Microsatellites are robust molecular markers for exploring genetic variability in all organisms, now widely used in plant systematics and evolutionary biology. It is sometimes a drawback, however, that de novo isolation and characterization to obtain working primers for as yet unexamined species are prohibitively time-consuming. Recently, high-throughput sequencing technologies have had a significant impact on many different types of genomic research; they also provide new opportunities for locating microsatellites in non-model and not-previously-sequenced organisms. Here we report results of high-throughput sequencing to identify a large number of microsatellites from six distantly related angiosperm plant species (Acer pseudosieboldianum, Drimys confertifolia, Dystaenia takesimana, Hepatica nobilis var. japonica, Myrceugenia fernandeziana, Rhaphithamnus venustus). Using a combination of a total (non-enrichment) genomic library and small-scale pyrosequencing (1/8 of a PicoTiterPlate) with multiplexing option, we determined 2.57 to 6.23 M bp of sequence (10,411 to 21,021 reads) in the genomes of the six species and identified 63 to 284 microsatellites that had flanking regions suitable for primer design; we also obtained 1 to 29 microsatellites located in the chloroplast genome. We tested PCR amplification, reproducibility, and level of polymorphism of 42 primer pairs for Acer pseudosieboldianum and 24 for another five species using eight individuals from two populations. As a result, between 46% and 83% of markers were clearly genotyped, and between 17% and 63% efficaciously detected polymorphisms within populations. This method is cost- and time-effective in comparison with traditional approaches involving cloning and sequencing.