We are now in a position to understand why there are no koti glyphs in G. The reason is that the text primarily describes time from a lunar perspective. Gb1-6 has a position equal to half the cycle, 236 = 8 * 29.5 = 472 / 2: Gb1-6 (236) Gb1-7 If sun had been the topic, then Gb1-6 would instead have been a koti glyph, I suggest. In other words, koti should not be used when the calendar is governed by the moon. Furthermore, I guess koti should only be used when referring to the sun. For example seems Hb3-6 to indicate a break in time caused by some other measurement: Hb3-4 Hb3-5 Hb3-6 (753) 251 3-6 is not necessarily connected with sun, as the Mamari moon calendar proves (with twice 36 glyphs). 753 is the ordinal number counted from Ha1-1, but presumably we should count from Ha1-4, in which case the day becomes number 250. It is clear that Hb3-6 is the last glyph, not the first of something new, because the 1296 glyphs in H will be found divided in two parts with Hb3-6 as the last glyph of the first part: 940 352 Hb9-48 (1) Hb3-6 (942) Hb3-7 Hb9-47 (1296) 942 = 3 * 314 = 2 * 471 354 = 12 * 29.5 No regular koti glyph appear among the 354 glyphs of the second part, none refers to the sun. Mercury and Venus are not the sun: Hb9-33 Hb9-34 Hb9-35 Hb9-36 Hb9-37 Hb9-38 Hb9-39 Hb9-40 Hb9-41 Hb9-42 Hb9-43 Hb9-44 Hb9-45 Hb9-46 Hb9-47 Hb9-48 Hb9-49 Hb9-50

Counting from Ha1-4 instead of from Ha1-1 can be explained as in the preceding glyph dictionary page: 63 is 36 reversed. At the end of side b there should be 63 glyphs to indicate that a season counted by 36 (or 360) has reached to its end.

Hb9-47 is glyph number 645 + 412 + 47 = 1104 counted from Ha1-4, and 189 glyphs then remain to the end of side b:

	190			1102
Hb9-48		Ha1-3 (1296)	Ha1-4		Hb9-47 (1104)
192 / 3 = 64			1104 / 3 = 368

With Hb9-48 as the first day of the cycle, one possible interpretation is to say that Venus rules the time outside those 368 days of a regular calendar. And the end of this regular calendar comes with Jupiter flying away (manu rere), i.e. these 368 days refer to a solar calendar.

But the end of side b is outside this solar calendar.

Koti in Wednesday has number 9-36, which can be read as 'end' (9) of the first half (36) - referring both to the week and to the solar year. But it seems possible to refer to the end of any cycle, cfr 36 as half 72 in the Mamari moon calendar.

9-36 should therefore be equal in meaning to 63. Moon has 29 (two nine) and sun 19 (one nine) to be more specific. 'Nine' is more general.

9-39 as the first glyph of Jupiter could point to a year with 390 days.

I will add at the end of the page:

The division of the glyphs in H in two parts at Hb3-6 is not the only one to keep in mind, we have earlier found another such:

	190			1102
Hb9-48		Ha1-3 (1296)	Ha1-4		Hb9-47 (1104)
192 / 3 = 64			1104 / 3 = 368

64 days are here 63 + 1, maybe to inform the reader about 36 (or 360) being reversed at the end of side b, and that a new one (1) will succeed.

We can speculate that 8 times 36 (368) indicates the perfect (8) cycle of the sun (36), and that these 368 days end with manu rere in Thursday (the day of 'Jupiter', the last 'garment' of sun).

2 more months are needed in the calendar, in order to reach 8 + 2 = 10. These can be expressed by 63, reversed 36, with 1 added. 1296 / 3 = 432 days are covered by the H text. 432 - 368 = 2 * 32 = 64. The beginning comes immediately beyond the end of 'Jupiter'.