A large-scale survey of the Central Arizona-Phoenix Long-Term Ecological Research study area was conducted in which many kinds of samples were collected (e.g., plant, insect, soil). The plant samples were of a smaller size than typical herbarium accessions due to limitations in personnel, storage space, available plant materials, and time. The vouchers were identified in the field if possible. The vast majority of the rest were identified in the herbarium morphologically after the survey; less than 5% of the vouchers remained unidentified. Of these, a subset were subjected to molecular analysis using sequences of the nuclear ribosomal ITS region and to a BLAST search of sequences in GenBank to find closely related taxa. The close relatives along with the unknown specimens were subjected to maximum parsimony analysis to identify the potential phylogenetic relationships of the unknown specimens. Through this analysis nine unknown plant accessions were identified to family, tribe, genus, and species. Once this level of identification was established, some specimens were re-examined morphologically and compared with potentially related taxa in the herbarium to confirm or improve identifications. We outline an efficient method for surveying and collecting plant vouchers for a large sampling area. We also demonstrate that one can combine morphological and molecular data in the identification process to produce more complete datasets.