Materials and Methods

This study is based on samples of hadrosaur teeth representing a latitudinal gradient that extends from northern Alaska to west Texas (Figure 1). The sample of dinosaur teeth was gathered from the Kikak-Tegoseak Quarry in northern Alaska (Fiorillo and Gangloff 2003; Fiorillo et al. 2010). This quarry is within the Prince Creek Formation, a largely fluvial rock unit that is based on biostratigraphy and radiometric dating and straddles the Campanian-Maastrichtian boundary (see discussions in Fiorillo 2006a; Fiorillo and Gangloff 2000, 2001; Gangloff et al. 2005; Fiorillo et al. 2009; Fiorillo et al. 2010), though the quarry itself is considered to be Early Maastrichtian in age (Fiorillo et al. 2010). These specimens are housed at the Museum of Nature and Science (formerly the Dallas Museum of Natural History).

Wear patterns on these Alaskan hadrosaur teeth were compared to wear patterns on samples of hadrosaur teeth found in more southern latitudes and include samples from southern Alberta, Wyoming, and west Texas. These southern samples are from Campanian, Maastrichtian, or Campanian/Maastrichtian aged rock units. Thus they are broadly correlative with the Prince Creek Formation (Table 1).

These southern samples were gathered from specimens housed at the Museum of Nature and Science, the Texas Natural Science Center (formerly the Texas Memorial Museum), and the Denver Museum of Nature & Science. Despite the institutional name changes for the first two organizations, the historic acronym has been retained in each case. Given that both the Museum of Nature and Science and the Denver Museum of Nature & Science both use the acronym DMNH, in this paper specimens from these two institutions are differentiated by the Denver Museum of Nature and Science specimens are listed here as DMNH [Denver]. The acronyms for specimens from these three institutions, respectively, are DMNH, TMM, and DMNH [Denver]. Table 1 is a list of taxa used in this study, the rock unit from which they were collected, and their general locality.

Horner et al. (2004) noted that wear is virtually continuous across the dental battery of hadrosaurs, thus providing the prediction that there would be little variance in microwear patterns among the teeth of a single tooth row. Specimens were obtained so that teeth could be examined across hadrosaurian tooth batteries (TMM 42314-1, TMM 40484-15, TMM 42315-1). These specimens are from southern Alberta and housed at the Texas Natural Science Center. Each tooth battery was examined to determine variation in microwear patterns across individual batteries to validate the use of isolated teeth as a proxy for microwear of individual hadrosaurs. Because adult hadrosaur maxillae do not fit into the chamber of scanning electron microscopes, epoxy replicas had to be generated. The tooth batteries were molded using Coltene Whaledent, and the subsequent casts were made using Tap Plastics 4 to 1 epoxy. Twelve teeth from three maxillae were examined.

Twenty-three isolated, unassociated teeth were examined. The lack of association among teeth provides a maximum number of individuals represented in the sample. These isolated teeth were collected from sites in Alaska, Wyoming, and Texas.

The LEO (now part of Zeiss) scanning electron microscope, model 1450 VPSE, housed in the Department of Geosciences at Southern Methodist University was used for this study. The wear surfaces of the teeth were examined at magnifications up to 1000X. The scanning electron microscope typically employed the backscatter detector. The parameters used to examine specimens are shown in Table 2. Analysis of patterns was facilitated by using Microware 4.02 developed by Ungar (2002).

Many parameters for assessing the significance of microwear are focused on patterns found in various mammal teeth where tooth structure is more clearly understood. The significance of some features is sometimes interpreted in relation to mammalian enamel microstructure (Maas 1991). The scale of magnifications used approached those used to study dental microstructure (Sander 1999; Maas 1991). Therefore regular patterns that appeared in either the dentine or the enamel were attributed to microstructure and were not included in the microwear analysis (Figure 2 and Figure 3).

Two parameters used effectively in determining resource partitioning in dinosaurs are the presence of pits, and scratch or striation orientation (Fiorillo 1998). In addition to the presence of pits, I examined the percent incidence of pits (sensu Ungar 1994), which is the number of pits divided by the total number of features multiplied by one hundred. Microware 4.02 (Ungar 2002) provides the length of the mean striation vector, which is a measure of the concentration of the orientations of the long axis of each linear feature. Shorter vectors indicate more random distributions of features (where r = 0 or approaches 0), and longer vectors indicate more preferred distribution (where r = 1 or approaches 1).

Comparative studies of microwear have focused on patterns of enamel wear (Kay and Covert 1983; Teaford and Walker 1984; Teaford 1985; Teaford and Byrd 1989; Walker and Teaford 1989; Walker et al. 1978; Taylor and Hannam 1987; Rivals and Deniaux 2003; Hotton et al. 1997; Fiorillo 1998, 2006b, 2008; King et al. 1999; Goswami et al. 2005; Schubert and Ungar 2005). Wear on the enamel and dentine of the ornithischian teeth is considered here in detail. Although comparable studies are lacking for wear patterns on dentine surfaces, wear patterns preserved on the dentine are discussed here in the context of associated wear on the enamel of the teeth. With respect to relative wear on dentine versus enamel in the teeth of small theropods, it was noted that the softer dentine exhibited relatively coarser wear (Fiorillo 2006b, 2008). A similar pattern of relative wear was expected with these hadrosaur teeth.

It was not a prerequisite for individual teeth to have fully developed wear facets to be included in this study. All scanning electron microscope images were taken with the wear surface approximately perpendicular to the electron beam.