A comparison is made between the different models of the biomass size spectrum proposed by a number of authors. Though superficially dissimilar, the models are all mathematically compatible if the differences in their underlying assumptions are taken into account. The simplest model does not consider the complexities of food webs over food chains, somatic growth, or the continuous nature of the size spectrum. Comparison with the more complex models, however, shows that these omissions do not seriously affect the slope of the size spectrum. For example, one model predicts that the effects of somatic growth and reproduction cancel if cohort biomasses remain relatively constant as the cohorts mature. If growth rate is related to body size in an allometric relationship and reproduction is ignored, then another model gives a slightly different slope (higher by roughly 0.03). If the same assumptions are used in both models, however, they give compatible results. Some simple equations are suggested for routine application in size spectrum analysis of biomass and production data.