Welcome Locality Data Entry        | Links | About Oonopidae | Outreach | Sign In 
Home | Project Description | Investigators | Species Pages |
Home >> Project Description >> Methods
Project Summary | Project Details | Methods | Institutional Participants | Individual Participants | Literature Cited
Crucial to the success of the team in meeting its goals are collections, both existing and to be acquired by new fieldwork, and the cyberinfrastructure that will allow us to work collaboratively despite the distances separating us. Fortunately, substantial resources already exist in both areas.
COLLECTIONS. Over recent decades, many arthropod students (not just arachnologists) have been involved in thorough sampling of litter faunas around the globe. Because oonopids are relatively small animals (ca. 1--4 mm in total length), they are most easily taken by mass collecting techniques (such as pitfall traps, Berlese funnels, and Winkler extractors). These mass collecting techniques have the added advantage that, especially when litter is concentrated by sifting prior to being processed, large samples often result. Having a dozen specimens of the same species, collected at the same place and time, is an enormous help, and not just in assessing inter- versus intra-specific variability. Good samples allow some specimens to be thoroughly studied by both compound and scanning electron microscopy, without fear that the animals will afterward be represented only by disarticulated parts. Because many of these samples, especially from more recent years, have been captured and stored in 95% ethanol, the existing material is often also useful for DNA sequencing.
The US collections are worldwide in scope, as are those in England, Finland, and France. Collections at the Field Museum are especially strong in litter material, because of that institution's long-running Berlese sample sorting program; the same is true of the Univ. of Turku collection, which includes abundant litter material collected in the Indo-Pacific, and the Instituto Butantan, which includes extensive pitfall and Winkler samples from a transect of over 40 Atlantic forest and cerrado areas extending from northeastern to southern Brazil.
Because substantial litter samples already exist for most geographic areas, our new field work will concentrate instead on acquiring additional canopy samples; it will be easy enough to secure the additional litter taxa needed (e.g., for sequencing) in the process of those canopy-focused trips. We plan a total of 12 collecting expeditions summing to over two year's worth of person-days in the field. Existing, newly sorted, and newly collected specimens will likely total over 75,000.
CYBERINFRASTRUCTURE. We are fortunate that one of the first cadre of PBI awards, the Schuh-Cassis plant bug PBI, has the American Museum (AMNH) as the primary base for its information technology activities. This allows us to capitalize on the substantial progress both their group, and the spider ATOL team, have already made in implementing much of the cyberinfrastructure our investigators will need. We envision four major Internet components:
1. World Spider Catalog. The oonopid section of the World Spider Catalog (Platnick, 2006) will provide the taxonomic framework to which all other information will be related. Our first task will be to move the data from the current version of the text catalog into the SQL-compliant database that serves the plant bug PBI website. This conversion will be accomplished by Robert Raven, using protocols well-tested during his recent conversion of the text files (from version 5.0) into the SQL-compliant database that currently serves both Species 2000 and GBIF.
2. Specimen Database and Interactive Mapping. The plant bug PBI project has already implemented a specimen database designed to allow data entry and retrieval, and interactive locality mapping, via a web-browser interface. That approach will enable all our far-flung participants to create and access a single, common resource of specimen-level information, and to produce maps reflecting all the available information. Only minor programming changes will be needed to adapt this solution to our needs, and funds are budgeted for that programming. The plant bug database incorporates unique specimen identifiers, with unique, matrix-coded barcode numbers that are placed on the insect pins and can be read and entered through visual inspection or by scanner. Oonopids will require a lot-based approach that assigns one identifier (and label) per vial rather than per individual specimen (one vial may contain many specimens taken at the same place and time, and the association, for example, of males with females, constitutes important information, just as with the lots in a fish collection). Where needed, a suffix will be added to the vial number to reference a particular specimen (e.g., one used for imaging or sequencing). This approach is accommodated in the existing database, which also allows batch loading of unique vial identifiers, a feature that expedites the entry of multiple records from a single collecting event.
3. Image database. We will provide Internet access to digital images of whole specimens (including types) and their morphological structures, including all the light and scanning electron micrographs, and scanned versions of all the illustrations, produced as part of this project. Images will be linked to the specimen database entries for the specimens used, as well as to the catalog entries for the taxa, and will be managed using the data structure developed by Martín Ramírez for the spider ATOL project . Images are assigned to body regions and standard views, enabling users to select and compare features across groups (for example, all the ventral views of male palpal endites supplied by any participants for any taxa), and will be linked to cells in the descriptive database. Digital imaging equipment needs are detailed in the Budget Justification.
4. Descriptive database. The substantial tasks of sorting and describing the available oonopid specimens, and the collaborative approach needed to accomplish them, demand that each participant be able to enter and access descriptive data, using a shared format, via the project's website. Single users have access to tools such as Delta for this purpose, but those tools do not yet scale to handle multi-user situations. Participant Xinping Wang and AMNH database specialist Mark Breedlove will design and implement a multi-user, Internet-accessible database, using TDWG's Standard Descriptive Data and additional formats, to accept descriptive information from us all, and output combined tables that can be used in programs like Delta, Intkey, and Lucid. A prototype of this database can be found on Xinping's coelotine spider website.
Our Steering Committee will establish a common format for oonopid diagnoses, descriptions, and other documentary information that will serve as a taxonomic template for all team members. The goal will be to enter maximal information in a highly structured format that will allow direct use of that information in formal descriptions for printed publications, species web pages, matrices for phylogenetic analysis, and in interactive keys, while also accommodating information from prior taxonomic studies (such as that of Saaristo, 2001, which will be taken as a starting point in establishing the common format).
Although some team members have experience with Delta, none of us have worked collaboratively via such an IT structure before; indeed, so far as we are aware, no similar multi-user approach to acquiring basic taxonomic information has been designed and successfully implemented. However, the prior publications of our participants show similar approaches to structuring taxonomic information, as well as substantial histories of co-authorship, and we anticipate little difficulty in implementing a standardized descriptive format for this project.
If an individual systematist were to tackle the oonopids, task ordering would be fairly obvious: gather the available collections, sort them until a putatively monophyletic group becomes apparent, then pull out all members of that group, work up those specimens, return to the unsorted residue, and re-start the process. To achieve the same end collaboratively will require substantial coordination of parallel, rather than sequential, efforts.
Our initial approach will focus on four fronts: cyberinfrastructure, specimen sorting, existing taxon redescriptions, and new fieldwork. Each participant will have primary responsibilities in one of these areas. During the first year of the project, the programming needed to implement our multi-user specimen, image, and descriptive databases will be completed. During that same year, all previously collected oonopid specimens will be borrowed; in some cases, we will have to sort specimens out of Berlese residues that are either totally unsorted or sorted only to the level of spiders (much of that work will be done by training undergraduates to recognize oonopids, a relatively simple task). The available specimens will be divided by geographic region, and assigned to appropriate team members who will do a first-pass classification, separating each region's species into (approximately) monophyletic genera. Also during that same initial year, other team members will borrow, redescribe, and image as many type specimens of existing names as possible; many of the existing early descriptions are accompanied by no, or no useful, illustrations, and those names can be placed only by examination of the type material. Obviously, the top priority here will be redescriptions and imaging of the type species of each of the available generic names; museum visits will be scheduled as needed. And finally, fieldwork will commence, concentrating initially on areas from which few canopy samples are available.
At the end of this process, ca. one year into the project, the participants will need to meet for two weeks, to work out a synonymy scheme for the results of the preliminary sorting (e.g., that Harvey's Genus A equals Saaristo's Genus B and includes the type species of Oonops). Via this process, it should be possible to recognize some putative genera that have multi-continental distributions, others that are more narrowly endemic, and some tentative monophyletic groups of those genera, and also to associate the available names with those putative genera.
On the basis of these results, participants will then redistribute the workload, so that each monographic team member has an appropriately sized, and at least potentially monophyletic, genus or group of genera to revise. Specimens will be re-routed to the appropriate workers, and the results of that redistribution should make the gaps in our knowledge very obvious, providing direction for the remainder of the fieldwork program and a focus for DNA sequencing.
Although these plans may seem ambitious, and certainly involve more coordinated efforts than working taxonomists are used to, our team members have substantial histories of successfully managing and completing large, collaborative efforts, including major biodiversity surveys (e.g., Platnick , Griswold , Gillespie ) and PEET projects (e.g., Platnick , Hormiga , Sierwald ). Details on the proposed fieldwork, taxonomic efforts, sequencing, and phylogenetics are in the Management Plan, along with personnel and organizational outlines.
The American Museum of Natural History, New York, in collaboration with
The California Academy of Sciences, San Francisco   The University of California, Berkeley
The Field Museum of Natural History, Chicago   George Washington University, Washington DC
©Copyright 2006-2014. All Rights Reserved.
This material is based upon work supported by the National Science Foundation under Grant No. DEB 0613754. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.