Kannan et al. [J. Rheol. 46, 979-999 (2002)] studied the problem of fiber spinning for polymers that are largely atactic in nature, within a thermodynamic framework developed by Rao and Rajagopal [Z. Angew. Math. Phys. 53, 365-406 (2002)], where significant crystallization does not take place. Here, we develop a model within the same framework that can take into account flow-induced crystallization and the anisotropy of the crystalline part of the sernicrystalline polymer. We find that the predictions of the model agree very well with experimental data. (C) 2005 The Society of Rheology.