas et al. (1987) Daoulas et al. (1987) Rosecchi et al. (1993)  75  76  Growth parameters of Greek freshwater fishes, Kleanthidis, P.K., Stergiou, K.I.  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. Daoulas, C., Koussouris. T., Psarras, T., 1987. Ecology and Possibilities of Fisheries Management of the Artificial Lake of Kremasta. Special Publication National Center of Marine Research. (Greece), 12 [in Greek, with English abstract]. Economidis, P.S., Sinis, A.I., 1986. Situation taxinomique et comparaisons des Aloses (Pisces, Clupeidae) provenant des lacs Volvi et Vistonis (Grèce). Description d’une nouvelle sous-espèce: Alosa caspia vistonica. J. Nat. Hist. 20, 723-734. Economou, A.N., Daoulas, C., Economidis, P., 1991. Observations on the biology of Leuciscus ‘svallize’ in the Kremasta reservoir (Greece). Hydrobiologia 213, 99-111. 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, 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 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., 2000. Fishbase 2000: Concepts, Design and Data Sources. ICLARM, Los Baños, Philippines. Gayanilo Jr, F.C., Sparre, P., Pauly, D., 1994. The FAO-ICLARM Stock Assessment Tools (FISAT): User’s Guide. FAO Computerized Information Series, Fisheries, Rome. 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 abstract]. Kevrekidis, T., Kokkinakis, A.K., Koukouras, A., 1990. Some aspects of the biology and ecology of Knipowitschia caucasica (Teleostei: Gobiidae) in the Evros Delta (north Aegean Sea). Helgol. Meeresunters. 44, 173-187. Kokkinakis, A., 1992. Comparative Study of the Biology and Dynamics of the Fish Chalcalburnus chalcoides macedonicus Stephanidis, 1971 (Pisces: Cyprinidae) of the Systems Volvi and Vistonis. University of Thessaloniki, Thessaloniki, Greece, Doctorate Thesis [in Greek, with English abstract]. Leonardos, I., 1996. Population Dynamics of Toothcarp (Aphanius fasciatus Nardo, 1827) in the Mesolongi and Etolikon Lagoons. University of Thessaloniki, Thessaloniki, Greece, Doctorate Thesis [in Greek, with English abstract]. Leonardos, I., Kokkinidou, A., Agiannitopoulos, A., Giris, S., 2000a. Population structure and reproductive strategy of an endemic species Scardinius acarnanicus (Stephanidis, 1939) in two W. Greece Lakes (Lysimachia and Trichonis). In: Proceedings of the Ninth Hellenic Ichthyological Congress, 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.  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), 3847. Neophitou, C., 1993b. Some biological data on tench (Tinca tinca (L.)) in Lake Pamvotida (Greece). Acta Hydrobiol. 35(4), 367-379. 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, Perca fluviatilis (L.) in the lake of Ag. Vasileios. Thalassographica 1(3), 245-265 [in Greek]. Papageorgiou, N., 1979. The length-weight relationship, age, growth and reproduction of the roach Rutilus rutilus (L.) in Lake Volvi. J. Fish Biol. 14, 529-538. 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, C.N., Vlachos, C.G., 1983. The age, growth, and reproduction of brown trout (Salmo trutta fario) in the Aspropotamos stream. Acta Hydrobiol. 25/26(3/4), 451-467. Pauly, D., 1998. Tropical fishes: patterns and propensities. J. Fish Biol. 53 (Suppl.), 1-17. Politou, C-Y., 1993. Biology and Dynamics of the Fish Alburnus alburnus (L., 1758) in Lake Koronia. University of Thessaloniki, Thessaloniki, Greece, Doctorate Thesis [in Greek, with English abstract]. Psarras, Th., Barbieri-Tseliki, R., Economou, A.N., 1997. First data on the feeding and biology of reproduction of Salaria fluviatilis. In: Proceedings of the Fifth National Symposium on Oceanography and Fisheries, 15-18 April 1997, Kavala, Greece, pp. 261-264 [in Greek, with English abstract]. Rafail, S.Z., 1973. A simple and precise method for fitting a von Bertalanffy growth curve. Mar. Biol. 19, 354-358. Rosecchi, E., Crivelli, A.J., Catsadorakis, G., 1993. The establishment and impact of Pseudorasbora parva, an exotic fish species introduced into Lake Mikri Prespa (north-western Greece). Aquatic Cons. Mar. Freshw. Ecosys. 3, 223231. Sinis, A., 1981. L’Autoecologie de l’Espèce Endémique Alosa (Caspialosa) macedonica (Vinciguerra) (Pisces: Osmeridae), du Lac Volvi. Aristotle University of Thessaloniki, Doctorate Thesis (in Greek). 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, 453465. 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 Risso, 1810) from Trichonis Lake (Greece). In: Proceedings of the Fifth National Symposium on Oceanography and Fisheries, 15-18 April 1997, Kavala, Greece, pp. 257-260 [in Greek, with English abstract]. Tsimenidis, N., 1976. The relationship between fish length and the length of the operculum for the carp in Lake 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.  78  Growth and population trends in Grand Canyon native fishes, Walters, C. et al.  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  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  SWCA Inc., 56 West 400 South St., Suite 201, Salt Lake City, Utah 84101, USA; Email: rvaldez@swca.com  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 releases from Glen Canyon Dam (GCD) may have caused declines in native 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. discobolus). Presumption of continued decline has prompted expensive proposals to restore more favorable physical habitat conditions for these fishes, by altering operation of Glen Canyon Dam (warm water 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 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). While abundances may have dropped initially after GCD was filled, there is little evidence to support 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 distribution monitoring, and 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].  1  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D.  79  index sampling for juvenile and adult densities using various fishing gears. For most sampling sites, the size structure data show evidence of continued recruitment, and no clear trend toward larger body sizes (older fish) as would be expected from non-recruiting adult populations. There are no clear trends in the density sampling data, though changes in methods and sites make long-term comparisons suspect. Markrecapture estimates have been obtained since the late 1980s for the humpback chub ‘subpopulation’ spawning in the Little Colorado River; these estimates are highly variable and show no consistent trend. Low ratios of juveniles to adults in size-frequency samples have been cited as evidence of low recruitment (Valdez and Ryel, 1995), but this ratio comparison is not valid since juveniles and adults are collected by different methods, in different habitats, with unknown differences in sampling rates (proportions of total population collected by each method). This paper shows that tag recovery data from the 1990s are consistent with the hypothesis that chub, flannelmouth, and bluehead sucker populations are relatively stable, and are not consistent with an assumption of continued rapid population decline. The analysis is based on substantial samples of fish that were tagged and subsequently (0.5-6.8 years later) recovered for growth and survival estimation. From changes in size of these fish and the assumption that fish grow according to von Bertalanffy curves, we can back-calculate an apparent age at tagging for each individual, and construct an age distribution for each tag sample. Assuming the tagged fish were a representative sample of the population age structure (at least for older fish), the age distributions can then be compared to expected age distributions under various hypotheses about recruitment success and rate of population decline. Specifically, the age distribution data give estimates of apparent annual mortality rate, and these estimates are surprisingly close to mortality rate predictions from growth parameters (Pauly, 1980) though lower than rates estimated from tag-recapture models. Were populations declining during the 1980s, we would expect considerably higher apparent survival rates (relatively older fish and fewer young ones).  METHODS The tag-recovery data used in this analysis are from a variety of tagging programs carried out by contractors to the Grand Canyon Monitoring and Research Center (Arizona Department of Fish and Game, Arizona State University, U.S. Fish and Wildlife Service) mainly during 1991-94, archived on a database at Arizona State University. We also analyzed much smaller data sets collected by Arizona Department of Fish and Game during the 1980s (ADFG, 1987), but do not report those results separately because sample sizes were only large enough for growth analysis and growth parameter estimates were very similar to those from the larger, more recent data set. Though fish were tagged and recovered throughout the Grand Canyon from Glen Canyon Dam to Lake Mead, most of the tagging and recovery was 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 (at least 6 months of growth) and 148 of 184 bluehead sucker records (again at least 6 months growth). The very 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 (Hilborn and Walters, 1992): La = L∞·(1-e-K(a-ao))  … 1)  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 fish tagged at lengths LS are recovered after time periods T at lengths LR, Fabens (1965) showed that the von Bertalanffy model can be written as: LR = LS + (L∞-LS)·(1 - e-KT)  … 2)  Growth and population trends in Grand Canyon native fishes, Walters, C. et al.  600  Age information from length at tagging  500 400 Length  This equation provides a nonlinear regression model for estimation of L∞ and K given a set of LR, LS 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 this method have been suggested to account for individual variation in the L∞, K parameters, most notably an extension of the Fabens model that accounts at least for variation in L∞ (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 individual fish. In this method, the growth observation Gi = LR - LS for each individual i is assumed to be Gi = (L∞ + Ei - LS) · (1 - e-KT), which is the Equation (2) prediction, but with a mean-zero normally-distributed deviation Ei in the asymptotic size for individual i from the population asymptotic average size L∞, i.e., L(i)∞ = L ∞ + Ei.  300 200 100 0 0  5  10  15  20  Age 600  Additional age information from length at recovery and time to recovery  500 Length  80  400 300 200 100 0 0  5  10  15  20  Age  Additionally, for chub we constrained the estimates Figure 1 Illustration of how tag-recapture data can provide improved estimates of fish age. Individual of L∞, K to pass through independent estimates of variation in growth curves implies that age is highly age-1 length L1 based on scale analysis and analysis uncertain given only length at first capture (top panel). of modal progression in seasonal juvenile length But given length at second capture and time until that frequency samples (chub estimate from Valdez and capture, the individual’s growth curve and hence age Ryel, 1995). For flannelmouth suckers, we were can be determined much more accurately, at least for unable to use such a constraint to improve the younger fish. 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 likely 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∞). Wang et al. (1995) suggest that variation among individuals in K should generally be quite 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, LR data and K using the following relationships derived from Equations (1) and (2), provided the population K and L∞ are known: 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 given 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  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D.  Pa = vaNa/(ΣvaNa)  81  … 5)  where va is an age-specific vulnerability to sampling. That is, Pa is the vulnerable number vaNa of age a fish in the population, divided by the total vulnerable population size summed over ages, ΣvaNa. Based on inspection of the data and discussions with experienced Grand Canyon scientists about how fish shift their distributions and become vulnerable to sampling as they grow, we assume that vulnerability va increases asymptotically from 0 to 1.0 with age, according to a sigmoid function. For convenience, we assume the following form: va = av/(ahv+av)  … 6)  where ah is age at 50 % vulnerability (age where v = 0.5) and the power parameter v represents steepness of the sigmoid function (high v values imply a steeper, more ‘knife-edged’ vulnerability curve around the age ah). Luckily, we found the results presented below to be largely insensitive to choices ah and v. Thus, and for maximum likelihood analysis, we elected to assume va = 0 for ages 0 and 1, and va = 1 for ages 2 and older (4 and older for bluehead sucker); this simplification can cause modest underestimation of mortality rates and hence favor the hypothesis that population size has been declining (see below). In a stable population with annual age-independent survival rate S (= e-M, where M is the annual instantaneous natural mortality rate), it is easy to predict the Na of Equation (5). These should be related according to Na = N1Sa-1 = N1e-M(a - 1)  … 7)  where N1 is average age 1 recruitment. We can take N1 = 1 without loss of generality since absolute numbers cancel in the calculation of P in Equation (5). In contrast, for a population that is growing or declining exponentially at annual rate r (Nt+1 = Nter), the Na should be related by Na = Na-1Se-r = Na-1e-M - r  … 8)  i.e., abundances at successive ages should contain effects of both survival rate and population growth. The apparent survival rate (Na/Na-1, Pa for high a) from an age composition sample will be higher than S if the population is declining (r<0). If we examine only data for ‘fully recruited’ ages (va = 1, a≥amin), eqs. (7) and (8) imply that the stable age proportions should vary as Pa = e-(M+r)(a - amin)(1 - e-M - r) if age is treated as a discrete variable, or Pa = e -(M+r) (a - amin) (M + r) if age is treated as continuous (here the terms 1 - e-M - r and M+r represent the sum and integral of Na over ages). For initial analyses of age vulnerability patterns, we defined a simple sum of squares criterion to compare goodness of fit of alternative hypotheses about Pa to the estimated age proportions pa: SS = Σa(pa - Pa)2  … 9)  We then used a simple search procedure (Solver in Microsoft Excel) to search for estimates of annual survival rate S and the vulnerability parameters ah and v that would minimize SS, with Pa calculated using the stable population ‘null’ predictions of Na from Equation (7). This approach allows definition of at least ranges of S that would predict sample age proportions anywhere near the estimated ones, and it indicated that the assumption of ‘knife-edge’ selection (va = 1, a≥amin) is reasonable for the chub and flannelmouth data. S estimates much higher than obtained by other methods would imply r<<0, i.e., the sample pa had come from a declining population (or some unknown problem with those independent estimates of S). S estimates much lower than from other methods might indicate an increasing population, but might also indicate some decrease in vulnerability of larger, older animals to tagging or recapture (failure in monotonic vulnerability assumption, Equation (6)).  Growth and population trends in Grand Canyon native fishes, Walters, C. et al.  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).  No m easurem ent errors, no K variation 40 Estimated age (years)  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 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.  35 30 25 20 15 10 5 0 0  10  20  30  40  True age (years)  Measurem ent errors, no K variation 40 Estimated age (years)  82  35 30 25 20 15 10 5 0 0  10  20  30  40  True age (years)  RESULTS  Measurem ent errors, high K variation 40 Estimated age (years)  A key advantage of the likelihood approach is that it allows use of Bayesian assessment methods to provide probability distributions for the parameters, integrating uncertainty over sample variation in individual growth rates and over a range of possible estimates for the auxiliary data L1. This method gives a more conservative assessment of statistical uncertainty (wider statistical limits) than would simpler approaches like bootstrapping, since there is no simple way to include uncertainty about L1 in such procedures. Standard 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 would 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 parameters as well.  35 30 25 20 15 10 5 0 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, especially for older fish, caused by having length measurement errors with 4 mm standard deviation 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 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).  Results from fitting the growth data are presented in Table 1 and Figures 3-4. Despite much obvious variation in measured growth rates, the maximum likelihood method 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  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  Bluehead sucker  600 Length (mm)  Length (mm)  500 400 300 200  Predicted  100  Observed  0 0  5  10  15  20  25  30  35  400 350 300 250 200 150 100 50 0  Predicted Observed 0  5  10  15  20  25  30  Apparent age (years)  Apparent age (years)  Length (mm)  Flannelmouth sucker 800 700 600 500 400 300 200 100 0  Predicted  Figure 4 Fits of the von Bertalanffy growth curve to length at first capture vs. apparent age at that capture, for Grand Canyon fishes.  Observed 0  2  4  6  8  10  12  14  Apparent age (years)  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  84  Growth and population trends in Grand Canyon native fishes, Walters, C. et al.  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 Humpback Flannelmouth Bluehead procedure chub sucker 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 Age regression 0.11 0.22 0.40 Z (S = e-Z) 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 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.  Posterior probability  Posterior probability  Apparent mortality rate  0.3  HBC FMS BHS  0.0 0  0.2 0.4 Instantaneous rate (Z)  0.6  0.6 HBC FMS  0.3  0.0 200  BHS  300  400  500  Asymptotic length (mm)  Growth K parameter  0.6  HBC FMS BHS  0.3  0.0 0.1  0.3  0.5 K  Asymptotic body length  Posterior probability  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 estimates are close to estimates of M from the Pauly (1980) equation for all species.  60  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 Appendix 1. HBC = humpback chub, FMS = flannelmouth sucker, BHS = bluehead sucker.  Fishes in Databases and Ecosystems, Palomares, M.L.D., Stergiou, K.I., Pauly, D.  85  Humpback chub  Bluehead sucker Observed  0.12  Proportion of sample  Proportion of fish  0.14  Predicted  0.1 0.08 0.06 0.04 0.02 0 0  5  10  15  20  25  0.2 Observed  0.15  Predicted 0.1 0.05 0 0  30  5  10  15  Proportion of sample  Flannelmouth sucker 0.1  Observed  0.08  Predicted  0.06 0.04 0.02 0 0  5  10  15  20  25  Age (yrs.)  25  30  35  Figure 6 Apparent age distributions at age of first tagging for Grand Canyon fishes. Curves show age distribution expected for stable population with annual 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.  Humpback chub age proportions by year of tagging, including all fish with >0.5 yr time to recapture  Humpback chub age proportions by year of tagging  0.16  0.16 1991, n=380  0.14 0.12  1993, n=539 1994, n=59  0.1  1991, n=912  0.14  1992, n=667  Proportion of fish  Proportion of fish  20  Age (yrs.)  Age (yr.)  0.08 0.06 0.04 0.02  1992, n=1227  0.12  1993, n=998 1994, n=203  0.1 0.08 0.06 0.04 0.02  0  0 0  5  10  15  20  Apparent age (years)  25  30  0  5  10  15  20  25  30  Apparent age (years)  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  86  Growth and population trends in Grand Canyon native fishes, Walters, C. et al.  than we have estimated. In fact the opposite appears to be the case for flannelmouth suckers; juveniles rearing in shoreline/backwater refuge areas appear to have lower first year growth rates than predicted by the von Bertalanffy model. It is obviously frustrating to be unable to make a more definitive statement about population trends, but uncertainties about natural survival rates and age-selectivity in sampling make this impossible. However, it is clear from the distributions of apparent ages that all three populations were achieving at least some recruitment well after construction of GCD; most of the fish tagged were almost certainly less than 20 years. old, implying they were recruited well after construction of GCD. That is, growth of tagged fish implies that adult size distributions cannot be interpreted simply as closed cohorts of old fish left over from healthy populations prior to GCD, though such closed cohorts apparently still exist in a few locations. It is understandable that Grand Canyon policy has been to avoid killing any native fishes for direct analysis of age and growth. There will likely be continued reliance on analysis methods based on tag recovery and 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 (<200 mm) in the tagging programs even if tagging these fish causes a relatively high mortality risk; and (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 reliable index of changes in mortality and recruitment rates. Age estimates vary widely among individuals of any adult length interval, i.e., growth rates vary widely among individuals so that length alone is a very poor 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.  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. 0CS-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 tagrecapture 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)  88  Growth and population trends in Grand Canyon native fishes, Walters, C. et al.  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 eliminated from the likelihood function above by assuming mean length at age 1, L1 is measured exactly so K = -log(1 - L1/L∞). But uncertainty in the independent estimate of L1 can be easily incorporated 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 small σ1 (L1 known perfectly) becomes equivalent to removing K from the estimation by calculating it as log(1 - L1/L∞). 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.  APPENDIX 2. MONTE 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 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 simulation tests revealed a number of potential biases. First, the maximum likelihood estimates of L∞ are biased slightly (5%) downward, and K given L1 is biased upward by the same relative amounts when 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 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. Sixth, measurement errors of the magnitude apparent in the data can badly bias the estimates, unless the data are ‘censored’ 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.  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 rate at which oxygen is extracted via the gills and transported into the body of fish limits, at any time, their 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 fish grow. Emphasis is given to the fact that it explains some of the observed direct and indirect effects of pond aeration on fish growth and related 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).  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].  20  90  Lake and pond aeration on fish growth, Pauly, D.  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 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.,  ... 1) 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 -  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  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  ... 5)  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 Figure 1 Relationships between an index of the gill area (a’ in increasing the level of activity required to Equation 2) and the growth performance of 35 species of teleosts secure the required food) will have the and two sharks as expressed by their value of Φ (see Equation 4; effect of reducing W∞ (Figure 2B). This adapted from data in Pauly 1979). 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 amount of food required to produce a given amount of fish flesh. This is related to a concept commonly used in fisheries science, the food conversion efficiency (K1) of Ivlev (1966) through l/FCR = K1, the latter being defined, for any time interval by: K1 = growth increment/food consumption  ... 6)  Various approaches exist to relate K1 and growth (e.g., Paloheimo and Dickie, 1966), or more precisely equation (6), with the VBGF as represented in equation (3). One of these was presented by Temming (1994a), i.e., K1 = A·(1 -W/W∞)l-d  ... 7)  where K1 is the food conversion efficiency at weight W, W∞ is as defined in equation (3), d is set at 2/3, as implied in the version of the VBGF presented here [there are other version of the VBGF, incorporating other, more realistic values of d, see Pauly, 1981; Temming, 1994b; and Essington et al., 2001], and A is a factor constrained between zero and one, and expressing the fraction of the ingested food that is available for protein synthesis. It can be expected that A will be related to the nitrogen content of the food (Pandian and Marian, 1985).  92  Lake and pond aeration on fish growth, Pauly, D.  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 oxygenlimited. 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).  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.  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  Common carp  Cyprinus carpio  Silver carp  Artistichthys nobilis  Bighead carp  Hypophthalmichthys molitrix  Japanese eel  Anguilla japonica  Tilapia Channel catfish  Cichlidae Ictalurus punctatus  Hybrid carp  Ctenopharyngodon idella X  Aristichthys nobilis a) food consumption and/or conversion  Location  Szarvas, Hungary Szarvas, Hungary Szarvas, Hungary Lukang, Taiwan Singapore Alabama, USA Florida, USA  Increase of: Cona -  Growth -  Survival -  -  -  -  -  -  √  Harvest √  Profits √  Source Abdul Amir (1988)  -  √  Abdul Amir (1988)  -  -  √  Abdul Amir (1988)  √  -  -  √  Anon. (1988a)  √  √ √  √ √  √  √ √  -  √  √  -  -  Anon. (1988b) Hollerman and Boyd (1980) Shireman et al. (1983)  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 assume a lake with a warm epilimnion, a cool (10°), well-oxygenated mesolimnion and a small, oxygenfree 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 stratification 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 feeding forays into 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 K(10°C) = 0.5 year-1, one can also estimate the value of K corresponding to W∞(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, and thus to reexamine as an example, the work of Coutant (1985, 1987, 1990), which documents a habitat – Chesapeake Bay – then in need of destratification, i.e., where striped bass – especially the large ones, for which oxygen supply was a problem – had become unable to grow and feed in certain areas of their original habitat.  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 curves resulting from the two scenarios in the text.  CONCLUSION The theory presented above, based on first principles and easily verifiable assumptions, can be easily developed to provide a comprehensive framework both for interpreting aeration experiments and for predicting potential effects of aeration in various water bodies. Without such development of the above theory, or of a modification thereof, the observed impacts of aeration on ponds and lakes will continue to be perceived as isolated facts, and not as elements of a wellarticulated system of principles allowing strong inferences and rapid advances.  94  Lake and pond aeration on fish growth, Pauly, D.  ACKNOWLEDGEMENTS 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 Pond Aeration’. The book project fell through, but this contribution was rescued and included here, only minimally updated (3 new references), as I feel it has not, 10 years later, lost any of its pertinence, notably for the interpretation of growth data in FishBase.  REFERENCES Abdul Amir, A., 1988. Aeration proves its worth with carp. Fish Farmer-Internat. File 2(1), 4-5. 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. 65-68. Fisheries Centre, University of British Columbia. Anon., 1988a. The miracle of Taiwan's eel culture. Aqua-O2 News 1(1), 4. Anon., 1988b. Veteran nets greater tilapia yields. Aqua-O2 News 1(2), 6. 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, 6372. Carlander, K.D., 1969. Handbook of Freshwater Fishery Biology. Vol. 1. The Iowa State University Press, Ames, Iowa. Carlander, K.D., 1977. Handbook of Freshwater Fishery Biology. Vol. 2. The Iowa State University Press, Ames, Iowa. 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. Coutant, C.C., 1985. Striped bass, temperature, and dissolved oxygen: a speculative hypothesis for environmental risk. Trans. Amer. Fish. Soc. 114(1), 31-61. Coutant, C.C., 1987. Thermal preference: when does an asset become a liability? Environ. Biol. Fishes 18(3), 161-172. Coutant, C.C., 1990. Temperature-oxygen habitat for freshwater and coastal striped bass in a changing climate. Trans. Amer. Fish. Soc. 119, 240-253. Essington, T.E., Kitchell, J.F., Walters, C.J., 2001. The von Bertalanffy growth function, bioenergetics, and the consumption rates of fish. Can. J. Fish. Aquat. Sci. 58, 2129-2138. Fast, 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, 419475. 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.  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.  Haggan, N., Beattie, A., Pauly, D. (eds) (1999) Back to the future: reconstructing the Hecate Strait ecosystem. Fisheries Centre Research Reports 7(3):65pp.  Pitcher, T., Chuenpagdee, R. (eds) (1993) Decision making by commercial fishermen: a missing component in fisheries management? Fisheries Centre Research Reports 1(2):75pp.  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.  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) Massbalance 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.  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 humanmade 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.  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.  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.  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.  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.  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.  Pauly, D., Palomares, M.L.D. (eds) (2002) Production systems in fishery management. Fisheries Centre Research Reports 10(8):28pp.  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.  Haggan, N., Brignall, C., Wood, L. (eds) (2003) Putting fishers’ knowledge to work conference proceedings. Fisheries Centre Research Reports 11(1):504pp.  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.  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.  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.  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.  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.  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.  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 2202 Main Mall, Vancouver, BC V6T 1Z4 Canada  Phone: 604 822-2731; Fax: 604 822-8934 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.52383.1-0074760/manifest

Comment

Related Items