. Bulletin of the Museum of Comparative Zoology at Harvard College. Zoology. Z*-^. Figure 7. Representative shells for samples depicted in same positions on Fig. 6. At bottom line, left to right: locality 230 (typical C. bendalli); locality 316 (convergent C. bendalli from Pongo Carpet); locality 253 (intermedi- ate shell from The Crossing); locality 254 (C, abacoense); "C. chrysaloides" from Grand Bahama. Above: locality 204 (C. bendalli from Grand Bahama). tion) are C. bendalli from Grand Bahama. The third axis divides samples of C. hen- dalli (the mottled morphotype) into its two
. Bulletin of the Museum of Comparative Zoology at Harvard College. Zoology. Z*-^. Figure 7. Representative shells for samples depicted in same positions on Fig. 6. At bottom line, left to right: locality 230 (typical C. bendalli); locality 316 (convergent C. bendalli from Pongo Carpet); locality 253 (intermedi- ate shell from The Crossing); locality 254 (C, abacoense); "C. chrysaloides" from Grand Bahama. Above: locality 204 (C. bendalli from Grand Bahama). tion) are C. bendalli from Grand Bahama. The third axis divides samples of C. hen- dalli (the mottled morphotype) into its two isolated areas. The matrix of factor scores (Table 2) permits us to infer the basis of separations in Figure 6 (consult the matrix of mean values—Table 1—for the raw information). Only three variables score highly on the first axis. (This serves as a dimension of reference for the mottled morphotype, C. bcmlaUi. Samples of C. ])endaUi load strongly upon it, and weakly upon the second axis—see Figs. 6-7.) Not surpris- ingly, these three variables are the ribbing measures 4-6. (Mottled samples always have much weaker ribs than ribby samples, but the ribs are always far more numerous in mottled samples; all our ribbing measures are counts.) No other variable so con- sistently separates C. bendalli from C. abacoeyise. The second axis, with it high loadings for C. abacoense (the ribby mor- photype), contains high scores for most measures of final size and whorl size. (The scores are negative in this case. The sign is of no particular importance, since it only indicates the direction of the reference vec- tor. The pattern of scores and loadings would not change if the vector pointed 180° in the opposite direction, thus reversing all the signs.) To some extent, this suite of high scores only mirrors the distinction by ribbing made on the first axis. Since refer- ence vectors are normalized, a small number of ribs must lead to a greater contribution to the vector from othe
Size: 1179px × 2119px
Photo credit: © Book Worm / Alamy / Afripics
License: Licensed
Model Released: No
Keywords: ., bookauthorharvarduniversity, bookcentury1900, booksubjectzoology
