*Bostock, Peter. Queensland Herbarium, Meiers Rd, Indooroopilly, Qld 4068. http://www.ozemail.com.au/~pbostock

Botanical Latin Translation.

TRANSLAT and LATIN are DOS programs which can assist in the translation of Botanical descriptions and diagnoses from Latin to English and vice versa. The programs will run under MS DOS versions 3.3 to 7 and Microsoft Windows 3.1x, Windows 95 and NT3.51 (service pack 5).

Botanical Latin, as practised over the past 300 or so years, is an artificial language, its words based on classical and medieval Latin and Greek, and its construction more or less on classical Latin. Latin is a highly inflected language, with a supposed regular construction according to classical grammars. In fact, the regular construction is artificially derived, and has more exceptions than regular rules. Nevertheless, botanical Latin is ideally suited to computer programming, so long as a colloquial English output is not required.

TRANSLAT and LATIN work by searching dictionaries (databases) of adjectives, verbs, nouns and pronouns, applying each number, gender, case etc in turn until a match is found. The programs check for comparative, superlative and diminutive adjectives, prefixes such as semi- or bi-, phrases such as 'toto caelo' (the whole [sky]), verbal derivatives such as participles and gerunds, and contractions such as 'm' (meter), c. (circa). TRANSLAT reads a text file of botanical Latin, and produces a file containing English meanings of the words in the input file. LATIN requires a nominative singular (present indicative 1st person singular for verbs) Latin word, and declines that word based on the user's requirements; it also remembers previous grammatical constructs, eg. prepositions that govern accusative, and will default the case, number and person accordingly.

Boston, Tony. Environmental Resources Information Network, Environment Australia.

Mapping and Modelling.

There are many examples of desktop mapping and GIS products on the market today. Finding the right product for a particular type of user or intended use is not always easy. Users range from the individual systematics researcher through to a number of users sharing a corporate mapping facility for a large institution. An overview of the most popular vector based desktop mapping products, MapInfo Professional, ArcView, and GeoMedia is provided. The features, functionality, ease of use, target market and cost of these packages will be documented with recommendations on the applicability of these products for users of biological data. The impact of the web on GIS and desktop mapping tools is also explored and the web-enabled versions of these products, namely MapInfo ProServer, ArcView Internet Map Server and GeoMedia Web Map are reviewed. These products allow maps and GIS functionality to be made available on the Internet.

A summary of modelling methods and tools for the prediction of the spatial distribution and abundance of species is provided. Five broad categories of modelling methods are recognised (Austin 1994), namely heuristic, statistical, rule-based, terrain and scenario models. Modelling tools used at ERIN including BIOCLIM, ANUClim, and GARP are described in some detail. Online (Internet) access to such modelling tools is described along with recent and future developments which allow Internet users to submit their own data to ERIN modelling facilities.

The impact of migration of Australian spatial data to the Geocentric Datum of Australia (GDA) from the Australian National Spheroid (ANS) by the year 2000 is discussed with particular reference to impact on geocoding of biological data.

Reference: Austin, M.P. (1994) Data capability, Sub-project 3, Modelling of landscape patterns and processes using biological data. Division of Wildlife and Ecology, CSIRO, Canberra.

*Brooker, Ian, John Connors and Andrew Slee, Centre of Plant Biodiversity Research, GPO Box 1600, Canberra, ACT, 2601.

EUCLID.

EUCLID is the first comprehensive interactive key to the identity of all the 310 eucalypts in that part of southern Australia defined by New South Wales, Victoria, Tasmania and south-eastern South Australia. EUCLID is designed for use by anyone with the most basic knowledge of plants. The user chooses the characters that are available. Absent or uncertain features are simply by-passed.

The traditional word key is supplemented by character illustrations which are "live" and may be selected without recourse to word descriptions. For final checking, most species are illustrated by a collage of colour pictures comprising the tree in its natural environment, the habit, bark, leaf venation, buds, flowers, fruits, and a map showing the natural distribution.

The database includes extra information on common names, nomenclature, synonymies, an English language description for each species, notes on related or similar looking species and how they may be distinguished, selected references to other works on eucalypts and also uses for the species.

Brooks, Alan. K. Information Technology Development Officer, National Herbarium of New South Wales, Royal Botanic Gardens, Mrs Macquaries Road, Sydney NSW 2000.

Specimen databases - today and into the future.

Specimen databases first appeared on the Herbarium time scale in Australia in the 1970's. The evolution of specimen databases since then has been steady. Today most Herbaria have stable systems that address some of the needs of specimen collections and related research.

In this presentation four main aspects of specimen databases will be considered:- hardware and software platforms (e.g. KE's Texpress for large systems, Access for smaller PC systems) for database systems; the What? and Why? questions of database existence; databases vs information systems, where to next?; interchange standards, a recipe to avoid doubling up (or worse) on data entry - the expensive bit for herbaria with large collections.

Specimen databases have at times been seen as an end instead of an essential tool. We must start to maximise the investments we have in data and, as with any research program, make sure we address the questions we originally set out to answer.

HISCOM has successfully interchanged data between major Australian Herbaria and some overseas institutions. This represents large savings in re-entering data and also provides much higher quality of data by reducing levels of error. In the future HISCOM will continue on distributed systems across the internet - the "virtual National Herbarium of Australia".

*Brooks, Alan. K. Information Technology Development Officer, National Herbarium of New South Wales, Royal Botanic Gardens, Mrs Macquaries Road, Sydney NSW 2000.

The National Herbarium of New South Wales Information System prototype: turning data into useable information.

Abstract not available.

Chapman A.R. Western Australian Herbarium, Department of Conservation and Land Management, Como, Western Australia, 6152.

Taxonomic descriptive data.

This paper discusses considerations for the electronic capture and management of taxonomic descriptive data. Descriptive taxonomic information is usually presented as a synthesis of data drawn from a variety of sources. Morphological data is extracted from specimens, existing literature or other specialist knowledge bases and combined with nomenclatural data, distribution and ecological information, as well as photographic images, drawings and maps.

Software employed to aid in the compilation of this data can range from a simple word-processor or database to specialist applications. These specialist applications can be categorised into a number of functional types such as data editors, data managers, menu shells and interactive keys. In this paper a range of programs exemplifying the various functional software types is examined.

Taxonomic descriptive database software can be broadly classified into those that comply with and work from the DELTA (Descriptive Language for Taxonomy) data standard and those which are independent of the standard. Adoption of the DELTA standard for codifying taxon descriptions allows the data to be more flexibly utilised by a range of programs, however, criteria for the selection of an appropriate methodology are dependent on the desired functionality, computer literacy and most importantly the expected target audience.

Taxon images are an increasingly important component of any electronic presentation of descriptive information and issues such as image capture, storage, management, identification and file format must be fully considered as the task is expensive in terms of time to source, scan and prepare the image, and in terms of storage of intermediate and final image files. Taxon and character illustrations and maps are commonly presented as images and so experience the same constraints.

Methods for disseminating taxonomic descriptive data are changing rapidly, particularly as Internet access and usage become common-place. Interactive identification and information retrieval via the World Wide Web is being actively explored in a number of ways. However, issues of accuracy, currency, custodianship and the role of the institution in the long-term maintenance and integration of data must be considered.

*Chapman A.R. Western Australian Herbarium, Department of Conservation and Land Management, Como, Western Australia, 6152.

A prototype for an integrated flora information system.

The need to integrate the gamut of electronic flora data in a simple, informative and readily available manner has coincided with the rise of the World Wide Web as a major information medium. The WA Herbarium has developed a working prototype to investigate the timely delivery of current flora information.

This prototype builds onto the herbarium's specimen and census databases, which are constructed in Knowledge Engineering's TEXPRESS database management system, to integrate taxon images, species distribution maps and protologues into a simple but powerful flora information system available throughout the departmental intranet.

*Chapman A.R. Western Australian Herbarium, Department of Conservation and Land Management, Como, Western Australia, 6152.

An Interactive Key to the Western Australian Flora.

A project to score fundamental descriptive data for over 11,000 taxa occurring in the state is nearing completion. It has been developed jointly by the Wildflower Society of WA, the WA Herbarium and Kings Park and Botanic Garden as a reworking of Beard's Descriptive Catalogue of WA Plants, 2nd edition, (1970). The project has used the DELTA methodology to code the raw data and present the results in hard copy and interactive key formats, which will be demonstrated.

Croft, Jim R. Australian National Herbarium & Australian National Botanic Gardens, GPO Box 1600, Canberra, ACT 2601, AUSTRALIA

Taxonomic and nomenclatural data.

Reliable taxonomic and nomenclatural data is pivotal to any study or assessment of environment and its biota. Yet despite almost a quarter of a millennium of modern scientific endeavour in this area we still do not have a single agreed protocol for documenting this activity for all life forms.

The data sorting, linking, and retrieval power of computers lends itself to processing the very large number of names and synonyms (and variations of these) in an efficient and readily accessible way. Yet, despite this, there is no common, shared or universal computer application to handle this, even for a single discipline.

Moreover, not only do biologists not have access to a single application, they do not have access to the standard authority file which should form the backbone of such an application. In some cases there is no agreement on what should be the authority files, in some cases the data is not available for unrestricted general use, in some cases the data simply does not exist in computer accessible form.

As a result, there have been a number of attempts to design databases to handle the arcane intricacies of biological taxonomic and nomenclatural practice, and to compile and endorse appropriate authority files. Many of these efforts have proceeded independently, largely to meet in-house institutional requirements, sometimes in competitive isolation from similar efforts in other institutions. There have been several attempts at collaboration and sharing resources, but in most cases the task of balancing the common need with local demands and practice has proven too difficult.

In some instances, the design of taxonomic databases has been part of a lager specimen database project, and the complexity of the organism name component has suffered at the hands of pragmatism and overall project priorities. Similarly problems have arisen with the authority file data needed to populate these taxonomic applications.

The lack of a common computing and communications platform for biological collections data has ensured a proliferation of solutions to a single problem. Nevertheless, there are a number of good database designs and applications for biologists to consider, and the choice of these will depend on the task to be performed and the way in which the data is to be used.

While it is most unlikely that biologists will ever use a single agreed computing platform, a current trend towards distributed platform-independent applications and data delivery based on agreed standards is very encouraging.

Croft, Jim R. Australian National Herbarium and Australian National Botanic Gardens, GPO Box 1600, Canberra, ACT 2601, AUSTRALIA

Towards a Virtual Australian Herbarium.

Declining resources and increasing demands on herbaria and museums in Australia and overseas are hindering the ability of these institutions to meet routine operational demands and maintain acceptable standards of identification and curation of the collections.

While attempting to serve local constituencies and meeting the requirements of local funding agencies, institutions are, in part, duplicating work being done, or already completed, at other sites around the world.

Logic suggests that collaborative effort and shared responsibility for common biological information would be the most efficient use of diminishing resources, both at the national and global scale. Yet, despite the apparently obvious nature of this ideal, it has been difficult to achieve. There have been a number of national and international attempts at collaborative biological database projects, but to date, none have been spectacular in their success.

With its relatively small number of States and Territories and its manageable number of significant biological collections, nearly all of which are government funded, and its reasonably sophisticated computing and communications infrastructure, Australia is ideally placed to achieve a nationally coordinated management of and access to its biological collections data. Peak national bodies such as Council for Heads of Australian Herbaria (CHAH) and CHAFC and their subcommittees such as HISCOM are well placed to facilitate such a project.

In the field of botany, a distributed database, or shared common database of collections data would lead to significant economies in the areas of database design, development and maintenance of authority files, data entry, exchange of collection information, access to a wider range of verifiable specimen-backed information, and so on. Past efforts to achieve this with common hardware and software platforms between institutions have failed and will continue to fail for a variety of technical and political reasons.

A contemporary approach using the Internet and distributed database technologies, given commitment and good will on the parts of participating institutions and their client groups, has a high likelihood of success.

An international project on plant name data involving the Harvard University Herbaria, the Royal Botanic Gardens Kew and the Australian National Herbarium may provide a promising model on which an Australian virtual herbarium could be built. Using the technology developed as part of this project and extending it to a wider range of collections information, data from Australia herbaria could be maintained and interrogated as a combined unit.

*Croft, Jim R. Australian National Herbarium and Australian National Botanic Gardens, GPO Box 1600, Canberra, ACT 2601, AUSTRALIA

The redevelopment of the Australian Plant Name Index.

Abstract not available

*Crosby, Trevor K., John S. Dugdale ¹ and Dominic J. Thoreau Landcare Research, Private Bag 92170, Auckland, New Zealand and ¹Private Bag 6, Nelson, New Zealand. [E-mail: CrosbyT@landcare.cri.nz]

Images of primary type specimens of New Zealand Lepidoptera on the Web and CD-ROM: new access to collection information.

The Internet offers new opportunities in delivering image and text information about specimens in biological collections to users. As part of a larger project examining methods of providing on-line access to specimen information, identification-quality images and collection information for 1800 primary type specimens (name bearers) of New Zealand's indigenous moths and butterflies (Lepidoptera) are now accessible via the Internet (URL http://www.marc.landcare.cri.nz/).

A CD-ROM version of the project has now been made, and the same Internet browser programs are used to view the information. This will allow independent computer users having a CD-ROM drive to be able to view and print the information without requiring on-line access.

For the first time anyone in the world can check on-line or on CD the identity of approximately 90% of New Zealand's Lepidoptera species. The barrier of distance has been largely eliminated, as a user can view an image of a name-bearing specimen irrespective of whether the name-bearer itself is part of The Natural History Museum collection in London, or it is housed in the New Zealand Arthropod Collection in Auckland or another New Zealand collection.

Side-by-side comparisons of images can be made on computer monitors, allowing users to check for differences between related species, or to confirm that different names are synonymous and refer to the same species. The images and collection information can be printed as required.

Systematists will always need to examine the physical specimens in their detailed studies, but information in a 'virtual collection' similar to this will be of value to many users who wish to make a first assessment of the identity of their specimens.

Funding Source: New Zealand Foundation for Research, Science, and Technology under contract no. C09401.

*Dauncey, E.A. Jackson, M. Rayner T.G.J. and Shah-Smith, D.A. Royal Botanic Gardens, Kew, Richmond, Surrey TW9 3AB, United Kingdom.

A computerised identification system for poisonous plants and fungi.

World-wide each year there are many thousands of cases of human poisoning by plants and fungi; Poisons Centres report that they account for up to 11% of calls received. Such cases are difficult to manage because the identity of the plant or fungus is often not known or they are incorrectly identified. In such instances inappropriate treatment may be administered. In addition, staff time and hospital resources are wasted if the plant or fungus is not toxic as is frequently the case.

Since 1990 the Royal Botanic Gardens, Kew and the Poisons Unit of Guy's & St Thomas' Hospital Trust have undertaken joint ventures to provide solutions for the prevention and treatment of plant and fungal poisoning. A central activity has been the design and publication of interactive software to enable people who are unfamiliar with plants and fungi to identify them quickly and accurately. The British Computer Society awarded this innovative software one of its three top prizes in 1994. Designed primarily for use by medical professionals in emergency situations, the first product to use this software, Poisonous Plants in Britain and Ireland, was published on CD-ROM in 1995. Its companion disc, Poisonous Fungi in Britain and Ireland, is due to be published in late 1997.

The identification system is comprised of three parts:

1. a dataset consisting of five files largely based on the DELTA (DEscription Language for TAxonomy) format, including a list of distinguishing characters and a list of taxa or 'suspects';

2. an 'identification engine' coded in C and written specifically for the project by a UK company, System Simulation Ltd.;

3. a highly visual user interface written by the Computing Section, Royal Botanic Gardens, Kew in Visual Basic to run in Microsoft Windows.

The interface prompts the user to identify a plant or fungus by answering non-technical questions, many of which are illustrated by graphic images. According to the user's answer the engine recalculates the number of 'suspects' and selects the next 'best' question to achieve an identification by asking the least number of questions. The selection of the 'best' question is made using an information theoretic measure based on the ID3 algorithm. A second algorithm, the minimax game tree algorithm, is also available.

When sufficient questions have been answered to reduce the number of suspects to five or fewer these can be viewed. Identifications are visually confirmed by the examination of photographic images, many of which show close-ups of important morphological characters. There are up to ten images for each suspect. The user is then able to access and print toxicological information and morphological descriptions.

The software is generic and, with the incorporation of an appropriate data-set, it can be used to identify the poisonous plants and fungi of other countries; a version for German plants, Die Giftpflanzen Deutschlands, will be published in early 1998. It can also be used for many other applications, for example, poisonous snakes or general Floras or Faunas.

*Dauncey, E.A., Jackson, M. and Kirkup, D.

[The Royal Botanic Gardens Kew's Index Kewensis on CD ROM: a index to protologues of the Worlds higher plant species brought into the electronic era].

Abstract not available.

Hastings, Anne.

Digital Technology and its role in Biological Imaging at CSIRO, Entomology.

This presentation will demonstrate the use of a flatbed scanner and a video camera mounted on a microscope to collect raw digital image data.

It will show how commercially available graphic software packages like "Adobe Photoshop" and "Apple Video Player" can then be used to grab, enhance, change and create accurate biological images.

Our first electronic publication was "Beetle Larvae of the world". The second, "Beetles of the World" is a companion volume that we are still working on. Both these projects use interactive identification software called INTKEY. "Beetles of the World" will enable any beetle to be identified at least to its family. In some cases an identification to a genus can also be made.

There are two types of image used in "Beetles of the World". Taxon images and Character images.

Taxon images are images of whole beetles that confirm or reject an identification. They come from a wide variety of sources.... coloured paintings, black and white drawings, black and white and coloured photographs and 35mm slides. Some are taken from very old publications like Lacordaire's "Genera des Coleopteres" published in the middle of the 19th century. Taxon images are the ones most often scanned on a flatbed scanner. The scanned images are enhanced in "Photoshop" by using powerful colour controls. This enables colours to be adjusted to match the original colours in paintings, drawings and photographs. Shadows are added and the beetle image can be further enhanced by using lighting and sharpening/blurring filters.

Character images are a mixture of both scanned images and video grabs of insect parts like legs, antennae or wings that show diagnostic features separating groups. Multiple video grabs can be made of an insect part by focusing up and down. These video images are then be pieced together in "Photoshop". The final image is a fully focused one. Interesting new software, which automates this process, is beginning to appear. For example, "Auto-Montage" is being used by the Department of Entomology at the Natural History Museum in London.

Video cameras have low resolution. Sometimes under certain circumstances you can grab 752 x 576 pixels but mostly it will be a maximum of 640 x 480 pixels. This limitation can be overcome by using the magnification of a microscope to your own advantage.

As a spin off from "Beetles of the World" we are beginning to learn about colour printing for both publication and presentations. Images published on the beetle CD-ROM will have quite low resolution but we have scanned all the original images at high resolution. We have archived these for future developments. Some of the results we have been getting by making different types of hard copy from our higher resolution files are suprisingly good and I think new doors are beginning to open.

John Lawrence is the author of the data for "Beetles of the World" and Anne Hastings is creating the biological images. Mike Dallwitz, Eric Zucher and Toni Paine are the team responsible for INTKEY.

*Houston, W.W.K. Australian Biological Resources Study, GPO Box 636, Canberra, ACT 2601 [Email: keith.houston@dest.gov.au]

PLATYPUS: A database package designed for taxon based, faunal work.

PLATYPUS is a comprehensive, relational database running in a Windows environment. It will manage taxonomic, geographic and ecological information on a regional or world-wide basis.

The PLATYPUS layout, built-in logic and context sensitive Help will promote and greatly assist the accurate documentation, management and dissemination of a wide range of information. Output reports, such as checklists, bibliographies, checklist statistics, and a comprehensive catalogue style format, may be exported to word processing and desktop publishing packages.

Data quality is ensured by rigorous built-in logic and comprehensive data validation that follows the rules and recommendations set out in the International Code of Zoological Nomenclature. World geographical areas, ecological keywords with definitions, museums, journals and standard phrases are listed but may be user-defined.

PLATYPUS is available commercially. Visit: http://www.ento.csiro.au/platypus/platypus.html

Jeans, Kevin Publisher Academic & Reference, CSIRO Publishing, 150 Oxford Street, Collingwood VIC 3066; email: kevin.jeans@publish.csiro.au; http://www.publish.csiro.au

[Publishing systematics in a multi media future].

Traditional publications such as books and journals will continue to play an important role in the foreseeable future. Journals will undergo a substantial change but books will continue to serve a function. New methods of publishing information, such as CD-ROMs, World Wide Web, email and specific software applications will complement traditional methods of publishing. Electronic publishing of information will offer enormous opportunities, many of which have not been fully utilised as yet. However in the development of software applications, unless assistance is provided to nurture their blossoming, Australia will once again lose its lead of developing truly innovative software applications.

*Lander, N.S. & Chapman, A.S. Western Australian Herbarium, Department of Conservation and Land Management, Como, Western Australia, 6152.

Interactive identification and information retrieval with DELTA and Intkey.

The DELTA format (DEscription Language for TAxonomy) is a flexible method for encoding taxonomic descriptions for computer processing. It has been developed over the last 25 years by Richard Pankhurst (now at RBG, Edinburgh) and Mike Dallwitz (CSIRO Division of Entomology, Canberra) and has been adopted by the International Taxonomic Databases Working Group (TDWG) as a standard for data exchange.

DELTA-format data can be used to produce natural-language descriptions, keys, classifications (cladistic and phenetic), and systems for interactive identification and information retrieval. Many packages by various authors have been written implementing the DELTA standard. The software developed by the DELTA team in Canberra, CONFOR and DELFOR for data management and INTKEY, implement the most current DELTA standards.

INTKEY, available in DOS, Windows 16-bit and 32-bit versions, is a sophisticated package for interactive identification and information retrieval. Many major applications have been developed using INTKEY, and the Western Australian Herbarium has created a number of products which can be demonstrated to illustrate INTKEYs functionality.

INTKEY and other DELTA-based softwares are now being engineered to take advantage of the increasing availability of CD and Internet technology to allow simple access to available biological information.

*Lander, N.S. Western Australian Herbarium, Department of Conservation and Land Management, Como, Western Australia, 6152.

MAX - personal computer software for managing species data.

MAX is a utility that facilitates the editing and maintenance of species-related data while maintaining a link to a master list of species names. Users can specify their own data fields, or choose a predefined format which functions as an electronic collecting book.

MAX allows users to record survey data or specimen collections with simple and reliable access to the latest taxon names, thereby minimising data entry and coding errors for names. MAX can be customised with the addition or substitution of the species master list and so can be used in other geographical areas for which there is an authoritative census and for organisms other than plants.

MAX can export data in a range of formats, most notably the Australian HISPID3 data exchange format, thereby making data available to all herbaria supporting this standard. It has been written in Borland's Delphi, a rapid application development environment, as a 32-bit application for Windows 95 or Windows NT.

*Lander, N.S. Western Australian Herbarium, Department of Conservation and Land Management, Como, Western Australia, 6152.

An Interactive Key to the Western Australian Threatened Flora.

A project to score comprehensive morphological and ecological descriptive data for Western Australia's 300 threatened conservation taxa. The project uses the DELTA methodology to code the raw data and present the results in both hard copy and interactive key formats. INTKEY 4 is used as the preferred software for providing interactive identification and information retrieval and its functionality will be demonstrated here.

Mackay, D.A. School of Biological Sciences, Flinders University, GPO Box 2100, Adelaide, S.A. 5001.

A review of current developments in software for statistical analysis.

Current developments in software for statistical analysis will be reviewed, concentrating on the broad analytical capabilities of such packages, the graphics capabilities of current software and the implementation of recent statistical methods and programming methodologies in software packages. This review will emphasise the present and likely future utility of statistical software to practising systematists and other biologists, rather than presenting a feature-by-feature comparison of current brand names.

*Mackay, D.A. School of Biological Sciences, Flinders University, GPO Box 2100, Adelaide, S.A. 5001.

SPLUS: an example of a statistical software package.

Abstract not available.

*Maslin, Bruce, Western Australian Herbarium, Department of Conservation and Land Management, Como, Western Australia, 6152.

"Wattle": an information system for sustainable utilisation of Acacia.

"Wattle" is an electronic information system for Acacia which is being developed through co-operative arrangements between a number of institutions nationally. When completed "Wattle" will elegantly deliver information essential for effective conservation and Landcare programs and facilitate sustainable, multipurpose utilisation of Acacia. The system comprises a powerful multi-access key which enables species to be quickly, easily and accurately named, irrespective of whether specimens are in flower or fruit; even sterile material can often be named. The key also permits species to be selected according to user-defined environmental, biological or utilisation criteria. "Wattle" will also provide comprehensive textual information on each taxon, complemented by colour photographs, line drawings and distribution maps. The text will include a description each taxon and information on their common name, synonymy, ecology, distribution, biology, utilisation potential and silviculture. The demonstration will comprise a subset of "Wattle" which will include about 70 taxa from the Kalannie-Goodlands-Jibberding Landcare Districts in the northern wheatbelt region of Western Australia. The presentation will be delivered via LucID (the data having been captured in DELTA).

Morrison, David A. Department of Environmental Biology & Horticulture, University of Technology Sydney, Westbourne St, Gore Hill, NSW 2065.

Phylogenetic software.

There are innumerable computer programs available for phylogenetic analysis. The evaluation to be presented can only cover a few of these, and will thus concentrate on those that are most commonly used. Not all of these programs provide all of the features that could be used in phylogenetics, and so the evaluation will be sub-divided into three logically-distinct areas:- tree building, tree assessment, and tree use. The evaluation will include appraisal of:- completeness of features, ease of use, calculation speed, and documentation.

Cladistic analysis involves an attempt to reconstruct evolutionary history. This process entails firstly building a cladogram to represent the phylogenetic tree. A number of different tree-building methods/criteria have been proposed, including parsimony, maximum-likelihood, spectral analysis, and distance methods. The implementation of these methods will be evaluated for the following programs:- PAUP*, Phylip, Hennig86, SplitsTree and Spectrum.

The cladograms then need to be assessed. There are three different types of assessment that can be carried out on cladograms:- assessing the magnitude of the phylogenetic signal; assessing the robustness of the monophyletic groups; and comparing the structure of multiple cladograms from the same taxa. The implementation of these methods will be evaluated for the following programs:- PAUP*, Phylip, MacClade, AutoDecay, RASA, RnA, RandomCladistics (and by default SplitsTree and Spectrum). Finally, the cladogram is not necessarily an end in itself, as the tree can be used as the basis of comparative biology. Comparative biology includes studies such as biogeography, host-parasite relationships, and molecular evolution. The implementation of these ideas will be evaluated for the following programs:- Component, TASS, CAIC.

*Nelson, Gareth and Pauline Ladiges School of Botany, University of Melbourne.

TASS: a tool for interpreting cladograms.

Geographic paralogy - analagous to the molecular phenomenon - has gone unrecognised in cladistic biogeography. It is evidenced by duplication or overlap in geographic distribution of taxa related by a particular node of a cladogram or organisms. Geographically paralogous nodes increase basally, therefore nonrandomly, in cladograms generally, such that most nodes of complex cladograms of organisms are geographically paralogous. A novel algorithm, implemented in a preliminary MS-DOS program (TASS), reduces a more or less complex cladogram of organisms to one or more subtree (area cladogram) that is paralogy free (subtree analysis). Geographic data associable with informative nodes of subtrees appear to be the only data relevant to cladistic biogeography; such data can be represented as either components or three items in a matrix for parsimony analysis.

*Simon B.K.¹, W.D. Clayton², P.D. Bostock¹, D. Sharp³ and W.A. Smith^{1 1}Queensland Herbarium, Meiers Road, Indooroopilly, Qld 4068; ² Royal Botanic Gardens, Kew, Richmond, Surrey TW93AB, England; ³ Botany Department, University of Queensland, St. Lucia, Qld 4067

(presented by P.D. Bostock¹)

Interactive keys to the grasses of the Moreton Pastoral District, south-eastern Queensland.

Members of the grass family (Poaceae) are some of the most difficult plants to identify. They have a specialised terminology associated with their reproductive and to a lesser extent vegetative characters. Because of the significant economic importance of grasses and that they are a major component of the flora of Australia there is a constant demand for researchers and students to be able to identify them quickly and reliably. Although published dichotomous keys to grasses are available for many areas of Australia these are often difficult to use for the non-specialist. Reasons are the restrictive manner in which these keys are constructed, and the applicability of the key to specific material at a particular stage in the life-history of the plant, for example the frequent requirement of fertile material. Interactive keys by computer, in which plants may be identified by the interrogation of their attributes from a database, are easier to use than dichotomous keys. Two of the commonly used interactive keys in Australia are the INTKEY component of the DELTA package in both msdos and Windows versions (Dallwitz 1993) and LUCID, a multi-access key in a Windows environment, produced by the Cooperative Research Centre for Tropical Pest Management at the University of Queensland. We intend to compile an interactive key to the grasses of Queensland, enhanced by scanned images of taxa (from illustrations, 35 mm slides and herbarium specimens), distribution maps and character states. This will include 932 taxa and 1100 characters. The present interactive set of the grasses of the Moreton Pastoral District of south-eastern Queensland (401 taxa and 733 characters), in both INTKEY and LUCID formats, is being tested as a sample for the complete project.

*Thiele, Kevin. c/- Centre for Plant Biodiversity Research, GPO Box 1600, Canberra ACT 2601.

Lucid - Contemporary Identification Tools for Biology.

Lucid is a new windows (TM) based interactive identification and taxon information package. It combines powerful identification and interrogation tools with a straightforward and intuitive interface - Lucid keys are easy to "play" with. A point-and-click interface based on lists of (or images of) states is used to describe a specimen to the key. The key then returns a list of taxa that match the description. Once an identification has been reached the user may access an open-ended cluster of text, graphic, sound and video files linked to the taxon. Lucid keys are created in a Builder module, also with a point-and-click interface, which makes it simple for taxonomists to build a distributable Lucid key. Version 1 of the Player is released; Version 1 of the Builder is due for release shortly. The Lucid Player and Builder will both be demonstrated and discussed using the Euclid key to eucalypts of south-eastern Australia, and new features under development for Lucid Version 2 outlined.

*Wilson K.L. Royal Botanic Gardens, Mrs Macquaries Road, Sydney, NSW 2000.

The IOPI Global Plant Checklist in action.

The provisional Global Plant Checklist has been available on-line since June through the IOPI home page (http://iopi.csu.edu.au/iopi/). It started with 4 regional datasets (for Australia, Europe, North America north of Mexico, and Peru), plus two edited family datasets (those for Casuarinaceae and Magnoliaceae). Other datasets are being added as (limited) resources allow. The structure and use of the Checklist will be demonstrated, and planned ways to involve botanists in editing and adding data will be discussed.