The use of transfection in the study of the biology of malaria parasites has been limited due to poor transfection efficiencies (frequency of 10(-6) to 10(-9)) and a paucity of selection markers. Here, a new method of transfection, using non-viral Nucleofector technology, is described for the rodent parasite Plasmodium berghei. The transfection efficiency obtained (episomal and targeted integration into the genome) is in the range of 10(-2) to 10(-3). Such high transfection efficiency strongly reduces the time, number of laboratory animals and amount of materials required to generate transfected parasites. Moreover, it allows different experimental strategies for reverse genetics to be developed and we demonstrate direct selection of stably and non-reversibly transformed, fluorescent protein (FP)-expressing parasites using FACS. Since there is no need to use a drug-selectable marker, this method increases the (low) number of selectable markers available for transformation of P. berghei and can in principle be extended to utilise additional FP. Furthermore the FACS-selected, FP-expressing parasites may serve as easily visualized reference lines that may still be genetically manipulated with the existing drug-selectable markers. The combination of enhanced transfection efficiency and a versatile rodent model provides a basis for the further development of novel tools for high throughput genome manipulation.