Hb1-222 and Pb10-2:

As a rule the left part of GD36 is thicker than the right part, while the right part is ending in a point. This assymmetry makes it possible to use its mirror image as a sign, perhaps for of looking backwards.

In some texts the difference between the left and right parts are not expressed by the left part being thicker, e.g. (Hb1-222 respectively Pb10-2):

But the right part clearly anyhow is the end part of the bent bough. In these two examples we can see this by the general appearance of the glyphs; the point is at right - even though in the first example also the left part is pointed.

Bb9-19 and Ba9-30:

If GD36 is inverted, will the pointed end still be at right? It could be mirrored in two possible ways: By placing the mirror vertically, then first mirror the glyph and after that rotate the mirrored glyph 180^o. Or the inversion could be done just by placing the mirror horizontally, getting the inverted glyph directly.

In the first instance the vertical mirror will shift the point of the branch from right to left, and then by rotating the mirrored glyph the point of the branch will return to the right, though now upside down. I.e. the same result as if we simply used a horizontal mirror; the point remains at right.

Examples of inverted glyphs indicate that there might be two variants: The point of the branch at right and the point of the branch at left (Bb9-19 respectively Ba9-30):

The point still at right can be explained as the result of a horizontally placed mirror, the point shifted towards the left can be explained as the result of simply rotating the glyph 180o around. The latter variant seems to be a more natural way of handling a glyph to be inverted.

As to the meaning of the sign of inversion in GD36, we may possibly think of the fundamental female trait of being receptive, a vessel. To invert the female bough would then become its opposite, 'closed'. Perhaps it would also be possible to use as a sign of 'empty'.

In Bb9-19 maybe the meaning is 'empty':

and then Bb9-20 could signify 'full'.

In Ba9-30 we should instead note the 'broken' character and interpret the glyph from that standpoint.