An electron microscope study has been made of the tail of the bandicoot spermatozoon (Perameles nasutaGeoffroi). The shape of the sperm head makes it possible to define dorsal, ventral, right and left surfaces on the head. The orientation of structures in the middle piece and tail relative to these head surfaces permits identification of the corresponding middle piece and tail surfaces. Dorso-ventral and transverse axes may be assigned to the spermatozoon. Major and minor axes are recognizable in transverse sections of the middle piece and tail. The tail contains two groups of fibrils, the axial filament complex and the peripheral fibrils. The axial filament complex consists of two central fibrils lying in the transverse axis of the head which are surrounded by an axial ring of nine fibrils not equidistant from each other. The axis of symmetry of the axial filament complex is in the dorso-ventral axis of the spermatozoon. The much thicker peripheral fibrils, of which there are nine, are found at varying but considerable distances from the axial ring of fibrils. They also are not equidistant from one another. Two of them differ from the other seven in shape and lie on the minor axis of the tail; four of the remaining seven lie on the dorsal half and three on the ventral half of the tail. Seven of the peripheral fibrils are connected with the homologous members of the axial filament complex by connecting laminae, which consist of laminar filaments passing between the two types of fibrils. One lamina has only one set of filaments. The other six have two sets of parallel filaments. The tail fibrils are enclosed by a spiral sheath which shows pronounced thickenings at both ends of the minor axis of the tail. Vacuoles, arranged in a characteristic way, are located within these thickenings. Thin processes connect the spiral sheath with two of the fibrils in the axial filament complex. A trilaminar membrane separates the spiral sheath from the tail sheath, which is an electron-transparent structure bounded by a single membrane. The possibility of establishing a functional relationship between the tail structures and sperm movement is discussed. The morphological evidence suggests that the tail beat is in the dorso-ventral plane.