The first of the pages from 'this link':

I cannot remember if I earlier in the glyph dictionary pages explicitly wrote that autumn equinox was where the year ended. I hope not.

With Gb5-6 as the 360th day of the year, 68 days beyond Gb3-1, summer could end somewhere beyond autumn equinox, 90 - 68 = 22, i.e. ca one month beyond. But we do not know where in our own calendar Gb8-30 is located - probably not at January 1. (Or rather - we should flip our calendar around because we are on Easter Island - probably not at July 1. But it is easier to think about solstices and equinoxes in our own well know calendar.)

Considering those 86 days from the beginning of their year to the beginning of their summer (at Ga4-3), and comparing with the mirror number 68 at the other end of their summer (beginning counting from Gb3-2), we can count 86 - 68 = 18, and then 86 - 9 = 77 = 68 + 9. The difference between January 1 and December 21 is also 9 days. If we go south of the equator and look into a mirror we will find January 1 arriving 9 days earlier than December 21. Given that spring equinox arrives a quarter after winter solstice (which is not exactly true), we need to count 9 days + a quarter to go from January 1.

But spring equinox south of the equator is autumn equinox and the other way around. Therefore I should have written July 1 and June 21 instead. The principle remains the same. The G calendar can begin 9 days earlier than June 21, i.e. at June 12.

The explorers arrived to the island on Maro 1 and it took them 10 days to construct a house and plant yams. My calculations would lead to them departing from Haga Te Pau 2 days later. Yet they seem to have left 25 days later. These 25 may be just a symbol for the sun (5 * 5).

Hanga Takaure would then arrive 86 - 9 = 77 days beyond 'January 1'. If we count with 30-day months, we will locate Hanga Takaure to 'March 17', close to spring equinox.

Summer would end 86 + 94 + 112 = 292 days beyond winter solstice, i.e. (292 - 9) / 30 = 9 months 13 days beyond 'January 1', around 'September 13'.

The year would end (292 + 68 - 9) / 30 = 11 months 21 days beyond 'January 1', around 'December 21'. G could count with Gb8-30 being at winter solstice.

Next page (in the series from 'this link'):

112 in summer has Gb1-6 in its middle, but 112 in takaure has Gb8-30 at its end. We cannot count with 180 days between them even if Gb1-6 would stand at summer solstice, because we don't know what happens during those 112 days in takaure.

With Ga4-3 at spring equinox and Gb3-1 at autumn equinox, which we have concluded, then Gb1-6 will come later than summer solstice. The distance 206 days is greater than those ca 180 expected between the equinoxes.

(86 + 94 + 56 - 9) / 30 = (236 - 9) / 30 = 7 months and 17 days from 'January 1'. Gb1-6 lies in the second half of 'August'.

Gb1-6 is the opposite in every respect to Ga1-26:

236 - 27 = 209 = 180 + 29. Ga1-26 could be located in the middle of 'January', 7 months before Gb1-6. Evidently some problems remain to be solved.

What if those 112 glyphs in takaure are extracalendrical? 471 - 112 = 359. Then symmetry demands also those 112 glyphs in summer to be extracalendrical: 359 - 112 = 247 = 8 months and 7 days. 9 * 28 = 252 = 247 + 5.

Next page:

24 is a good period number for the situation when a great sun season ends. In the following 26 glyphs there may be 'turbulens' according to the wing in Ga5-14. In Ga6-15 the 'eating' maybe is at the back side (tu'a), in the past, and the bulging stomach could say 'satisfied'.