On the other hand, the glyphs in the upper part of all four pages 21 to 24 face to the right, and, as already set out in detail, are practically all written in _reverse position_ as regards their prefixes, etc. And so also does the Eb-glyph in the day-columns we are now considering face to the right. These columns, unlike those on page 21, which include all of the 20 day-signs, only include 5 of the day-signs: Kan, Lamat, Eb, Cib and Ahau; Eb being the only non-symmetrical one of these. We have thus quite strong evidence, especially as provided by the position of the prefixes, for a right to left reading, opposed by the direction of this katun-number series--if it be one. In Egyptian writing, of course, the direction of the reading changes with the facing of the figures. To return now to the columns themselves, all the day-signs in any one column have each the same red numeral, so that we have: 8 Cib, 8 Ahau, 8 Kan, 8 Lamat, 8 Eb; and so on. The red numerals to each column also decrease by 2 towards the right, pari passu with the blue numerals. If we read each column downwards, it will form a closed circuit or round, returning into itself, with intervals of 104 days, from 8 Cib to 8 Ahau, etc., and again from 8 Eb back to 8 Cib. But if we next try to go to the next column, the series breaks, for from 8 Eb to 6 Lamat is only 76 days. We get a like break whether we read upward or downward, or right to left. Taking the columns separately then, the entire series (whether made up of 13, 20 or any other number of columns) cannot be made to read in one regular series, with a constant interval between the successive days of the whole. But, if we restore two columns, making 13 columns, and then read horizontally _across_, either right to left, or left to right, one line after another, the first day of the second line follows the last of the first, and after going through the whole 65 terms, we return again from the last of the last line to the first of the first--always with a constant interval. In other words, this section could be written around a wheel. If we read left to right, the distance from (10 Kan) to 8 Cib, etc., is 232 days; 232×65=15,080. Or if from right to left,[33-*] the interval from (12 Lamat) to 1 Cib, etc., is 28 days; 28×13 = 364, ×5 = 1820. That both of these products are multiples of 260 is a truism, and cannot in any way require us to see a tonalamatl reckoning as the basis of this passage. Nor is each separate day-column a tonalamatl in fifths, as so often found.