According to Gemelli[44] the south was denoted by a "blue field," and the symbol Tochtli; east by a red field, and the symbol Acatl; the north by a "yellow field," and the symbol Tecpatl, and the west by a "green field," and the symbol Calli. In this plate we have precisely the colors he mentions, red in the east, and yellow in the north, but green is at the south, and blue at the west. Sahagun remarks[45] that "at the end of fifty-two years the count came back to _Cetochtliacatl_ (one-Rabbit-Cane), which is the figure of the reed dedicated to the east, which they called _Tlapcopcopa_ and _Tlavilcopa_, nearly towards the fire or sun."[46] This language is peculiar and important, and indicates that he had a Mexican painting similar to the plate now under discussion before him, in which the year symbols were at the _corners_ instead of at the _sides_. On this supposition only can we understand his use of the term "_Cetochtli-acatl_," and the expression "nearly towards the fire," &c. His use of the term "fire" in this connection undoubtedly indicates red. His language is therefore in entire harmony with what we find on this plate. According to Gemelli and Chavero the element _earth_ was assigned to the south; in this plate, in the right space inclosed by the green loop, we see the great open jaws representing the earth out of which the tree arises. From a careful examination of this figure, so frequently found in this and other Mexican Codices, I am convinced it is used as the symbol of the grave and of the earth. The presence of this symbol and of the figure of death in this space, as also the figures of the gods of death and the under world in the corresponding space of the Cortesian plate, strongly inclined me for a time to believe that this should be considered the north, as in the Aztec superstitions one class of the dead was located in that region; but a more thorough study leads me to the conclusion that these figures are intended to represent the earth and to symbolize the fact that here is to be found the point where the old cycle ends and the new begins. I will refer to this again when I return to the description of the Cortesian plate. All the authorities, except Boturini, refer the year Tecpatl or Flint to the north, which agrees with the theory I am advancing, and in the lower left-hand corner we find in the red circle the figure of a flint, which according to my arrangement applies to the north, represented by the yellow loop.