556 THE INSCRIPTIONS AT COPAN. 
This peculiar grouping of the 29 and 30 day lunar months on pages 51 to 58 
of the Dresden Codex into higher lunar periods of never more than 6 months and 
never less than 5 immediately suggested that Glyph C of the Supplementary Series, 
which was known to treat of the moon, and the coefficient of which was never 
found to be higher than 6, was in fact a positional indicator in the same sort of a 
lunar group, the bar-and-dot coefficients from 1 to 6 indicating the position of 
Glyph A in a 5 or 6 month lunar period, an explanation which has since met with 
general acceptance. 
But this did not explain why the Maya, either in the lunar calendar on pages 
51 to 58 of the Dresden Codex, or in Glyph C of the Supplementary Series, should 
have grouped the lunations into larger periods of 5 and 6 lunations each. The first 
to suggest the true explanation of this characteristic, or at least so far as pages 51 
to 58 are concerned, was Meinshausen, who in 1913 showed that the totals of days 
on pages 51 to 58 of the Dresden Codex very closely agree with the intervals be- 
tween eclipses of the sun and moon: 
“Tt will occur to everyone who observed the eclipse of the sun last summer (1912) what 
an impression such a phenomenon of nature makes; therefore the fact will be easily under- 
stood that among the ancient civilized peoples who were much occupied with astronomy 
there were none which has failed to leave conspicuous records of such events. When, there- 
fore, starting from this fact, I searched the Dresden Codex for eclipses of the sun and moon, 
I was nevertheless not a little surprised instead of the expected slight and brief remarks to 
find, besides others, a long chapter that seemed to me to deal with such phenomena. 
careful study of the periodicity of the eclipses of the sun and moon soon brought me to the 
conclusion that my assumption, in spite of the contrary explanation of Professor Forste- 
mann of this part of the manuscript, was wholly correct. The following comparisons of the 
periods between eclipses of the sun on the one hand and the moon on the other with the 
numbers of the Codex leaves no doubt that the latter actually arose from the observation 
of such appearances. 
“Eclipses of the moon ordinarily recur for the whole earth, as appears from the follow- 
ing table, after 177 days, but occasionally after 502 days. If 502 be divided by 177 the 
remainder is 148. The Codex shows exactly these numbers. The agreement is, moreover, 
of such sort that a coincidence appears to be excluded.” 
Meinshausen then gives a series of lunar eclipses which took place in the years 
1778 to 1811 (first table, page 557), and a series of solar eclipses which took place in the 
years 1775 to 1808 (second table, page 557), as recorded in the Berlin Astronomical 
Almanacs. ‘The first three columns in each table describe the date of the eclipse, 
viz, (day) 20, (month) 7, (year) 1778. ‘The fourth column shows the differences in 
days between two successive eclipses; the fifth column, the total number of days 
from the beginning of the series to each succeeding eclipse; and the last column, the 
numbers in the Dresden Codex. When the latter disagree with the corresponding 
totals in the fifth column by more than 1 day, they are inclosed in parentheses. 
In the first table, which shows the lunar eclipses, out of a total of 51 num- 
bers quoted from the Dresden Codex, 23 or 45 per cent. agree exactly; 12 or 24 per 
cent. are I day off; 2 or 4 per cent. are 2 days off; and only 14 or 27 per cent. are 
28, 29, and 30 days off. Meinshausen suggests that the reason there are any dis- 
agreements at all between the actual eclipse periods and the manuscript is due tothe 
fact that not all the eclipses are visible at any one point on the earth’s surface, and 
consequently that the remainder of 148 in the Codex was always placed arbitrarily 
immediately before an observed eclipse. 
The agreements in the second table, that showing solar eclipses, are even more 
significant. Here, of the 69 numbers quoted from the Dresden Codex, 28 or 40 
per cent. agree exactly; 26 or 38 per cent. are I day off; 2 or 3 percent. are 2 days off; 
and only 13 or 19 percent. are 28, 29, or 30 days off. These agreements are again so 
1Meinshausen, 1913, p. 221. 
