PubMed1993
The first birds probably evolved from a line of theropod dinosaurs in the late Jurassic or early Cretaceous. The "trees down" theory proposes that avian ancestors were arboreal, whereas the "ground-up" (cursorial) theory suggests that they were terrestrial, and ran and jumped for prey. We present suggestive evidence from reptilian and avian female reproductive biology that supports the arboreal theory, although neontological evidence can never authenticate a paleontological event. The "law of follicular constancy" is an empirical observation that the number of ovulations per female per ovulatory cycle (instantaneous fecundity, or IF) is the same regardless of the amount of ovarian tissue present. In vertebrates with an IF of two or more, surgical removal of one of the paired ovaries (unilateral ovariectomy, or ULO) leads to a doubling of ovulations from the remaining ovary (compensatory ovarian hypertrophy, or COH), this ovary cycling at the same frequency as it did before surgery. In vertebrates that produce one egg alternately from each ovary (an IF of one), however, ULO leads to a form of COH in which the remaining ovary still ovulates one egg at a time but twice as frequently. In most birds, only the left ovary is present; it ovulates a single egg every 1-2 days until the species-specific clutch size is reached. Inasmuch as this avian ovulatory pattern is similar to an accelerated version of that occurring after ULO in a species that alternates ovulation, we propose that birds evolved from dinosaurs with such a pattern. A great majority of extant reptiles with an IF of one are either anoline or gekkonid lizards, and many of these ovulate several times a year. Furthermore, most species in these tropical groups are arboreal. Even considering phylogenetic constraints within anoline and gekkonid lizards, we propose that correlations of arboreality and an IF of one in these groups are implied as adaptive relationships and represent ecological parallelisms. Therefore, we propose that the ancestors of birds were arboreal. Furthermore, they probably were of small size, as are all lizards with an IF of one.