The emergent properties of the collection of species in a natural community, such as diversity and the distribution of relative abundances, are influenced by both niche-based and neutral (stochastic) processes. This pluralistic view of the natural world reconciles theory with empirical observations better than does either a strictly niche- or neutrality-based perspective. Even so, rules (or rules of thumb) that govern the relative contributions that niche-based and stochastic processes make as communities assemble remain only vaguely formulated and incompletely tested. For example, the translation of non-random (non-neutral) ecological processes, which differentially sort among species within a community, into species-compositional patterns may occur more influentially within some demographic subsets of organisms than within others. In other words, the relative contributions of niche vs. neutral processes may vary among age-, size-, or stage-classes. For example, non-random patterns of mortality that occur among seedlings in a rain forest, or among newly settled juveniles in communities of sessile marine communities, could be more influential than non-random mortality during later stages in determining overall community diversity. We propose two alternative, mutually compatible, hypotheses to account for different levels of influence from mortality among life-cycle stages toward producing non-random patterns in organismal communities. The Turnover Model simply posits that those demographic classes characterized by faster rates of turnover contribute greater influence in the short-term as sufficient mortality gives rise to non-random changes to the community, as well as over the longer-term as multiple individuals of a given fast-turnover demographic class transition into later classes compared to each individual that ratchets from a slow-turnover starting class into a later class. The Turnover Model should apply to most communities of organisms. The Niche Model, which posits that niche-based processes are more influential in some demographic classes relative to others, may alternatively or additionally apply to communities. We also propose several alternative mechanisms, especially relevant to forest trees, that could cause dynamics consistent with the Niche Model. These mechanisms depend on differences among demographic classes in the extent of demographic variation that individual organisms experience through their trait values or neighborhood conditions.