Teeth in jaws are generally graded by a useage of “heterodonty” versus one of “non-heterodonty.” In some groups, teeth that are actually identical in two spectra (form and size) are termed isodont (literally, “same tooth”) and could reasonably pass for one another without any confusion. The opposite of isodont dentitions has been called “heterodont,” (literally, “different tooth”) but this is used in two ways.
First, “heterodonty” has been used when teeth differ in size from one another, or when size is extreme and form is slightly different; the term is generally applied to the size-based variation.
Second, “heterodonty” has been used when teeth do not necessarily differ in size, but are different in form. In this case, their morphology can be differentiated without recourse to a scale bar.
In the first case, using a basic example, the extant crocodilian Alligator mississippiensis has teeth of different essential size, from small, large, and extremely large teeth along the tooth row. Many reptilian dental series follow a gradient “small-large-small” morphology , which agrees with the first case definition of “heterodonty.” This morphology does not agree with the second case definition, however, in which case a taxon with highly divergent morphology but with similarly-sized teeth in the dental series would not be “heterodont.” It should be noted that when dealing with mammals, “heterodonty” is used for both size based and morphology based differentiation, including and especially in the human jaw; where one can regard the size, rather than the form, of the molar against the premolar (termed “anisodonty”) based solely on the relative width of the crowns [n1]. Conversely, differentiating morphology into incisor, canine, bicuspid and molar (as well as the scaling morphological variation between “true” molars and any postcanine premolar) in mammals results in a size-disregarding complex where teeth are graded based first on morphology then on root structure, which ignores the crown entirely. In Sauropsida (as opposed to the derived mammalian Synapsida), we lack this differentiability without choosing particular taxa (e.g., Ceratopsidae, which have two-rooted maxillary and dentary teeth). Thus, a better determinant is needed.
Above, a figure represents the two axes of “heterodonty” as used historically and currently. These two axes are intended to show that both determinants may be in effect in a given dental series, and are not exclusive to one another. Rather than use Alligator mississippiensis, one may then turn to Crocodilus niloticus, in which the size spectrum is more xtreme, while the morphological spectrum increases.
So the question here revolves around the use of the term, and how to distinguish functional size differentiation from morphological differentiation in the absence of size. One of the best ways to do this is to explicate the various morphologies that abound in dentition, and discriminate them based on size, and to determine overlap between the two datasets. It is clear that this has not yet been done. I mention this here not because I think I have the answer, but because it is a particularly seemingly important question that has gone unresolved in the paleontological and zoological literature. This is based most certainly on the fact that non-mammal dentition is described in an herpetological manner, where the dentition is generally all the same, and virtually never regionally or morphologically differentiated, even when they occur on different bones. That issue alone (the Reptile vs Mammal teeth quandary) deserves its own research goal, but is interelated to the question of differentiation and the terminology useful to describe it.
What is clear, however, is that “heterodonty” is a term to describe two different axes of variation, and it alone is not sufficient to assess both of them. We certainly need another term to account for one of these axes, but it requires assessing what the term “heterodonty” should apply to, if it should apply to either one of them alone. It may even be that “heterodonty” is correctly applied in the general, but when it is, it should not be opposed by a term “isodonty,” where a spectrum with “heterodonty” at one end is expected.
 Auffenberg, W. 1981. The Behavioral Ecology of the Komodo Monitor. (University Presses of Florida, Gainesville.)
[n1] This term in particular refers to upper versus lower molariform teeth, however.