1880.] 



G. Thibaut — On ilie Suryaprajnapti. 



Ill 



that there are two of each kind, express ourselves as if there were 

 only one. To proceed. 



Tlie astronomic-chronological period on which the system of the 

 Siiryaprajiiapti is based, is the well-known quinquennial yuga or cycle 

 with which we have long been acquainted from the Jyotisha Vedaiiga. 

 The same cycle is described in the Garga Samhita as we see from 

 the extant fragments of the latter work, and we learn from Varaha 

 Mihira's Panchasiddhantika that it likewise formed the fundamental 

 doctrine of a Paitamaha Siddhanta which, according to Varaha Mihira's 

 judgment, was one of the more important Siddhantas known at his 

 time. It is alluded to and rejected in a few words by Brahmagupta 

 in the dushanadhyaya of the Sphuta Brahma-siddlianta. Eeferences 

 to this cycle are met with in the early history of Buddhism. Whether 

 the so-called Vedic literature is acquainted with a cycle of this nature 

 is doubtful.* It will not be necessary to dwell in this place at length 

 on the constitution of the yuga; it will suffice to state that it is based 

 on the assumption of five sidereal revolutions of the sun being exactly equal 

 in duration to sixty-seven periodical revolutions of the moon and to sixty- 

 two synodical months, while one complete revolution of the sun is supposed 

 to be performed in three hundred and sixty-six days. That a cycle of this 

 nature based as it is on an utterly wrong assumption could maintain itself 

 for a considerable time as it manifestly has done is a matter for legitimate 

 wonder, and does not find a parallel in the history of chronological systems 

 among any other civilized nation. At the end of one yuga already the 

 quantity of the error induced by the mistaken estimation of the length of 

 the solar year amounts to nearly 5 x f = 3f days, the accumulation of 

 which quantity after the lapse of a few yugas could not escape the atten- 

 tion, we should think, of even the most careless observers. The matter 

 would indeed lie altogether differently if a conjecture (or as it stands we 

 might almost say, an assertion) of Colebrooke referring to this point had 

 been verified. He — after having given an account of the manner in which 

 the Jyotisha- Vedanga manages to maintain harmony between civil and 

 lunar time — continues " and thus the cycle of five years consists of 1860 

 lunar days or 1830 nycthemera, subject to a further correction, for the 

 excess of nearly four days above the true sidereal year : but the exact 

 quantity of this correction and the method of making it, according to this 

 calendar, have not yet been sufficiently investigated to be here stated." 

 The fact is that of this correction which Colebrooke considered so indis- 

 pensable, that he speaks of it as being actually found in the Vedanga, no 



* The question referred to in tlie text cannot be discussed hero. The writer 

 hopes shortly to find an occasion fully to treat it elsewhere. 



