Charles Explorer logo
🇬🇧

Blending inheritance revisited: Simulation of selection in systems with non-particulate inheritance

Publication at Faculty of Science |
2015

Abstract

The old hypothetical model of blending inheritance based on the assumption of non-particulate, non-genetic coding of traits was previously identified as problematic. If the information is carried by the traits themselves, it can be lost because of the averaging of parental qualities in posterity.

This objection against Darwinism was later overcome with the rise of Mendelism. However the inheritance of cultural variants is probably non-particulate and we still observe a cultural evolution.

We investigated the nature of evolutionary processes in systems with non-particulate inheritance through computer simulations. In presented models we consider theoretical population consisting of individuals exhibiting a quantitative trait.

We compute the individual fitness as a difference between the trait value and previously set optimum. Surviving individuals pair at random and give rise to the next generation.

Average trait value of offspring is equal to parental average and individual descendants are normally distributed around the parental average. Small proportion of them can therefore exhibit higher/lower trait value than both of their parents.

We believe that this is a reasonable assumption. We find the model, where each offspring shows exactly intermediate trait value, implausible.

The results of simulations show that blending inheritance may not constitute a problem for selection and that initial variability provided by "promising individuals" can be sufficient even in very homogenous populations. We also argue that in unstable adaptive landscapes gradual evolutionary processes based on non-particulate inheritance can generate a punctuated equilibria pattern.

The presented simulations thus may offer important insight into evolution of non-genetic systems.