Morphological evolution of the carnivoran sacrum

Abstract

The sacrum is a key piece of the vertebrate skeleton, since it connects the caudal region with the presacral region of the vertebral column and the hind limbs through the pelvis. Therefore, understanding its form and function is of great relevance in vertebrate ecomorphology. However, it is striking that morphometric studies that quantify its morphological evolution in relation to function are scarce. The main goal of this study is to investigate the morphological evolution of the sacrum in relation to its function in the mammalian order Carnivora, using three-dimensional (3D) geometric morphometrics. Principal component analysis under a phylogenetic background indicated that changes in sacrum morphology are mainly focused on the joint areas where it articulates with other parts of the skeleton allowing resistance to stress at these joints caused by increasing muscle loadings. In addition, we demonstrated that sacrum morphology is related to both the length of the tail relativised to the length of the body, and the length of the body relativised to body mass. We conclude that the sacrum in carnivores has evolved in response to the locomotor requirements of the species analysed, but in locomotion, each family has followed alternative morphological solutions to address the same functional demands.

Keywords: 3D geometric morphometrics; Carnivorans; evolution; locomotion; sacrum.

FIGURE 1

Sacrum connections with different body regions such as the presacral and caudal regions of the vertebral column, as well as with the hind limbs trough the lumbosacral, sacrocaudal and sacroiliac articulations, respectively, in a model of Canis lupus, as an example

FIGURE 2

2Anatomy of the carnivoran sacrum represented in a specimen of Canis lupus as an example. Both the skeletal components of the carnivoran sacrum and the insertion areas of the main sacral muscles in ventral (a), lateral (b) and dorsal (c) views are shown

FIGURE 3

Representation of the 28 landmarks digitised in a sacrum of Canis lupus in dorsal (a), ventral (b), caudal (c), lumbar or anterior (d) and lateral (e, left) views. In (e), it is also shown a latero‐lumbar view (right) for a better explanation of how landmark 19 was digitized. The anatomical criteria followed for landmark digitalisation is in Table 2. Landmarks 20–22 (not shown) of the left side are the symmetric with of 17–19 of the right side, respectively

FIGURE 4

Phylogenetic background and both meristic and morphological variation of the sacrum. (a) Tree topology used in comparative analyses (modified from Nyakatura and Bininda‐Emonds, 2012). (b) the number of sacral vertebrae in each taxa (meristic variation). (c) the lateral and dorsal views of the sacra belonging to one species of each family show morphological variation

FIGURE 5

Multivariate regression analysis between sacrum shape and the centroid logarithm. Note that the regression is strongly influenced by the presence of ursids and pinnipeds (specimens with a logarithmic centroid size higher than 5.4)

FIGURE 6

6Results of the principal components analyses performed on the morphometric data of the sacrum. (a) Scatter plot illustrating the position of different species on the first two principal components. The phylogeny from Nyakatura and Bininda‐Emonds (2012) is plotted in the morphological space. Arrows indicate outliers within locomotor groupings. The convex hull of the terrestrial category is not represented as terrestrial taxa is not grouped. (b) Main phenotypic variation corresponding to the minimum and maximum of the two first PCs. The 3D model of the sacrum used in the morphing of the shape changes accounted for by each PC correspond to C.lupus

FIGURE 7

Box plot showing the tail length relativized to body length and the body length relativized to body mass in the different locomotory categories included in our sample. For the raw values of tail length, body length and body mass, see Table 3. The vertical line inside each box is the median. Box length is the interquartile range (IQR) and shows the difference between the 75th and 25th percentiles. Horizontal bars enclose values of 5–95%

FIGURE 8

Multivariate regression analysis between sacrum shape (regression score) and Relative Tail Length (a) and bivariate regression analysis between Relative Tail Length and Centroid size, log‐transformed (b)