In Tahua there is a sofisticated use of viri. The text has 4 viri glyphs:   Aa5-7 Aa8-26 Ab1-1 Ab7-26 Although quite similar they are nevertheless not exactly alike. Ab1-1 is harmoniously drawn, but shorter than Ab7-26. Aa8-26 has a thick bottom 'tail' end, whereas Ab7-26 contrariwise has a thick upper 'tail'. Aa5-7 and Aa8-26 are drawn as if deformed. Neither the top part nor the bottom part in Aa5-7 is thick, and the glyph is slightly bent forward, as if it was old. The distance measured in glyphs from Aa8-26 to Ab1-1 is 60 and between Ab1-1 and Ab7-26 the distance is 520 glyphs. Let us instead say that 60 is measured beyond Aa8-26 up to and including Ab1-1 (which, of course, also must give 60): 59 520 Aa8-26 Ab1-1 Ab7-26 - 580 = 20 * 29 - The viri glyph type is connected with 29, and the distance between Aa8-26 and Ab7-26 is 20 * 29 = 580. A definite proof of viri being associated with 29 is delivered when we complete the cycle and measure from Ab7-26 up to and including Aa8-26: 752 Ab7-26 Aa8-26 754 = 26 * 29 We recognize the numbers (20, 26, 29, 46) from the structure of the K calendar. The number of glyphs in Tahua is (20 + 26) * 29 = 1334. But the 4th viri, Aa5-7, what function has it?

Also the ordinal numbers in Aa5-7 suggest a minor position compared with the other 3 viri glyphs. Obviously important is Ab1-1, but also Ab7-26 carries a powerful number: 7 * 26 = 182 (associating to half a year, or - if each day needs two glyphs - a quarter).

Next page:

The great hau tea in Aa4-70 probably indicates the 'center of summer'. The 'roof' is open at left as if to let in the light. The line at left of the vertical midline is not straigth and presumably indicates 'spreading out'.

It is the last glyph of its kind and therefore probably located at 'noon':

Aa4-63 is glyph number 314 from Aa1-1 (as if half a year has been reached). There is a 10-glyph sequence in which Aa4-70--71 are located:

I have redmarked pairs of glyphs because in this game the middle probably is defined from a total with an even number.

Counting ahead a further 314 to reach 628 makes us arrive at Aa8-43:

Henua ora in Aa8-45 and Aa8-52 indicate that some kind of end has been reached. But 1334 = 628 + 628 + 78, and the text maybe tells about two cycles around the year. Alternatively each day needs two glyphs, in which case 314 at Aa4-63 means that 'noon' has been reached after a quarter - twice 314 is needed to cover a cycle of half a year and Aa1-1 will be located at the beginning of the 2nd quarter.

The internal parallel ought to have 314 at Ab5-1:

Counting backwards from Ab5-1 it implies a beginning at Ab1-12, but Rei at Ab1-13 suggests the new season begins there:

Ab1-1	Ab1-2	Ab1-3	Ab1-4	Ab1-5	Ab1-6	Ab1-7
Ab1-8	Ab1-9	Ab1-10	Ab1-11	Ab1-12	Ab1-13

The expressive forward design of Ab1-13 makes it unique among the Rei in the Tahua text:

Counting 628 forward from Ab1-13 we reach Ab8-60:

Ab8-54	Ab8-55	Ab8-56	Ab8-57	Ab8-58	Ab8-59	Ab8-60
Ab8-61	Ab8-62	Ab8-63	Ab8-64	Ab8-65	Ab8-66	Ab8-67
Ab8-68	Ab8-69	Ab8-70	Ab8-71	Ab8-72	Ab8-73	Ab8-74
Ab8-75	Ab8-76	Ab8-77	Ab8-78	Ab8-79	Ab8-80	Ab8-81
Ab8-82	Ab8-83	Ab8-84

To summarize the tentative results based on a hypothesis of two glyphs per day and therefore twice 314 for each quarter:

The number of glyphs in line a is 628 + 42 = 670 and the 'extra' 42 arrive at the end (blackmarked). In line b there are 628 + 36 = 664 glyphs, or rather - as we can see now - 12 + 628 + 24 = 664 glyphs. Instead of 42 and 36 there is a more harmonious 12, 24 and 36.

Strangely, the central (redmarked) 314-glyphs (Aa4-63 and Ab5-2) are of different types. A simplifed table illustrates it better, and then we can see that all 4 of the 314-glyphs are different:

Alternatively we let side b come first (according to the Metoro way of reading Tahua):

 

This looks better with a natural order between 12, 24 and 42. Beyond 42 arrives 12 and their sum is 54 (= 3 * 18) - but so it is also if we consider side a as the first side.

The possibility still remains, however, that we should start counting at Ab1-12 instead of at Rei. I think the pictures in the 314-glyphs tell me so:

 

25 is not a bad number in the circumstances. To have the 'black' numbers as odd is as it should. 42 will then not stand alone but be added to 11 in order to reach 314 - 9 * 29.