The Fossil Rodent Midden Record in the Americas: Metagenomics and other novel opportunities

Fossil rodent middens are amalgamations of plant and animal remains embedded in cemented blocks of crystallized urine and preserved for millennia in aridland caves and rock shelters. Since 1960, >2000 fossil packrat (Neotoma) middens have been analyzed and archived from western North America. Since 1994, another 1000 middens produced by other rodents (Abrocoma, Phyllotis) have been studied from western South America. Middens are among the world's richest sources of plant and animal remains amenable to myriad laboratory analyses that have been used in clever ways to investigate biological responses to climate variability and change over the last 50,000 years. In past decades, the community of midden researchers has dwindled, slowing adoption and progress in midden aDNA analysis. Only a handful of genetic studies have been attempted. Ancient DNA in middens was used to identify the dung, diet, and Pleistocene biogeography of a rare rodent in northern Chile (Kuch et al. 2002 Mol Ecol) and of an extinct ground sloth preserved in southern Argentina (Hofreiter et al. 2003 Quat Res). Ancient DNA from a midden in Tiburón Island in the Gulf of California showed that a unique haplotype of desert bighorn sheep inhabited this land-bridge island 1500 years ago, and that its successful introduction in 1975 represents an 'unintentional rewilding' (Wilder et al., 2014 PLOS ONE). Ancient DNA from middens in the Grand Canyon, USA provided evidence of papillomavirus (PV) infection and long-term codivergence with packrats over the last 27,000 years, the oldest-known PV sequence (Larsen et al. 2018 Virus Evol). High-throughput sequencing and metabarcoding have been reported from South African, Australian ((Murray et al., 2012 Quat Sci Rev), Chilean (Diaz et al. 2017 Global Change Biol), and now North American (Moore et al., In Review) middens. Middens in the Americas offer the chance to genetically profile entire communities through time and space, and to reconstruct time-lapse molecular phylogeographies for individual species. Keys to future research will be revised field and laboratory sampling protocols, thorough understanding of midden aDNA taphonomy, accelerated barcoding of modern plant taxa in midden areas, easy access to existing physical and data archives, and a new generation of paleoecologists trained in genomics.