In the West-Kalimantan area (in Indonesia and Borneo), the rubber-based smallholder system is a traditional agroforest system called "Jungle rubber". This extensive system is well-adapted to local ecological and economic conditions. It represents 73% of Indonesian rubber production. Using a participative approach the SRAP project (Smallholder Rubber Agroforestry Project. ICRAF-CIRAD) has implemented on-farm experimentation in order to improve rubber based agroforestry systems by increasing "Jungle rubber" productivity. Taking into account the advantages of agroforestry practices, the project developed several clonal rubber based systems named RAS (Rubber Agroforestry Systems), which have allowed smallholders to increase labour productivity and yields threefold with limited initial investments and maintenance costs. The inter-rows are planted with fruit trees, timber trees, a combination of fast-growing trees and cover crops or intercrops. This enables production and income diversification. In these systems. the rubber tree is in competition with spontaneous vegetation growing in the inter-row but studies of growth and production show that there are no negative consequences for rubber productivity. Spontaneous vegetation in the inter-rows results in plant biodiversity that it is important both to quantify and qualify. These unplanted species are used for different purposes by the local population and include some potentially marketable species. The objective of this paper is to present the results of a Study conducted in 2001 on the level of biodiversity in two specific RAS systems. The transect methods implemented in "RAS" inter-rows, was used to quantify the emerging biodiversity (number of species and their identity). This type of transect was implemented in plots of different ages to understand the dynamics and the ecology of the spontaneous vegetation. 4 villages and 23 rubber agroforest plots in the West-Kalimantan province were selected and data was collected