This study was carried out as part of a contract funded by the Scottish Office to determine whether approaches to risk assessment, based on current recommendations, were adequate in determining the possibility of an adverse environmental effect of introduction of GMOs to the environment. The studies used a representative soil bacterium, Pseudomonas fluorescens, genetically modified by either chromosomal (FAC510) or plasmid (pUCD607) insertion of lux genes for bioluminescence. The starvation state is considered an important stress factor that bacteria added to soil may face. The ability to recover from this state, when substrates and nutrients become available, may be considered essential for the success of an introduced organism, and may ultimately determine the ecological effects of a GMO. The recovery response to the GMO constructs from carbon starvation was investigated in pure culture studies, and upon addition of starved cells to sterile soil. Pure culture studies clearly demonstrated a starvation-survival disadvantage of both the GMO constructs, as determined by growth and respiratory activity (dehydrogenase activity) when resuscitated from periods of starvation of up to 21 days. Correlation between dehydrogenase and luminescence activity was investigated. The lack of correlation between the two processes may have pointed to a change in maintenance energy costs as a consequence of genetic modification. Determination of the starvation-survival response in sterile soil was inconclusive, due to high variability and limitations to, and uncertainty in, cell extraction and culture.