Comment 1: for some amphibians our best information is often 'has not bred since date x', because many species are only visible when they breed and we do not know how long they can live between breeding episodes.

Comment 2: sampling must not only include expected range, but also expected habitats. Most amphibians have a complex life history meaning that both aquatic and terrestrial habitats must be sampled.

Comment 3: failure to hear advertisment calls would be just as significant as absence of a sighting.

Comment 1: some work has been done with indirect evidence; eg. woodpecker holes. But this is unacceptably tenuous.

Comment 2: states that the history of rediscoveries is built around long absences (or even not so long absences but people have not looked at the evidence) during which it is "feared" that extinction has occurred, almost always by office-workers (rather than fieldworkers) who are charged with documenting these things. So, apart from actually looking at the right time and in all the possible places, it is actually really important to assemble all the data available, the better to establish what those times and places are.

Comment 3: the environmental threat criterion used for fish could profitably be added for birds. In New Zealand, areas of suitable habitat lack taxa solely because of presence of introduced species.

Comment 1: plant distribution might be called into play for herbivorous species.

Comment 2: cannot think of other sub-criteria to help refine this other than the capture and dissemination of a whole lot more primary data about many more taxa.

Comment 1: a missing criteria could be complete eradication of the habitat; eg. drying a desert spring, clearing an oasis, transforming a karst tower into cement, etc.

Comment 1: Criteria relating to EED need to be adaptable to the organisms being considered. Criteria for birds and mammals would be different to those for cryptic taxa. A threatened species listing committee on which I work has used the following definition for adequacy of survey:

ADEQUACY OF SURVEY

1. Surveys should encompass the range of available techniques and sampling should have been conducted throughout the possible (as distinct from the known) geographic range of the taxon.

2. Surveys should be conducted in a range of seasonal and environmental conditions consistent with the biology of the taxon.

3. Taxa confined to geographically restricted or specialised habitats may require less time or effort to survey than those occurring in more widespread habitat.

4. In the case of taxa known from very few specimens, a fact pertinent to the nomination, the nominator should demonstrate that all available collections have been examined. (This is very good point; MacPhee and Flemming (1999) also drew attention to hypodigms in their paper).

Comment 2: Sometimes archaeological and ethnographical data can be useful - ancient art, tools, etc. People of the past usually reflect their interactions especially with hunting mammals."