Figure 1.
Variables measured for morphometric analysis.
Length, width, and total shearing crest length were measured on upper second molars. Morphology varies with phylogeny; examples shown here are A) Rangifer tarandus (tooth in situ), B) Perameles nasuta (tooth cast), and C) Dorcopsis hageni (tooth cast). Tic marks are 1 mm.
Figure 2.
Microwear features tallied during analysis.
Large features show up dark on a light field (A); small scratches and small pits show up light on a dark field. (B and C) Dark field versus bright field illumination is achieved by adjusting the light source angle (see also Figure S1 in Supporting Information S1). Scale bars are 0.5 mm.
Figure 3.
Microwear patterns of marsupial and placental herbivores plotted according to feeding guild (A) and phylogeny (B).
Note, the partial separation between marsupials and placentals in B is only based upon the absence of obligate hard object feeders (and, thus, of elevated pit counts) among marsupials in the data set. The three dietary groupings in A are significant (p<0.01). Marsupial and placental herbivores of comparable guilds are statistically indistinguishable (p = 0.84).
Figure 4.
The relative frequencies of microwear variables within feeding guilds.
A) Fine scratches (Sf), B) Small Pits (Ps), C) Coarse Scratches (Sc) D) Large pits (Pl), E) Total scratches, F) Total pits. Box plots show the median (center line), interquartile range (boxes), 1.5 times the interquartile range (whiskers), and outlier points. Discriminant analysis using the four independent variables (Sf, Sc, Ps, Pl) as dependents is depicted: G) all five feeding guilds included. Functions 1 and 2 are plotted and account for 71.8% and 24.6% of the variance, respectively. Wilks’ Lambda = 0.252, Chi-Square = 250.51, p<0.001. The discriminant function coefficients are Sf0.039/0.578, Sc0.499/0.688, Ps0.845/−0.416, Pl0.071/−0.166. F) Herbivores groups only (grazers, browsers, hard-object feeders). Functions 1 and 2 are plotted and account for 75.1% and 24.9% of the variance, respectively; Wilks’ Lambda = 0.194, Chi-Square = 192.80, p<0.001. The discriminant function coefficients are Sf0.005/0.379, Sc0.324/0.844, Ps0.822/−0.330, Pl0.270/−0.276. HO = Hard-Object Feeders, B = Browsers, G = Grazers, I = Insectivores, C = Carnivores. Hulls surround each dietary group.
Figure 5.
Microwear results of insectivores (A) and carnivores (B) plotted with herbivorous taxa.
Marsupial and placental insectivore groups are not significantly different (p = 0.65), nor are hypercarnivores (flesh eaters) and bone carnivores (chew and consume bone) (see Table S5 in Supporting Information S1 for details). Regardless of feeding preferences, both flesh and bone specialists consume both types of food, depending on availability, the season, the animal’s status in social groups, and other factors, resulting in a lack of further differentiation.
Table 1.
Microwear differences between feeding guilds.
Figure 6.
The relative frequencies of body mass and Shearing Crest Score (SCS) within feeding guilds.
The distribution of A) body mass (BM) and B) Shearing Crest Score (SCS) among all specimens and feeding guilds is shown. Box plots show the median (center line), interquartile range (boxes), 1.5 times the interquartile range (whiskers), and outlier points. C) Discriminant function analysis using BM and SCS as dependent variables. HO = Hard-Object Feeders, B = Browsers, G = Grazers, I = Insectivores, C = Carnivores. Function 1 (75.2%) and Function 2 (24.8%) are largely synonymous with BM/Ps and SCS, respectively. Wilks’ Lambda = 0.686, Chi-Square = 51.76, p<0.001. The discriminant function coefficients are BM−0.157/0.989, SCS0.994/0.116, Ps0.845/−0.416, Pl0.071/−0.166. D) Bivariate plot, BM vs. SCS, showing polygonal dietary morphospaces. Hulls delineate dietary groups.
Figure 7.
Flow chart representing the path taken toward the assignment of feeding guild using the three different analyses (body size, tooth morphology, LDM microwear) in succession.