208 Shinzo Shinjo 



calendar by the intercalation of seven months in nineteen years, seems 

 to have been effected almost without a failure. We even suspect the 

 existence in tliat early period of a definite calendar system, by means 

 of which the intercalation may have been mechanically effected. 



It seems quite certain, however, that in the latter half of the 

 Chan kuo age, a certain calendar system, called Clinan yii U ^jl^M, 

 had been founded and was afterwards in general use, until the adop- 

 tion of the T'ai cJi'u U :i:^M C^^' ''Fi^'^t calendar system") in 104 

 B.C., under the reign of the Eemperor Wu-ti j^^. In the Glman yil 

 calendar system, it is assumed that 



the length of a year = 365.25 days, 



the length of a month= 29.53085 days, 



making 



29.53085x235=365.25x19, 



that in the year 366 B.C., 



the seasonal epoch N-ch'un jj^^, the ^4 part of a year preced- 

 ing the vernal equinox, and considered at that age as the 

 ideal beginning of a year, 

 the new-moon (conjunction), considered as the ideal beginning 



of a month, 

 the day cTiia-yin ^%, considered as the first in the sexagesim- 

 al system, 

 and the dawn, considered as the ideal beginning of a day, 

 — all these happened just to coincide with one another, that this 

 coincident epoch ought to be taken as the natural origin of the 

 calendar system, and that hence the year 366 B.C. was to be counted 

 as the year cMa yin ^J|; the first year of counting in the sexagesi- 

 mal system. 



From evidence from variour sources, we assume the epoch of 

 foundation of the Ohuan-yil calendar system to have been somewhere 

 about the time 360-330 B.C. As a sidelight on this matter, I may 

 quote a passage from the book of Mencins ^^, a collection of say- 

 ings of the great sage Mencius, who is believed to have been active 

 about 300 B.C. "The heaven is high, the stars are remote; yet if 

 we proceed on the right way in the search for truth, we should be 

 able to find out the epoch of solstices even thousands of years hence." 



^^M^o ^mzm^^o ^^mWCo 1"m;tH5g, Br^Mi5:-ife.o 



This passage is especially interesting to us, as showing first that 

 about the time of Mencius, a difinite calendar system must have been 



