This trial was made up of 17 provenances of Prosopis, one provenance of Acacia nilotica and one provenance of Leucaena leucocephala. The Prosopis provenances included P. cineraria (Pakistan and Yemen, four provenances), P. flexuosa (Chile, two provenances), P. glandulosa var. torreyana (Mexico, three provenances), a landrace of P. juliflora with presumed local origin (India), P. pallida (Peru, three provenances), P. tamarugo (Chile, three provenances) and four provenances from Mexico and Chile with unknown species identification. The trial was established in 1989 with a spacing of 3x3 metres. It was fertilised twice, and irrigated several times during the first years after planting. The intensive tending means that recommendations from the trial should only be used under similar circumstances. The trial was assessed at an age of five years in 1992. Different growth parameters were measured and subjected to analyses of variance and multivariate analyses.
The differences between species were significant for all the analysed variables. P. tamarugo and P. flexuosa had a very low survival, whereas the rest of the provenances had survivals ranging from 35 % to 95 %. Of the new introductions, P. pallida seemed the most promising. However, the overall best performance was found in the local provenances and landraces (A. nilotica, L. leucocephala and P. juliflora). The fastest growing provenances had increment rates in basal area of 1 m2 ha-1 y-1, corresponding to a dry weight production of approximately 3 t ha-1 y-1. Differences within species were not significant.