Plant exploration in the Pacific Islands during Taihoku Imperial University period (1928-1945) based on herbarium data

Survey of natural resources in the Southeast Asia and the Pacific is one of the most important missions of Japan during its colonization period since late nineteen century. Tens of thousands of specimens were collected in these regions mostly by two Japanese botanists in the early 20th century, Ryozo Kanehira and Takahide Hosokawa. They also represented the two major periods of plant surveys in Taiwan, led by the Department of Forestry of Government Research Institute and the Department of Botany of Taihoku Imperial University (TIU, now National Taiwan University, NTU), respectively. However, these specimens at NTU were never properly inventoried and largely neglected by taxonomists. This study is intended to collect the data from TAI herbarium at NTU and compile a database for future researchers interested in the Pacific flora. More than 4,700 specimens, including 407 types, collected from the Pacific islands were databased at the TAI herbarium (NTU). The specimens were mostly collected by Hosokawa from Micronesian islands, including Palau has 602 species, 1644 specimens, Pohnpei has 307 species, 1054 specimens, Mariana has 288 species, 777 specimens, Kosrae has 216 species, 484 specimens, Chuuk has 189 species, 463 specimens, and Yap has 198 species, 340 specimens. The top five families for these collections are Rubiaceae, Orchidaceae, Gramineae, Leguminosae, and Cyperaceae. These specimens were inventoried into a database, representing 145 families, 573 genera, and 1087 species of vascular plants. The results were compared with other flora of Micronesia, and here is a major difference found in the ferns on genus number. A comparison for the plants in the island of Pohnpei was conducted with a 2010 plant list, there are 87 species only present in our database, out of a total of 278 species, but there are 551 species in 2010 plant list, of which 360 species added as new records. We speculate these differences might be due to the bias of collection and /or new records on Pohnpei. The results will definitely provide valuable information for the studies of Micronesian flora.