In the 2007/2008 southern summer season a stratified random selection of travellers to the Antarctic were sampled for propagules on their way to Antarctica or sub-Antarctic islands. This file lists the plant seeds that were found in the samples.
Identification of the seeds was done mainly by comparing the seeds (or more often photographs of the seeds) with photographs of seeds in seed-atlases and ... in databases on the web (see the list below). Because often we had only a single specimen of a specific seed morphotype, we did not use any destructive methods (e.g. making cross-sections of the seed). All seeds have been stored, so they are available for further study.
For each identification a confidence level was given on a 4-point scale (0 = no identification available; 1 = low confidence in identification: it may be the taxon listed, but it would not be surprising if it was not; 2 = moderate confidence: we think it is the taxon indicated, but we may be wrong; and 3 = high confidence = we are convinced it is the taxon indicated).
Sometimes it was not possible to see if something was a seed or not. Whenever we had serious doubts about something being a seed, it was not counted as such. This way we may well have discarded (figuratively: all material has been kept) some seeds, but this will result at most in a somewhat conservative estimate of the propagule load of the samples. Equally we have discounted seeds that were seriously damaged, and thus not viable. Again in general we were fairly conservative in this matter.
All seeds were grouped in groups that were morphologically different (morphotypes), and for which we suggest they are different species (or groups of closely related species) . All morphotypes were given a unique number. Most seeds were identified more or less independently by several people. Subsequently differences in identification were checked and discussed, until some consensus was reached. Where no consensus was reached, identification was given at the taxonomic level where we agreed, and lower levels were given as unknown. For quite a number of seeds we did not arrive at an identification even at the family level.
Resources used for seed identification
Botha C (2001) Common weeds of crops and gardens in South Africa. Ark grain crops institute. Potchefstroom
Cappers R T J, Bekker R M, Jans J E A. (2006) Digital seed Atlas of the Netherlands. Barkhuis Publishing. Groningen.
Corner, E. J. H. (1976). Seeds of Dicotyledons. Cambridge University Press, Cambridge.