We determine the observing sites for eclipses of large magnitude recorded in ancient Chinese chronicles from 200 BCE to 900 CE, by adopting the difference between terrestrial time and universal time, ΔT, given by Morrison & Stephenson (2004). The records of solar eclipses with large magnitude are divided into four groups in accordance with the historical variations of the capital cities of ancient Chinese dynasties. We determine areas in which all the eclipses in each group, with an eclipse magnitude larger than a certain threshold value, could be observed. We find that these areas coincide with the historical capitals, which agrees with the general idea that the solar eclipses were observed at the capital of each dynasty. This result also veries the ΔT values during the period from 100 BCE to 400 CE, during which historical records of eclipses are so rare that the ΔT values can only be obtained by interpolating the long-term data. Moreover, we show that the eclipses described by the term Ji in East-Asian history are not all total eclipses; their mean magnitude is 0:96 ± 0:04. We find that complementary expressions, such as dark daytime and appearance of stars during the eclipse, strengthen the possibility that eclipses described by the term Ji were total. We also provide quantitative definitions for expressions such as `being not complete and like a hook', `being almost complete', `visibility of stars during the eclipse', and `darkness during an eclipse.' The literal meanings of these expressions are in agreement with the recent physical modeling of sky brightness during total eclipses provided by K¨onnen & Hinz (2008).