Although the hypothesis from K of 192 days for the 1st half year survived a test against G, the prediction about niu arriving at the beginning of each half year proved difficult to come to grips with. Although at the beginning of the 1st period - immediately following Rei - a shape like niu appears in Ga2-29 it is not niu but the black and living king (Kioe Uri), not a theoretical construct with open perimeter (as in Ka3-16):

In the absence of coconut trees on Easter Island it is as if the king had taken upon himself to be that mythical tree.

And then I have listed 7 niu glyphs in my catalogue, none of which has the normal shape:

 

Ga2-29	Ga3-13	Ga4-26	Ga6-5	Gb2-28	Gb3-8	Gb7-25

They all (not only Te Kioe Uri) appear to be 'glyph plays', not the normal niu type.

An important discovery was the resemblance between Gb3-8 and Aa6-14:

Aa6-14 (472)	niu
Gb3-8 (298)

This discovery then led on to finding Ga3-13, which seems to mark the end of the takaure season:

Gb3-7	Gb3-8 (298)	Gb3-9	Gb3-10 (300)

3
	Ga3-10	Ga3-11	Ga3-12	Ga3-13 (72)	Ga3-14	Ga3-15	Ga3-16

At Hua Reva old sun has vanished, and ariki in Gb3-7 (with 6 limbs as if an insect) has to realight the 'fire':

Gb3-3	Gb3-4	Gb3-5 (295)	Gb3-6
Gb3-7	Gb3-8 (298)	Gb3-9	Gb3-10 (300)

Gb3-7 has position 297 = 3 * 99, as if Venus was involved (in triplicate). The frontal 'arm' in Gb3-8 is not open as the one in the past - the new season has begun. In Ga3-13 the mirror image is seen, the frontal 'arm' is open - the season of takaure has ended. Its length - measured in glyphs - is 13 * 19 = 247.

	245 = 9 * 49		225 = 9 * 25
Gb3-8 (298)		Ga3-13 (72)
247 = 13 * 19

Maybe it is a symbolic number which cannot be translated into days? But 365 - 247 = 118 = 2 * 59 suggests it is meant to be days.

247 indicates a year with 365 days, a solar year. Therefore it is understandable that there can be two different Hanga Takaure:

sun	moon
Ga3-13 (72)	Ga1-1 (472)

The distance between them is exactly 400 glyphs (with sun coming first). We are moving forward!

Is it then possible to find the moon equivalent to Gb3-8? 298  + 400 = 698, and then 698 - 472 = 226:

sun	moon
Gb3-8 (298)	Ga8-23 (226)

In Gb3-8 there may be 5 + 4 = 9 'feathers (alternatively 6 + 5 = 11), in Ga8-23 there are 14.

3 * 8 = 24 and 8 * 23 = 184. A number play could be involved: 3-8 reversed into 8-3 and then 2 added as a mark of the moon.

More relevant is to look at the glyphs around Ga8-23. The preceding Ga8-22 indicates how ragi now is at the top, after the previous 'person' has fallen straight down head first:

Ga8-16	Ga8-17	Ga8-18	Ga8-19	Ga8-20
Ga8-21	Ga8-22	Ga8-23	Ga8-24	Ga8-25
Ga8-26	Gb1-1	Gb1-2	Gb1-3	Gb1-4

Ga8-22 suggest Te Pei. But Te Pei arrives 10 glyphs later:

Ga8-22 (226)	Gb1-7 (236)

Does it mean there are 10 days between which are outside the regular calendar?