The Catalog of Fishes

William N. Eschmeyer

Department of Ichthyology
California Academy of Sciences

Introduction

The primary way that information about plants and animals is stored is by taxonomic categories, typically species. Since many participants in the workshop are not that familiar with taxonomy, Stan and I agreed that it would be helpful to provide some information in this area first.

It is important to understand what taxonomy is, why a hierarchical classification is useful, and why classifications and names change, often frequently, thereby making it difficult to keep track of information about plants and animals.

First, some definitions:

Taxonomy:

the study and practice of naming and classifying organisms; as done by taxonomists.
 

Systematics:

the classification and study of organisms with regard to their relationships. It is a little broader and includes more activities than taxonomy.
 

Taxon (plural: taxa):

a taxonomic unit, such as a species or a genus.

I am a fish taxonomist, so I will first try to explain taxonomy and the process of creating taxa, using some examples from fishes. Then I will tell you about how we built our databases in terms of both our strategies and our products. Then I have some information on the cost and value of such databases. I'll give you some rough estimates of what this project cost in today's dollars. The figures not very precise but they're in the ballpark, and we may not have any estimates at all for projects like this in other groups, so I think they will be useful. I have also included an appendix that summarizes the various fields in each database, the query forms we use to get into the databases, and some of the outputs that we've found useful.

Taxonomy and Nomenclature

Discovering, Describing, and Naming New Species

Taxonomists typically work in universities and natural history museums, some work in federal and state agencies, but there is actually no restriction on anyone's ability to participate in scientific taxonomy and describing new species; one can be an amateur. In Ichthyology, some new scientific names have been introduced by amateur aquarists. I work in a natural history museum where a lot of our work entails describing species and higher categories to show relationships among species.

Discovering and describing new species is not that difficult, but it is time consuming (Fig. 1). It begins when a taxonomist becomes aware that a specimen, or group of specimens, doesn't match any of the species that are already known to science. This fish (Fig. 2), for example, appeared in a Japanese aquarium as a single specimen that two of my Japanese colleagues couldn't identify. They knew it was a kind of scorpionfish, which is my area of specialty, so they sent it to me and allowed me to co-author the paper with them. We took a chance and described a new species from this one specimen, but that's a little unusual.

Figure 2. Rhinopias argoliba Eschmeyer et. al (1973), as featured on the cover of a popular magazine.

To begin the real work, we assemble specimens, which may require doing more field work (the fun part) or borrowing specimens from other natural history museums. The photograph below (Fig. 3) shows what the red fish looked like when it came to me. Preserving specimens makes them less attractive, but permanent collections are very important because they enable us to compare new discoveries against material and decisions people made in the past.

Figure 3. Preserved specimen (holotype) of Rhinopias argoliba Eschmeyer et al. 1973. Deposited in the permanent collection of the Enoshima Aquarium, Fijusawa, Japan (EA No. 1). 129 mm. S.L.

We study variation by collecting data on specimens, and we define species by making piles. When I said that in talk to museum docents a hand went up and the person asked, "What's a pile?" Well, a pile is a group of individuals that you think all belong to the same species. There are no fixed rules you can use to determine where the boundaries between species are. You simply compare individuals and look for similarities and differences, and particularly for gaps in variation that is otherwise continuous. You also try to rule out male-female differences and geographic differences. In fishes, collecting data means counting things like vertebrae and fin-rays, and taking measurements of body parts, and this is often very tedious. Figure 4 shows some of the data an Indian colleague and I collected when we described a new species of stonefish.

Figure 4. Counts and measurements for the type specimens of Synanceia alula Eschmeyer & Rama Rao (1973). Measurements in mm., percent standard length in parentheses.

After you've examined this variation and sorted the data in lots of different ways, you end up with piles of specimens. Then you have to go back into the literature and to the type specimens to find out which of the previously described species in that taxonomic group—a genus or a family, for example—match the piles you've got. Usually in fishes you have a pile or two left over, and these are the new species that you have to describe and name. The International Code of Zoological Nomenclature tells you how to apply existing names to the piles you have, but it doesn't tell you whether or not your piles are valid species. That's a scientific or biological question, and the code only deals with nomenclature, not biology. The rules of zoological nomenclature were codified in 1843, and scientists all around the world have followed them ever since. It's one of the few areas of total world cooperation—with a couple of exceptions.

Type specimens are very important in this process of assigning names to species because, in a sense, they carry the names with them. Imagine that you put all your fish out on a table, including the type specimens, and sorted them into piles. Some piles would contain just one type; others would contain several types, and still others wouldn't contain any. When there's just one type in the pile, it's easy; that pile gets the name associated with that type. If the pile contains more than one type, you consult the rules of nomenclature to determine which name you have to use, and usually it's the name that was published first. If a pile doesn't contain a type, it may need both a new name and a new type specimen to go with that name. A name doesn't have to be fixed by a type to be usable, but types do make it a lot easier for the next worker to figure out which name applies to which species. So nowadays, when we name new species, we typically designate types and deposit them in institutions that hold permanent collections, like natural history museums. The locality where the primary type was collected is called the type locality. That's very important and I'll talk more about type localities later.

Scientific names are latinized. Traditionally they came from Greek and Latin nouns and adjectives, but they can be virtually any combination of letters. Taxa can be named after a person (e.g., smithi), or a place (e.g., hawaiiensis). In the old days when people knew Greek and Latin, they would coin a name that somehow described the species. In the case of the red fish I showed you earlier (Figs. 2 & 3), a member the genus Rhinopias, we gave it the specific name argoliba, which means 'white teardrop.' The full species name is then Rhinopias argoliba. The new stonefish I described with Dr. Rama Rao has fewer pectoral fin rays than other members of the genus Synanceia, so I looked up some names that might be descriptive. I chose alula, which is a Latin feminine noun meaning 'little wing,' and made the full name Synanceia alula (Fig. 5). So these are both instances of following tradition and choosing a name that describes the species.

Figure 5. The distribution and etymology of Synanceia alula. From Eschmeyer & Rama Rao, 1973.

Having sorted out the species and the names that go with them, there are still a few more steps. One is to go back through the literature and find all of the names and descriptions that have ever been applied to your species, whether as a part of formal nomenclature or through simple identifications. In our taxonomic publications we typically list all of these names in a synonymy immediately after the heading of a taxonomic description (Fig 6). These lists are very important because they tell people what names to search under when they're looking for more information about that taxon. The synonymy in our stonefish paper, for example, indicates a photograph of another specimen had been published earlier, but the authors of that paper had not recognized it as a new species and identified it instead as Synanceia verrucosa.

Figure 6. The synonymy and type designations for Synanceia alula. From Eschmeyer & Rama Rao, 1973.

Other important components of a taxonomic work are: a key to the species, which helps other people identify specimens you didn't have the chance to examine yourself, including new material (Fig. 7); written descriptions of the species, including diagnostic characters; and a drawing or a photograph of the species (Fig. 8).

Figure 7. Key to the genera and species of Synanceia. From Eschmeyer & Rama Rao, 1973.

Figure 8. A pen and ink drawing of the head (dorsal view) in the holotype of Synanceia alula. From Eschmeyer & Rama Rao, 1973.

Once all of this work is finished, we publish these contributions in scientific journals, for the most part. Figure 9 shows a typical publication that names new taxa. This is the paper where we described the new stonefish Synanceia alula. Once the names are published they get picked up and used by our colleagues, who write other papers and books such as The Fishes of the Bahamas or The Fishes of Australia.

Figure 9. Title page of a typical paper describing new species.

Classifications

Phylum: Vertebrata 
  Class: Osteichthyes 
    Order: Perciformes 
      Family: Xiphiidae    
        Genus: Xiphius 
          Species: Xiphius gladius 

Xiphius gladius Linnaeus 1758 
Common name: Swordfish  

A species name consists of two words, the generic name and the specific epithet; Xiphius gladius, for example. Linnaeus came up with this system in 1758, and it caught on. By convention there can only be one genus in animals named Xiphius. Unfortunately, there can be a genus Xiphius in botany as well, so valid homonyms between the two kingdoms is a problem we'll have to take up at some point. A specific epithet, e.g., gladius, can be used over and over again in different genera, but there can be only one genus Xiphius and the combination of the two names must also be unique.

Species are placed into successively higher taxa, which creates a hierarchy. The main categories are shown in Figure 10, but there can be many intermediate categories, such as superfamily, tribe, or subgenus. Subspecies are used in some vertebrate groups but not very often in fishes. The important thing about classifications is that they contain information. If two species are closely related, we put them in the same genus. Then, for example, if a pharmaceutical product is found in one species, the biochemists can go to the classification to learn about its nearest relatives and perhaps find a similar chemical or a better chemical. So the classification shows decreasing relatedness as you work your way up into higher categories. Species that are placed more closely together almost certainly have extensive similarities, including their DNA, physiology, behavior, etc., so there is a lot of predictive value in a classification.

Figure 11. Rhinopias aphanes Eschmeyer 1973. 230 mm. Lady Elliot Id., Great Barrier Reef. (Photo by J. E. Randall.)

This is another species of Rhinopias I got to describe from a single specimen (Fig. 11). It has this drastically different coloration, so it was obviously new, but on the basis of morphology, it's closely related to the red fish (Figs. 2 & 3) so it got placed in the same genus.

Since the 1960s, most taxonomists have adopted a different underlying theory to guide them in building classifications, one of proposing relationships based on advanced or derived features, and then creating higher categories from these diagrams that show relationships. Figure 12 shows a very simplified example. If we assume that A, B, and C are species today, this says that in the past there was an ancestral Species-1 that gave rise to two species: A, and the ancestor of B and C. This ancestor developed new features that are now found in its descendants B and C, but not in A. Then B and C diverged and acquired new features.

Figure 12. A hypothetical cladogram.

This shows us the first case of why species names change. One person may believe all three species belong in one genus, whereas another person may believe A is one genus, and B and C form another. We would have two different front end names (genera) and same last name.

Names change for many other reasons. One is the rule of priority. Many species have been described and named more than once. This is especially true for widely distributed species, large species, commercially important species, those that change markedly in their growth, or where there are pronounced differences between males and females. In fishes we have about 50,000 names that can be used, but there are only about 25,000 valid or actual species. So on average we've described each one two times. We still describe about 200 new species of fishes a year, and I suspect that we will end up with about 35,000 valid species before we are through.

How is the rule of priority applied? Let's suppose Stan describes a species and I describe a species. Then a third worker, such as Dr. Collette over here, who is also an ichthyologist, re-studies the group and concludes that the two species are the same. The rule of priority then says that we have to use Stan's name, because his name was published first, and mine becomes a junior synonym. This is simple enough, but a lot of information might have accumulated in the literature under my species name. This is especially true if both names were in use for a long time.

The rule of priority causes a lot of aggravation when a long-forgotten name is rediscovered and causes a widely-used name to become a junior synonym. So one of our problems is how to track the information that accumulated under the junior synonym. In 1989 the Rainbow Trout had both its generic and specific names changed. The genus changed to reflect what we know about its relationships to other trout and salmon, and the specific name changed because an older species from Kamchatka (i.e., an earlier name and description) was determined to be the same species. You can imagine the reams of information that were accumulated under the name Salmo gairdneri, which is now a synonym of Oncorhynchus mykiss. So that's a real problem of the system.

Another rule that causes species names to change is gender agreement between the parts of the name; if the specific name is an adjective, it must agree in gender with the generic name. If the species marmoratus, for example, is switched from a masculine genus to a feminine genus it becomes marmorata. Sometimes names are changed for technical reasons. As an example, no two species can have the same first and last name. So if marmoratus moved to a genus where there was already a marmorata, we have to give one of them a new name.

Useful information is frequently found under misidentifications. Yet another problem is that names are frequently misspelled. To alleviate these problems taxonomists make synonymies; a historical account of the names and misidentifications for each species. These are extremely valuable because they tell you the other names you'll have to use to find all of the relevant information about that species.

Higher classification in fishes is undergoing much study and rearrangement at this time; new families, subfamilies, and genera are being introduced on a regular basis. There is no standard classification. So the changing nature of classifications because of changing ideas about relationships, and the changing of names for the reasons I've outlined, make it almost impossible to know, for any given species, which names—i.e., species, genus, or even family names—we need to use to find pertinent information in the literature. So I think taxonomic databases and some new techniques may help us solve some of these problems.

Building The Catalog of Fishes

Strategy and Products

Now I'd like to show you briefly how we built the databases for fishes. We began by focussing on the genera of fishes, using the most complete sources, though some of these were rough. These were:

• The Genera of Fishes, produced by Jordan from 1918 into the 1920s;
• Fowler's Catalog of World Fishes, (Fig. 13) partially published in the Journal of the Taiwan Museum (not proofed by Fowler and contains many errors, so publication was discontinued);
• some 5" x 7" cards at the University of Michigan; and
• Zoological Record, which of course we went through year by year.

Figure 13. A page from Fowler's Catalog of World Fishes.

In addition to the published parts of Fowler's work, the Academy of Natural Sciences in Philadelphia had a manuscript of several thousand pages (Fig. 14) and a set of 3" x 5" cards (Fig. 15) that Fowler had compiled in the 30's and prior to that time. You can see from the card that he was automated in the early 1900s; he had rubber stamps made for the primary journals. If he had had a computer he would have done this many years ago.

Figure 14. A page from Fowler's unpublished manuscript.

Figure 15. A Fowler card.

We wanted to proof every original description—those first descriptions I mentioned with the red fish. We had to solve some technical nomenclature problems, and wrote an interpretation of the Code to explain our decisions. The Code is very difficult to use and doesn't contain many examples. After seeing many of the problems in fishes, however, I began to understand why they wrote it the way they did. The Code has changed significantly since 1843, and more than once, so different sets of rules apply to names published in different time periods. We used the current literature to get the current status for as many genera as we could, about 10,300 genera, and to put them in a classification.

We published The Genera of Fishes in 1990. The three basic parts are: 1) all the genera of fishes, in alphabetical order—we found that that was the best way to present the information; 2) the genera in a classification; and then 3) the literature cited.

For a typical genus (see Fig. 16) you get the genus name, author, date, pages, a unique reference number for the source publication, some technical information, and current status with additional references—a sort of guide to get you started. So there are 10,300 of those entries for generic names.

These were placed in classification, and we had some reasons for how we structured our classification—trying to keep related groups of genera together given a system where the higher classification is constantly changing.

For each reference we did our best to determine the exact date of publication. This is important because of the rule of priority, and it can be very difficult to determine. As the librarians will know, a lot of journals, even major ones, aren't published on the date printed on the cover. Even journals like the Annals and Magazine of Natural History have run months behind. When it says published in August of 1864, it really may have been published in 1865. In some cases publication may have covered a number of years. So we tried to find references that give publication dates for the individual sections, and thus the species as they occurred in those sections.

Then we received funding to do the species catalog. We used the same basic strategy—Zoological Record, from 1864 to present; Fowler's cards that went up to about 1905, Fowler's Fishes of the World—and we completed a rough draft in 1991. We had 55,000 records at that time, but a number of these were duplicates and these had to be cleaned up. Some early authors published in two places, sometimes three places, with different dates of publication. We got additional funding to do that in March of 1992. We needed to tie all of these records to the original publication, which is more difficult than it might seem, and we wanted to proof all the original descriptions, or at least as many as we could.

To support the proofing process, we wanted to have available at my institution all of the original descriptions. In some cases, there are only few copies still available. We didn't use interlibrary loan that much, so I spent a lot of time Xeroxing at the British Museum, the Philadelphia Academy, American Museum of Natural History, and Smithsonian. In my library, I have to put white gloves on to handle rare books; at the British Museum, those are not rare books. I even got to a point that I knew so much about Xerox machines that they would let me do it myself.

We checked the spelling, the original genus, the authorship, the reference and dates, checked the pages and figures, and we went a little further and tried to provide the type localities, which is something you might not need to do for just an authority file, but it's very helpful for the specialist, so we picked it up whenever possible. We wanted to document the current status of every species name, so we went through most of the journals for the last ten years, other recent sources, as well as important earlier monographs.

My average for proofing descriptions was about 100 a day. If there were a lot of species described in one source, it went very quickly, but if the 100 species were in 30 references, I had to find the references, I had to translate some of them, etc. If you do a hundred a day for a year, you're half through, and you turn around and do 100 a day for the next year.

We published the Catalog of Fishes in the same format as the Genera, with species arranged alphabetically, then in a classification, which we changed from the earlier classification, and then the references. We wanted to make this available electronically on line, and two years ago we put it on the Academy's Gopher server as a WAIS-indexed text file.

After we got the Catalog on line and met our obligations to NSF, I spent about two years improving the database by gathering information on type specimens. This is primarily a help for the specialist because type specimens are critical to taxonomic work and before this, there was no a guide to where these are. At the British Museum I found out that they had filled out little slips of paper when they moved the type specimens into caves during World War II. Each slip of paper gave the name and the catalog number of a type specimen. So I took all 15-20,000 of them back home with me and entered them into the database. I went back to the British Museum, checked my data against their other records, and found that they have holdings relevant to original descriptions (types of all kinds) for close to 10,000 species of fishes. What has not happened yet is to check my database against their actual specimens. At other museums, however, I was able to physically crawl around looking for the types, and checked them off when I found them.

The first product came out at 2,900 pages. We cut the point size down to 8.5 and 8.4, so it could have been even bigger. I was the only proofer and the manuscript was 1.5 miles long, so there are a lot of errors. We are continuing the project and we're hearing from people about errors already. (I heard about one today.) There are three pictures—all on the cover—so it's duller than a phone book. What you get is the second edition of the Genera of Fishes, the species, and species and genera in a classification, the literature cited, and the appendices, one of which is the interpretation of the code, expanded to include species. You also get a CD-ROM with the data in an application that runs on Windows 3.1, 95, and NT, as well as an electronic version of the printed format (Adobe Acrobat files).

The Databases

Figure 19 shows the main screen that provides access to each of the different databases: Species, Genera, Families (which is partially done), we skipped the fossils—that was not published—a Classification database, References database, Journals database, Museums database, and a Place-Names database. This is a menu to the various "seek" programs we use to get into the databases. For example, because we tied each species to a reference we can call up (seek) the reference and display all of the contained species, in order by page. This makes proofing go much more quickly. So we had a programmer who actually worked with us for a year to do the tailored programming.

Figure 19. The main screen for our taxonomic databases application.

Figure 20 shows the screen for editing a species record. It shows the genus, species, author, date, a unique reference number for the original description, whether it was a species or a subspecies, and the journal, page and figures. A lot of this is assigned automatically. The classification fills automatically if you can get a current genus. The status comment says this is a synonym of another species, and the status has a reference. We recorded the museum that holds the type specimens. We recorded type locality information as a string, but we tried to be very consistent and to have the country for every record, so we could query by countries, for example. (Of course there are problems with this. You find "Georgia" in Eurasia, a "Georgia" in the sub-antarctic, and a "Georgia" in this country.) We also have the location of secondary types.

Figure 20. Screen for editing species records.

A basic reference is shown in Figure 21. We segregated the first two author names for a variety of reasons, for searching and for authorship of species. We have the actual year published, and when we could determine it, the actual month and day of publication for priority reasons. We had to type in the title of the publication, but if the reference was a journal article, we could type in our acronym for the journal and pull in some of the other information automatically. Of course we had to add the volume, number, and pages for the reference to make the citation complete. And we marked electronically which database the reference applied to, so if we just wanted the references for genera we could just select the ones that had a yes in that box.

Figure 21. Screen for editing reference records.

Figure 22 shows an example of our Journal database. We recorded the the BIOSIS abbreviation to help with cross-linking and tracking, but we had to make up a lot of the journal abbreviations. We tended to use just the first word of each of the major words in the title.

Figure 22. Screen for editing journal records.

There was a publication that treated a lot of museum acronyms, but not all, so we had to make up the missing ones. There are about 400 museums that contain type specimens.

We didn't get too far with geographic place names, but we tried to be consistent, and it was very helpful to record the place names used with the very early fish localities. For example, in the 1800s they used "Surinam," but now this has an "e" on the end of it. There were no electronic gazetteers available when we started this, but that would have been very helpful.

Cost and Value of the Catalog

Based on certain assumptions I've estimated the cost was probably almost two million dollars. Grants paid 500,000 of that. That would be in today's dollars. In addition to the authors of the species section (Fig. 23), there were technicians, and two programmers. Doug Long worked for six months on the project, Missy Hoang worked 20-40% of her time for about 10 years, and I've worked 14 years on the project, some of the time at 100% of my time or nearly 100% of my time.

Figure 23. Authors of the species section in the Catalog of Fishes. From left to right: Douglas Long, Mysi Hoang, Carl Ferraris, and William Eschmeyer.

Based on the number of taxa and the degree to which we wanted to look at original descriptions and really get it right, the database is particularly valuable for ichthyologists. In the early years it used to take us months and months, if not years, to assemble the species names and descriptions that we would have to deal with in our area of specialty—like in scorpionfishes for me. Well now you can do that in an hour or two by selecting the family and downloading all the names and references. Because of the way we went about it, we found a number of species that specialists hadn't known about, even when they had been working on a family for 20 years. Dates of publication are taken care of, by and large. The original references—where the species was first published—are all done now. The location of type specimens is not complete, but certainly takes us a long way. So it greatly facilitates the research of ichthyologists.

We let the genera be used as an authority file in some projects, particularly for museum collections through the MUSE project. We've downloaded sections of the database for use in other projects and we expect to do more of that. For example, I worked with some Japanese colleagues who wanted the records for fishes that occur in Japan. They were hoping to get funding from their ministry of education to add information and pictures about the Japanese fish fauna. Because we don't have complete distributions recorded, we had to come at it a couple of different ways. It was easy to get all the records that have a type locality in Japan. Then we went through and flagged the Japanese authors and the Japanese journals and got all references tied to these, and then the species records tied to the references. After stripping out the duplicates, we got a pretty good list of the fishes of Japan, which they could take further.

There's another group, ICLARM in the Philippines, that's building a database called Fishbase, which is mostly fisheries data. They use the Catalog as their authority file, and they've digitized several thousand pictures of fishes, as well as distribution maps. So those are some examples of how the Catalog is being used.

It's also helpful for people working on fishes in a more peripheral way. Fish parasitologists love this because they often get a host name and know don't know anything about it. So they're buying the book, too.

For ecologists or anyone else working with fishes, particularly now that we have on-line searches, I think this will make some major changes in the way they do their work. In the old days you used to go do your study, and then look in the literature to find information that seemed relevant, put it in the introduction and make yourself look real smart. Now you can go look for the near relatives, do the searches on near-relatives, and this will almost certainly effect your experimental design before you start your project, rather than at the end. So I think that's something that biology students should be taught.

I think the 16,000 major taxonomic references we pulled together could represent the start of an international literature cited database. I can see a point not too far out where we could do away with the literature cited sections in the back of our publications; you would just give the author, date, and reference number of anything in the international database. We wouldn't have to repeat all that information in our papers. In addition to saving space and effort, it would improve accuracy. The literature cited is typically the last thing everybody prepares and there are often a lot of mistakes in that information.

Taxonomic names are important for tying biological data together. Do you want to see a distribution map, DNA sequences, pictures, these sorts of things? The fish data could be used in a lot of other databases, and there are some starts underway already. As I said, we're using subsets of this data for other projects, such as the Japan download.

So that's the project. I look forward to hearing from some of you. The botanists have always been better at managing taxonomic information than zoologists, and I'm looking for some new ideas about where I can take my project further.

Transcript of Discussion

Acknowledgements

This material is based on work supported in part by the National Science Foundation under Grant Nos.: 8416085, 8801702, and 9108603.

Biographical Information

Bill Eschmeyer received his B.S. in Fisheries from the University of Michigan in 1961. He received his M.S. and Ph.D. in Marine Biology from the University of Miami in 1967, where he studied the systematics of fishes. His first and only job after leaving Miami has been at the California Academy of Sciences in San Francisco, where he as been for over 30 years. He dedicated the first part of his career to the systematics of scorpionfishes, stonefishes, and their allies. For the last 14 years his main focus has been on databases of taxonomic information for ichthyology. That project is very recently completed and now published as the "Catalog of Fishes." This three volume set includes information on 53,000 species/subspecies, 10,000+ genera/subgenera, 16,000 references.

References

Appendix: Technical Overview of Taxonomic Databases Editor

Databases: Species, genera, families, fossil species, fossil genera, classification, references, journals, museums, place names. The fossil databases have the same structure as the recent ones, plus fields for geologic time period and parts, if not a whole specimen.

1. Species database.
   1. Fields: Genus, species, level (sp., subsp., var.), author, date, reference number, journal acronym, journal or book descriptor, citation (e.g., volume), page, plates. Fields to handle classification. Fields to handle synonymies. Availability of name. Large field for information on type catalogs, nomenclatural items, and current status with citations to author, date, reference number that document status. Fields for types and type locality. "Notes to us" field, and a "done" field. Common names field is not used at present. Plus we use tracking fields, such as vol./p. in Zoological Record.
   2. Links to other databases: References, place names, museums, genera, classification.
   3. Seek programs for entry: Species by author, original genus, species + author, species + genus, family number, journal acronym, journal name, reference number, current genus, append new record.
   4. Output programs: Alphabetical by species, by reference (short version), by reference (long version), first author, author (complete), original genus, journal, locality, by museum (primary types), museum (all types), by museum on cards, family/subfamily, family/subfamily (short version).
2. Genera database.
   1. Fields: Generic name, subgenus of, author, date, page, gender, classification, type designation, fields involved with status, remarks and memo field for nomenclatural items and status references, notes, done, reference number. Plus tracking fields for sources.
   2. Links: Classification, references (+ journal).
   3. Seek programs for entry: Genera by genus + author, author, by classification, reference number, append new record.
   4. Output programs: Genera by reference (short version), reference (long version), name, author, family/subfamily.
3. Families database.
   1. Fields: Family name, original level (e.g., fam. or subfam.), author, year, journal, classification fields, technical nomenclatural fields (such as type genus, author, date), fields for status, notes.
   2. Links: Classification, references.
   3. Seek programs for entry: By author, original level, original name, reference number, type genus, zoological record number/p.
   4. Output programs: Families by reference (long and short version), original name, author, journal, type genus.
4. Classification database.
   1. Fields: Name of taxon, common name, index name (to make index), age (fossil, recent or both), current status, classification fields, "tracking" fields.
   2. Links: Species, genera.
   3. Seek programs for entry: In phylogenetic order, by higher taxon name, update classification.
   4. Output program: Printout of classification.
5. References database.
   1. Fields: First author, initials, second author, initials, other authors, year, month and day, title, journal acronym, journal name, pages, plates, remarks to be published, notes, done, fields to limit to one project (e.g., only references dealing with genera).
   2. Links: Species, genera, families.
   3. Seek programs for entry: By author(s) + (optional) year, journal name, journal acronym, reference number, append record.
   4. Output programs: References by first author, by any author, by journal.
6. Journals database.
   1. Fields: Acronym, standard abbreviation, full name, library call number, notes, Biosis journal acronym.
   2. Links: Species.
   3. Seek programs for entry: By acronym, by journal name, append record.
   4. Output programs: journals by acronym, by abbreviated title.
7. Museums database.
   1. Fields: Museum acronym, full name of museum or institution, notes.
   2. Links: Species.
   3. Seek programs for entry: By acronym, by museum or institution name, append record.
   4. Output programs: Directory of museums, acronyms.
8. Place names database.
   1. Fields: Place name (string, specific to general), a usable name (yes or no), notes.
   2. Links: Species.
   3. Seek programs for entry: By place name, append new record.
   4. Output program: Geographic dictionary.

Other programs: Generation of bibliographies based on selected references (several versions). Use of "filters" can produce a variety of outputs, such as all species with a type locality in Brazil.