One of the main challenges for the future of Natural Rubber production is the scarcity of skilled manpower to tap the trees. The only way to cope with such issue is to reduce the tapping frequency. The key is the carbon supply to the latex producing tissues. With low tapping frequencies, the latex exported at each tapping day is higher than in traditional systems. Then the trees must mobilize huge amount of carbon at each tapping. Does the latex carbon come directly from the primary sources, the leaves where C is assimilated through photosynthesis, or from reserve pools as wood starch, or both? Knowing the actual C sources and knowing the pathways towards latex is then necessary to manage the tapping systems. Stable isotopes and especially 13C are widely used in plant science as tracers but studying their natural abundance also provides insightful information on tree physiology. We first compared the seasonal variations in natural abundance in tapped and untapped latex of RRIM600 trees to the variations in their leaves which vary according to climate and phenology. We showed that latex d13C was higher and varied much less than that of leaves in tapped trees. The lack of correlation between variations in d13C in leaves and in latex suggests that recent photosynthates are mixed in a large pool of stored carbohydrates that are involved in latex regeneration after tapping. We then did a field labeling of full crowns of 3y-old tapped rubber trees with 13CO2 to trace the carbon from its assimilation in the leaves to the tree sinks and particularly to latex. Such experiment, using a specifically designed chamber, has never been done on rubber trees. Three trees (RRIT 408) were labelled in June and three other in October. We sampled leaves, phloem, wood and latex to analyze their 13C content and determine the dynamics of carbon allocation from leaves to reserves and latex. The latex samples will be followed during one year. The presentation describes the methodology and pre