Within jawed vertebrates, pelvic appendages have been modified or lost repeatedly, including in the most phylogenetically basal, extinct, antiarch placoderms. One Early Devonian basal antiarch, , possessed pelvic girdles, suggesting the presence of pelvic appendages at the origin of jawed vertebrates; their absence in more derived antiarchs implies a secondary loss. Recently, paired female genital plates were identified in the Late Devonian antiarch,, in the position of pelvic girdles in other placoderms. We studied these putative genital plates along an ontogenetic series of ; ontogenetic changes in their morphology, histology and elemental composition suggest they represent endoskeletal pelvic girdles composed of perichondral and endochondral bone. We suggest that pelvic fins of derived antiarchs were lost, while pelvic girdles were retained, but reduced, relative to This indicates developmental plasticity and evolutionary lability in pelvic appendages, shortly after these elements evolved at the origin of jawed vertebrates.

During fish growth, short periods of rapid changes (morphological, behavioral, physiological, ecological), or thresholds, are interspersed with longer periods of slower development, the steps. This growth pattern is known as saltatory ontogeny. Thresholds delimit main developmental stages (embryo, larva, juvenile, adult) and are periods where modifications can lead to new life-histories, and ultimately to evolutionary novelties. We sought to determine whether saltatory ontogeny could be recognized in a jawed stem-gnathostome, the Late Devonian antiarch placoderm Bothriolepis canadensis, using an extensive size series (220 specimens: 5–220 mm in armor length). The small specimens of this series reveal a previously undocumented immature feature, a preorbital depression on the premedian and lateral plates of the dermal headshield. This depression is a plesiomorphic condition reputed to be absent in highly nested antiarchs such as B. canadensis, which possess instead a preorbital recess. Our objectives were: to describe the ontogenetic morphological changes of the preorbital area in B. canadensis, including the timing of disappearance of the depression using binomial logistic regressions, and to quantify growth allometries of the premedian and lateral plates, with linear and segmented regressions. We found significant segmented allometric patterns in the premedian plate, suggesting saltatory ontogeny in B. canadensis. Moreover, segmented patterns were congruent with the ontogenetic loss of the preorbital depression. An ecomorphological hypothesis is proposed to explain these simultaneous morphological and morphometric changes in B. canadensis. A similar hypothesis is extrapolated to interpret the innovation of the preorbital recess and loss of the preorbital depression during the antiarch phylogeny.

Morphological and developmental similarities, and interactions among developing structures are interpreted as evidences of modularity. Such similarities exist between the dorsal and anal fins of living actinopterygians, on the anteroposterior axis: (1) both fins differentiate in the same direction [dorsal and anal fin patterning module (DAFPM)], and (2) radials and lepidotrichia differentiate in the same direction [endoskeleton and exoskeleton module (EEM)]. To infer the evolution of these common developmental patternings among osteichthyans, we address (1) the complete description and quantification of the DAFPM and EEM in a living actinopterygian (the rainbow trout Oncorhynchus mykiss) and (2) the presence of these modules in fossil osteichthyans (coelacanths, lungfishes, porolepiforms and ‘osteolepiforms’). In Oncorhynchus, sequences of skeletal elements are determined based on (1) apparition (radials and lepidotrichia), (2) chondrification (radials), (3) ossification (radials and lepidotrichia), and (4) segmentation plus bifurcation (lepidotrichia). Correlations are then explored between sequences. In fossil osteichthyans, sequences are determined based on (1) ossification (radials and lepidotrichia), (2) segmentation, and (3) bifurcation of lepidotrichia. Segmentation and bifurcation patterns were found crucial for comparisons between extant and extinct osteichthyan taxa. Our data suggest that the EEM is plesiomorphic at least for actinopterygians, and the DAFPM is plesiomorphic for osteichthyans, with homoplastic dissociation. Finally, recurrent patterns suggest the presence of a Lepidotrichia Patterning Module (LPM).

Morphological and developmental similarities, and interactions among developing structures are interpreted as evidences of modularity. Such similarities exist between the dorsal and anal fins of living actinopterygians: (1) both fins differentiate in the same direction [dorsal and anal fin patterning module (DAFPM)], and (2) radials and lepidotrichia differentiate in the same direction [endoskeleton and exoskeleton module (EEM)]. To infer the evolution of these common developmental patternings among osteichthyans, we address (1) the complete description and quantification of the DAFPM and EEM in a living actinopterygian (the rainbow trout Oncorhynchus mykiss) and (2) the presence of these modules in fossil osteichthyans (coelacanths, lungfishes, porolepiforms and ‘osteolepiforms’). In Oncorhynchus, sequences of skeletal elements are determined based on (1) apparition (radials and lepidotrichia), (2) chondrification (radials), (3) ossification (radials and lepidotrichia), and (4) segmentation plus bifurcation (lepidotrichia). Correlations are then explored between sequences. In fossil osteichthyans, sequences are determined based on (1) ossification (radials and lepidotrichia), (2) segmentation, and (3) bifurcation of lepidotrichia. Segmentation and bifurcation patterns were found crucial for comparisons between living and extinct taxa. Our data suggest that the EEM is plesiomorphic at least for actinopterygians, and the DAFPM is plesiomorphic for osteichthyans, with homoplastic dissociation. Finally, recurrent patterns suggest the presence of a Lepidotrichia Patterning Module (LPM).