Figure 1.
Tooth measurements used in this study.
ADM, anterior denticles per millimetre; BW, basal width; CH, crown height; FABL, fore-aft basal length; and PDM, posterior denticles per millimetre.
Figure 2.
Qualitative morphotypes used to construct a priori categories within formations and the qualitative characters that define them.
A, Saurornitholestinae; B, Dromaeosaurinae; C, cf. Zapsalis; D, Dromaeosauridae; E, cf. Richardoestesia gilmorei; F, cf. Richardoestesia isosceles; G, cf. Pectinodon; and H, cf. Troodon. Qualitative characters: 1, posterior denticles apically oriented (that is, asymmetric denticles with a shorter apical side); 2, anterior denticles much smaller than posterior denticles; 3, posterior denticles rounded; 4, anterior denticles the same or slightly smaller than posterior denticles; 5, anterior denticles usually absent; 6, strong longitudinal ridges; 7, posterior denticles large and apically oriented; 8, posterior denticles are small and rounded; 9, anterior denticles are similar in size to posterior denticles or absent; 10, tall isosceles triangle shape; 11, posterior denticles very large and often rounded with apex of tooth frequently forming apical-most denticle; 12, posterior denticles are very large and apically hooked; and 13, anterior denticles are very large or absent. A, B, and H modified from Larson et al. 2010; C–F modified from Larson (2008); and G modified from Longrich (2008). Scale bars are 1 mm and correspond to images of crowns.
Table 1.
Individual tooth sample sizes of categories used in analyses by lithostratigraphic unit and qualitative morphotype.
Figure 3.
Canonical variate analysis of 1047 teeth in 32 categories distinguished by qualitative morphotype and chronostratigraphic unit.
The black star indicates the centroid of the dataset from which the relative orientations of the biplot rays in the upper right were calculated. Canonical axes 1 and 2 indicate the first two axes of maximum discrimination in the dataset.
Figure 4.
Example discriminant function analysis (DFA) canonical plots.
A, Saurornitholoestes langstoni (Dinosaur Park; gray) vs. Oldman Saurornitholestinae (black), no discrimination of categories (hit ratio = 69.44%, p = 0.1363). B, S. langstoni (Dinosaur Park; gray) vs. Atrociraptor marshalli (Horseshoe Canyon; black), discrimination of categories (hit ratio = 94.215%, p<0.0001). Centroids (with 95% confidence interval) and associated convex hulls are labelled. Canonical axes 1 and 2 indicate the first two axes of maximum discrimination in the dataset.
Table 2.
Hit ratios for each pairwise discriminant function analysis (DFA) of theropod tooth categories showing the percentage of correctly identified elements.
Figure 5.
Discriminant function analysis of type specimens versus referred isolated teeth.
A, Dromaeosaurus albertensis. B, Atrociraptor marshalli. Gray indicates holotype specimen teeth, black indicates referred isolated teeth. Associated convex hulls are marked. Canonical axes 1 and 2 indicate the first two axes of maximum discrimination in the dataset.
Figure 6.
Summary of quantitative morphotypes showing their stratigraphic ages.
Each tooth icon likely represents a distinct taxon with the indicated known range based on formation as observed in this study. A, Lancian Saurornitholestinae gen. et sp., UCMP 187036 (reversed); B, Pectinodon bakkeri; C, Lancian cf. Richardoestesia gilmorei, UCMP 120255 (reversed); D, Lancian cf. Richardoestesia isosceles, UCMP 187175 (reversed); E, Atrociraptor marshalli, TMP 2000.045.0035; F, Horseshoe Canyon Dromaeosaurinae gen. et sp., TMP 1999.050.0116 (reversed); G, Horseshoe Canyon cf. Troodon sp., TMP 2000.045.0024 (reversed); H, Horseshoe Canyon cf. R. gilmorei, TMP 2003.015.0002; I, Saurornitholestes langstoni, TMP 1995.147.0026; J, Bambiraptor feinbergi, AMNH FR 30556; K, Dromaeosaurus albertensis, TMP 1986.130.0211; L, Zapsalis abradens, TMP 1987.050.0008; M, Troodon formosus, TMP 1995.147.0025; N, Dinosaur Park cf. Pectinodon sp., TMP 2000.021.0001; O, Richardoestesa gilmorei, TMP 2000.019.0004; P, Oldman cf. R. gilmorei, 1987.080.0035; Q, Richardoestesia isosceles, LSUMGS 489∶6238 (reversed); R, Milk River Saurornitholestinae gen. et sp., UALVP 50531 (reversed); S, Milk River Dromaeosauridae gen. et sp., UALVP 48365 (reversed); T, Milk River Dromaeosaurinae gen. et sp., UALVP 49571; U, Milk River cf. Zapsalis sp., UALVP 49582; V, Aquilan cf. Richardoestesia gilmorei, UALVP 48157 (reversed); and W, Aquilan cf. Richardoestesia isosceles, UALVP 48279 (reversed). B modified from Longrich (2008); H modified from Larson et al. (2010); J modified from Burnham (2004); P modified from Sankey et al. (2002); Q modified from Sankey (2001); R–W modified from Larson (2008). Teeth scaled to matching FABL. [full page width].
Table 3.
Table. 3. Taxonomic identifications of small theropods with teeth in the formations used in this study based on holotype material, referred skeletal material, and the results of the current study.