During the years 1952-53 the author has examined systematically specimens of the ichthyologic fauna from the southern coast of the State of S.Paulo (Brazil), gathering material and data in order to make a revision of some fish families found in that region. Studying the family Atherinidae in the collection n. 7, of 6-18-52, from Cananéia, the author came across with a specimen of the genus XenomeIani ris, showing an evident deformation of the vertebral spine (Photo n. 1). The Instituto Oceanográfico not being equipped with a small X-ray apparatus that would allow to make the proper documentation to illustrate the finding the author took advantage of the Hollister technique of diaphanization. In this way the first observations were made by transparence. Following, as the diaphanized tissues would still impede a perfect visibility and the observation of small details was found to be impossible, a small portion of tissue that covered the vertebral spine was removed without injuring the piece, in order to facilitate the study. In Fig. 1, the author tried to give an ideia of the more deformed portion. It is composed of 6 vertebrae in which, together with small curvatures, deep morphological alterations of the hemal and neural spines can be observed. The 23 first vertebrae of the specimen are normal. The other suffered well pronounced deformations that the author shows in detail. The case seems identical to the one studied by Plehn (1924) and Steinböck (1928). However no data could be obtained that would enable to explain satisfactorily the causes of such deformations, and for this reason, the author limits himself to present a list of probable causes. The major part of the authors that have studied the subject attribute anomalies of this type to alterations of the muscular system and, in fact, this seems the most acceptable explanation. Nevertheless the author supposes that other causes have interfered in the evolutive process of the specimen, specially traumatisms suffered in the larval or post-larval period, being of such nature that were capable of altering the myodynamics of the metameric musculature, without affecting, however, the development of the specimen.