Beschreibung: Iron fertilisation experiments offer a unique tool to investigate biological systems in situ and the scope of reaction by manipulating one parameter. During two in situ iron fertilisation experiments, carried out in contrasting seasons (EisenEx early austral spring 2000 and EIFEX late austral summer to early fall 2004) in the Atlantic sector of the Southern Ocean, no clear seasonal impact on the phytoplankton built up and main group distribution could be observed. In both experiments chlorophyll biomass increased threefold and maximum concentrations were around 3 \mug/l. Diatoms, identified by marker pigment were the main beneficiary of the iron fertilisation; however, microscopic observations showed a different diatom species succession in both experiments. Small differences were observed in the other phytoplankton groups including phototrophic dinoflagellates, pelagophytes, cryptophytes and chlorophytes. These groups increased in the spring EisenEx experiment immediately after iron fertilisation, while in the fall EIFEX experiment most other phytoplankton, except for phototrophic dinoflagellates, constantly decreased during the experiment.A significant difference between experiments was observed in the phaeopigment concentration which was threefold higher during EIFEX than in EisenEx, suggesting a large role for zooplankton in the fate of the phytoplankton bloom. Additionally during EIFEX, a strong shift of chlorophyll degradation products indicated a change in the zooplankton community over the course of the experiment. Another pronounced difference between the two fertilisations was the high phytoplankton biomass below 100 m towards the end of EIFEX. The diatom marker fucoxanthin and chlorophyll a could be followed in the water column to a depth of 4000 m and a low ratio of phaeopigments to chl a indicated that the export of fresh material most likely originated from the iron fertilised patch. This was corroborated by a similar diatom composition at both the surface and deep water column, as well as in fluff layers above the sediments. The longer observation period during EIFEX (37 days) as compared to EisenEx (21 days) could explain why a clear export signal was detected during the former and not during the latter experiment. Further explanations for the decline in the EIFEX bloom could include iron limitation towards the end of the experiment due to a different fertilisation strategy or the end of the growth season. Possible causes and implications of export from iron-induced phytoplankton blooms will be discussed.