My interest in brain evolution of fish

Teleost brain evolution



This research was conducted by Robert Huber and Moira J. van Staaden in collaboration with Les S. Kaufman (Boston Univ.), and Karel F. Liem (Harvard Univ.)

The species assemblages of cichlids in the 3 largest African Great Lakes are among the richest concentrations of vertebrate species on earth. The faunas are broadly similar in terms of trophic diversity, species richness, rates of endemism and taxonomic composition, yet are historically independent of each other. Hence, they offer a true and unique evolutionary experiment to test hypotheses concerning the mutual dependencies of ecology and brain morphology.

We examined the brains of 189 species of cichlids from the three large lakes Victoria, Tanganyika and Malawi. A first paper demonstrated that patterns of evolutionary change in cichlid brain morphology are similar across taxonomic boundaries as well as across the three lakes (van Staaden et al., 1995). Here we report a close relationship between the relative sizes of various brain structures and variables related to the utilization of habitat and prey. Causality is difficult to assign in this context; nonetheless, prey size and agility, turibidity levels, depth, and substrate complexity are all highly predictive of variation in brain structure. Areas associated with primary sensory functions such as vision and taste, relate significantly to differences in feeding habits. Turbidity and depth are closely associated with differences in eye size and large eyes are associated with species that pick plankton from the water column. Piscivorous taxa, and others that utilize motile prey, are characterized by a well developed optic tectum and a large cerebellum compared to species that prey on molluscs or plants. Structures relating to taste are well developed in species feeding on benthos over muddy or sandy substrates.

The data militated against the existence of compensatory changes in brain structure. Thus enhanced development of a particular function is generally not accompanied by a parallel reduction of structures related to other modalities. Although genetic and environmental influences during ontogeny of the brain cannot be isolated, this study provides a rich source of hypotheses concerning the way the nervous system functions under various environmental conditions and how it has responded to natural selection.
check out the brains of some representative cichlids
Xenotilapia ornatipinnis
shallow sand
sand-sifting insectivore
Maravichromis lateristriga
shallow vegetation
sand-sifting insectivore
Chilotilapia rhodesii
shallow mud
oral-crushing molluscivore
Simichromis diagramma
shallow rock
Labeotropheus fuelleborni
shallow rock
Protomelas similis
shallow vegetation
higher plants
Maravichromis orthognathus
Bathybates minor
Haplochromis nyereri
shallow rock



This research was conducted in collaboration with M. Kent Rylander (Texas Tech Univ.)

The size of seven neural structures was compared in 51 species of Notropis, Pteronotropis, Cyprinella, Luxilus, Lythrurus, and Hybopsis, and related to the turbidity of the species' habitat. This last parameter was assessed for each species by personal communication with 42 ichthyologists. To control for size differences among species, all analyses were performed on the residuals from a regression of each character on standard length. Principal components analysis (PCA) of the residuals produced four significant PC-axes that together explained 65% of the total variation represented in the original variables. The size of brain structures concerned with vision, olfaction, and gustation was correlated with habitat turbidity. Two-way Analyses of Covariance (ANCOVAs) revealed significant differences between species in the size of all structures. Sexual dimorphism was exhibited by olfactory bulb and cerebellum, and significant two-way interactions (species vs. sex) were detected for the telencephalon, optic lobes, cerebellum, vagal lobe, and the eye. Cluster analysis indicated that neither similar turbidity preference nor shared phylogeny is alone sufficient to explain the observed differences in brain morphology.

Significant differences in stratification and size of the visual layers of the optic tectum were found between three clear-water minnows (Notropis amabilis, N. boops, Cyprinella venustus) and three turbid-water minnows (N. atherinoides, N. bairdi, and C. lutrensis). Correlations among a variety of neural structures suggested the importance of stratum marginale (SM), stratum opticum (SO), and stratum fibrosum et griseum superficiale (SFGS), stratum griseum centrale (SGC) and stratum periventriculare (SPV) in vision, of stratum album centrale (SAC) and SGC for olfaction, and of SPV for the processing of acoustico-lateral information.

Quantitative analysis of the optic nerve of minnows using light- and electron microscopy demonstrated that anatomical characteristics of the visual system are closely related to habitat turbidity. Species in the genera Notropis and Cyprinella inhabiting predominantly clear water had larger eyes and almost twice as many optic nerve fibers compared to minnows of turbid habitats. No differences were detected in the thickness of myelination, the axon diameter profile, or the number of optic nerve fibers per retinal area, indicating that the relative number of fibers, as well as their anatomical characteristics, are similar in all species and independent of habitat turbidity. It is therefore hypothesized that quantitative differences in the number of visual elements available for sampling and processing in the retina, optic nerve, and optic tectum are sufficient to account for presumed differences in visual performance.


Special Feature:

Here are some fish links:

Back to my personal homepage, my crayfish work, my stats page, or my work in spatial analysis.

last modified: 9/17/98