The relationship of extrudate swell occurring in capillary rheometer and extruder is the aim of this study. Three assumptions used are: (1) the molten polymer is sheared and given the largest deformation at the entrance of the die; (2) the retraction of an elastic part occurs during the molten flow in the die, and (3) the total retraction of the melt appears at the exit of the die. The largest deformation at the entrance of the die depends linearly on shear rate and the retraction in the die depends on both shear rate and die dimensions (diameter and length), which is defined as a retraction function. Seperation of these two effects on the largest deformation and the retraction function is proposed by defining a extrudate swell rate function, K as (square root of (X)2-1)/Yw, where X is the extrudate swell ratio and Y is the shear rate. This indicates that K is a function of shear rate, die dimension and melt properties, similar to the retraction function. Extrudate swell of natural rubber compounds mixed with carbon black calcium carbonate were examined with a capillary rheometer and an extruder at processing temperature of 100 oC. The die used had various sizes varying from 1.5 mm - 5.5 mm in diameter and 16 mm - 30 mm in length while shear rate is varied from 1-2000s-1. The results show that the emperical extrudate swell rate function K is equal to AYwn, where n is a constant varying from -0.70 to - 0.89 depending on the die length. An agreement between the extrudate swell rate functions (K) obtained either from the capillary rheometer and or with the extruder is discovered. Therefore, extrudate swell behaviour observed in laboratory equipment as a capillary rheometer can be used to explain the behaviour in an industrial extruder.