Nasal bones in mammals arise in connective tissue dorsal to the nasal capsule. These bones are noticeably reduced and modified in species whose skulls have undergone major shape transformations (e.g., Cetacea) or where a maneuverable proboscis has developed (e.g., Sirenia, Proboscidea, Mirounga, Tapirus, Nasalis; Stark, 1967; van der Klaauw, 1948-1952). Bats of the genus Chiroderma (5 species) seem to defy these generalizations as the nasal capsule in these bats is unroofed by bone, and a wide cleft extends from the frontal area to the tip of the nasal cartilage. Chiroderma seems a likely candidate to join the sirenian Dugong as the only mammals to lack nasal bones. The dorsal surface of the rostrum of Chiroderma (Fig. 1) was described by Dobson (1878:534) as having a "nasal aperture continued backwards by an oval slit, which reaches as far backwards as a line drawn between the most anterior parts of the bony orbit." He felt that this slit represented "a deep cleft between the nasal bones" that "probably becomes closed in some species when the animal is fully grown" (Dobson, 1878:531-532), suggesting the nasal bones grow medially with age. Unfortunately, this belief was based upon a taxonomic misunderstanding. Dobson accepted uncritically Peters' (1866) assignment of Phyllostoma pusillum Wagner to the genus Chiroderma. This species, with nasals that meet at the midline as in other bats, now is placed in the genus Vampyressa. Because Dobson examined only two other skulls of Chiroderma, Peters' misallocation led him to believe this rostral cleft closed with age. Examination of any reasonably large series of Chiroderma, though, would indicate that this cleft remains open even in old individuals. Perhaps because of his misunderstanding of the ontogeny of the nasal cleft, subsequent workers have not agreed with Dobson's view that nasal bones are present in Chiroderma. Miller (1907:157), for instance, described the nasals of Chiroderma as "apparently absent, their place occupied by an emargination." Most subsequent workers have followed Miller's interpretation; Goodwin and Greenhall (1961) go so far as to grant Chiroderma the common name "Bats without Nasal Bones." Those who have followed Miller's interpretation that nasal are absent in Chiroderma seem to have overlooked his uncertainty on the issue. In a footnote (1907:157), he wrote "I have seen no skulls sufficiently immature to show whether the nasals are absent or not." Because the cranial bones of bats co-ossify early in postnatal development, the sutures between bones are not visible in any but the youngest individuals. Without examining preor neonatal specimens, it is not possible to resolve the disagreement between Dobson's view that nasals are present in Chiroderma, and Miller's cautious view that they are absent. I have examined two mid-term fetuses of Chiroderma and find Miller's caution well taken: Chiroderma does possess nasal bones. The two fetal specimens I examined are from fluid-preserved female Chiroderma villosum taken in Costa Rica and Chiapas, Mexico, and stored in the collections of the University of Michigan Museum of Zoology (numbers 111474 and 114456, respectively). The fetuses were carefully skinned then cleared and stained for cartilage and bone using Wassersug's (1976) alcian-alizarian technique. For comparative purposes, I also examined a series of 24 fetuses of Sturnira lilium from Costa Rica, and 16 of Uroderma bilobatum from Panama. Both series were from fluid-preserved females in the UMMZ collection, and represent a range in relative age from embryos just beginning cranial ossification to near-term fetuses. These specimens were cleared and stained in the same way as were the fetal Chiroderma. Both Chiroderma fetuses are mid-term, as judged by vertebral and cranial development. All vertebrae are at least partly ossified in both centrum and neural arch, and the latter have begun to fuse at the dorsal midline. The cranial bones, except for the nasals, are well advanced in development. Although the main roofing bones do not meet at the midline, the frontals contact the parietals, which are fused with the squamosals, and the paired interparietals are in contact with the supraoccipitals. The deciduous teeth are erupted, and conical apices of permanent canines and molars are visible. Both the medial wall of the orbit and the auditory capsule are unossified. The ossifications I identify as nasals are situated between the rostral edge of the frontal and the caudal end of both maxilla and premaxilla. One individual (UMMZ 111474) has two small nasal ossifications on each side (Fig. 2E); the other (Fig. 2F) displays one large center of ossification almost connecting the frontal and maxilla on the right side and two nearly fused ossifications in February 1984 163