The case of random fiber breakage during processing is modeled by the cutting of combed cotton sliver. Experimental fiber length data are taken from sliver subjected to calibrated degrees of breakage by cutting at various gage lengths. As different levels of breakage are inflicted, the corresponding length distributions are found to exhibit distinct trends. A fiber-breakage model of cotton fiber length is used to predict changes in the shape of the length distribution and to interpret the experimental results. Comparison of experimental data with prediction from the theory suggests that certain systematic features of fiber length distribution might be used to index the degree of breakage which has occurred in a specific cotton sample of unknown processing history.