From George Beccaloni, The Natural History Museum, London, 26 April 2002 My initial thoughts are as follows:- Producing a Web-based database containing fairly detailed taxonomic info. and host data for World butterfly taxa would be very achievable within say 3-4 man years (I am already some way towards doing this). It would also be possible to include basic distributional data as this is present on the NHM index cards (see below). Given perhaps 8-10 man years it should be possible to produce a complete synonymic Web-based catalogue of World butterfly taxa, similar to the catalogue Malcolm Scoble et al. produced for geometrid moths (see the Orthoptera Species File (OSF) http://viceroy.eeb.uconn.edu/Orthoptera for a possible Web-based model). I think that such a database is a necessary foundation for the project you are proposing and would be of enormous benefit to Lepidopterists, other biologists and naturalists. A rough breakdown of the work which I think would be needed to produce the database is as follows:- 1) TAXONOMIC DATA. Most of this info. is available either in existing Paradox databases in the NHM (e.g., Bridges' catalogues) or is contained in the NHM systematic card index to World Lepidoptera names. The latter is now in partial electronic form (i.e. the cards have all been scanned and by the end of the year they will be fully indexed) and it will in the near future be available over the Web. The strategy I suggest to produce a database is:- a) to get as much of the info. present on the scanned images of the index cards into a database using OCR techniques (which are being developed by our colleagues at the University of Essex); b) to check the OCR data firstly against various databases I have produced of author names, journal titles etc and then visually against the electronic card images; c) to replace most of the info. for genus-group names with the data in Brian Pitkin & Paul Jenkins' 'Butterflies & Moths of the World : Generic Names and their Type-species''; d) to go through Zoological Record and add all names (and associated info.) published since 1980; e) to check all infrasubgeneric names electronically against existing taxonomic catalogues (eg. Bridges); f) to manually check all infrasubgeneric names against the original descriptions (and possibly scan in all the original descriptions at the same time) [this alone might take perhaps 4 man years to do!]; g) to check relevant parts of the information in the database against the systematic lists being produced by Lamas, Mielke, Robbins, & Callaghan (for Neotropical species) and Ackery, Smith & Vane-Wright's "Carcasson's African Butterflies" (for Afrotropical species) [hopefully all of these will be available in electronic form], plus other recent revisions and systematic lists. This will enable a final check to be made of the basic data and will enable the current combinations of the names to be determined. 2) DISTRIBUTIONAL DATA. Basic info. is present on the index cards for most World butterflies and this could be captured during stage 1d. Locality data for the type specimens would be captured during stage 1. Obtaining detailed distributional data from museum specimens in World collections would be a massive undertaking.... 3) HOST INFO. A large amount of this has already been compiled into database form by myself, Gaden Robinson, Phil Ackery and students, but more data input would be necessary for certain World regions. It might take 1-2 additional man years to do a really thorough job. 4) TYPE SPECIMEN PHOTOGRAPHS. At a guess I would say that it would take 1-2 man years to obtain a fairly good set of electronic photos of type specimens of World butterflies. The NHM has many existing slides of 'types' which could be scanned and Gerardo and others have a lot of photos of types which they might be willing to lend so that they too can be scanned. If one were to start from scratch and photograph all the types in the NHM and other collections in a standardized fashion then this would be a huge undertaking (not least because of the difficulty of simply locating the types in the collections concerned)..... 5) WEB SITE. I guess it might take someone up to 6 months to produce a good web site to host the above data. Ideally there should be a facility whereby trusted experts can modify and add to the data contained within the site (although all suggested changes and additions should be vetted by an editor first before they are implemented). Hope to hear from you soon. Cheers, George .......................................................................................................................................... Dr George Beccaloni Research Entomologist Entomology Dept. The Natural History Museum Cromwell Rd South Kensington London SW7 5BD