Bentho-pelagic divergence of cichlid feeding architecture was prodigious and consistent during multiple adaptive radiations within African rift-lakes

Abstract

Background: How particular changes in functional morphology can repeatedly promote ecological diversification is an active area of evolutionary investigation. The African rift-lake cichlids offer a calibrated time series of the most dramatic adaptive radiations of vertebrate trophic morphology yet described, and the replicate nature of these events provides a unique opportunity to test whether common changes in functional morphology have repeatedly facilitated their ecological success.

Methodology/principal findings: Specimens from 87 genera of cichlid fishes endemic to Lakes Tanganyka, Malawi and Victoria were dissected in order to examine the functional morphology of cichlid feeding. We quantified shape using geometric morphometrics and compared patterns of morphological diversity using a series of analytical tests. The primary axes of divergence were conserved among all three radiations, and the most prevalent changes involved the size of the preorbital region of the skull. Even the fishes from the youngest of these lakes (Victoria), which exhibit the lowest amount of skull shape disparity, have undergone extensive preorbital evolution relative to other craniofacial traits. Such changes have large effects on feeding biomechanics, and can promote expansion into a wide array of niches along a bentho-pelagic ecomorphological axis.

Conclusions/significance: Here we show that specific changes in trophic anatomy have evolved repeatedly in the African rift lakes, and our results suggest that simple morphological alterations that have large ecological consequences are likely to constitute critical components of adaptive radiations in functional morphology. Such shifts may precede more complex shape changes as lineages diversify into unoccupied niches. The data presented here, combined with observations of other fish lineages, suggest that the preorbital region represents an evolutionary module that can respond quickly to natural selection when fishes colonize new lakes. Characterizing the changes in cichlid trophic morphology that have contributed to their extraordinary adaptive radiations has broad evolutionary implications, and such studies are necessary for directing future investigations into the proximate mechanisms that have shaped these spectacular phenomena.

Figure 1. Anatomical landmarks examined.

1 = Tip of the anterior-most tooth on the premaxilla; 2 = Tip of the anterior-most tooth on the dentary; 3 = Maxillary-palatine joint (upper rotation point of the maxilla); 4 = Maxillary-articular joint (lower point of rotation of the maxilla); 5 = Articular-quadrate joint (lower jaw joint); 6 = Insertion of the interopercular ligament on the articular (point at which mouth opening forces are applied); 7 = Posterio-ventral corner of the preopercular; 8 = Most posterio-ventral point of the eye socket; 9 = The most anterio-ventral point of the eye socket; 10 = Joint between the nasal bone and the neurocranium; 11 = Posterior tip of the ascending process of the premaxilla; 12 = Dorsal-most tip of the supraoccipital crest on the neurocranium; 13 = Most dorsal point on the origin of the A1 division of the adductor mandibulae jaw closing muscle on the preopercular; 14 = Most dorsal point on the origin of the A12 division of the adductor mandibulae jaw closing muscle on the preopercular; 15 = Insertion of the A1 division of the adductor mandibulae on the maxilla; 16 = Insertion of the A2 division of the adductor mandibulae on the articular process.

Figure 2. Morphological disparity relative to lake age.

(X = LT species; • = LM species; ★ = LV species). The trendline depicted does not denote a significant relationship between a lake's age and the morphological disparity of its cichlids.

Figure 3. PC score plot of the All Lakes data with ecomorphological groupings.

X = LT species (blue); Tropheini from Lake Tanganyika = large X (red); • = LM species (black); ★ = LV species (red); A = long jawed predators; B = pelagic fishes with large eyes placed near gracile, protrusile jaws; C = Hard biting fishes from Lake Malawi. D = Hard biting fishes from Lake Tanganyika. The key to the species can be found in Table S2.

Figure 4. Pictorial descriptions of the shape variation described by PC1 in each of the four datasets.

Plates A–L display pairwise comparisons of cichlids whose head shapes are strongly separated along PC1, but which are otherwise very similar (they have similar scores on other axes). The paired species from each dataset occupy the same row, and are immediately followed by a deformation grid that depicts the shape transformation associated with the PC1 axis in question. 1st row = All Lakes, 2nd row Lake Tanganyika, 3rd row = Lake Malawi, 4th row = Lake Victoria. A: Labeotropheus fuelleborni. B: Bathybates fasciatus. C: All Lakes PC1 deformation grid. D: Spathodus sp. E: Bathybates fasciatus. F: LT PC1 deformation grid. G: Labeotropheus fuelleborni. H. Tyrannochromis macrostoma. I: LM PC1 deformation grid. J: Neochromis nigricans. K: Pyxichromis parorthostoma. L: LV PC1 deformation grid. Scale bars = 1 cm.

Figure 5. Pictorial description of the shape variation described by PC2 in the All Lakes dataset.

The horizontal lines describe the relative extent of the distributions of the variation present in each lake along PC2. Top line = LV, Middle line = LM, Bottom line = LT. Plates A and B depict specimens whose head shapes are strongly separated along PC2, but which are otherwise very similar (they have similar scores on other axes). Only specimens from the All Lakes dataset are depicted. A. Lobochilotes labiatus. B. Trematocara nigrifrons. C: All Lakes PC2 deformation grid. Scale bars = 1 cm.

Figure 6. Scree plots of eigenvalues for the PC axes derived from PCAs of the individual lakes datasets.

A = Lake Tanganyika. B = Lake Malawi. C = Lake Victoria.