|Select image to enlarge|
Geometry of leg development.
a. Mature, left 2nd-leg disc (~150 μm across ). Lines are folds. The central circle is the 'end knob'. Connections to larval skin (stalk) and CNS (nerve) have been cut.
b. Longitudinal section (BM and PM = basement and peripodial membranes; CE = columnar epithelium). Transverse lines are cell boundaries. The black cell is one of several larval neurons that innervate Keilin's organ [834, 2287]. These neurons invaginate with the disc (a piggybacking that is ancient in dipterans ). Their dendrites (not shown) stay attached to larval skin through the stalk , while their axons project to CNS (a route later taken by adult axons). Adepithelial cells are omitted.
c. Idealized fate map of left 2nd-leg disc. Segment primordia are the coxa (Co), trochanter (Tro), femur (Fem), tibia (Tib), basitarsus (Ba), and tarsal segments 2-5 (unmarked), with claws mapping centrally. The sternopleura (Stpl) is part of the body wall (cf. Fig. 4.4). AC and PC (shaded) are A and P compartments. Spokes denote bristle rows, whose spacing is estimated: it may not be so uniform (cf. e). As shown below for the tarsus, the annuli telescope out (toward the viewer) as the covering PM (not shown; cf. b) peels away . In this schematic the tarsus also rolls clockwise ~90°. Compass directions (D, V, A, P) refer to prospective or actual adult axes (cf. d, e).
d. Anterior face of left 2nd leg (~2 mm long). There are 2 macrochaetes (distal tibia) and numerous chemosensory (curved) bristles on the tibia and tarsus. EB (edge bristle) is at the D midline. Tarsal rows 5-8 are marked on compass below.
e. Basitarsus (whole surface) drawn as if slit along D midline and spread flat. Most bristles have a bract (triangle) and align in rows (numbers at top). Five chemosensory bristles lack bracts and reside between rows, as do 3 sensilla (circles at 3.5, 5.5, & 8.5). Between rows 1 and 8 is a lawn of hairs. Row-1 and -8 bristles are peg-shaped and darker; others are tapered and lighter. Aside from bristle thickness and pigmentation, other useful landmarks include bristle lengths and intervals -- both of which increase linearly from V to D (graph at top). Indeed, the symmetry is so precise  that it was surprising when the A/P compartments were found to be offset from this D-V mirror plane by a few cells ('gaps' below) [1800, 2449]. Row-1 bristles can reside in either AC or PC (cf. Fig. 3.9c).
Panels a and b were traced from  and  respectively (see [1774, 2144] for nerve-stalk asymmetry, [1516, 2287] for folding details, and  for region-specific cell shapes); c is adapted from [1807, 3807]; and e is based on data in [1801, 1803].
N.B.: For reasons explained in Fig. 4.4h legend, some authors orient the axes (c) with the D pole pointing into the stalk (at 12 o'clock vs. ~1 o'clock here). Claws (c) actually map more dorsally [1812, 3807, 4140]. Numbering of rows (e) follows original nomenclature . The numbers are backwards in [837, 1039, 2533, 4159]. Also, bristles do not really sprout from discs (e): they start to differentiate ~1 day after pupariation.