Click on image to view larger version.

Fig. 3. Atro, like ft, is required for PP and in the R3 cell. (A)
Section of a Atro35 mutant clone and (A') diagram of
PP. The clone is marked by the absence of pigment. White arrow indicates a
wild-type ommatidium, non-autonomously affected by the clone. The strong
photoreceptor specification defects make it impossible to assess the polarity
of most ommatidia inside the clone (yellow arrow). (B-B'') Close-ups of
three ommatidia from Atro35 clones with mosaic R3/R4 pair
and schematic representation of such mosaics (C). The pigment granules (PG;
marking the wild type) are always present on the R3 rhabdomere (black arrows)
and never on the R4 cell in mosaic ommatidia. (D) Section through the eye of
fly expressing Atro under sev control. Several ommatidia
appear to have lost their chirality and have adopted a symmetric form of the
R3/R3 type (green arrows). (E) Section through the eye of a viable combination
of two different Atro mutations
(Atro35/Atro11). Several ommatidia appear to
have lost their chirality and have adopted a symmetric form of the R4/R4 type
(blue arrows). Some ommatidia have extra inner photoreceptors (yellow
arrow).