The pure glyph in Ka3-7 may be regarded from another angle. It has now been inferred that the calendar in K begins not around spring equinox (as earlier was thought) but presumably around winter solstice. Furthermore, Rei glyphs will then not appear at the equinoxes, where mago (in Ka4-14) respectively the 'humpback' (in Kb2-16) are located - Rei glyphs perhaps are connected only with the solstices. The 1st period in the calendar begins with a Rei glyph (appropriately with ordinal number 15):

1
	Ka3-15	Ka3-16	Ka3-17	Ka3-18	Ka3-19	Ka3-20	Ka3-21

26 (half 52) glyphs beyond mago (in Ka4-14) another Rei is found (12 periods beyond the 1st period) probably at the beginning of the 2nd half year:

13
	Kb1-7	Kb1-8	Kb1-9	Kb1-10	Kb1-11
	23	24	25	26	27
				destroyed glyphs follow here
	Kb1-12	Kb1-13	Kb1-14

Most interesting, though, is yet another Rei located 26 glyphs before the one in Ka3-15:

Ka2-1	Ka2-2	Ka2-3	Ka2-4	Ka2-5	Ka2-6	Ka2-7	Ka2-8
36	35	34	33	32	31	30	29
						etc
Ka2-9	Ka2-10	Ka2-11	Ka2-12	Ka2-13	Ka2-14
28	27	26	25	24	23

The implication is that Ka2-10 is initiating a sort of calendar characterized by another type of end glyph for each period, three of which are:

Ka3-1	Ka3-2	Ka3-3	Ka3-4
Ka3-5	Ka3-6	Ka3-7	Ka3-8
Ka3-9	Ka3-10	Ka3-11	Ka3-12

Ka3-8 has no hand sign, which possibly should be interpreted to mean that there is no person present - the ghost has gone to Hiva, and the pure elbow ornament also says so.

The newly found calendar, with 36 glyphs, probably has this type of end glyph because sun is absent and we now have to again revise the model for the calendar which begins with Ka4-15. Rei glyphs may, after all, be also at the equinoxes.