A study of the population dynamics of the endemic liana millettia puguensis (gillet.,) papillionaceae in the pugu forest reserve, tanzania

Studies on the population dynamics of Millettia puguensis were carried out in Pugu Forest Reserve. These included studies on seed-bank dynamics, as well as herbivory, mortality and recruitment of Millettia puguensis seedlings / ramets and the distribution and abundance of the species in relation to variations in altitude, topography, canopy openness and differences in the characteristics of the host tree species. Seed bank studies showed that Millettia puguensis was completely absent in the seed bank of the Pugu Forest Reserve. The mean mortality of Millettia puguensis seedlings and ramets was highest under dense canopy conditions (44-55%) and significantly higher (P<0.001) than it was under mortality showed that these varied with variations in the degree of canopy openness. Pathogens caused the highest seedling mortality followed by spider webs in dense canopy stands. Herbivory caused the highest mortality in open and sparse canopy stands. Millettia puguensis was observed to reproduce mainly by ramets whose mean survival was found to be higher than that of seedlings regardless of season or nature of canopy openness. The preponderance of asexual reproduction in Millettia puguensis was most probably associated with the fact that most of its flowers do not set fruits.The number of ramets and seedlings recruited per month was significantly higher (P<0.001) in open and sparse canopy stands compared to dense canopy stands for all topographic localities during the period immediately after the long rains, tailing off during the dry season to the lowest point in November before picking up again during the short rains. Millettia puguensis was observed to occur in greatest abundance at an altitude range of 110-160m.a.s.1. with other altitudes showing lower abundances of the species except the altitude range of >260-310m.a.s.1 (metres above sea level) which was observed not to be inhabited by the species. Densities of Millettia puguensis increased significantly (P<0.05) with an increase in the degree of canopy openness at different altitudes. The steep slopes of the study area were observed to support the highest densities of Millettia puguensis followed by valley bottoms and the ridge tops came last. With regard to the size-class distribution the results showed that the highest proportion of Millettia puguensis individuals were in the 0-5cm stem size class with a general decline of the proportion of individuals with increasing stem size of all the tree species recorded in all the study plots only 30.21% were observed supporting Millettia puguensis. There was a weak positive correlation between the gbh of the host tree species and number of individuals of Millettia puguensis supported (r = 0.159, 0.05<P<0 .025). The mean number of individuals of Millettia puguensis supported by the various tree species was significantly positively correlated with the bark texture of the supporting tree species (r = 0.9484, n = 129). Tree height class which supported the highest number of individuals of Millettia puguensis was that of 8 1 m followed by the >14m canopy height and lastly by the height class.