Theory for normal stresses in slits and capillaries

It is argued that experimental results for the flow of certain elasticoviscous liquids through slits and capillaries are consistent with those obtained by other means in as much as they indicate a second normal stress difference which is negative and smaller than the first normal stress difference. Existing interpretations of these experimental results, which predict a large positive second normal stress difference, are in error because they assume that the centreline pressure at the exit of the capillary is zero.