BACKGROUND: Experimental research on the effects of homeopathic treatments on impaired plants was last reviewed in 1990. OBJECTIVES: To compile a systematic review of the existing literature on basic research in homeopathy with abiotically stressed plants using predefined criteria. METHODS: The literature search was carried out on publications that reported experiments on homeopathy using abiotically stressed whole plants, seeds, plant parts and cells from 1920 to 2010. Outcomes had to be measured by established procedures and statistically evaluated. Using of a Manuscript Information Score (MIS) we identified those publications that provided sufficient information for proper interpretation (MIS?5). A further evaluation was based on the use of adequate controls to investigate specific effects of homeopathic preparations and on the use of systematic negative control experiments. RESULTS: A total of 34 publications with abiotically stressed plants was identified, published between 1965 and 2010. The 34 publications described a total of 37 experimental studies. Twenty-two studies included statistics, 13 had a MIS?5, 8 were identified with adequate controls and 4 with negative control experiments. Significant and reproducible effects with decimal and centesimal potencies were found, including dilution levels beyond Avogadro's number. One experimental model was independently assessed by another research team and yielded inverted results compared to the original trial. CONCLUSIONS: Abiotically stressed plant models seem to be a useful approach to investigate homeopathic basic research questions, but more experimentation and especially more independent replication trials are needed. Systematic negative control experiments should be implemented on a routine basis to exclude false-positive results.