Redescription of Pseudotropheus livingstonii and Pseudotropheus elegans from Lake Malaŵi, Africa Author Stauffer Jr, J. R. Author Konings, F. Author Ryan, T. M. text Zootaxa 2016 4154 2 169 178 journal article 10.11646/zootaxa.4154.2.4 ee339baa-0fd3-4bc3-8a4e-b903c8d00615 1175-5326 255176 B7D82FB6-A77A-49B3-AF05-87DB9B2F9A44 Pseudotropheus elegans Trewavas 1935 ( Fig. 3 ) Pseudotropheus elegans Trewavas 1935 Metriaclima elegans .— Stauffer et al . 1997 Pseudotropheus livingstonii (non-Boulenger).— Konings 2007 Pseudotropheus sp. ‘acei’.— Konings 2007 Material examined. Pseudotropheus elegans , BMNH 1935.6.14.127, holotype , 85.4 mm SL, Chilumba Bay , Lake Malaŵi ; PSU 11394, 12 specimens, 43.4–63.1 mm SL, Chitande Island , 12º 23.764'S 34º 15.275'E , Lake Malaŵi . Diagnosis. Pseudotropheus elegans is distinguished from all other members currently in Pseudotropheus , except P. williamsi , by a pale yellow to hyaline dorsal fin and by the absence of distinct vertical bars below the dorsal fin. Most species of Pseudotropheus either have distinct bars below the dorsal fin or a dorsal fin with black pigment. It is distinguished from P. williamsi by the absence of two horizontal lines of black dots on the flank. Description. Principal morphometric ratios and meristics for holotype and population from Chitande Island in Table 1 . Medium-sized mbuna, ovoid body (mean BD 31.6% SL) with greatest depth at about 6–7th dorsal spine. Dorsal body profile with gradual curve downward, more acute towards caudal peduncle; ventral body profile slightly convex between pelvic fins and base of rays of anal fin with upward taper to caudal peduncle. Dorsal head profile round, with continuous curve between interorbital and dorsal-fin origin; horizontal eye diameter (mean 36.7% HL) greater than preorbital depth (mean 16.3% HL); eye (along horizontal axis) in anterior half of head; snout straight; jaws isognathus; tooth bands with 4 rows in upper jaw and 3–4 rows in lower; teeth in anterior outer row bicuspid with posterior lateral teeth primarily unicuspid, and teeth in inner rows tricuspid FIGURE 3. The holotype of Pseudotropheus elegans (BMNH1935.6.14.127). Dorsal fin XVII–XIX (mode XVIII) and 8–9 (mode 9). Anal fin III and 8–9 (mode 8). First 4–5 dorsal-fin spines gradually longer posteriorly with fourth spine about 1½ times length of first; last 13 dorsal-fin spines increasingly longer posteriorly with last spine longest, about 2 times length of first; soft dorsal fin with subacuminate tip, third or fourth ray longest, to approximately ¼ length of caudal fin. Anal-fin spines progressively longer posteriorly; 3rd or 4th ray longest, to base of caudal fin in both sexes; 1–3 small yellow spots on posterior part of anal fin. Caudal fin subtruncate to slightly emarginate. Length of pelvic fin to first spine of anal fin. Pectoral fin short and paddle-shaped, length to vertical line through base of 10–11th dorsal-fin spine. Flank scales ctenoid with abrupt transition to small scales on breast; 32–33 lateral-line scales; cheek with 3 rows of small scales; caudal fin with tiny scales to ¼ length; no scales on other fins. Gill rakers on first ceratobranchial 10–12 (mode 11). Recently captured fish with gray head, white gular region, and black opercular spot. Laterally gray ground coloration; scales with green outline; breast and belly gray. Dorsal fin pale yellow to hyaline. Caudal fin with two ventral rays and membranes black; remainder clear with faint white spots. Anal fin black with white marginal band; 1–3 yellow ocelli in rayed portion. Pectoral-fin rays clear; pelvic-fin rays black anteriorly, remainder clear. Coloration of female similar to male. Remarks. The holotype of P. elegans and a population of this species collected from Chitande Island were compared morphologically to the holotype of P. livingstonii and a population of P. livingstonii collected from Cape Maclear. A plot of the sheared third principal components of the morphometric data against the first principal components of the meristic data ( Fig. 4 ) showed that the holotype of P. livingstonii grouped within the minimum polygon clusters and the 95% confidence ellipses formed by the meristic and morphometric data from those individuals collected at Cape Maclear. A plot of the sheared second principal components against the sheared third principal components ( Fig. 5 ) showed that the holotype of P. elegans grouped within the minimum polygon cluster and the 95% confidence ellipses formed by the morphometric data from those individuals collected at Chitande Island . FIGURE 4. Sheared third principal components (morphometric data) plotted against the first principal components (meristic data) of the holotype (x) and specimens of Pseudotropheus livingstonii from Cape Maclear (■) and the holotype (+) and specimens of P. elegans from Chitande (o). The minimum polygon clusters are bounded by 95% confidence levels. FIGURE 5. Sheared second principal components (morphometric data) plotted against the sheared third principal components (morphometric data) of the holotype (x) and specimens of Pseudotropheus livingstonii from Cape Maclear (■) and the holotype (+) and specimens of P. elegans from Chitande (o). The minimum polygon clusters are bounded by 95% confidence levels. The first principal component (size variable) of the morphometric data explained 96% of the observed variance, the sheared second principal component explained 25% and the third sheared principal component explained 15% of the remaining variance. Variables that had the highest loadings on the sheared second principal components of the morphometric data were preorbital depth (-0.51), distance between the posterior insertion of the anal fin and dorsal origin of the caudal fin (0.38), and the distance between the posterior insertion of the dorsal fin and the ventral origin of the caudal fin (0.27). Variables that had the highest loadings on the sheared third principal components of the morphometric data were snout length (-0.63), vertical eye diameter (0.43), and head depth (0.31). The first principal component of the meristic data explained 39.8% of the variance. Variables with the highest loadings on the first principal components of the meristic data were tooth rows on the upper jaw (0.40), pored scales posterior to the lateral line (0.36), and anal-fin rays (0.38). The HRCT scans permitted us to critically analyze the shape of the skull of the type specimens of P. livingstonii , P. elegans , and M. lanisticola , a sand-dwelling species of Metriaclima . We found that the angle that the ethmo-vomerine bloc makes with the parasphenoid is much more acute in Pseudotropheus than in Metriaclima . The angle in the holotype of P. livingstonii is 57°, and in that of P. elegans is 58°. The ethmo-vomerine bloc/ parasphenoid angle in the holotype of M. lanisticola is 48°.