562 THE INSCRIPTIONS AT COPAN. 
Since the first two elements are identical with two of the elements of Glyph C, 
and because, when present, Glyph D always immediately precedes Glyph C, it 
seems reasonable to infer that the two characters are closely connected in meaning. 
R. K. Morley has pointed out by far the most important characteristic of this 
glyph, namely, that when Glyphs E and D both occur in the same text, as 1s fre- 
quently the case, Glyph D never has a coeficient and Glyph E always has one. 
The latter may be only 1, it is true, as on Stela 8 at Naranjo, but wherever Glyphs 
E and D are both present the latter never has a coefhicient. 
This characteristic very strongly suggests that Glyph D contains within itself 
the data for Glyph E also, that is, that Glyph D could be expanded into Glyphs D 
and E at will by attaching its coefficient to Glyph E, and either omitting Glyph D 
altogether or recording it without a coefhicient. This may possibly explain why 
Glyph E is present only in 30 per cent. of the texts under observation. When it was 
desired to record them both, however, perhaps to fill in a space on the monument, 
then Glyph D was recorded without a coefficient, its coefficient going over to Glyph 
E. In fact, whatever these two glyphs may mean, it is evident that they are very 
closely connected, if not indeed actually synonyms, since we get all three com- 
binations of them possible: (1) Glyph D by itself; (2) Glyph E by itself; and (3) 
both together; but what we never get is D with a coefficient when E 1s present. 
Glyph E (figure 79, w—b’), as the writer has already shown,! is probably reduci- 
ble to a number of days. It is composed of but two constant elements, the moon- 
sign (the same variant as in Glyph A) and a series of numerical coefficients ranging 
from ito1g. Since the variant of the moon-sign is the same as in Glyph A, it must 
have a numerical value of 20, and since the coefhicients attached to it are always 
either to the left or above (see figure 79, w—b’) they are doubtless to be regarded as 
multipliers, so that the glyph can stand for as low as 20 days (1. ¢., I X 20) or as 
high as 380 days (i. ¢., 19 X 20). Furthermore, at Yaxchilan it is sometimes modi- 
fied by the same superfix as the cycle-sign is in the great-cycle glyph, that is to say, 
by an element that multiplies it by 20, making it 400.2 And in one case at least 
(Stela 24 at Naranjo) this is modified by a bar-and-dot coefficient on its left as high 
as 18, making it equal to 7,200 days (1. ¢., 18 X 400) under this assumption, and with 
7,600 days (1. ¢., 19 X 400) as its possible maximum. 
However probable it now appears that Glyphs D and E are synonymous and 
that both may be reduced to a number of days, we are yet entirely in the dark as to 
what determined the number of days they appear to record. Any suggestion that 
might be brought forward at this time would be purely speculative, but in closing 
the presentation of these two glyphs the writer wishes to note that he regards it 
as highly probable that this number has something to do with the accompanying 
Initial Series in each case, possibly indicating the distance therefrom, either before 
or after some specific observed eclipse phenomenon. 
We come next to the last glyph of the Supplementary Series, Glyph X (see 
figure 79, c’-j’) which is the sixth character from the left or the third from the 
right. ‘This sign is the least understood of all the characters of the Supplementary 
Series, 1s the most variable of all (hence its name Glyph X), and in importance 
ranks with Glyphs A and C. In the other glyphs of this count we have been able 
to establish essential elements or characteristics which remain constant, no matter 
through what changes the accompanying coefhcients may pass. But in Glyph X 
for the first time we reach a glyph which possesses no element common to all of its 
examples, but on the contrary which passes through a number of changes. Happily 
it appears possible to classify these, at least roughly. The commonest element in 

1Morley, 1916, pp. 384-387. 2Tbid, 1915, pp. 117-119. 
