future science group 535 ISSN 1759-7269 10.4155/BFS.12.47 © 2012 Future Science Ltd Among the many species in the Jatropha genus, only one species, Jatropha curcas L. has been exploited in agri- culture [1,2] . Termed the physic nut, jatropha has been cultivated for biodiesel production, and its seed cakes have been used as organic fertilizer [3,4] . As jatropha seeds are rich in hydrocarbon molecules, it is used as a straight vegetable or as raw material for the transesterifi- cation process to produce a variety of products including biodiesel, glycerine and animal feed [2,5] . In Kenya, there is large climatic variation in terms of rainfall, tempera- ture and soils – factors which have major effects on crop growth and yield. Hence, the yield and oil content of jatropha, like other crops, is likely to be affected by cli- matic factors, and in the absence of a breeding program, the yields are likely to remain low [6,7] . Several years ago, a few genotypes collected from diverse parts of the country were being domesticated; there was then a need to study their adaptation and performance. Observations indicate that the mean per- formance of these genotypes vary from one location to another location, a sign of strong genotype × envi- ronment (G × E) interactions [1,2,8,9] . In analyzing G × E interactions, one is able to characterize the available genotypes and deduce which ones have a minimal variance for yield across different environments and can, therefore, be considered stable [8–10] . The Eberhart and Russell model defines a stable geno- type as one that has unchanged performance regardless of the variation in the environment [6] . Upon regression of the mean yield of each genotype onto the environmen- tal index, a stable genotype will have no deviations from the general response to environments and, therefore, produces a predictable performance. The environmental index is constituted from the mean yield of all genotypes in each location minus the grand mean of all genotypes in all locations. The expectation is that a stable genotype produces a mean square that has a regression coefficient (b i ) of 1 and a deviation from the environmental index (S 2 di ) of 0. For Kenya to fully exploit the potential of jatropha as a biofuel, there is the need to evaluate the stability of performance of the current genotypes and determine their G × E interactions. Materials & methods Germplasm collection Jatropha accessions were collected from different agro- ecological areas of east Africa, with some procured from Madagascar (three), India (one) and Mexico (four). Adaptation of Jatropha curcas L. in the agroecological environments of Kenya: genotype × environment interactions analysis Kahiu Ngugi*, Alex Nabiswa & Josiah Kinama Background: Jatropha curcas L. has the potential for becoming an important feedstock for biodiesel and bioenergy in Kenya. The objective of this study was to evaluate the effects of genotype × environment interaction and to determine the stability of performance of the currently grown genotypes. Methodology: A field trial consisting of 49 genotypes was laid out in a lattice design of two replications in two contrasting agroecological environments, namely Thika and Kibwezi, for 2 years. The Eberhart and Russell stability method was used to measure the performance of yield components of the 49 genotypes. Results: Environmental variance influenced the performance of genotypes for all the traits measured and genotype × environment interactions were important in determining their performance. RESEARCH ARTICLE Department of Plant Sciences & Crop Protection, Faculty of Agriculture, College of Agriculture & Veterinary Sciences, University of Nairobi, PO Box 29053-00625, Kangemi, Nairobi, Kenya *Author for correspondence: E-mail: kahiu.ngugi@yahoo.com Biofuels (2012) 3(5), 535–543 For reprint orders, please contact reprints@future-science.com