Two successive mechanisms have been described in perichondral ossification: (1) in static osteogenesis, mesenchymal cells differentiate into stationary osteoblasts oriented randomly, which differentiate into osteocytes in the same site; (2) in dynamic osteogenesis, mesenchymal cells differentiate into osteoblasts that are all oriented in the same direction and move back as they secrete collagen fibers. This study is aimed at testing the hypothesis that the ontogenetic sequence static then dynamic osteogenesis observed in the chicken and in the rabbit is homologous and was acquired by the last common ancestor of amniotes or at a more inclusive node. For this we analyze the developmental patterns of Pleurodeles (Caudata, Amphibia) and those of the lizard Pogona (Squamata, Lepidosauria). We processed Pleurodeles larvae and Pogona embryos, prepared thin and ultrathin sections of appendicular bones, and analyzed them using light and transmission electron microscopy. We show that static osteogenesis does not precede dynamic osteogenesis in periosteal ossification of Pleurodeles and Pogona. Therefore, the null hypothesis is rejected and according to the parsimony method the ontogenetic sequence observed in the chicken and in the rabbit are convergent. In Pleurodeles and Pogona dynamic osteogenesis occur without a previous rigid mineralized framework, whereas in the chicken and in the rabbit dynamic osteogenesis seems to take place over a mineralized support whether bone (in perichondral ossification) or calcified cartilage (in endochondral ossification). Interestingly, in typical dynamic osteogenesis, osteoblasts show an axis (basal nucleus-distal endoplasmic reticulum) perpendicular to the front of secreted unmineralized bone matrix, whereas in Pleurodeles and Pogona this axis is parallel to the bone matrix.