In almost all the rivers of Formosa, we see numerous terraces (Fig. 17). The writer studied the terraces of the Takkiri River. Lying in the northeastern part of Formosa, it is about 45 km in length. It rises in the high mountains of Nankotaizan (3797m) and Gokanzan (3416m), and flows into the Pacific Ocean. The, terraces along the Takkiri are distributed as shown in Fig 1, the high ones of which have a relative height of 100m in its lower course and over 3003m above sea level in its upper course. The low terraces have relative heights of less than 100m. The larger ones along the Takkiri are mostly those formed by changes in the river course. There were two depositions; first, the high terraces were formed, followed by the low ones (Fig 7). The writer obtained a fossil (Celtis sp.) near the base of a high terrace and also striated gravels from that deposit (Fig 7) . The main terraces along the. Takkiri are as follows:
Name Above sea level (m) Relative height (m) Remarks Rotien. 155 120 Fig 3 Browan 392 192 Fig 5 Doyon 750 380 Fig 7 440 70 Daoras 710 230 Profile resembles Fig 7 Kumoheru 980 300 Fig 8 Siipau 980 380 Fig 9 Umezono 1100 380 Fig 10 Tausai 1420 430 Fig 11
Fig 13, 14, 15 show the Takkiri terraces projected over the profile of that river. From the projection, the writer believes tat deposition took place as shown in Fig 16. It is possible to conceive that depo-sition Occurred, resulting in two groups of terraces, high and low, in all the other rivers of Formosa. There seems to be no special reason for the development of the Takkiri terraces. The main accu-mulation terraces in Formosa with large relative height are as follows Name of rivers Name of terraces Above sea level (m) Relative height (m) Girandaktisui Menotan 1300 220 Daidakusni Piy ahaw 1060 200 Takkiri Tausai 1420 480 Tansui Teirikk 800 200 Taian Tengu 900 200 Taiko Sikayo 1650 100 Taito Makanaji 1200 100 Dakusui Trock 1520 180 Muso 1820 Sobun Tappo 980 140 Simotansui Magatun 909 150 Takkuf akura 1030 242 Regardless of the direction, rivers that take their rise in high moun-tain ranges have large deposition, and the development of terraces begin upstream. In those mentioned above, the cause of depostion may be either the subsidence of the whole of Formosa or climatic changes. As to the latter, it is reported that there are eirges at Nankotaizan and glacial relic flora and fauna in the high mountains of Formosa. To explain the depostion, climatic change is highly pro-bable. The cause of the revival of erosion may be also both crustalup he aval and climatic changes.

View full abstract