One of the antecedents of FishBase was the work and vision of Walter Fischer, FAO, who inspired experts throughout the world to collaborate on the production of FAO’s first set of Identification Sheets (Fischer 1973) and their numerous successors, and to publish, through FAO’s Species Identification and Data Programme, an extremely useful series of FAO Species Synopses and FAO Species Catalogues (Fischer 1976). Walter Fischer also perceived the need for a global database of basic information on the exploited fish and invertebrates of the world, and this led to the development of FAO’s SPECIESDAB database (Coppola et al. 1994, see below).

Daniel Pauly had followed these developments with keen interest: he had been, since the days of his field work in Indonesia, in the mid-1970s, a user of FAO products, and he knew their worth, especially for work in the tropics. He had assembled a card-index of most of the population dynamics data then available for fishes and, inspired by Walter Fischer’s vision, he suggested, in 1987, that these data should be transferred to a standardized and continuously updated database which he intended to use for his own research and to make available to others through what was then known as the ‘ICLARM Software Project’ (Pauly et al. 1995).

He discussed this idea with Rainer Froese, then at the Institut für Meereskunde, Kiel, Germany, who was exploring the capabilities of computers and video systems in general and artificial intelligence (AI) in particular for identification purposes and who had just finalized an expert system for the identification of fish larvae (Froese and Schöfer 1987; Froese 1988, 1989, 1990; Froese et al. 1989, 1990a, 1990b; Froese and Papasissi 1990).

The FishBase idea was first proposed by Daniel Pauly in ICLARM’s 1988 five-year plan (ICLARM 1988), already with a widened scope, as follows:

"The information gap [presently hobbling] on tropical fisheries probably cannot be bridged using [only] classical means, such as maintaining extensive libraries, encouraging interlibrary loans and electronic data exchange. Rather it can be expected that shortage of funds for such classic activities will become increasingly problematic, and hence increase the isolation of scientists working on tropical resources from the mainstream of their science and from reference materials.

It is proposed to alleviate this problem by developing a self-sufficient database implemented on standard microcomputers [...] which would provide key-facts and information extracted from the literature. It would largely replace stock assessment text books. The database would constitute an ‘expert system’ (an artificial intelligence type information system in which commands or queries can be made in simple English).

Rainer Froese subsequently tried to implement such a system in the AI programming language PROLOG. However, when he realized that this would entail the handling of possibly more than 1,000 variables at the source code level, he discarded this option and, rather, reviewed the relational databases available at that time (dBase, FoxBase, Clipper, Paradox, Oracle, Btrieve, Ingres). He found that these databases were either limited, demanded a lot of programming, could not be distributed without royalties, or were not really meant for PCs. It was by chance that he came across DataEase, a little known DOS database software that combined relational power with exceptional ease of use.

When Rainer Froese was invited by Daniel Pauly to visit ICLARM in the end of 1988, he brought with him the basic design of what was to become FishBase, implemented in DataEase. This design was fine-tuned, table by table, field by field, in a series of meetings with ICLARM scientists Daniel Pauly, Roger Pullin, Ambekar Eknath, Astrid Jarre and Maria Lourdes D. Palomares. Also ICLARM programmers Felimon Gayanilo, Jr. and Mina L. Soriano had a critical look at the design of the database, and, after long discussions, agreed that:

• using a commercial relational database software was a better approach than programming the system from scratch; and

• DataEase would be a good choice for prototyping FishBase until a better software was found (Froese et al. 1988).

In January 1989, Daniel Pauly and Rainer Froese visited FAO, Rome, to coordinate efforts on FishBase and SPECIESDAB (Coppola et al. 1994), a database conceived by Walter Fischer (see above), implemented in dBase by Rino Coppola, and compiled by Nadia Scialabba. SPECIESDAB contained scientific and vernacular names as well as basic, ecological and fisheries information on the species covered in the FAO Species Catalogues. Work on SPECIESDAB had started in 1986. It already covered all catalogues then published so far. The visit led to a Letter of Agreement signed on 15 November 1989 between ICLARM and FAO. It stated that ICLARM and FAO would collaborate in the development of FishBase and would both be entitled to distribute it. This agreement gave FishBase a firm footing and probably helped in attracting the first grant.

Following an initiative of Rainer Froese, the European Commission supported the project in October 1989 with a first grant that allowed the hiring of an additional research assistant (Crispina Binohlan) for data encoding (also Susan M. Luna was assigned full time to the project while Belen Acosta returned to her previous assignment), the purchase of computer equipment (ICLARM’s first Local Area Network), and another visit of Rainer Froese to ICLARM in December 1989, to supervise data entry and to write a larger proposal for funding by the European Commission. This funding was granted and, in September 1990, FishBase started as one of ICLARM’s major projects under the Directorship of Daniel Pauly, with Rainer Froese as Project Leader.

Soon after the start of full-time data entry it became clear that the distinction between ‘commercial’ and ‘other’ fishes was arbitrary, and the original ‘goal of 2,500 species’ to be ‘provided on diskettes’ (see above) was changed to include all finfish species, with CD-ROM as the medium of distribution.

Gabriella Bianchi, who had worked previously with FAO’s Species Identification Programme and had authored and edited several major publications on tropical fishes, stayed with the FishBase Team for two weeks in August 1992. She highlighted the problem of synonymous names that had been entered from older references; she also reviewed the MORPHOLOGY table, which was modified following her suggestions. Overall, she concluded that "the database appears to be well structured and easy to use and understand. However, for many of the 6,000 species already entered, information is still limited."

FishBase received a second review by Kent Carpenter, project partner from FAO (1990-1995). Kent Carpenter spent two weeks (23 June – 8 July 1993) with the FishBase staff, and had a critical look at the information we had entered on the two families for which he is the world expert, i.e., the Caesionidae and the Lethrinidae. He pointed out that we had no mechanism in place to ensure that information and nomenclature from ‘primary’ sources (e.g., family revisions done by world experts, such as the authors of FAO Species Catalogues) always take precedence over other sources and are not changed unless in agreement with the experts. This criticism applied mainly to information that had come from ‘secondary’ sources such as checklists prepared by fisheries departments, faunal studies done at a time when insufficient taxonomic information was available or have become outdated because taxonomic information has improved substantially, and faunal studies not done by experts.

We accepted that criticism and started thinking about ways to achieve the required level of quality. The project made an effort to use the latest revisions for as many families as possible to update the SPECIES, SYNONYMS, STOCKS, COUNTRY and MORPHOLOGY tables. Species and families that were updated according to such revisions were marked, to alert encoders and users of their special status. Species still based on other sources were also marked as such. The bulk of the species has now been updated.

Fish are important to humans in numerous ways, leading to different types of information being available about their biology, distribution, etc. After three years of work we found we had started more mini-projects (= tables) to accommodate this diversity of information than we could possibly fill and keep updated on a permanent basis. Thus, on 9-10 September 1993 the FishBase Team retreated to a beach resort in Anilao, Batangas (south of Manila), to take stock. At the end of two days we had sorted out the wishful from the necessary and streamlined the latter further by an estimate of what each team member could actually achieve in the remaining year before the first release of FishBase. A number of tables were discarded or shelved (AQUARIUM, BREEDSYS, COMPETITORS, ECOREF, ECOSYSTEM, EGGNURS, FRYNURS, GAZETTEER, LARVNURS, MUSEDAT, SHARKMORPH). Others were maintained but with less emphasis (DISEASES, DISREF, OXYGEN, SPEED, OCCURRENCES, GILL AREA, EGGDEV, VISION). In hindsight, this workshop enabled us to overshoot by only two weeks, the deadline for the first release of FishBase on CD-ROM (September 1994), at least as far as data validation was concerned.

Through these early years, preliminary versions of FishBase were installed in many Research Institutes all over the world. However, this installation process also showed the limitations of the DataEase software for creating a royalty-free product.

The DataEase run-time module was difficult to create and limited in its functionality. A slightly better module would have cost twice as much in royalties per user than the current FishBase CD-ROM. Also, as of September 1994, there was no DataEase version that would run from a CD-ROM. Since the PC market was moving towards the Microsoft Windows interface, we decided that FishBase should also make use of that new standard. In mid-1993 we reviewed the available Windows databases (Microsoft Access, Paradox, Foxpro and SuperBase) and decided to use Microsoft Access, mainly because we had the impression that it would be the one requiring the least programming. Programmer Portia Bonilla started recreating the many FishBase tables and procedures under Microsoft Access in December 1993, but it was not before September 1994, i.e., a few weeks before the first release, that we were confident enough to transfer permanently all data to Microsoft Access (see ‘FishBase and Microsoft Access’, this vol.).

ICLARM’s Coastal and Coral Reef Resource Systems Program (CCRRSP) of which FishBase was the largest single project was reviewed in April 1994 by ICLARM’s Program Committee and by two external reviewers, T.J. Pitcher and J.J. Polovina. The reviewers wrote concerning FishBase: "Scope is huge. Will be very powerful tool and we support transfer to Windows Access system to enable flexible searches. Need to acknowledge that first release may have errors and should openly solicit revisions."

One of the early assumptions of the project was that microcomputer hardware, particularly for mass storage, would develop fast enough to hold huge amounts of data at the time of the first FishBase distribution. This turned out to be true and in August 1994, we were able to purchase a first-generation CD-ROM recorder, a one-gigabyte harddisk and a multimedia recording package for altogether US$8,000. In September 1994, we cut ICLARM’s first CD-ROM (a FishBase Demo disk) and in December 1994, we started in-house production of the complete FishBase plus several other ICLARM software on CD-ROM.

Cutting individual CD-ROMs in-house is one thing, mass-producing 100 or 1,000 copies is another. Requests for FishBase soon outstripped our production capabilities and we had to look for other options. At the time, there was only one commercial CD producer in the Philippines, but unfortunately with no experience in CD-ROM production. It took another considerable effort to overcome a series of annoying problems until, on 6 April 1995, we received a packet with 130 copies of what we called FishBase 100, the first mass-produced version of FishBase to be distributed to collaborators and a few early buyers. Thus, after five years, sweat and tears (but no blood), we had finally turned a vision into a product.

In September 1995, we produced 1,000 copies of FishBase 1.2 which were thereafter widely distributed and which helped to broaden our base to more than 160 collaborators and more than 400 recipients. An analysis of these first recipients showed the following distribution: Universities 36%, Governments 14%, Private Sector 14%, International Research Centers 8%, Museums 7%, Individuals 6%, Non-government Organizations 5%, Libraries 4%, United Nations and their specialized organizations 4%, and Donors 3% (c.f. with Fig. 1 for usage of FishBase). Thus, although FishBase had reached the foreseen range of users, its main target group (Government Fisheries Departments) was underrepresented. This analysis was confirmed by the fact that only 36% of the recipients were from developing countries. It seemed that additional measures were needed to reach the intended audience (see the ‘ACP Training Project’, below).

FishBase 1.2 was reviewed by R.A. McCall and R.M. May in Nature, Vol. 376:735, 31 August 1995 (see also Froese and Pauly 1995). Under the title More than a seafood platter the authors concluded: "In short, FishBase draws together and makes accessible a huge amount of information about fish and fisheries, which was previously buried in the ‘gray literature’ of reports from fisheries institutes or working parties. [...] Perhaps most important, and certainly closest to the authors’ hearts, it will benefit developing countries, where the lack of comprehensive libraries is often keenly felt."

FishBase 1.2 was also positively reviewed by K. Matsuura (1995) in the Japanese Journal of Ichthyology, Vol. 42 (3/4): 342-343. This review (in Japanese) strongly urged Japanese biologists to contribute to FishBase.

On 1-10 October 1995, the FishBase Team organized an FAO-ICLARM-MSI-NORAD Workshop, devoted to the creation of an FAO identification guide to the marine living resources of the Western Central Pacific. During that workshop, 35 renowned fish taxonomists each spent one full day on a close inspection of the information that we had gathered for their respective fish families. The FishBase Team noted all their suggestions and comments, and we marked all records that they had reviewed as checked by the expert. This close contact helped us to understand better the viewpoint of taxonomists and gave us a more secure feeling of our achievements and remaining challenges. It also led to many new friendships and continuing collaboration.

In another form of recognition, FishBase was invited to be one of the Global Species Databases that contribute to a global index of all known species, a project of the Species 2000 Federation (Bisby and Smith 1996; Bisby 2000). ICLARM now hosts the Philippine Office of Species 2000, tasked to produce and update the Species 2000 Annual Checklist on CD-ROM (see also www.sp2000.org).

In June 1996, we produced 1,000 copies of FishBase 96. The suffix ‘96’ was chosen to indicate our intention to produce annual updates of FishBase.

FishBase 96 presented the first fully tested version of FishBase, thanks largely to the excellent review process organized by Maria Lourdes D. Palomares (see ‘Bugs, Blanks and Errors’, this vol.). It had a much improved user interface, more and better pictures, the first graphs (see ‘Graphs in FishBase’, this vol.), a ‘Quick Identification’ routine, and it covered 15,000 species of finfish.

FishBase 96 reached about 1,000 users, won us many new collaborators, and helped to attract the ACP-EU grant (see below) that supported further improvement and distribution from 1997 to 2000. Due to increased contacts in preparation for this project, the number of users in developing countries had already increased to 47%, up from 36% in FishBase 1.2.

In April 1996, the Program Committee of the ICLARM Board of Trustees reviewed ICLARM’s role in database development. It noted that a minimum of US$70,000-80,000 per year was needed for long-term maintenance of databases such as FishBase. It recommended a continuing role of ICLARM in database development.

A review in Aquaculture (Rowell 1997) commended the size and scope of FishBase 96, but deplored the ‘many gaps and inconsistencies’. It used the herring as example, a temperate species that is extraordinarily well researched and has not received much attention in FishBase. It was correctly noted that in the REFERENCES table, the keyword ‘farming systems’ overlapped with the keyword ‘aquaculture’ (the former has meanwhile been removed). The review concluded: "It is a truly impressive undertaking which will, as the wrinkles are ironed out, become an increasingly useful tool for its target audience".

FishBase 96 was reviewed in the Journal of Fish Biology 50(3): 684-685 by R.J. Wootton (1997). He criticized the poor binding of the FishBase 96 manual (true, compared with the binding of this volume) and the fact that "for taxa with which one is familiar, important sources of published information have not yet been tapped." In addition, he pointed out that "the method of bringing together information from different tables to create new combinations is not transparent." The problem of incomplete information is discussed in the chapter ‘Bugs, Blanks and Errors’. The many new boxes in the FishBase 97 book, and expanded in subsequent versions, providing background on the new graphs should¾ at least partly¾ have taken care of the latter problem. The review concludes: "Overall, the importance of this database, if it can be progressively expanded, is incalculable."

A review in Reviews in Fish Biology and Fisheries 7(3): 374-375 (Turner 1997) criticized the lack of freshwater fish photos, problems with maps, errors on Lake Malawi fishes, incomplete checklists, and uneven coverage of genetics. It concludes that for the fields of freshwater fish biology, inland fisheries and conservation, evolutionary biology, and behavioral ecology "much is omitted and what is presented is strewn with far more errors than I have ever seen in any printed reference book intended for use by scientists". We fixed the reproducible errors, moved the warning ‘incomplete’ from the footer to the header of the respective checklists, and decided to continue nevertheless.

A review in the New Zealand Journal of Marine and Freshwater Research 31:281-285 (Francis 1997) commends the goal of FishBase to provide key information on tropical fishes. It noted that references had not been used consistently, e.g., for creating checklists of countries or islands. It noted the limited usefulness of FishBase for temperate countries such as New Zealand. It concludes that "this is a good product that will get much better. [……] scientists should consider becoming collaborators of the project and help it to improve and expand". We completed the mentioned checklists, improved information for New Zealand, and started a very fruitful collaboration with the author.

In 1996, it had become clear that the Internet and especially its Worldwide Web was there to stay and would revolutionize the dissemination of information. Making key information on fish as widely available as possible is what FishBase is all about, and thus moving FishBase to the Web was the obvious thing to do. However, while it was possible to query a few tables and display the result in a WebPage (as done with FishBase by David Gee in the context of Species 2000), recreating and testing hundreds of MS Access 2.0 forms for use on the Web was a task well beyond the capabilities of the FishBase Team of 1996. Rather, we thought it would be wiser to wait for Microsoft or another company to create tools that would automate the transfer from MS Access to the Internet.

Meanwhile, a FishBase homepage was created by Tom Froese and published on the Internet in August 1996. This preliminary homepage featured some background information, some nice photos by J.E. Randall, an interactive demo with FishBase screens, the full FishBase glossary (2,500 terms), and the full FishBase 96 Book (179 pages).

In May 1997, we hired John Falcon to become the first FishBase Webmaster, to update and to develop the homepage continuously and eventually to make all of FishBase available on the Net.

In the framework of the special support the European Union gives to associated African, Caribbean and Pacific (ACP) countries ICLARM signed in December 1996 a project agreement with the Commission of the European Union on ‘Strengthening fisheries and biodiversity management in ACP countries’. The duration of the project was four years, i.e., to December 2000. The project entailed the establishment of regional training nodes in ACP countries, and the gradual building of a functional web of regional and trans-regional cooperation, using modern communication facilities (Vakily et al. 1997a, 1997b).

Training focused on the role biodiversity plays in the assessment of the status of aquatic ecosystems. A major effect of the training was expected to be the gradual building of national biodiversity databases on fish in the ACP countries. To this end, FishBase would serve both as a source of existing information and as a tool to be used as a structuring element in the collection of biodiversity data. Ultimately, the project aimed to contribute to an increased awareness among fisheries researchers and managers in ACP countries of the importance to conserve biodiversity for sustainable use of aquatic resources.

In December 1996, Jan Michael Vakily was hired as Training Coordinator of the ACP Project, supported by Research Assistant, Grace T. Pablico. Five regional outposts were established in Africa (Namibia, Senegal and Kenya), the Caribbean (Belize) and the Pacific (New Caledonia) and two-week training courses for mainly fisheries scientists from the region were conducted at these nodes. Feedback from these courses was used to improve FishBase data, routines and interface. In January 2000, former Steering Committee member Boris Fabres replaced Jan Michael Vakily as Network Coordinator of the project.

To advise the European Commission and to guide and assist the Project Team in executing the 5 million ECU Fisheries Biodiversity Management Project, the Commission of the European Union had invited the following persons as members of the ACP Steering Committee: Dr. Cornelia Nauen, Belgium, Chair; Dr. Tim J. Adams, New Caledonia; Dr. Eduardo Balguerias, Spain; Mr. Amadu Bailo Camara, Guinea Bissau; Mr. Boris Fabres, Trinidad and Tobago; Prof. Guy Fontenelle, France; Mr. Thomas W. Maembe, Tanzania; Dr. Jean Calvin Njock, Cameroon; Dr. John Tarbit, United Kingdom (later replaced by Dr. Helge Paulsen, Denmark); and Dr. Ben van Zyl, Namibia.

The first meeting of the Steering Committee was on 3-5 June 1997 in Manila. After an in-depth introduction to an early version of FishBase 97, they concluded: "The Steering Committee recognized the excellent quality of the work carried out so far by the team. This has led to an extremely useful product." They confirmed the course of the project to cover all fish and to assign them to all countries and large ecosystems. They especially supported the efforts to establish a strong link between FishBase and Ecopath (see Box 21). Four meetings of the Steering Committee were convened between 1997 and 2000, during which continued support was given to the efforts of the Project team and its goal to make FishBase as useful as possible for fisheries and biodiversity management.

FishBase 97, released in November 1997 covered more than 17,500 species and contained many more and improved pictures, much improved annotated checklists, more occurrence points and thus better maps, many more graphs, a yield-per-recruit analysis routine applicable for the over 1,000 species for which we then had growth parameters, a tool to compare and analyze growth parameters (AUXIM), and more data for more species. Due to the increased number of pictures (about 12,000) FishBase 97 came on two CD-ROMs.

FishBase 98 was released in late 1998 on two CD-ROMs. It had been transferred to MS Access 97 and therefore required Windows 95, a Pentium processor, and at least 16MB of RAM. The database covered over 20,000 species and the FishBase book had grown to 293 pages. Major improvements were the inclusion of Eschmeyer’s (1998) Catalog of Fishes databases and a number of new graphs analyzing, e.g., FAO catch data. The number of pictures had grown to over 15,000, and the number of references to over 12,000. With this release the number of registered FishBase users grew to 1,623 in 149 countries.

The two main objectives of FishBase 99 were to produce a version with French help text and book, and to pass the 23,000 species threshold. These objectives were achieved with the December 1999 release, which came on three CD-ROMs mainly because the number of pictures had increased to 17,000 and the number of fish collection records had increased to 300,000. A new ‘Key Facts’ form provided ‘best estimates with error margin’ for essential management parameters such as length at first maturity and length at optimum yield. The number of references used had grown to 16,000. The number of registered users of FishBase CD-ROMs grew to 1,800 in 154 countries.

The French-language book documenting FishBase 99 was based on a translation cum update of the FishBase 98 book, by Nicolas Bailly of the Muséum National d’Histoire Naturelle, Paris, and Maria Lourdes Palomares, of the FishBase Project, with the support of a number of Francophone colleagues, notably the co-editor of the present volume.

A draft of this version was made available to the Francophone participants of the Fourth Training Course on Fisheries and Biodiversity Management in the Context of the ACP-EU project held in Dakar, in 12-23 April 1999, and their feedback was incorporated into the final version. Also, FishBase 99 was presented at one of the preparatory meetings to the Francophone Summit held in Moncton, New Brunswick, Canada, 10-13 August 1999, and devoted to: "New Tools, New Approaches for the Sustainable Management of the Marine Environment". In view of the utility of a French version of the FishBase book in making the database itself accessible to Francophone scientists throughout the world, the participants of the meeting included among their recommendations to the Heads of Francophone States to "make available all databases and information of global utility (for example ‘FishBase’, now translated into French)".

This provides a strong support for the plan by the Muséum National d’Histoire Naturelle to maintain the French version of FishBase, including a Francophone web site (see below The FishBase Consortium).

Over the years, the data accumulated in FishBase reached a level where they allowed scientific studies that could not have been done otherwise. A first, widely recognized example is the article of Pauly et al. (1998) in Science 279:860-863 which used a combination of FAO catch data with trophic data available in FishBase to demonstrate that the world fisheries were ‘Fishing down the food web’. Other published examples are an article comparing the ‘living fossil’ Latimeria chalumnae with modern fishes (Froese and Palomares 2000), and a set of empirical equations to estimate important management parameters for all fishes (Froese and Binohlan 2000). A number of exciting new studies, e.g., on generating Lindeman pyramids from the ecosystem-related data in FishBase have started in 2000.

In the course of 1998 it became clear that there would be no magical tool to translate our existing FishBase user interface from MS Access to the Internet, mainly because the increased response time often characteristic of on-line use asked for a completely different design philosophy. In March 1999, we therefore hired Meynard Gilhang as Webmaster (and replacement for John Falcon) and Eli Agbayani as web database programmer, to create a new web interface for eventually all FishBase tables, including graphs and reports. We settled for Cold Fusion as web server software.

In mid 1999, the Swedish Centre for Coastal Development and Management of Aquatic Resources (SWEDMAR) was tasked to perform a mid-term review of the ACP Training Project and of FishBase. The review team (Lars Hernroth and Roger Lindblom) concluded that the FishBase Team was highly competent and that progress made so far was very impressive. It pointed out that the coverage of several FishBase tables was still incomplete, thus limiting the usefulness of FishBase as a management tool. It stressed the need to continue the FishBase effort and to find a permanent home for it, independent of project-based funding.

Following up on the SWEDMAR review, the European Commission contacted several European institutions to explore their interest in taking FishBase on as one of their permanent activities. At an extended Steering Committee meeting in Brussels in March 2000 the following institutions were present and declared their interest: Muséum National d’Histoire Naturelle, Paris; Musée Royal de l’Afrique Centrale, Tervuren; Swedish Museum of Natural History, Stockholm; Institut für Meereskunde, Kiel. Together with ICLARM and FAO these institutions agreed to form a consortium to jointly continue the FishBase effort. At a subsequent meeting in November 2000 at FAO, Rome, the Consortium was formally established and the Fisheries Centre of the University of British Columbia, Canada, was accepted as 7^th member of the Consortium, represented by Daniel Pauly.

FishBase 2000 comes on four CD-ROMs, to accommodate over 25,000 fish species, over 25,000 photos, and over 600,000 fish collection records. It uses MS Access 2000 as database engine and user interface, but it may well be the last edition to do so as the development and maintenance of two different user interfaces is too demanding. Future CD-ROM (or better DVD-ROM) versions of FishBase are planned to be copies of the Internet version.

Over the years, the FishBase Team grew to include a post-doctorate fellow/research scientist (Maria Lourdes D. Palomares), more research assistants (Susan Luna, Crispina Binohlan, Armi Torres, Liza Agustin (later replaced by Christine Casal), Pascualita Sa-a, Emily Capuli, Rodolfo B. Reyes, Jr., Cristina Garilao), an artist (Roberto Cada later replaced by Rachel Atanacio), a sequence of programmers (Dominador Tioseco, Portia Bonilla, Alice G. Laborte, Ma. Josephine France Rius, Eli Agbayani), and a project secretary (Maria Teresa Cruz). The project also maintained temporary outposts (two years each) in Malawi (Department of Fisheries, Emmanuel Kaunda, Dennis Tweddle), Ghana (Institute of Aquatic Biology (IAB), Mamaa Entsua-Mensah), the Philippines (University of the Philippines, Marine Science Institute (UP-MSI), Emily Capuli) and Peru (Universidad Nacional Agraria La Molina (UNALM), Jaime Mendo) to ensure that FishBase would meet the needs of prospective users in national programs.

A number of volunteers joined the FishBase Team at different times, Magnus Olsson Ringby from Sweden, Sari Kuosmanen-Postila from Finland, Analyn Palomares, Ilya and Angela Pauly, Henry Angeles, Neil Del Mundo, Tom Froese, Jayson McArthur, Drina Sta. Iglesia from the Philippines, Anne Johanne Dalsgaard from Denmark and Shen-Chih Wang from Taiwan.

With the start of the ACP Training Project (see above), the team was joined by Training Coordinator, Jan Michael Vakily (replaced in late 1999 by Boris Fabres as Network Coordinator, his assistant, Grace T. Pablico, and Webmaster, John Falcon (replaced by Meynard Gilhang). The existing FishBase could never have been assembled without substantial input from collaborators all over the world (Fig. 2).

Notably, FishBase acts as a host to databases that continue to be maintained and updated by the contributing institutions, with or without inputs from FishBase staff.

Outstanding contributors are:

• FAO’s database SPECIESDAB (Coppola et al. 1994) added about 800 commercially important species to FishBase and thus helped FishBase move fast ahead in the early stages of the project. Also, SPECIESDAB was used to check data such as scientific names, FAO names, FAO areas, etc. prior to the first release;

• FAO’s database on species introduction (INTRO) prepared by Robin Welcomme helped cover all internationally introduced species (Welcomme 1988);

• the contribution by W.N. Eschmeyer of his GENERA database which was included in FishBase 1.0 allowed standardization of all generic names and higher taxa (Eschmeyer 1990). Since 1998, FishBase also contains the databases underlying Eschmeyer’s (1998) Catalog of Fishes;

• Thurston and Gehrke’s OXYREF database, which provided the largest collection of respiration experiments (Thurston and Gehrke 1993);

• the International Game Fish Association’s World Records database (IGFA 1994);

• Craig Hilton-Taylor who made the 2000 IUCN Red List data available for inclusion in FishBase (Hilton-Taylor 2000);

• Guy Teugels of the Musée Royal de l’Afrique Centrale (MRAC), who provided a copy of CLOFFA IV as WordPerfect file and encouraged us, on behalf of the editors, to use all the information in the CLOFFA series (Daget et al. 1991);

• The Musée Royal de l’Afrique Central (MRAC), which made their collection database available through FishBase;

• Jean-Claude Hureau of the Muséum National d’Histoire Naturelle (MNHN), who provided a preliminary set of records from the fish collection database GICIM (Hureau 1991). The Museum and ICLARM signed a Memorandum of Agreement on 12 October 1993 to make all GICIM records available in FishBase and to collaborate on the further development of both databases;

• Ed Houde, who provided his unique database on larval dynamics for distribution through FishBase (Houde and Zastrow 1993);

• the Programa de Ecología Pesquerías y Oceanografía del Golfo de México (EPOMEX), of the Universidad Autónoma de Campeche, then led by A. Yañez-Arancibia, which expressed its interest in FishBase rather early, and provided the project through its newsletter, Jaina, a medium for reaching out to colleagues in Mexico and other Latin American countries (see Pauly and Froese 1992). One EPOMEX scientist, Cristina Bárcenas-Pazos, entered ecotoxicological data into a table created for the purpose (see the ‘ECOTOXICOLOGY table’, this vol.). Also, EPOMEX received a grant from a national donor for collaboration with FishBase to improve the coverage of Mexican/Latin American species;

• Ransom A. Myers, previously with the Department of Fisheries and Oceans, Canada, who made his unique database on recruitment time series available through FishBase;

• Roland Bauchot and his colleagues at the Université Paris VII, who supplied their database on fish brains;

• FAO, for their data on catches and aquaculture production, and also for pictures and other information from the Species Identification and Data Programme, now led by Pere Oliver, made available for distribution through FishBase;

• John E. Randall, who made over 10,000 slides of Indo-Pacific and Caribbean fishes available for inclusion in FishBase.

Bisby, F.A. 2000. The quiet revolution: biodiversity informatics and the Internet. Science 289:2309-2312.

Bisby, F.A. and P. Smith. 1996. Species 2000: indexing the world’s known species. Project Plan Version 3. 44 p.

Coppola, S.R., W. Fischer, L. Garibaldi, N. Scialabba and K.E. Carpenter. 1994. SPECIESDAB: Global species database for fishery purposes. User’s Manual. FAO Computerized Information Series (Fisheries) No. 9. FAO, Rome. 103 p.

Crawford, S.S. 1997. Development of a global fish database. Environ. Biol. Fish. 50:231-234.

Daget, J., J.-P. Gosse, G.G. Teugels and D.F.E. Thys van den Audenaerde, Editors. 1991. Checklist of the freshwater fishes of Africa (CLOFFA). Vol. IV. Institut Royal des Sciences Naturelles de Belgique, Bruxelles, Belgium, Musée Royal de l’Afrique Centrale, Tervuren, Belgium and Institut Français de Recherche Scientifique pour le Développement en Coopération – ORSTOM, Paris. 740 p.

Eschmeyer, W.N. 1990. Catalog of the genera of recent fishes. California Academy of Sciences, San Francisco. 697 p.

Eschmeyer, W.N., Editor. 1998. Catalog of fishes. Special Publication, California Academy of Sciences, San Francisco. 3 vols. 2905 p.

Fischer, W., Editor. 1973. FAO species identification sheets for fishery purposes. Mediterranean and Black Sea (fishing area 37). FAO, Rome. pag. var.

Fischer, W. 1976. The FAO species identification sheets programme: a common task for ichthyologists and fishery workers. Rev. Trav. Pêches Marit. 40(3/4):568-569.

Francis, M.P. 1997. FishBase 96 CD-ROM, compiled by R. Froese and D. Pauly. New Zealand J. Mar. Freshwat. Res. 31:282-284.

Froese, R. 1988. The use of quadratic discriminant functions in connection with video-based measurements for identification of fish larvae. ICES C.M. 1988/L:11: 8 p.

Froese, R. 1989. Computer-aided approaches to identification. II. Numerical taxonomy. Fishbyte 7(3):25-28.

Froese, R. 1990. Moderne Methoden zur Bestimmung von Fischlarven. Universität Hamburg. Doctoral Thesis. 260 p.

Froese, R. and C. Binohlan. 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. and C. Papasissi. 1990. The use of modern relational databases for identification of fish larvae. J. Appl. Ichthyol. 6: 37-45.

Froese, R. and D. Pauly. 1995. EC Project reviewed in Nature. EC Fish. Coop. Bull. 8(3):4-5.

Froese, R. and M.L.D. Palomares. 2000. Growth, natural mortality, length-weight relationship, maximum length and length-at-first-maturity of the coelacanth Latimeria chalumnae. Environ. Biol. Fish. 58:45-52.

Froese, R. and W. Schöfer 1987. Computer-aided identification of fish larvae. ICES C.M. 1987/L:23: 10 p.

Froese, R., F.C. Gayanilo, Jr. and M.L. Soriano. 1988. ICLARM fish database project: report I. ICLARM FishBase Project Document No. 1, December 1988. 160 p.

Froese, R., I. Achenbach and C. Papasissi. 1990a. Computer-aided approaches to identification. III. (Conclusion). Modern databases. Fishbyte 8(2): 25-27.

Froese, R., K.-G. Barthel, W. Welsch, M. Rölke, C. Schubert, B. Hermann, S. Mees, D. Schnack and J. Lenz. 1990b. Development of an underwater video system for recording of ichthyoplankton and zooplankton. ICES C.M.1990/L:90: 5 p.

Froese, R., W. Schöfer, A. Röpke and D. Schnack. 1989. Computer-aided approaches to identification of aquatic organisms: the use of Expert Systems. Fishbyte 7(2):18-19.

Hilton-Taylor, C., Compiler. 2000. 2000 IUCN Red list of threatened species. IUCN, Gland, Switzerland and Cambridge, UK. 61 p.

Houde, E.D. and C.E. Zastrow. 1993. Ecosystem- and taxon-specific dynamic and energetics properties of fish larvae assemblages. Bull. Mar. Sci. 53(2):290-335.

Hureau, J.C. 1991. La base de données GICIM. Gestion informatisée des collections ichtyologiques du Muséum, p. 225-227. In Atlas Preliminaire des poissons d’eaux douce de France. Conseil Supérieur de la Pêche, Ministère de l’Environment, CEMAGREF et Muséum National d’Histoire Naturelle, Paris.

ICLARM. 1988. ICLARM five-year plan (1988-1992). International Center for Living Aquatic Resources Management, Manila, Philippines. 114 p.

IGFA. 1994. IGFA world records. International Game Fish Association, Pompano Beach, Florida. 40 p.

Matsuura, K. 1995. FishBase: a biological database on Fish. Japan. J. Ichthyol. 42(3/4):342-343.

McCall, R.A. and R.M. May. 1995. More than a seafood platter. Nature 376(6543):735.

Nelson, J. 1994. Fishes of the world. 3^rd ed. John Wiley and Sons, New York. 600 p.

Pauly, D. and R. Froese. 1992. FishBase, base de datos computarizada sobre los peces: uso potencial para América Latina. Jaina 3(4):11-13.

Pauly, D., F.C. Gayanilo, Jr., R. Froese and V. Christensen. 1995. Software tools for management of tropical aquatic resources, p. 90-102. In G.T. Sakagawa (ed.) Assessment methodologies and management. Proceedings of the World Fisheries Congress, Theme 5. Oxford & IBH Publishing, New Delhi.

Pauly, D., V. Christensen, J. Dalsgaard, R. Froese and F. Torres Jr. 1998. Fishing down marine food webs. Science 279:860-863.

Rowell, T.W. 1997. Book reviews – FishBase 96: concepts, design and data sources. R. Froese and D. Pauly (eds.), ICLARM, Manila, 1996, 179 pp. Aquaculture 154(1):87-91.

Thurston, R.V. and P.C. Gehrke. 1993. Respiratory oxygen requirements of fishes: description of OXYREF, a data file based on test results reported in the published literature, p. 95-108. In R.C. Russo and R.V. Thurston (eds.) Fish physiology, toxicology, and water quality management. Proceedings of an International Symposium, Sacramento, California, USA, 18-19 September 1990. US Environmental Protection Agency, USA.

Turner, G.F. 1997. Book reviews – FishBase 96: concepts, design and data sources. R. Froese and D. Pauly (eds.), ICLARM, Manila. Rev. Fish Biol. Fish. 7(3):374-375.

Vakily, J.M., R. Froese, M.L.D. Palomares and D. Pauly. 1997a. European Union supports project to strengthen fisheries and biodiversity management in African, Caribbean, and Pacific (ACP) countries. Naga, the ICLARM Q. 20(1):4-7.

Vakily, J.M., R. Froese, M.L.D. Palomares and D. Pauly. 1997b. Fisheries and biodiversity management: how can ACP countries meet the challenge? EC Bull. 10(1):4-8.

Welcomme, R.L., Compiler. 1988. International introductions of inland aquatic species. FAO Fish. Tech. Pap. No. 294, 318 p.

Wootton, R.J. 1997. Review of FishBase 96. J. Fish Biol. 50(3):684-685.