The morphology and structure of the olfactory organ of Cynoglossus semilaevis Gunther are described. The oval olfactory sacs on both sides differ in size and in the number of lamellae, With those on the abocular side having smaller sacs and fewer lamellae than those on the ocular side. On the ocular side, the average ratio of sac length to eye diameter is 2.1 （i.e.〉1） with an average of 91 lamellae, while on the abocular side, the values were 1.7 （i.e.〉1） and 69, respectively. In addition, the surface morphology varies in different parts of the lamella. The frontal part, near the anterior nostril, is a non-sensory margin with cilia-free epidermal cells. Within this is an internal ciliated sensory area, which is intercalated with ciliated receptor cells and a few ciliated non-sensory cells. Additionally, some dense ciliated non-sensory cells make up a non-sensory area, which also contains cilia-free epidermal cells distributed in patches. In the rear of the olfactory sac near the posterior nostril, the lamellae differ in morphology from those of the frontal olfactory sac but are similar in having few ciliated receptor cells. In other words, the surface of the lamellae in the rear part of the olfactory sac is mainly non-sensory. At present, four types of lamellae （~ E IlIand IV） have been recognized in relation to the pattern of the sensory epithelium. In this study, the frontal and rear lamellae resembled types I and IV, respectively, but are referred to as types r and IV because they are slightly less developed. Data on the ratio of length of lamellae to eye diameter, number of lamellae and the type of surface pattern of the lamellae show that the development of the olfactory system of C. semilaevis facilitates prey capture.
We conducted behavior experiments on the roles of vision, olfaction, and the lateral line in feeding in tongue sole Cynoglossus semilaevis Gtinter. Chemical cues from the prey alone could not elicit any searching or attacking behavior by the fish; only when moving foods were presented, the chemical nature of the foods affected feeding. In addition, we found that normal tongue sole did not accurately recognize moving artificial prey of different shapes. Neither chemosense nor vision plays an important role in prey detection. Reactions to moving objects and the response of the oropharyngeal cavity to different foods showed that the （eyeless） tongue sole use mainly the lateral line for prey detection and capture. Gustation in the oropharyngeal cavity is essential for the fish to swallow its prey.
