Genetic diversity, inbreeding and mean kinship
Consistent with our predictions, observed levels of genetic diversity were significantly lower in King Island scrubits than in all Tasmanian scrubtit subpopulations apart from the Tasman Peninsula. Interestingly, the samples from the Tasman Peninsula exhibit the similar within-population mean kinship values as those on King Island. Relatively lower levels of genomic diversity are observed in many island vertebrate populations, particularly on small islands as shown in northern quolls Dasyurus hallucatus (von Takach et al. 2022) and black-footed tree rats (Djintamoonga) Mesembriomys gouldii (von Takach et al. 2023). Individual inbreeding coefficients and mean kinship measures were also higher in the King Island scrubtit than in the Tasmanian scrubtit but, against our predictions, inbreeding measures (FIS) were lower for the Colliers Swamp and Nook Swamps subpopulations than for all Tasmanian scrubtit subpopulations. FIS detects non-random mating in the most recent generation (Waples, 2015), so our lower-than-expected estimates for these two King Island subpopulations could simply mean that these individuals are not mating with close relatives. However, other results — notably the reasonably high individual inbreeding coefficients calculated from the modified Visscher’s method (Figure 5) — suggest these unusually low FIS values may be explained by small and / or sex-biased sampling and therefore requires further study.
Phenotypic evidence of defective traits in small, inbred populations is likely to occur but not always noticed in wild systems (but see e.g. Roelke et al. 1993 and Harrisson et al. 2019). Although unknown at this time if it is a defective trait or not, over 50 % of King Island scrubtits we sampled exhibited crown baldness (Figures 5 & S3). Baldness was present in both sexes and all three King Island subpopulations, but was not recorded in any Tasmanian scrubtits. The probability of baldness was negatively associated with multi-locus heterozygosity, but the relationship disappeared when we restricted the analysis to the King Island population. It is possible that baldness may be due to non-genetic effects such as endemic parasitism, disease, inter/intraspecific aggression linked to low habitat availability or an ageing population (Lachish et al. 2012; Thys et al. 2017; van Velden et al. 2017). However, baldness was similar in all affected individuals (Figure S3) and we found six candidate SNPs that may play a role in the development or expression of this trait. One of these SNPs is associated with the DOCK11 gene, a gene that has been linked to early feather development in chickens (Figure S10; Table S7; Cheng et al. 2018). Further investigation into the phenotype-genotype associations may be able to determine the genetic and physiological pathways leading to baldness.