Fast and singular muscle responses initiate the startle response of Pantodon buchholzi (Osteoglossomorpha).

The startle response of Pantodon buchholzi, the African butterfly fish, is a complete or incomplete ballistic jump resulting from abduction of the pectoral fins. This study analyzed the neuromuscular basis for such a jump by recording in vivo electromyograms (emgs) from the muscles of abduction, the muscularis abductor superficialis (MAS) and the muscularis abductor profundus (MAP). The motor neurons innervating the MAS muscle were localized by retrograde transport of biocytin. The latency between stimulus and the evoked emg in the MAS was less than 5 ms; the latency of the MAP was about 6.5 ms. A single emg was recorded per jump. High speed video demonstrated that onset of a startle movement began within 10 ms of the onset of fin abduction. The emg associated with this movement is short (<2 ms) and followed by a variably-shaped, slower and smaller potential of 10-30 ms duration. The brief period between stimulus and startle response of Pantodon suggests a Mauthner neuron-related response, only with the behavior occurring in the vertical plane. The MAS may act only in a startle response, whereas the MAP might have a role in other behaviors. Elicited jumping habituates after a single trial. Electrophysiological evidence is presented indicating that the innervating motor neurons are suppressed for seconds following a stimulus. The neurons innervating the MAS are located at the medullary-spinal cord junction and possess an average radius of approximately 17.9 mum. These fish have been historically described as 'fresh water' flying fish. As a single emg occurs per startle response, repetitive pectoral activity generating flying cannot be supported. Pantodon 'flight' is ballistic.

Identification of visual pallial telencephalon in the goldfish, Carassius auratus: a combined cytochrome oxidase and electrophysiological study.

A strategy based upon a comparative decrease in bilateral symmetry of cytochrome oxidase (COX) histochemistry following unilateral eye enucleation was used to identify the primary visual area in the area dorsalis of the telencephalon of the goldfish, Carassius auratus. The lateral zone of area dorsalis (Dl) at about the level of the anterior commissure exhibits such a bilateral difference. A parallel decline in the symmetry COX reactivity was observed in the associated part of the central zone (Dc). Electrophysiological activity using extracellular techniques confirmed the visually-driven activity of neurons in these areas. Lesions confirmed the loci in the lateral zone of area dorsalis, including both its dorsal and ventral parts. Single- and multi-unit recordings exhibited a variety of responses to different light stimuli. Single unit latency measures proved not to be a reliable measure of target areas. Responses habituated to stimuli repeated within 5 s and were only reliably evoked with intervals greater than several seconds.

Coherence in nervous system design: the visual system of Pantodon buchholzi.

One of the more unusual visual systems of the Actinopterygii is that of Pantodon buchholzi (Osteoglossomorpha: Osteoglossidae). Its adaptations associate neuroanatomy at different levels of the visual system with ecological and behavioural correlates and demonstrate that the visual system of this fish has adapted for simultaneous vision in air and water. The visual field is divided into three distinct areas: for viewing into the water column, into air, and for viewing the aquatic reflection from the underside of the water surface. Cone diameters in different retinal areas correlate with the differing physical constraints in the respective visual field. Retinal differentiation between the aquatic and aerial views is paralleled at different levels of the central nervous system. A diencephalic nucleus receives both direct and indirect (tectal) afferent input from only the aerial visual system and a specific type of cell in the optic tectum is preferentially distributed in the tectum processing aerial inputs. Distinctions within a single sensory system suggest that some behaviours may be organized according to visual field. For Pantodon, feeding is initiated by stimuli seen by the ventral hemiretina so the anatomical specializations may well play an important role as elements in a feeding circuit.

Morphologic characterization of the ice worm Mesenchytraeus solifugus.

Ice worms occupy a unique position in metazoan phylogeny in that they are the only known annelid that completes its life cycle in ice. The mechanism(s) associated with this adaptation are likely to occur at different levels, ranging from modification of their metabolism to changes in morphology. In this study, we examined specimens of Mesenchytraeus solifugus by scanning electron microscopy (SEM) and transmission electron microscopy (TEM) in an effort to identify morphologic structures that may aid in its glacial habitation. We report that M. solifugus contains an elongated head pore at the tip of its prostomium, numerous sensory structures, and differentially oriented setae that curve abruptly at their distal end.

Defining sameness: historical, biological, and generative homology.

Current debate concerning homology arises from three different research interests-phylogenetics, character evolution, and generative pathways. Phylogenetic homology focuses on descent of the character from a common ancestor. Biological homology addresses character evolution and diversification. Exceptions to the general case complicate these two approaches: historically and biologically homologous characters may be produced by different generative pathways, and minutely similar characters produced by the same generative pathways may have a sporadic phylogenetic distribution. We suggest that for studies of comparative developmental biology, new descriptive terms are needed to distinguish similar structures that result from the same generative pathways from those that result from different generative pathways. The terms syngeny, meaning "same genesis", and allogeny, meaning "different genesis", allow the acknowledgement of sameness at the generative level and can be used in combination with the terminology of historical homology and biological homology to describe any given character.

Optomotor response of Anableps anableps depends on the field of view.

Anableps anableps (Cyprinodontiformes) inhabits the niche at the water surface such that its cornea is bisected by the water surface. Consequently, its visual field encompasses simultaneous views into air and water by ventral and dorsal retina, respectively. The optomotor response (OPM) of Anableps was elicited by a moving stimulus pattern in either one or the other environment. Using four related visual displays, we found that this fish exhibits a classical OPM response when presented with suprathreshold flow-fields in its aerial visual field. It lacks an OPM response to the same flow-field when presented in its aquatic visual field, although it may respond by exhibiting optokinetic nystagmus (OKN) and non-OPM motor activity. We conclude that the neurological circuit for the teleost OPM in Anableps operates only for the aerial view and is probably connected to only the ventral retina.

Expression of cytochrome oxidase in hair cells of the teleost utricle.

Ultrastructural variation in some cytoplasmic organelles and synaptic structures is one characteristic distinguishing the types of hair cells in the teleost ear. In this study, we explored differences in mitochondria by analyzing mitochondrial reactivity for cytochrome oxidase (COX) in hair cells of the teleost utricle. The reactivity for COX within mitochondria in the subcuticular compartment directly beneath the cuticular plate differentiated among hair cells in utricles of three teleost species, Carassius auratus, Pantodon buchholzi, and Astronotus ocellatus. Mitochondria in the subcuticular region of hair cells in the striola reacted intensely. Within juxtastriola and extrastriolar hair cells near the striola, mitochondria reacted at a lowered intensity than in striolar hair cells. Subcuticular mitochondria of extrastriolar hair cells located distant from the striola reacted negligibly. The reactivity of mitochondria in other cytoplasmic compartments did not provide similar evidence for distinguishing among teleost hair cells. Mitochondria within intraepithelial branches of the eighth nerve terminals in the different utricular regions reacted to COX histochemistry commensurate with their respective presynaptic hair cells. Branches of sensory afferent neurons innervating striolar hair cells displayed a dense COX reaction. Sensory afferents innervating the extrastriolar hair cells did not display many mitochondria at synapses nor, when present, was the staining as dense. The presynaptic side of the hair cell-afferent nerve synapse usually, but not always, contained reactive mitochondria. The presynaptic side of the efferent nerve-hair cell synapse did not necessarily contain mitochondria. Mitochondria filling the cytoplasm in a type of juxtamacula cell revealed uniformly dense COX reactivity.

An atypical diencephalic nucleus in actinopterygian fishes: visual connections and sporadic phylogenetic distribution.

Nucleus rostrolateralis is a distinctive diencephalic nucleus in some ray-finned fishes. In the osteoglossomorph Pantodon, it is a large ovoid nucleus with visual system connections. A topologically and cytoarchitectonically similar nucleus has been found in six species of non-osteoglossomorph fishes, two species of gars and four euteleosts. Of the latter, two are ostariophysans of the genus Danio, one an atherinomorph, and one a notothenioid percomorph. A variety of characteristic similarities can be found in nucleus rostrolateralis among all these species. The present study reports on these similarities in Danio and in the euteleost Xiphophorus as compared with the nucleus in Pantodon. While this nucleus has a phylogenetic distribution that might imply convergent evolution, the high degree of similarity in its features across species strongly suggests that its genetic basis may be the same despite the lack of phenotypic homology.

Visual connections of the atypical diencephalic nucleus rostrolateralis in Pantodon buchholzi (Teleostei, Osteoglossomorpha).

Among the many thousands of teleost fish species, a diencephalic nucleus rostrolateralis (RL) has been identified in only six widely divergent species. In one of these, Pantodon buchholzi, its retina projects to both RL and the optic tectum, the latter in a visuotopic manner. The ventral part of the retina beneath a horizontal black-pigmented septum connects to the dorsomedial optic tectum and nucleus RL. The dorsal part of the retina connects to the lateroventral optic tectum and little, if at all, to RL. Using DiI tracing, the position of RL in the optic pathway of this fish has been directly demonstrated. Cells in the stratum periventriculare of the dorsomedial optic tectum contribute to the afferent input of RL (bilaterally); cells in the ventrolateral tectum do not. RL is also reciprocally connected with the torus longitudinalis and may project to three nuclei of the preglomerular complex. Ganglion cells in the retina that project to RL are sparsely distributed throughout the ventral hemiretina compared with ganglion cells that project to the optic tectum. Since this fish is an obligatory surface feeder, the neuroanatomical connectivity of nucleus RL in P. buchholzi suggests a role in the fish's visual identification of targets for feeding behavior.

Hair cell heterogeneity in the goldfish saccule.

A set of cytological studies performed in the utricle and saccule of Astronotus ocellatus (Teleostei, Percomorphi, Cichlidae) identified two basic types of hair cells and others with some intermediate characteristics. This paper reports on applying the same techniques to the saccule of Carassius auratus (Teleostei, Otophysi, Cyprinidae) and demonstrates similar types of hair cells to those found in Astronotus. Since Carassius and Astronotus are species of extreme taxonomic distance within the Euteleostei, two classes of mechanoreceptive hair cells are likely to represent the primitive condition for sensory receptors in the euteleost inner ear and perhaps in all bony fish ears.

Two types of sensory hair cell in the saccule of a teleost fish.

Previous investigations have demonstrated significant ultrastructural differences in hair cells located in various regions of the utricle of the oscar, Astronotus ocellatus. In this study, we used TEM and SEM to examine cells from the central and marginal regions of the saccule to determine if similar hair cell ultrastructural differences exist in this endorgan. Based upon ultrastructural characteristics, central saccular cells closely resemble utricular striolar cells while marginal cells are intermediate in ultrastructure between striolar and extrastriolar cells of the utricle.

Tectal projection to an unusual nucleus in the diencephalon of a teleost fish, Pantodon buchholzi.

Nucleus rostrolateralis, a newly identified nucleus, has been found to date in only three species of ray-finned fishes, two of which are osteoglossomorphs. It is relatively large and well developed in only one of the osteoglossomorphs, Pantodon buchholzi, in which it receives a relatively sparse, primarily contralateral, input from the retina. The present report describes a relatively intense, bilateral projection from the optic tectum to nucleus rostrolateralis.

Heterogeneity of sensory hair cells in a fish ear.

The ultrastructure of sensory hair cells in the utricle of the cichlid fish, Astronotus ocellatus, the oscar, was studied by transmission electron microscopy of serial ultrathin sections from different regions of the epithelium. Two distinctly different types of hair cell were found, one located in the striolar region of the epithelium and the other in the extrastriolar region. Striolar hair cells have a well-defined perinuclear cisterna located just below the nucleus, and large perinuclear mitochondria. Synaptic bodies of striolar cells are small and located in clusters, while those in extrastriolar cells are relatively large and individually dispersed. The extrastriolar hair cell closely resembles the amniote type II hair cell. On the basis of these data, and consistent with earlier studies, it appears that the striolar hair cells closely resemble amniote type I hair cells in many significant ways. Thus we have called them type I-like cells. The extrastriolar hair cells appear to be typical of eighth nerve mechanoreceptors commonly described for fish and closely resemble the amniote type II hair cell.

Sensory hair cells of a fish ear: evidence of multiple types based on ototoxicity sensitivity.

Sensory hair cells from the striolar region (striolar hair cells) of the utricle and the lagena of the ear of a teleost fish Astronotus ocellatus (Cuvier) ear are sensitive to gentamicin sulphate, an ototoxic drug. In contrast, sensory hair cells from outside the striolar region (extra-striolar hair cells) are not sensitive to gentamicin. These data, combined with results from studies showing different ultrastructural features and different immunoreactivity to a calcium binding protein, S-100, lead to the suggestion that there are distinguishable types of hair cells in these endorgans. These results add to the increasing evidence that classifying the sensory hair cells of fish ears only as the traditional 'vestibular type II' may be inadequate for properly understanding structure and function of the fish ear.

Retinal projections in the freshwater butterfly fish, Pantodon buchholzi (Osteoglossoidei). I. Cytoarchitectonic analysis and primary visual pathways.

The freshwater butterfly fish, Pantodon buchholzi, is a member of the most primitive radiation of teleosts. The retinofugal projections were studied in this fish with autoradiographic and horseradish peroxidase (HRP) methods, and the cytoarchitecture of the retinorecipient regions in the diencephalon and pretectum was analyzed with Bodian-, cresylecht-violet- and acetylcholinesterase-reacted sections. The rostral diencephalon of Pantodon contains a large retinorecipient nucleus, not previously identified in any other fish, i.e. nucleus rostrolateralis. Other nuclei that are described correspond to those previously recognized in other species. The majority of retinorecipient nuclei are positive for acetylcholinesterase, particularly those in the pretectum, as has been found in other species of teleosts. Most of the retinofugal fibers decussate in the optic chiasm. Some fibers project via the axial optic tract to preoptic nuclei and a region in the rostral hypothalamus. Fibers leave the medial optic tract to terminate in nucleus rostrolateralis and in dorsal and ventral thalamic nuclei, accessory optic and tubercular nuclei, periventricular and central pretectal nuclei, and sparsely in the deep tectal fascicle and terminal field. Dorsal optic tract fibers project to the dorsal accessory optic nucleus, superficial and central pretectal nuclei, and superficial and deep tectal layers. Ventral optic tract fibers project to the superficial pretectum, accessory optic nuclei, posterior tuberculum, nucleus corticalis in the central pretectum, and superficial tectal layer. Fibers that remain in the ipsilateral optic tract project to most of the targets reached by contralaterally projecting fibers. A few fibers in the contralateral medial optic tract redecussate via the posterior commissure to reach the ipsilateral periventricular pretectum. No labeled retinopetal cells caudal to the olfactory bulb were identified in any of the HRP cases.

Retinal projections in the freshwater butterfly fish, Pantodon buchholzi (Osteoglossoidei). II. Differential projections of the dorsal and ventral hemiretinas.

Pantodon buchholzi, the freshwater butterfly fish, is a member of the Osteoglossomorpha, the most primitive of the four major teleost radiations. The projections of fibers originating in the dorsal and ventral hemiretinas in Pantodon, as determined with autoradiography, are reported here. Fibers originating in the ventral hemiretina reach their targets through the axial, medial and dorsal optic tracts. Fibers that originate in the dorsal hemiretina reach their points of termination by way of the axial, medial and ventral optic tracts. Projections of the various tracts to preoptic, thalamic, tubercular, pretectal and tectal regions, as described in the previous study of total retinal projections, were verified. The retinal projections to the preoptic, thalamic and tubercular nuclei do not map topographically. Ventral hemiretinal fibers are mapped, however, onto the dorsal part of the nucleus pretectalis superficialis pars parvocellularis, the rostral part of the dorsal accessory optic nucleus, the entire nucleus pretectalis periventricularis pars ventralis and the dorsomedial portion of the optic tectum. Ventral hemiretinal fibers also supply most if not all the retinal innervation to the central pretectal nucleus. In contrast, dorsal hemiretinal fibers are mapped onto the ventral part of nucleus pretectalis superficialis pars parvocellularis, the entire dorsal accessory optic nucleus and the ventrolateral portion of the optic tectum. The dorsal and ventral hemiretinal projections to the tectum about at a cytoarchitectonically recognizable point, indicating that no discontinuity is present in the retinal connectivity with the tectum. The pars parvocellularis of nucleus pretectalis superficialis is a simple, unfolded, and nonlaminar structure in Pantodon. This structure contrasts markedly with the more complex, folded structure of the nucleus in the majority of other examined teleosts. The orientation of the projections from the dorsal and ventral hemiretinas onto this nucleus in Pantodon is congruent with that seen in other fishes only after a schematic unfolding of the nucleus in these fishes.

Spatial and morphological differentiation of trigger zones in afferent fibers to the teleost utricle.

A morphological correlate of the trigger site (the locus of action potential initiation) was identified in afferent axons of the utricle in the ear of two species of teleost fish. These sites were identified by the ferric-ferrocyanide (Prussian blue) cytochemical procedure and they were correlated with the geometries of afferent intraepithelial arbors as visualized by means of a silver stain. The intraepithelial arbors of afferent fibers show regional distributions that correlate with axon diameter and Prussian blue staining. Afferent axons with diameters greater than 4-5 microns only innervate the striola regions of the epithelium and terminate as one of two distinct types of intraepithelial arbors. Afferent axons with diameters smaller than 4 microns are ubiquitously distributed throughout the epithelium. Arbors that stained by Prussian blue within the utricular epithelium are restricted to the striolar regions. These arbors possess nodal-like membrane in different branches as postsynaptic membrane. Afferents that innervate hair cells in the extrastriolar epithelial regions stained with Prussian blue only at the extraepithelial terminal heminode. The postsynaptic membrane of these afferents is passive or dendritic-like.

S-100 immunoreactivity identifies a subset of hair cells in the utricle and saccule of a fish.

Certain hair cells of fish exhibit strong immunoreactivity to an S-100 antibody. By their spatial locations in the utricle and saccule, these hair cells appear to possess a relatively short kinocilium and a roughly ovoid cell shape. In the utricle, these cells are predominantly located in the striola. In the saccule, these cells are found within the central area of the epithelium. In both of these epithelia the strongly immunoreactive hair cells coincide with the locations of hair cells possessing F1 ciliary bundles.

Variations in receptor cell innervation in the saccule of a teleost fish ear.

Using transmission electron microscopy, we quantitatively analyzed the afferent and efferent synapses on 67 sensory hair cells along the saccular epithelium of the oscar (Astronotus ocellatus), a cichild fish with a non-specialized ear. The synaptic profile (number of afferent and efferent synapses per cell) varied considerably among cells. The number of synapses per hair cell ranged from three to 24, and all but six of the 67 hair cells had both afferent and efferent synapses. Statistical analysis showed that the synaptic profiles did not significantly vary anywhere on the saccular epithelium except at the edges. There, hair cells had significantly fewer efferent synapses than hair cells in other epithelial regions. This statistical variation in efferent synapse distribution in different epithelial regions corresponds with the lengths of ciliary bundles in these regions. The synapses on hair cells showed a regional specificity in position. In all cells synapses were never located more apically than the top of the nucleus. On hair cells towards the periphery, the most apical synapse on the hair cells tended to be afferent. On more centrally located cells, the most apical synapse was efferent in 92% of the cells.