Elongational and shear flow experiments on solutions of polybutadiene in dekalin and polyacrylonitrile in dimethyl formamide have been examined. It has been found that by using the total strain in elongation rather than strain rate the data obtained at different rates of extrusion and take‐up rate in the elongational flow experiments can be correlated. The use of the concept of total strain can be shown to allow the data to be considered in terms of constant macromolecular entanglement. Moreover, by plotting the variation of the normalized elastic modulus in elongation, Ḡ n =G/Ḡ 0 , against the normalized elongational viscosity, η̃ n =η̃/3η 0 , it becomes possible to obtain the total strain at which the elongational response of a polymer solution changes from mainly viscous to mainly elastic. The results have been considered in terms of a set of constitutive equations developed by Marrucci et al. Remarkable similarity exists between their theoretical predictions and the experimental results, in elongational flow, on the above solutions. These are discussed on the molecular level and it is proposed that the anomalous situation that exists between the proposal that chain entanglement decreases and the change in response of the fluid from viscous to elastic as elongation progresses, can be resolved. This is achieved by postulating the existence of a network of very‐long‐lifetime entanglements between the macromolecules.