The Steering Committee (see Management Plan) will establish a common format for descriptions, diagnoses, and other documentary information on taxa ("taxon treatments"). This will then serve as a taxonomic template for all team members. The aim will be to satisfy the following criteria: 1) provide the maximum information possible per taxon in a highly structured format; 2) allow direct use of information in print publications, on the Internet, in phylogenetic analysis, and in interactive keys; and 3) accommodate information from prior taxonomic studies.

We envision two possible alternatives for structuring information in taxon treatments. One approach is to use an "off the shelf" program such as DELTA, created at the CSIRO, Canberra, to capture character information on taxa then generate descriptions and identification keys from it. The alternative is to create structured XML-coded templates for higher taxa, genera, and species. XML includes the desirable features of having widely accepted standards and support through open source and commercial tools, as well as presentation of the same documents in different systems. Furthermore, the XML-coding approach will allow for structuring and direct incorporation of information for newly described taxa as well as for most information in previously published "taxon treatments" and allow for precision searching of data within those treatments. We propose to develop the Taxon-X Schema that derived from an analysis at the AMNH of taxonomic publications and their included taxonomic treatments. Minimum elements in the schema would be: taxon_name (an available scientific name); authority (details of publication of a taxon name); synonymy (references to uses of a name or its synonyms); diagnosis (key diagnostic features); descriptions (detailed descriptive information); materials examined (information on specimens studied); biotic associations (host associations, e.g., plants); keys (identification aids); and figures (illustrations of morphology or possibly tabular information). Within the descriptions themselves a tag set will be developed to code for detailed character information found in the Orthotylinae and Phylinae.

Publications of project personnel indicate remarkably similar approaches to the structuring of taxonomic information, as well as a history of co-authorship. Thus, we envision little difficulty in standardizing the format for descriptions. Funds are budgeted for the necessary programming. The timeline for implementation will be the middle of the first grant year.

Descriptions of New Taxa: Descriptions of new taxa will be entered directly into the taxon-treatment format discussed above.

Descriptions of Previously Described Taxa: Because all genera and suprageneric taxa will be involved in the phylogenetic analysis for the group, revised diagnoses/descriptions of them will be prepared. Wherever feasible, published descriptions will be digitized and made machine readable (see Budget Justification for cost analysis) with the elements of taxon treatments being XML-coded as part of the process, thus minimizing the need for redescription of taxa. Our team has the linguistic skills to translate descriptions where necessary (see Table 5).

Terminology: We will utilize standard terminology by extending the glossary of Schuh and Slater (1995) to cover structural elements found in the Orthotylinae and Phylinae, such as details of male and female genitalia. The glossary will be on the project web site.

Illustrations: Illustrations will be an integral part of the project database. We will incorporate habitus illustrations of all species, using proven technology applied by Schuh (2001) and Cassis et al. (in press). Most of the imaging will be done by the technical assistants of PIs Schuh and Cassis (see also Management Plan and Budget Justification).

Morphological illustrations will be crucial in interpreting descriptions and for use with the interactive keys. A large body of published material already exists, particularly as line drawings of the male genitalia and scanning micrographs of pretarsal structures, setal types (including trichobothria), and scent-gland evaporatory structures. We will incorporate these in the project database. We will add new illustrations, using all available technological approaches.

Histological Data: The histological skills of postdoctoral investigator Christiane Weirauch offer the opportunity to develop information crucial to understanding scent-gland structures in the Miridae. Co-PI Cassis (1995) has demonstrated the classificatory power of these structures, but they are still underutilized and not well understood anatomically and histologically.

Table 4 provides a breakdown of the higher taxonomic groupings that may be recognized during this study, as well the numbers of described genera and species, predicted numbers of new genera and species, and numbers of taxa that have been recently revised. In the last category are groups we anticipate can be incorporated into the project with minimal effort for character coding and creation of taxon treatments.

Table 5 (see Management Plan) indicates the number of years each investigator will be sup-ported and the percentage of effort to be devoted, revealing a total of 40 investigator years. Dividing the total taxa (593 genera + 5355 species = 5948) by number of investigator years (40) gives an average of 148 taxa per year each of our research personnel would need to process.

Experience and level of research support will vary among the investigators and thereby influence their ability to complete work on taxa assigned to them. We have been conservative in estimating effort on the part of the senior investigators, knowing that their institutional responsibilities will be greater than those of remaining personnel. Considering the strength of experience of our senior investigators, the number recently revised taxa, and the size of our team, we view the per-investigator load to indicate feasibility of project completion as described.

Assignment of Tasks: Tentative assignments of taxonomic responsibility are given in Table 4. Final assignments will be made by the Steering Committee, based on knowledge and exper-tise of the individual investigators in combination with the complexity of groups under study.

In line with the PBI proposal guidelines, we will perform a generic and suprageneric phylogenetic analysis on a global scale, with the ultimate aim of producing a total evidence analysis of relationships incorporating DNA sequence data for as many taxa as possible. In doing so it will be our aim to produce the most informative classification possible. We believe there are strong arguments for the use of parsimony methods in any such undertaking (Farris, 1979, 1983; Siddall, 2002). Software is now widely available that will analyze large data sets of the type to be assembled for this project (Nixon, 1999). This holds true for morphological and sequence data. Relevant software packages, routines, and utilities, include NONA, PIWE, TNT (Goloboff, 1999), POY (ftp.amnh.org/pub/molecular/poy), and WinClada, all available at the AMNH through site licenses, with TNT and POY running on a massive cluster computer. We have allocated funds to purchase software for use by all investigators on a local basis.

Members of the project team have substantial experience in the area of phylogenetics. For example, Schuh (2000c) has published a textbook on the subject and has taught Principles of Systematics for the City University of New York over the last decade. All of our senior investi-gators have published phylogenetic analyses for various groups of the Miridae and for the Heteroptera more broadly, some including combined morphological/sequence data sets (Cassis, 1995; Cassis and Silveira, 2001, 2002; Cassis and Moulds, 2002; Cassis et al., in press; Henry, 1991, 1997; Schuh, 1974, 1976, 1984, 1991; Schuh and Schwartz, 1995, Schwartz, 1984, 1994; Schwartz and Foottit, 1999; Wheeler et al., 1993; Yasunaga, 1997, 1998). Finally, members of our project team will consult with AMNH Division of Invertebrate Zoology colleagues J. M. Carpenter, M. E. Siddall, and W. C. Wheeler, each of whom has made significant contributions to the overall theory of phylogenetic analysis, combined analysis of morphological and sequence data, and analysis of non-protein coding sequences.

Data on morphology will be documented through the descriptive process, imaging, and illustration. We anticipate that establishment of a common format for taxon treatments will lead to the production of a "first-pass" character matrix for the generic and suprageneric taxa in the Orthotylinae and Phylinae. We will emphasize acquisition of new character information through techniques including scanning electron microscopy, high-resolution digital imaging, and histology, among others. This matrix will undergo refinement as additional character information becomes available and as character definitions are revised and updated.

Wheeler and Schuh have already gathered extensive sequence data on Heteroptera, including the Miridae. We will take advantage of every opportunity to acquire specimens preserved for DNA sequencing as a way of broadening that sample. Whereas Wheeler and Schuh collected approximately 3200 base pairs per taxon (16s, 18s, 28s, Co1), we anticipate doubling the number of bases per taxon, using primers made available as a result of NSF-funded "A Tree of Life" investigations on the Arachnida at the AMNH. To the greatest extent possible, PhD candidates and postdoctoral investigators will be trained to perform the DNA sequencing, broadening their technical expertise and overall contribution to the project.