422 Mr. Marsden on the Era of 
At the expiration of the third year, the parts amounted to 14 ; 
of the fourth year, to 25 ; and of the fifth, to 36 ; when the 
intercalation was again applied, and a balance of 6 parts car- 
ried on. From this it will be underflood in what manner the 
fractional exceedings of each year were combined and difpofed 
of through the l'ucceeding years of the cycle ; and it will be 
neceflary only further to remark, that, when the aggregate of 
the fractions falls fhort no more than 2 or 3 parts of the 
number of 30, they ftill add the intercalary day, and deduct 
the deficiency from the excefs of the following year, which, 
in the courfe of one cycle, takes place only three times. At 
the end of the 29th year, the accumulated fractions, amount- 
ing exaflly to 30, are commenfurate with the intercalation 
then applied ; and the excefs of the 30th, or lafl year, is ac- 
counted for in the firfl intercalation of the fucceeding period. 
The operation would doubtlefs have appeared more methodical, 
if the firfl intercalary day were not to have been added till the 
end of the third year, and the eleventh, or lafl, till the end 
of the 30th year or termination of the cycle. From this confi- 
deration fome commentators have been led to difient from the 
more general idea, as above given, and to fuggefl, that the 
embolifm is in fa£t applied fo foon after the commencement 
of the cycle, as the yearly accumulation of the fra&ional 
parts exceeds the fum of half a day, or twelve hours, and 
that it accordingly is made to take place at the end of 
the fecond year, becaufe the fradlions then amount to 17 b. 
36% or 22 parts in 30; at the end of the fifth year, becaufe 
they then amount to 25 ; and at the end of the feventh year, 
to 1 7 parts ; keeping thus as near as poffible to the mean divi- 
sion of time, by applying the compenfation before it is fully 
wanted. The effedl, however, is in both cafes the fame, 
and 
