Semipallium marybellae Raines, 1996
RAINES, B. K. 1996. Description of Semipallium marybellae n. sp. (Bivalvia: Pectinidae) from Guam, with notes and bivariate statistical comparison to Semipallium tigris (Lamarck, 1819). La Conchiglia, 28 (279): 20-30, figs. 1-15. [p. 21, figs. 1-13]
1996 Semipallium marybellae Raines, 1996
B. K. Raines, 1996, figures 1-13.
|
«Description
Shell medium in size, average height 43 mm; oblique, convex, higher than wide, rather thin and light weight for its size. Valves semitransparent especially near umbo; left valve less convex than right; ventral margin of mature specimens well developed, truncated; byssal notch rather small; auricles very unequal; umbonal angle about 90 degrees. Exterior of valves with 9 pronounced radial ribs; secondary radial riblets covering entire shell, both on ribs and in interspaces. Shagreen microstructure present on both valves; pattern on left valve of random blotches and speckles of bright red and reddish brown tones on cream background; yellow staining present on alternating ribs. Right valve paler than left with only fine speckles of coloration on cream background. Interior surface glossy white with yellow staining varying in intensity with exterior pattern showing through; pinkish coloration sometimes present proximal to one or both auricles on left valve. Hinge weakly developed; ctenolium present and active with 4-6 low, narrowly spaced teeth. Type material
All type material was live collected except paratypes 1 and 8. Type material has been deposited with the Santa Barbara Museum of Natural History, Santa Barbara, California (SBMNH), with the Academy of Natural Sciences, Philadelphia, Pennsylvania (ANSP), and the Zoological Museum of Amsterdam (ZMA). The collocation of the type material is given in Table 1. Type Locality
Luminao Barrier Reef, 13° 26' N, 144o 37' E, and Asan Point 13° 28' N, 144° 42' E, Guam. Etymology
The name given to this new species is in memory of the author's late mother, Mary Raines 1940-1991. She was a beautiful lady, and a beloved wife and mother. Distribution and Habitat
At present, S. marybellae is known only from two locations on Guam; on the outer reef areas of Luminao Barrier Reef and the reef off Asan Point. Live specimens were found both free swimming and byssally attached to substrate, living in and under coral rubble at a depth of 20-30 m. Discussion
S. marybellae was placed in the genus Semipallium because its physical characteristics are best represented by the description of that genus. It could also have been placed in the genus Complicachlamys, but the validity of this genus is in question. Some researchers have used the two genera interchangeably, though Complicachlamys is probably a junior synonym of Semipallium (Grau 1959, pp 120-121). The closest congener to S. marybellae is S. tigris, which has also been collected within the same general habitat from Asan Point, Guam. S. tigris, however, occurs throughout the Indo-Pacific region, and differs significantly from S. marybellae in numerous conchological features. S. tigris is generally larger, up to 60 mm in height, and has 9-11 fold-like ribs with approximately 95 secondary radial riblets per valve which typically display pronounced granulation, whereas Semipallium marybellae attains a height of 50.4 mm and has only 9 true costae with approximately 114 secondary radial riblets per valve which are relatively smooth. Similar shagreen microsculpture is present on both species. The exterior coloration of S. tigris consists of dark reddish brown to black blotches, which form a recognizable pattern of concentric wavy bands on a white background (Fig. 5). S. marybellae is more colorful than S. tigris and lacks a set pattern (Fig. 1-4). In addition to red, S. marybellae also occurs in exterior color variations of tan and flesh tones or orange and brownish tones, both on a cream background (Fig. 2). S. tigris tends to be less inflated and the ventral margin is less developed than S. marybellae (Figs. 6-9). The auricles of S. tigris are proportionally different in size and shape than in S. marybellae, and the size of the byssal notch is nearly twice that of S. marybellae (Figs. 10 & 11). In cross section, the ribs of S. marybellae are semi-circular and pronounced (Fig. 8). The riblets are evenly spaced and of equal height, with the two outer riblets of each rib standing longitudinally perpendicular to the rib giving it a squared appearance. The interspaces are slightly wider than the costae and somewhat flattened through the middle. In contrast, the ribs of S. tigris in cross section appear low and depressed (Fig. 9). The riblets are evenly spaced, but are of increasing height towards a central riblet on each rib. This central riblet is more elevated than the rest and heavily serrated giving the ribs of S. tigris a more triangular appearance. The interspaces are flattened centrally and are proportionally much wider than the ribs. The interior coloration of S. tigris is very similar to the new species with the exception of the dark brown staining along the hinge line, typical of S. tigris but absent in S. marybellae (Figs 10 & 11). The ctenolium of S. tigris consists of 5-6 widely spaced teeth, which are pronounced and well developed (Fig. l3). The ctenolium of S. marybellae consists of 4-6 narrowlyspaced teeth, which are low and less developed (Fig. 12). The hinges of both S. tigris and S. marybellae are weakly developed. To verify the measurable differences in characteristics between the two species were not merely a chance, a bivariate statistical analysis was performed. The system of measurement and the ratios selected were based on Waller (1972, pp. 230-232), with some minor modifications (Fig. 14 & Table 2). The formulas and applications used for the analysis were per Imbrie (1956, pp. 214-252). Ten samples of each species were used for the analysis. The samples of S. tigris were selected from various locations throughout the Indo-Pacific region, including Guam. The assumptions for the analysis were as follows: the samples of S. tigris, regardless of geographic locality, would be considered as one population, and the samples of S. marybellae would represent the other population. The two populations were considered to be separate and distinct species. In order for the second assumption to be true, the statistical discrimination results for the univariate analysis must meet at 99% level of significance; given the number of samples, the T-distribution value must be greater than 2.878. In addition, the statistical discrimination results for the bivariate analysis must meet at 95% level of significance; given this, the Zdistribution value must be greater than 1.960. The univariate analysis was comprised of fourteen separate measurements taken from each sample. The means, standard deviation, standard error and the coefficient of variation were determined for each set of measurements. The results of the T-distribution tests demonstrated that differences found in eleven of the fourteen sets were statistically significant (Table 3). The bivariate analysis was comprised of ten randomly selected ratios derived from the univariate analysis. The mean, standard deviation, standard error and the coefficient of correlation were determined for each set of ratios. The coefficient of relative dispersion about the reduced major axis was determined for only those sets of ratios which were found to be statistically significant. The results of the Z-distribution tests demonstrated that differences found in six of the ten sets were statistically significant (Table 4). Scatter diagrams were also developed for those six sets of ratios to provide a visual display of the results (Fig. 15). The results of the discrimination tests demonstrated that many differences in the individual characteristics of the two species are statistically distinguishable. After reviewing the data from the bivariate analysis, it was determined that treating the samples of S. tigris as a single population did not skew the results, with the possible exception of one anomaly. The anomaly appears as a lack of correlation in the height/ventral margin ratio, AM/VVl. Since the AM/VVI ratios of the two populations did not exceed the minimum value required during this discrimination test, the overall findings were not affected. If the test had demonstrated that there was a statistical significance, the results of that test would have been rejected. In Table 4, the ratios, means, standard deviations, standard errors and coefficient of correlations are expressed as numerical values; the coefficient of relative dispersion about the reduced major axes are expressed as percentages; and the Z-distribution tests are expressed as numerical values. Other ratios could have been selected for the bivariate analysis providing varied results. The intent, however, of the analysis was to determine, if any of the measurable differences were statistically significant. The results of the analysis support the argument that these are two separate species. In addition to Semipallium tigris, S. marybellae was examined and compared with other allied pecies, and the findings are discussed briefly herein: S. amicum (E.A. Smith, 1885) is smaller in size and has eight broadly rounded ribs with very deep, narrow interspaces; S. barnetti Dijkstra, 1988, is also smaller in size and has seven to nine low irregular ribs; S. crouchi (E.A. Smith, 1892) is similar in size and has nine ribs, but lacks the pronounced radial riblets; S. dianae (Crandall, 1979) differs from S. marybellae by the same characteristics as S. crouchi; S. luculentum (Reeve, 1853) is similar in size has nine shallow irregular ribs; S. fulvicostatum (Adams & Reeve, 1850) is smaller in size and has up to ten costae, which are lower and lack the pronounced radial riblets; S. kengaluorum Dijkstra, 1986, has eight ribs, and is now considered as a junior synonym of Complicachlamys dringi (personal communication, Mr. Henk Dijkstra); C. drlngi (Reeve, 1853) is similar in size, but is narrower and has eight ribs. All of the allied species differ in coloration and lack the the well developed ventral margins of Semipallium marybellae. In addition, the auricles of these allied species differ proportionally in size and structure from S. marybellae.» BRET K. RAINES, 1996
|