Criticizing adaptive landscapes, ecology and genealogy

Disentangling ecological vs. genealogical dimensions is a core task of hierarchy theory in evolutionary biology. As Eldredge repeatedly epitomized, organisms carry out (only) two distinct kinds of activities: they survive, and they reproduce. At the organismal level, the organism stays the same whether we consider it ecologically or genealogically - yet, differences can occur in what features we consider relevant, and what fitness measurement we use. At higher levels, the two dimensions diverge, realizing different systems. Reproductive (deme) may not coincide with ecological (avatar) population. Further upwards, along the ecological dimension, higher-level systems are grouped by energy- matter interconnection, whereas, along the genealogical dimension, higher taxa are assembled by relatedness. In Dobzhansky's (1937) use of the adaptive landscape visualization (Wright 1932), all living species are imagined as distributed on adaptive peaks which correspond to ecological niches in existing environments. Peaks are grouped forming genera and higher taxa (e.g., "feline", "carnivore" ranges), and geographic speciation is figured out - like adaptation - as movement on the landscape. In criticizing Dobzhansky's landscape, Eldredge wrote that species actually do not occupy ecological niches; demes don't, either; avatars do. I point out that neighborhood and movement need to be conceived separately in genealogical and ecological spaces. Indeed, ecology should be further split in at least two spaces: geographic and phenotypic/adaptive. Movement in one space may in fact result in stability in the other(s). I also comment on the adaptive landscape: technical limitations prevent it from being coherently used above the population level, even though as a metaphor. Finally, I emphasize the partiality of any landscape - based on the choice of relevant features and fitness components - and interpret partiality as the way of approaching complex multi- hierarchical structure in evolution.