(A) A stock population of a model organism is used to investigate the potential for adaptive multigenerational plasticity. The phenotypic distribution for the ecologically relevant trait of interest is illustrated, and it is assumed to be relatively constant in standard culture conditions. As a self-fertile hermaphrodite, laboratory strains of C. elegans are essentially isogenic. However, nongenetic variation exists, though its origins and extent are unclear. Rectangles represent nongenetic heritable variation within the population, with black and white shading indicating two different ‘epigenotypes’ affecting the trait. "Epigenotype" is analogous to "genotype" except it is based on something other than DNA sequence, such as small non-coding RNAs, post-translational histone modifications, or other heritable molecular or macromolecular entities. Epigenotype is depicted as singular and binary, though multiple variants likely affect complex traits, and they may exist in more than two states or vary continuously. (B, C) An experimental line from the stock is subjected to some novel stressful environment (e.g., pathogens, starvation) in the first generation (F0), and phenotype (e.g., survival, fecundity) is assessed in a subsequent generation (Fn, where n is as little as one generation for intergenerational effects and is at least three for transgenerational effects). A control line is maintained in parallel without stress exposure (not shown). A heritable difference in phenotype is observed, presumably accompanied by a change in epigenotype frequency (an increase in the black epigenotype). There are at least two general though non-exclusive interpretations for the mechanistic basis of this adaptation. With adaptive multigenerational plasticity (B), some white individuals in the experimental F0 population were induced to become the black type, and individuals of the Fn population inherited this black type and others the white type, but without selection in either case (lines between generations indicate inheritance). Phenotype may or may not be affected in the exposed generation, as indicated by the dashed line. With selection (C), black individuals in the F0 population contribute more offspring to subsequent generations, because the experimental treatment resulted in differential survival or fecundity for each type, resulting in an increase in the average amount of black in the Fn population. Notably, adaptive multigenerational plasticity and selection are not mutually exclusive. In fact, if the plastic response is heritable and adaptive, then the induced epigenotype may also be selected for. Nonetheless, ruling out selection on pre-existing variation as the sole cause of changes in phenotypic frequency is critical to demonstrating adaptive multigenerational plasticity. Furthermore, the case for such plasticity will be strengthened by identification of a causal source of heritable nongenetic variation and demonstration of its induction by the stress.