Fig. 3. white, brown and scarlet mutants have organelle counts that mostly do not differ from wild-type controls. All values are mean ± s.d. (N=4 flies). (A,B) Sizes of R1–R6 profiles. Relative to wild-type controls, no differences were detected in profile sizes, either in their cross-sectional area (µm²; A) or membrane perimeter (µm; B). (C,D) The number of synapse profiles per micrometer of membrane perimeter, counted as either tetrad (C) or feedback (D). (E) Number of synaptic vesicle profiles per R1–R6 profile. These are significantly higher in Oregon-R wild-type than all mutant R1–R6 (^*P<0.03, t-test). (F–H) Numbers of capitate projections (CP) in the same samples as C–E, seen as three profiles: single penetrating (F), shallow (G) and multiple headed (H). (For definition of types, see Materials and methods.) No significant differences were seen except in the number of multiple-headed penetrating invaginations, which was greater in the wild-type control than in either white or scarlet heads (^*P<0.05, t-test, in H). Some counts showed wide variation, which we attribute to our methods for sampling organelle profiles at relatively low frequencies. By chance, the variation appeared larger in wild-type than mutant values (D,G,H).