This study examines the oviposition response of a specialist weevil (Mogulones larvatus) to patches of its host, the noxious weed Paterson"s curse/salvation Jane (Echium plantagineum). We simultaneously examined the effect of patch size and plant density (and their interaction), on the recorded oviposition patterns. Our results show that oviposition first occurred on the largest patches with the highest number of plants. However, there was no significant effect of patch size or number of plants per patch at the end of the experiment. At this time the level of attack per plant was negatively correlated with plant density. This negative effect of density on the level of oviposition was not mediated by a reduction in plant size at higher densities. The pattern observed may indicate a risk-spreading strategy by females.

A survey was carried out on milk production and reproductive performance of dairy cattle: 24 farms, with a total of 900 animals and distributed in four agro-ecological zones, were visited every 15 days over 18 months. Cows were fed on natural pastures as the only source of feed, and animal performance was dependent on the season and exhibited a dramatic drop in dry spells. Numeric productivity indices integrating productive performance for settler's, multipurpose, crop-livestock integrated and modern farms were 0.56, 0.74, 0.69 and 0.63, respectively. Milk productivity was higher on modem farms (6.7 L/cow per day) than in the other systems, and higher with Holstein-Friesian cows (7.7 L/cow per day) than with indigenous cattle (1.8 L/cow per day) or crossbred animals (3.7 L/cow per day). This paper speculates on the opportunity to improve the genetic potential of indigenous cattle, concomitantly with the efforts to adapt exotic cattle to a mountainous equatorial environment.

In developing countries, animal products supply presents a major challenge in meeting the demand for the two next decades. Many researchers point out the necessity for a 'reasoned intensification', especially in agricultural areas, by integrating agriculture and livestock activities. But intensification only occurs on specific farms. How can one explain the different dairy developments or technical options between farms? An analysis of the diversity of dairy farming systems in Mbarara district (Uganda) has been conducted from a monthly crosssectional survey of a sample of 22 farmers, identified from a large household survey (183 dairy producers). The multi-table factorial analysis allows a thorough review of the interactions and points out some causal relationships between the development of the dairy systems and the social and technical management of the whole farm. This analysis underlines the key factors of intensification such as genetic improvement or market opportunities; but also the degree of intensification which is well correlated to the stage of family development. If the