In a previous contribution (Journal of the International College for Advanced Buddhist Studies国際仏教学大学大学研究紀要,vol.I, 1998, pp.176 ff.), I examined several problems connected to the precise identification of the Land of Anxi (Anxiguo安息國) as reflected in Chinese sources dating back to the Wei, Jin, Northern and Southern Dynasties 魏晉南北朝 (3^-6^
centuries C.E.). My investigation revealed that by the middle of the 6^
century, we witness the spread of a wrong identification of the Land of Anxi, which originally referred to the Parthian Empire, with Bukhara. The present study continues this line of research and casts further light upon the reasons behind this geographical confusion. In Chapter One of my study, I examine relevant documents regarding the Land of Anxi in the Former Han前漢and Later Han 後漢 Dynasties (3^ century B.C.E.-3^ century C.E.) as well as later historical records on the Land of An安國(=Bukhara) in the Sui隋Dynasty (581-618). It appears that the mistaken identification of the two regions, which is clearly seen in the Chapter on the Western Regions西域傅of the Suishu隋書, is not found in any source from the time of the Former Han Dynasty to the middle of the 6^
century. The confusion seems to appear rather suddenly sometime after the middle of the 6^
century. My data also reveals that while the name 'Land of Anxi'安息國is well-attested since the Former Han, the appellation 'Land of An'安國seems to have come into existence around the middle of the 6^
century. Chapter Two deals with the so-called 'Barbarians'胡人from the Western Regions who adopted the surname An安as reflected in Chinese sources from the Later Han to the Northern and Southern Dynasties (1^ century-6^
century). An analysis of the origin of these 'Barbarians' leads me to make some conjectures concerning the time when the Sogdians began to use the surname An. From the Later Han Dynasty to the Period of the Three Kingdoms三國(1^ century-3^ century), the surname An安was used by people original from the Parthian Empire. Later on, it appears that Sogdians from Bukhara started to adopt the same surname. Such a situation is clearly attested in 545 C.E. and may go back as early as to the latter half of the 5^
century. In Chapter Three, I conclude that the erroneous identification of Bukhara with the 'Land of Anxi' did not come from first-hand geographical information obtained by the Chinese about the Western Regions. Its cause seems to be connected with the fact that the Sogdians from Bukhara began to borrow the surname An for their convenience and declared themselves to be original from the `Land of Anxi'.
