Olfactory flow in the sea catfish, Ariopsis felis (L.): Origin, regulation, and resampling.

The olfactory epithelium of the sea catfish, Ariopsis felis, is found on a pinnate array of lamellae (the olfactory rosette) housed within a nasal chamber. The nasal anatomy of A. felis suggests an ability to capture external water currents. We prepared models from X-ray micro-computed tomography scans of two preserved specimens of A. felis. We then used dye visualisation and computational fluid dynamics to show that an external current induced a flow of water through a) the nasal chamber and b) the sensory channels of the olfactory rosette. The factors responsible for inducing flow through the nasal chamber are common to fishes from two other orders. The dye visualisation experiments, together with observations of sea catfishes in vivo, indicate that flow through the nasal chamber is regulated by a mobile nasal flap. The position of the nasal flap - elevated (significant flow) or depressed (reduced flow) - is controlled by the sea catfish's movements. Flow in the sensory channels of the olfactory rosette can pass through either a single channel or, via multiple pathways, up to four consecutive channels. Flow through consecutive sensory channels (olfactory resampling) is more extensive at lower Reynolds numbers (200 and 300, equivalent to swimming speeds of 0.5-1.0 total lengths s-1), coinciding with the mean swimming speed of the sea catfishes observed in vivo (0.6 total lengths s-1). Olfactory resampling may also occur, via a vortex, within single sensory channels. In conclusion, olfactory flow in the sea catfish is regulated and thoroughly sampled by novel mechanisms.

The functional nasal anatomy of the pike, Esox lucius L.

Olfactory flow in fishes is a little-explored area of fundamental and applied importance. We investigated olfactory flow in the pike, Esox lucius, because it has an apparently simple and rigid nasal region. We characterised olfactory flow by dye visualisation and computational fluid dynamics, using models derived from X-ray micro-computed tomography scans of two preserved specimens. An external current induced a flow of water through the nasal chamber at physiologically relevant Reynolds numbers (200-300). We attribute this externally-induced flow to: the location of the incurrent nostril in a region of high static pressure; the nasal bridge deflecting external flow into the nasal chamber; an excurrent nostril normal to external flow; and viscous entrainment. A vortex in the incurrent nostril may be instrumental in viscous entrainment. Flow was dispersed over the olfactory sensory surface when it impacted on the floor of the nasal chamber. Dispersal may be assisted by: the radial array of nasal folds; a complementary interaction between a posterior nasal fold and the ventral surface of the nasal bridge; and the incurrent vortex. The boundary layer could delay considerably (up to ~ 3 s) odorant transport from the external environment to the nasal region. The drag incurred by olfactory flow was almost the same as the drag incurred by models in which the nasal region had been replaced by a smooth surface. The boundary layer does not detach from the nasal region. We conclude that the nasal bridge and the incurrent vortex are pivotal to olfaction in the pike.

Olfactory flow in the sturgeon is externally driven.

Fluid dynamics plays an important part in olfaction. Using the complementary techniques of dye visualisation and computational fluid dynamics (CFD), we investigated the hydrodynamics of the nasal region of the sturgeon Huso dauricus. H. dauricus offers several experimental advantages, including a well-developed, well-supported, radial array (rosette) of visible-by-eye olfactory sensory channels. We represented these features in an anatomically accurate rigid model derived from an X-ray scan of the head of a preserved museum specimen. We validated the results from the CFD simulation by comparing them with data from the dye visualisation experiments. We found that flow through both the nasal chamber and, crucially, the sensory channels could be induced by an external flow (caused by swimming in vivo) at a physiologically relevant Reynolds number. Flow through the nasal chamber arises from the anatomical arrangement of the incurrent and excurrent nostrils, and is assisted by the broad, cartilage-supported, inner wall of the incurrent nostril. Flow through the sensory channels arises when relatively high speed flow passing through the incurrent nostril encounters the circular central support of the olfactory rosette, decelerates, and is dispersed amongst the sensory channels. Vortices within the olfactory flow may assist odorant transport to the sensory surfaces. We conclude that swimming alone is sufficient to drive olfactory flow in H. dauricus, and consider the implications of our results for the three other extant genera of sturgeons (Acipenser, Pseudoscaphirhynchus and Scaphirhynchus), and for other fishes with olfactory rosettes.