FISffES. 231 



nected with the brain by an elongate tractus olfactorms. The 

 optic thalamic are small, but the lobi inferiores on the ventral 

 surface of the twixt brain are well marked. The optic lobes 

 are strongly developed (only a single lobe occurs in Proto- 

 pterns'), and contain a large ventricle (epicoele). The optic 

 nerves cross in the teleosts ; a true chiasma is developed in 

 other groups (p. 61}. The eleventh and twelfth nerves are 

 lacking. 



The nasal organs only exceptionally have connection with the 

 cavity of the mouth. The eyes are without accessory glands ; 

 the lens is strongly convex, and is connected with the wall of the 

 posterior chamber by a structure, the falciform process, the distal 

 end of which is swollen into a bulb, the campanula Halleri, in 

 the walls of which muscles are developed so that this structure 

 plays a part in accommodation. There is no true Eustachian 

 tube in connection with the ear, and a stapes never occurs. 



The muscular system shows plainly its division into myo- 

 tomes, and there is no differentiation of layers of oblique muscles 

 in the abdominal walls. Epiaxial and hypaxial muscles (p. 109) 

 are well differentiated. 



The vertebral centres are either incompletely developed or 

 they are of the amphicoelous type ; the only exception being 

 found in Lepidosteus, where they are opisthoccelous, this con- 

 dition being brought about in the same way as in amphibia 

 (p. 139). Only trunk and caudal regions are differentiated. 

 Ribs, when present, are, except in elasmobranchs, modified 

 haemal arches. In Pofypterus both kinds of ribs occur (p. 144). 

 The skull is noticeable for the great development of the visceral 

 arches and their independence from the other parts. Almost 

 always the hyomandibular acts as a suspensor of the lower jaw. 

 The paired appendages are always fins, pectoral and ventral, but 

 occasionally one or the other of these may be degenerate. As 

 a rule these occupy positions, the pectoral at what would compare 

 to the shoulder, the ventral just in front of the anus ; but occa- 

 sionally the ventral fins may move forward to a position just 

 behind the pectorals (thoracic), or even in a line with these 

 (jugular). The pelvic girdle never enters' into connection with 

 the vertebral column ; the shoulder girdle rarely. The latter. 



