The nature of spontaneous mutations, including their rate, distribution across the genome, and fitness consequences, is of central importance to biology. However, the low rate of mutation has made it difficult to study spontaneous mutagenesis, and few studies have directly addressed these questions. Here, we present a direct estimate of the mutation rate and a description of the properties of new spontaneous mutations in the unicellular green alga Chlamydomonas reinhardtii. We conducted a mutation accumulation experiment for 350 generations followed by whole-genome resequencing of two replicate lines. Our analysis identified a total of 14 mutations, including 5 short indels and 9 single base mutations, and no evidence of larger structural mutations. From this, we estimate a total mutation rate of 3.23 · 10 210 /site/generation (95% C.I. 1.82 · 10 210 to 5.23 · 10 210 ) and a single base mutation rate of 2.08 · 10 210 /site/generation (95% C.I., 1.09 · 10 210 to 3.74 · 10 210 ). We observed no mutations from A/T / G/C, suggesting a strong mutational bias toward A/T, although paradoxically, the GC content of the C. reinhardtii genome is very high. Our estimate is only the second direct estimate of the mutation rate from plants and among the lowest spontaneous base-substitution rates known in eukaryotes. N EW mutations are the ultimate source of the genetic variation necessary for adaptation via natural selection. Moreover, the mutation rate has profound consequences for a myriad of disciplines, including conservation, genetics, medicine, and evolution. The rate of new mutations influences both the speed at which populations respond to natural selection and the rate at which fitness may decline due to inbreeding. The rate, strength, and sign of fitness effects of new mutations are critical parameters of models of the evolution and maintenance of sexual reproduction. Currently, however, characterization of the rates, causes, and effects of different kinds of mutations is lacking.Although mutation is fundamentally important for adaptive change, the vast majority of mutations influencing fitness are generally believed to be deleterious (Keightley and Lynch 2003; but see Shaw et al. 2002;Rutter et al. 2010). For this reason, most theory, especially in recombining populations, predicts that selection will drive mutation to lower and lower rates (Kimura 1967;Kondrashov 1995;Dawson 1999;Lynch 2008). Even when a mutator allele increasing the mutation rate arises and produces a beneficial mutation, recombination will tend to disassociate the mutator from the beneficial mutation, eliminating any benefit it may gain from hitchhiking (for example, Raynes et al. 2011). Although selection is expected to drive the mutation rate toward zero, the mutation rate remains detectably above zero.Relatively few studies have estimated the spontaneous mutation rate directly through the sequencing of a large fraction of the genome (see below). Accurate estimation of the mutation rate is made difficult by the very low rate of spontan...