Angela J. Roles

Molecular phylogenetics has benefited tremendously from the advent of next‐generation sequencing, enabling quick and cost‐effective recovery of whole mitogenomes via an approach referred to as ‘genome skimming’. Recently, genome skimming has been utilised to recover highly repetitive nuclear genes such as 18S and 28S ribosomalRNAgenes that are useful for inferring deeper evolutionary relationships. To address some outstanding issues in the relationships among Northern Hemisphere freshwater crayfish (Astacoidea), we sequenced the partial genome of crayfish species from Asian, North American and European genera and report the successful recovery of whole mitogenome sequences in addition to three highly repetitive nuclear genes, namely histone H3, 18S and 28S ribosomalRNA. Consistent with some previous studies using short mtDNAand nuclear gene fragments, phylogenetic analyses based on the concatenation of recovered mitochondrial and/or nuclear sequences recovered the Asian cambarid lineage as basal to all astacids and North American cambarids, which conflicts with the current taxonomic classification based on morphological and reproduction‐related characters. Lastly, we show that complete H3, 18S and 28S ribosomalRNAgenes can also be consistently recovered from a diverse range of animal taxa, demonstrating the potential wide utility of genome skimming for nuclear markers.

As the ultimate source of genetic diversity, spontaneous mutation is critical to the evolutionary process. The fitness effects of spontaneous mutations are almost always studied under controlled laboratory conditions rather than under the evolutionarily relevant conditions of the field. Of particular interest is the conditionality of new mutations-that is, is a new mutation harmful regardless of the environment in which it is found? In other words, what is the extent of genotype-environment interaction for spontaneous mutations? We studied the fitness effects of 25 generations of accumulated spontaneous mutations in Arabidopsis thaliana in two geographically widely separated field environments, in Michigan and Virginia. At both sites, mean total fitness of mutation accumulation lines exceeded that of the ancestors, contrary to the expected decrease in the mean due to new mutations but in accord with prior work on these MA lines. We observed genotype-environment interactions in the fitness effects of new mutations, such that the effects of mutations in Michigan were a poor predictor of their effects in Virginia and vice versa. In particular, mutational variance for fitness was much larger in Virginia compared to Michigan. This strong genotype-environment interaction would increase the amount of genetic variation maintained by mutation-selection balance.