Tropical Diversity (2024) 4(1): 1-12.

ISSN: 2596-2388

DOI: 10.5281/zenodo.10863712

RESEARCH ARTICLE

1

Free DELTA: An open-source system for processing taxonomic descriptions

Free DELTA: Um sistema de código aberto para o processamento de descrições taxonômicas

Mauro J. Cavalcanti1 * https://orcid.org/0000-0003-2389-1902, Edinaldo Nelson dos Santos-Silva2

https://orcid.org/0000-0002-3340-4541, Luis Geraldes-Primeiro2 https://orcid.org/0000-0002-3892-8969

1Ecoinformatics Studio, Caixa Postal 18123, CEP 20720-970 Rio de Janeiro, RJ, Brazil.

2Laboratório de Plâncton, Coordenação de Biodiversidade, Instituto Nacional de Pesquisas da Amazônia, Av.

André Araújo, 2936, Aleixo, CEP 69060-000, Manaus, AM, Brazil.

*Email: maurobio@gmail.com

Received: 28, November 2023 / Accepted: 31, January 2024 / Published: 2, February 2024

Resumo O sistema DELTA (DEscription

Language for TAxonomy) consiste de um formato

flexível para a codificação de descrições

taxonômicas e um conjunto de programas para o

manejo e organização da informação taxonômica,

como construção de chaves dicotômicas, produção

de descrições em linguagem natural e

identificação interativa. Todavia, o sistema

DELTA original foi baseado em um modelo de

software proprietário, que coloca limites à

utilização e expansão do sistema. O Projeto Free

DELTA é uma alternativa de software livre e

código aberto para um sistema genérico de

processamento de descrições taxonômicas, que

proporcione uma ferramenta de suporte acessível

para inventários sistemáticos de biodiversidade.

Palavras-Chave: identificação biológica, chaves

taaxonômicas, informática para biodiversidade,

formato DELTA

Abstract The DELTA (DEscription Language for

TAxonomy) system consists in a flexible format

for coding taxonomic descriptions and a set of

programs to handling and organizing taxonomic

information, as construction of dichotomous keys,

production of natural language descriptions, and

interactive identification. However, the original

DELTA system was based on a proprietary

software model, that puts limits to the use and

expansion of the system. The Free DELTA

Project is a free, open-source, software alternative

for a generic system to the processing of

taxonomic descriptions, which offers an

accessible tool to support systematic biodiversity

inventories.

Keywords: biological identification, taxonomic keys,

biodiversity informatics, DELTA format

Cavalcanti et al. (2024)

Free DELTA

2

Introduction

A critical issue for the entire world is

decreasing biological diversity, with detrimental

impacts for food production, pharmaceuticals

research, and the climate, which directly impair

economic systems and human societies (Wilson,

1992). A better understanding of biodiversity is

needed in order to decrease its current rate of loss

and then revert it in the future. The first fundamental

step is to know what exists and how to identify it,

what is of primary concern to the science of

biological systematics (Forey et al., 1994; Vane-

Wright, 1996). As biodiversity losses increase all

over the world, the need to increase the efficiency of

systematists and the usefulness of taxonomic data,

with better means of organizing them (Watson,

1971) has also become evident. The application of

information technology to the storage, retrieval, and

analysis of systematic data provides the primary

means of achieving this (Edwards & Morse, 1995).

The DELTA format (Dallwitz, 1980;

Dallwitz & Paine, 1992) is the most comprehensive

format developed for the electronic storage and

retrieval of taxonomic descriptive data which aimed

to be universally accessible to the biological

systematics research community. Software packages

based on the DELTA format, in turn, provide a

flexible and powerful set of tools for efficiently

handling complex and variable taxonomic

descriptive data and generating essential taxonomic

products like dichotomous keys and natural-

language descriptions (Watson et al., 1988; Dallwitz,

1993; Dallwitz, Paine, and Zurcher, 1993; Dallwitz,

Paine, and Zurcher, 1999; Askevold & O’Brien,

1994; Coleman et al., 2010). The DELTA system

has been demonstrated to be a valuable tool for

biodiversity surveys, where the precise and efficient

identification of organisms is of major importance

(Ellis, 1988; Allkin et al. 1992; Jarvie & Stevens,

1998; Sharkey, 2001).

However, a major problem hampering the

more widespread use of the DELTA system is that

much of the existing software for handling data in

DELTA format is closed source, available only in

binary executable form, and so not amenable to be

modified, upgraded, or ported to different operating

systems, for any parties outside the original

development team. This invariably leads to the end

of support for such software, when funding ceases

and the development team disband. In this paper, an

open source, cross-platform software package for

handling taxonomic descriptive data based on the

DELTA format is presented and described. This

package is developed after the GNU Project and the

Free Software Movement (Stallman, 1999), and

therefore assures the user community complete

freedom not only to run the package, but to modify it

to suit any special needs.

The DELTA System

DELTA stands for DEscription Language

for TAxonomy. The DELTA system consists of a

flexible format for coding descriptive taxonomic

information and a set of associated programs which

manipulate the data to produce natural-language

descriptions and keys, interactive identification and

Cavalcanti et al. (2024)

Free DELTA

3

information retrieval, and to convert data into

formats required by programs for phenetic and

cladistic analysis.

The DELTA format allows all kinds of

characters, both qualitative (binary or multi-state,

ordered or unordered) or quantitative (integer or

real, with units if needed). Comments are allowed

anywhere, and character dependency can be

described. Directives are used to control computer

processing of DELTA-coded data.

Although other generic formats for the

representation of biological descriptive data have

been proposed (White & Allkin, 1992; Maddison et

al., 1997), the DELTA format was especially

designed to provide a generalized system capable of

handling all the different kinds of data used by

taxonomists, as well as for facilities to directly

generate taxonomic products as dichotomous keys

and natural-language descriptions, and translate data

into many external formats required by programs for

classification and interactive identification

(Dallwitz, 1993).

The original DELTA system has been under

development since the mid 1970's by Mike Dallwitz,

Toni Paine, and Eric Zurcher at CSIRO Division of

Entomology, Canberra, Australia (Dallwitz, 1980).

Similar packages for processing of DELTA data

were also developed in the United Kingdom by

Richard Pankhurst of the Royal Botanic Gardens at

Edinburgh (Pankhurst, 1986), in The Netherlands by

Eric Gouda at the Botanic Gardens of Utrecht

University (Gouda, 1996), in Australia by Nicholas

Lander at the Western Australian Herbarium

(Lander, 1993), in Spain by Antonio Valdecasas at

the Museo Nacional de Ciencias Naturales (Bello et

al. 1996), and in Brazil by Mauro J. Cavalcanti at

Museu Nacional/Universidade Federal do Rio de

Janeiro (Cavalcanti, 1996). All these packages are

closed source, proprietary software, available only in

binary executables and restricted to a single

operating system. Later on, DELTA programs based

on the open source model were developed in

Germany by Gregor Hagedorn at the Institute of

Microbiology, Federal Biological Research Center

(Hagedorn 2001), in the United States by Michael

Bartley and Noel Cross at the Arnold Arboretum of

the Harvard University and then in Germany by

Dieter Neubacher and Gerhard Rambold at

Mycology Department, University of Bayreuth

(Neubacher & Rambold, 2005), in Italy by Claudio

Rivetti and Riccardo Percudani at University of

Parma (Rivetti & Percudani, 1999), and in Australia

by the Atlas of Living Australia (2014).

In 1988, DELTA was adopted by the

International Working Group on Taxonomic

Databases for Plant Sciences (TDWG) as an

internationally recognized standard for data

encoding and exchange. Combining traditional

taxonomic practices with modern information

technology, the DELTA system has since became

one of the most valuable and widely used

computerized tools for biodiversity research.

Many datasets have already been prepared

with DELTA, for flowering plant families, grasses,

sedges, legumes, beetles (adults and larvae), ants,

corals, plant viruses, etc. Several of these datasets

incorporate line, gray-scale and color illustrations,

and some are freely available on the Internet; at the

time of writing, there are 61 datasets in the DELTA

website (https://www.delta-

intkey.com/www/data.htm) and 28 datasets in the

NaviKey website

(http://www.navikey.net/webprojects.html).

Cavalcanti et al. (2024)

Free DELTA

4

The Free DELTA Project

The Free DELTA project was launched in

April, 2000 with the aim of creating a complete

cross-platform, free, open-source software system

for processing taxonomic descriptions coded in

DELTA format, following the decision by CSIRO

Division of Entomology to stop funding the

development of the original DELTA programs.

The Free DELTA package was designed

from to start to include a library for parsing DELTA-

format files plus all the utilities needed to process

taxonomic data coded in DELTA: format-

conversion, key-construction, distance-matrix

generation, and interactive identification programs,

as well as a specialized editor for handling

taxonomic descriptive data. However, although the

Free DELTA programs should be able to process

regular DELTA-coded files, they would not be

identical to the original DELTA system. Free

DELTA software would incorporate several

improvements, based on our experience with other

taxonomic computing systems. In particular, it

should offer a better treatment of quantitative data

(with automatic breakdown of real numeric

characters into classes and computing of descriptive

statistics for each), as well as more analytical and

format-conversion procedures.

The Free DELTA parser libraries should

provide a standard, versatile and extensible

collection of routines for handling DELTA datasets.

This set of routines is currently available in C++,

Python, and Object Pascal programming languages

and provide the basic support for the Free DELTA

software; furthermore, they are ready for

incorporation within individual application programs

or database systems as they are created by different

software developers. These routines are released as

complete, open source code. So, a user who needs a

different system for taxonomic data management is

able to develop it himself, or hire any available

programmer or company to develop it for him or

her, while retaining the benefits of a standard format

for encoding taxonomic data. The availability of full

source code also enables a user to tailor these

routines according to his or her more specific needs.

It is hoped that this will help to further expand the

DELTA user community, as well as to strengthen

the role of DELTA as a standard taxonomic data

coding and transfer system.

In July, 2000 the first DELTA parser library,

written in C++ by Denis Ziegler, with additional

contributions by Guillaume Rousse and Bastiaan

Wakkie, has been developed for the Free DELTA

Project.

In May, 2005 the Free DELTA project was

registered at SourceForge, the world's largest open-

source software development website, with the

largest repository of open-source code and

applications available on the Internet, and where the

Free DELTA website is now to be located on a

permanent basis.

In February, 2010 the PyDELTA parser

library, written in Python by the Mauro Cavalcanti

and Thomas Kluyver was released as yet another

contribution to the software infrastructure of the

Free DELTA Project; this library would later be

used as the basis for the development of the first

prototype of the Free DELTA Editor.

In February, 2013 the former CSIRO

DELTA Development Team released the

Cavalcanti et al. (2024)

Free DELTA

5

FORTRAN source code of the main “classic”

DELTA programs (CONFOR, KEY, and DIST),

which were then incorporated into the core of the

Free DELTA Project; the source code of these

programs was not modified by the Free DELTA

project, and new executable files were not generated.

In July, 2016 a prototype of the Free

DELTA Editor was developed as a cross-platform

desktop application written in the Python

programming language and using the PyDELTA

parser library. This tool was released on an

experimental basis in December, 2017.

Finally, in January, 2020 a production

version of the Free DELTA Editor was released to

the public, That is a much-improved version of the

experimental software previously developed,

entirely re-implemented in FreePascal/Lazarus as a

full cross-platform desktop application.

The Free Delta Editor

The Free Delta Editor (FDE) is a cross-

platform software package for editing DELTA data

files and running DELTA based programs

(CONFOR, KEY, DIST, INTKEY).

This tool provides an interactive, user-

friendly interface to DELTA-format files as a real

DELTA data entry/maintenance program. With it, it

is possible to insert or delete items, characters, and

states, without having to worry about the

consistency of the DELTA files. A database-like

interface makes it easy to setup and maintain the

main DELTA data files (ITEMS, CHARS and

SPECS), as well as several of the DELTA directives

files (TONAT, TOKEY, TODIS). A spreadsheet-

like visualization facility for item and character data

is also available (Figure 1).

Figure 1 - The Free Delta Editor database viewer (A) and data matrix viewer (B).

Cavalcanti et al. (2024)

Free DELTA

6

It also makes possible to start all DELTA

programs and create dichotomous keys for

identification, natural-language descriptions,

statistical reports, distance matrices, cluster and

ordination analysis, providing a viewer to look at

and print the results (output files) (Figures 3 and 4).

Figure 3 – Natural-language description viewer.

Cavalcanti et al. (2024)

Free DELTA

7

Figura 4 - Dichotomous key viewer.

The Free Delta Editor is written in

FreePascal/Lazarus, a free, cross-platform object-

oriented programming language for Rapid

Application Development. Therefore, the program is

platform-independent, and can run without

modifications on any PC compatible with the x86

architecture, under GNU/Linux, MacOSX, and MS-

Windows.

Its user interface resembles, but is not a

clone of, that of the CSIRO and ALA DELTA

Editors. A major difference is that in the Free Delta

Editor all views are read-only and there is only one

way of describing a taxon; the ‘Items’ and

‘Characters’ panes are read-only, and attributes may

be edited only in the ‘Value’ pane This is an attempt

to make the process of creating a descriptive

database more orderly and less confusing than it

seems to be in the case of the CSIRO/ALA Editors;

these editors also have a ‘Values’ pane, but, in

addition, attributes can be edited in the ‘Characters’

pane, which means that there is more than a way of

editing the character attributes, therefore making the

process prone to confusion for the inexperienced

user.

Another important difference concerns the

data file format: the Free Delta Editor uses

conventional zip-compressed files (with the .dtz

extension) to store all DELTA-formatted data. At a

minimum, these files must include the three basic

DELTA files (CHARS, ITEMS, and SPECS) and

can be created either by exporting an already

existing dataset from the CSIRO Delta or ALA

Delta programs and compressing them into a zip file

(changing its extension to .dtz), or by using the Free

Delta Environment Import option from the File

menu.

The Free Delta Editor provides for

clustering and ordination (using principal

coordinates) of the distance matrix computed by

DIST, using for that the powerful R free software

environment for statistical computing and graphics

(Ihaka & Gentleman, 1996). Therefore, R must be

installed on a users machine in order for performing

cluster and principal coordinates analyses with the

Free Delta Editor (Figure 5).

Cavalcanti et al. (2024)

Free DELTA

8

Figure 5 – Cluster analysis viewer.

The Free Delta Editor is initially available in

English, French, Spanish, and Brazilian Portuguese,

but any interested person can provide a translation

for her/his native language. No knowledge of

programming is required, just a freely available and

easy to use free software tools like POEdit or

Virtaal.

The current emphasis of the Free Delta

Editor is on the production of natural language

descriptions, dichotomous keys, and phenetic

analyses. Support for interactive identification is

quite limited and restricted to the MS-Windows

version, which simply calls the CSIRO DELTA

Intkey program, if installed. Support for Intkey is

not available in the GNU/Linux version. The

primary goal of the Free Delta Project in the near

future is the development of a new, open source,

cross platform interactive identification program

(‘Freekey’), similar, but not identical to, the current

CSIRO DELTA Intkey. This will be implemented as

a separate application, which will use the same data

files created and edited using the Free Delta Editor

(therefore no 'format conversion' will be necessary).

Currently, binary executables are provided

for MS-Windows (XP, 7, 8, 10) and GNU/Linux

systems (Debian); a Mac OS version is not

available; however, since the development tool now

used for the implementation of the Free Delta

software (FreePascal/Lazarus) is fully cross-

platform, it should be fairly simple to compile a

version for Mac's.

All Free DELTA software is released as free

software, under the GNU General Public License

(Stallman, 1999).

The Free Delta Editor applied to biodiversity

research

In a systematic study on Neotropical

microcrustaceans, the Free Delta Editor was applied

Cavalcanti et al. (2024)

Free DELTA

9

to create a large database of 2700 characters and 37

species in the copepod genus Notodiaptomus Kiefer,

1936(Copepoda, Calanoida, Diaptomidae). The

ability to store and process this volume of taxonomic

data has made the Free Delta Editor a unique

multiplatform software package among DELTA-

based programs.

Using the DELTA programs directives

(commands), it was possible to obtain automatic

description in natural language, dichotomous

identification keys for males and females, an

assisted interactive key with images, annotations,

and synonyms (Figure 6), as well as a character

matrix for the analysis of phylogenetic relationships

using the PAUP program (Swofford, 2003).

Undoubtedly, the possibility of managing these

types of data, in the scale and complexity, was

decisive for the systematic study of such a

morphologically diverse group of organisms as

microcrustaceans (see Geraldes-Primeiro, 2023).

Indeed, the versatility of character matrices

generated by DELTA-based programs extends to

their compatibility with various phylogenetic

inference software, such as PAUP (Swofford, 2003)

and TNT (Goloboff et al., 2008). This allows

researchers to seamlessly integrate their taxonomic

data into different analytical tools, enabling a

broader spectrum of analyses and enhancing the

utility of the generated datasets.

Tropical Diversity (2024) 4(1): 1-12.

ISSN: 2596-2388

DOI: 10.5281/zenodo.10863712

RESEARCH ARTICLE

10

Figure 6 – DELTA Intkey program running the interactive key for the species of Notodiaptomus. A. Initial

screen with the characters used (left) and taxa for identification (right); B. Secondary screen with the selected

character (top left) and remaining taxa (top right); C. Screen illustrating the states observed for the selected

character.

Cavalcanti et al. (2024)

Free DELTA

11

Conclusion

Recent advances in the application of

computer technology to taxonomy offer the prospect

of fast, large-scale character extraction and

automated identification of organisms using

computer vision and neural networks (Weeks &

Gaston, 1997; MacLeod, 2008; Gaston & O’Neill,

2004; La Salle et al., 2009). Coupled with the more

conventional methods of computer-aided

identification best represented by the DELTA

system, such techniques can provide a much-needed

boost in the productivity of taxonomists faced with

the biodiversity crisis. In this context, we expect that

the availability of a free, open source, cross-platform

software tool for the storage, retrieval, and analysis

of DELTA-coded taxonomic descriptive data will

prove a useful contribution to the present and future

generations of taxonomists worldwide. This further

justifies the importance of such initiatives as the

Free-DELTA software, which in addition to

ensuring speed and consistency in the management

of complex taxonomic data, provides greater

precision to achieve one of the most fundamental

logical precepts of science: reproducibility (Peng,

2011)..

Acknowledgments

We thank Drs. Mike Dallwitz (CSIRO

Division of Entomology [retired], Australia), Sami

Rabei (Mansoura University, Egypt), and Gleyson

Desidério (Instituto Nacional de Pesquisas da

Amazônia, Brazil) for their reviews of the

manuscript, providing corrections and useful

suggestions which much contributed to the

improvement of this paper.

References

Allkin, R., Moreno, N. P., Gama Campillo, L. &

Mejia, T. 1992. Multiple uses for computer-

stored taxonomic descriptions: keys for

Veracruz. Taxon 41: 413-435.

Askevold, I. S. & O’Brien, C. W. 1994. DELTA, an

invaluable computer program for generation of

taxonomic monographs. Annals of the

Entomological of America 87: 1-16.

Atlas of Living Australia. 2014. Open DELTA – A

port of the DELTA (DEscription Language for

TAxonomy) suite of applications into Java.

https://github.com/atlasoflivingaustralia/open-

delta.

Bello, E., Valdecasas, A. G.- & Elvira, J. R. 1996. A

Spanish and English DELTA Editor: EDEL

95. DELTA Newsletter 12: 24-26.

Cavalcanti, M. J. 1996. Designing a Graphical User

Interface for DELTA: Some considerations.

DELTA Newsletter 12: 2-4.

Coleman, C. O., Lowry, J. K. & Macfarlane, T.

2010. DELTA for Beginners: An introduction

into the taxonomy software package DELTA.

ZooKeys 45: 1-75.

Dallwitz, M. J. 1974. A flexible computer program

for generating identification keys. Systematic

Zoology 23: 50-57.

Dallwitz, M. J. 1980. A general system for coding

taxonomic descriptions. Taxon 29: 41-46.

Cavalcanti et al. (2024)

Free DELTA

12

Dallwitz, M. J. 1993. DELTA and INTKEY. In:

Fortuner, R. (ed.). Advances in Computer

Methods for Systematic Biology: Artificial

Intelligence, Databases, Computer Vision. The

Johns Hopkins University Press, Baltimore,

p. 287-296.

Dallwitz, M.J. & Paine, T.A. 1992 onwards.

Definition of the DELTA format.

https://www.delta-

intkey.com/www/standard.htm

Dallwitz, M.J., Paine, T.A. & Zurcher, E.J. 1993

onwards. User’s guide to the DELTA System:

a general system for processing taxonomic

descriptions. https://www.delta-

intkey.com/www/uguide.htm

Dallwitz, M.J., Paine, T.A. & Zurcher, E.J. 1999

onwards. User’s guide to the DELTA Editor.

https://www.delta-intkey.com/www/delta-

ed.htm

Edwards, M. & Morse, D. R. 1995. The potential for

computer-aided identification in biodiversity

research. Trends in Ecology and Evolution 10:

153-158.

Ellis, D. V. 1988. Quality control of biological

surveys. Marine Pollution Bulletin 19: 506-

512.

Forey, P. L., Humphries, C. J., & Vane-Wright, R. I.

(Eds.). 1994. Systematics and Conservation

Evaluation. Clarendon Press, Oxford.

Gaston, K. J. & O’Neill, M. A. 2004. Automated

species identification: why not? Philosophical

Transactions of the Royal Society, Biological

Sciences 359: 655-667.

Geraldes-Primeiro, L. J. O. 2023. Sistemática

Biológica de Notodiaptomus (Copepoda,

Calanoida, Diaptomidae). Tese de Doutorado.

Instituto Nacional de Pesquisas da Amazônia –

INPA, Manaus, Amazonas. Ministério da

Ciência, Tecnologia e Informação – Governo

Federal do Brasil. 662 pp.

Goloboﬀ, P. A., Farris, J. S., & Nixon, K. C. 2008.

TNT, a free program for phylogenetic analysis.

Cladistics 24: 774-786.

Gouda, E. J. 1996. TAXASOFT DELTA Editor.

DELTA Newsletter 12: 12-13.

Hagedorn, G. 2001. Making DELTA accessible:

databasing descriptive information. Bocconea

13: 261-280.

Ihaka, R. & Gentleman, R. 1996. R: A language for

data analysis and graphics. Journal of

Computational and Graphical Statistics5: 299-

314.

Jarvie, J. K. & Stevens, P. F. 1998. Interactive keys,

inventory, and conservation. Conservation

Biology 12: 222-224.

Lander, N. S. 1993. DELTA Menu System for

Windows. DELTA Newsletter 8: 10-12.

La Salle, J., Wheeler, Q., Jackway, P., Winterton, S.,

Hobern, D. & Lovell, D. 2009. Accelerating

taxonomic discovery through automated

character extraction. Zootaxa 2217: 43-55.

MacLeod, N. (ed.) 2008. Automated Taxon

Identification in Systematics: Theory,

Approaches and Applications. CRC Press,

Boca Raton, FL.

Maddison, D. R., Swofford, D. L. & Maddison, W.

P. 1997. NEXUS: an extensible file format for

systematic information. Systematic Biology

46: 590-621.

Neubacher, D. & Rambold, G. 2005. NaviKey – a

Java applet and application for accessing

descriptive data coded in DELTA format:

http://www.navikey.net.

Cavalcanti et al. (2024)

Free DELTA

13

Pankhurst, R. J. 1986. A package of computer

programs for handling taxonomic databases.

Computer Applications in the Biosciences 2:

33-39.

Pankhurst, R. J. 1991. Practical Taxonomic

Computing. Cambridge University Press,

Cambridge.

Peng, R. D. 2011. Reproducible research in

computational science. Science 334: 1226-

1227

Rivetti, C. & Percudani, R. 1999. WebDelta – A set

of Perl scripts for Web entry of DELTA

descriptions.

https://web.archive.org/web/20050227052425/

irisbioc.bio.unipr.it/webdelta/webdelta.html.

Sharkey, M. J. 2001. The All Taxa Biological

Inventory of the Great Smoky Mountains

National Park. Florida Entomologist 84: 556-

564.

Stallman, R. 1999. The GNU Operating System and

the Free Software Movement. In: DiBona, C.,

Ockman, S. & Stone, M. (eds.). Open Sources:

Voices from the Open Source Revolution.

O’Reilly Media, Boston, p. 31-38.

Swofford, D. L. 2003. PAUP*: Phylogenetic

Analysis Using Parsimony (*and other

methods). Sinauer Associates, Sunderland.

Watson, L. 1971. Basic taxonomic data: the need for

organisation over presentation and

accumulation. Taxon 20: 131-136.

Watson, L., Dallwitz, M. J., Gibbs, A. J., &

Pankhurst, R. J. 1988. Automated taxonomic

descriptions. In: Hawksworth, D. L. (ed.).

Prospects in Systematics. Clarendon Press,

Oxford, p. 292–304.

Weeks, P. J. D. & Gaston, K. J. 1997. Image

analysis, neural networks, and the taxonomic

impediment to biodiversity studies.

Biodiversity and Conservation 6, 263-274.

White, R. J. & Allkin, R. 1992. A language for the

definition and exchange of biological data sets.

Mathematical and Computer Modelling 16:

199-223.

Wilson, E. O. 1992. The Diversity of Life. Harvard

University Press, Cambridge, MA.

Vane-Wright, R. I. 1996. Systematics and the

conservation of biological diversity. Annals of

the Missouri Botanical Garden 83: 47-57.