KORSTBMANN.] MAYA CHRONOLOGY 477 



idea which Doctor Seler communicated to me in a letter of December 

 21j 1890. He wrote to me that in his opinion the years in the Dres- 

 den codex did not begin with Kan, Muhic, Ix, and Cauac, but with 

 Akbal, Lamat, Ben, and Ezanab; according to the corrected calen- 

 dar, therefore, the last days of the year must be Ik, Manik, Eb, 

 and Caban. But Kan, etc., still rank as the principal days, and the 

 years are designated by the first principal day encountered in them. 

 For instance, they are distinctly prominent as principal days in 

 Codex Cortesianus, pages 3a to 6a ; Troano codex, pages 33c to 32c and 

 23 to 20, and Dresden codex, pages 9b and 29c. 



New light next falls on number 4. The day XIII 20 (Akbal), 

 wherein the highest Aveek-day number is connected with the last 

 day of the series, is nothing more than the new year's day of the year 

 1 Kan. These periods of 91 days, therefore, arranged in groups of 

 four, are the 4X91 days which, following the day XIII Akbal, 

 make up the year 1 Kan, as, for instance, in the Dresden codex on 

 pages 32 and 64. In the series to be found on the latter page the sig- 

 nificance of the solar year is quite apparent, emphasized by the 

 singularly elaborated sign of the zero in the fourth and the eighth 

 terms of the series ; that is, at the close of the first and of the second 

 years. 



As Doctor Seler himself writes me, number 5 can also be simply 

 explained. For the Dresden codex, pages 25 to 28, does not treat of 

 the last two days of the year, but far more naturally of the last day of 

 the old and the first day of the new year. I must leave it to Doc- 

 tor Seler to establish his view by discussion of the pictures and 

 glyphs. 



Lastly, niunber (> also presents a more satisfactory aspect. For now 

 III 2; 13, 3d month is no longer called 3 Chicchan which is followed 

 by the thirteenth day of the third month, but far more simj^ly 3 Chic- 

 chan which is the thirteenth day of the third month. Tlie normal 

 date IV Ahau, 8, 18th month therefore really falls on the eighth day 

 of the eighteenth month and, in fact, as I have alwaA^s believed, in 

 the year 9 Ix, which, however, according to the new theory began 

 with 8 Ben. 



The next step is to attempt further conquests in this realm of 

 glyphs, starting from this firm basis of numbers and computations, 

 and the first thing to be done is to search for pictures which express 

 the conceptions of year, the change to a new year, the beginning of 

 the year, and the close of the year. As the serpent pictures have an 

 undeniable reference to periods of time, so the most perfect symbol 

 for the year, it seems to me, is a serpent forming a closed ring. Such 

 a serpent is found in Codex Cortesianus, page 3a, and inscribed 

 within it the numeral 18, which I am inclined to interpret as mean- 

 ing the eighteen months. Likewise in Codex Cortesianus, pages 



