The glyph type with features from both maitaki (left) and marama, (right), e.g. Eb7-38, may be - I have stated - positively correlated with the presence of tara glyphs. In the text of E (where my idea originated) it is easy to see that the statement probably is true. The only two instances are Eb7-38 and Eb8-29:
 
Eb7-36 Eb7-37 Eb7-38
Eb8-13 Eb8-14 Eb8-15 Eb8-16 Eb8-17 Eb8-18 Eb8-19
Eb8-20 Eb8-21 Eb8-22 Eb8-23 Eb8-24 Eb8-25 Eb8-26
Eb8-27 Eb8-28 Eb8-29 Eb8-30 Eb8-31 Eb8-32 Eb8-33
Eb8-34 Eb8-35 Eb8-36 Eb8-37 Eb8-38 Eb8-39 Eb8-40