What Is a Fox?

Three different ways to categorize living things: traditional terminology, biological classification, and cladistics

All of us try to make sense of the world around us, and a key part of that effort is naming the things we see. This allows us to talk about these things, to accumulate knowledge about them, and to compare experiences with other people. We feel much more in control when we are familiar with the things we see, and we have a better sense of how to react. It can be quite unsettling to see a large, unidentified creature roaming nearby — but the experience is quite different if you can confidently identify the animal.

However, we often find ourselves in a sort of muddled middle. You might end up saying “I think I saw a fox today, but I’m not sure.” Perhaps you only got a brief glimpse of the animal from a distance. Or perhaps you are not completely confident in your ability to distinguish a fox from other similar animals — especially if you are traveling in unfamiliar territory. This raises the question of “What is a fox?” What distinguishes a fox from all other animals?

The traditional criteria for distinguishing a fox from other similar creatures include the size of the animal, the shape of the body and the head, the presence of a thick brushy tail, the color pattern, and so on. However, traditional criteria are seldom universal; people simply needed to distinguish one type of local animal from all other local animals. Depending upon where you live, the local mix of animals might differ, which could mean that the criteria for distinguishing a fox from other animals might also vary.

We inherited the word fox from Old English, which means that we’ve been using the word for many hundreds of years, perhaps even a thousand. To put it another way, the word fox was in popular use long before anyone worried about a scientific definition. In fact, a lot of words we use to categorize natural things predate scientific terminology — including such words as fish, bird, fruit, and bug. But when these words are adopted for scientific use, they are given more precise definitions, which can easily conflict with the original popular meanings of the words.

For example, the word fish originally meant any animal that lives underwater. We can still see traces of this meaning in words such as shellfish and crayfish. The popular meaning has since narrowed to include only those creatures whose body shape conforms to our current stereotype of a fish. However, dolphins and whales have a similar shape — and as a result, small children often call them fish until taught otherwise. On the flip side, even adults may be reluctant to categorize seahorses or eels as fish because they don’t match their mental image of a fish — even though scientists classify them as fish.

So, if we assume that the word “fox” might now have a scientific definition — one that applies anywhere in the world — then what is that definition? What criteria do scientists use to distinguish between foxes and all other animals? To answer that question, we first need to take a look at the past and present state of biological classification. Then we can apply those concepts to the case of foxes.

Biological Classification

The ancient Greeks divided the world of nature into three large “kingdoms” — animals, plants, and minerals. This line of thinking remained deeply embedded in Western thought for millennia. In 1735, a scientist named Carl Linnaeus published a book that greatly expanded on this idea. He divided each of these three kingdoms into classes, which he subdivided into orders, then families, genera, and species — a hierarchy of six levels. He further refined this classification system in later editions of his books. The editions he published in the 1750s became the starting point for modern biological classification (which does not include minerals). We still use this system for dividing living things into groups and subgroups — although scientists long ago added a 7th level, called the phylum or division, just below the kingdom level. Interestingly, we still use some of the Latin species names that Linnaeus invented.

Our formal system of biological classification has now been in use for more than 250 years. During that time, the details have constantly evolved due to new discoveries about living things. But in recent decades, there has been a massive change in the criteria and methods that taxonomists use. Sixty years ago, we relied primarily on morphology (physical features) to categorize living things, and now we rely primarily on genomics instead. In other words, scientists compare the DNA of two creatures to identify the similarities and the differences. This has had a huge impact on the entire taxonomic system, at all levels of classification. Every year scientists continue to make updates to the scheme, affecting every level of the hierarchy.

Despite this long tradition of using seven levels of biological classification, the world of living things does not fall naturally into seven distinct levels. For the system to reflect reality, we would need dozens of levels. This has led to a never-ending debate on how best to categorize all the creatures of the world into seven fundamental levels. This problem has also inspired an alternative way of thinking called cladistics — a topic to which we will soon return.

Our Evolving Methods for Classification

When Linnaeus set up his classification system, the concept of biological evolution had no role. Two species could be considered “related” simply because they looked similar or had similar characteristics. Few people considered the idea that two related species might have evolved from a common ancestor. Therefore biological classification was considered to be an artificial system invented by the human imagination. Despite its artificial origin, the system was perceived as a helpful way to organize our knowledge of the world. Just as you and I might choose different criteria for organizing our kitchen drawers, the details of this classification system depended heavily on which criteria were chosen as the organizing principles.

For classifying animals, the main organizing principle was “body plans”. For example, vertebrates (animals with backbones and internal skeletons) have a body plan that is quite distinct from other animals (the invertebrates). Therefore all vertebrates were grouped together into a single high-level category (the phylum Chordata). However, the invertebrates are quite varied. A grasshopper has one body plan, a sea star has another, and an earthworm has still another. Each of these three examples needed to go into a separate high-level category. These high-level groups could be further subdivided according to increasingly subtle differences in the body plans and other details. For example, vertebrates have long been divided into fish, amphibians, reptiles, birds, and mammals — although current thinking is that birds and reptiles should be grouped together.

For classifying plants, a different organizing principle was chosen. Instead of grouping plants according to their body plans, they were grouped according to their reproductive structures — in other words, by their flowers or equivalent organs. If you have never studied botany, then this may seem an odd decision. After all, for most of us, our top-level criterion for categorizing plants is to divide them into trees versus smaller plants. Because the official classification system was based on reproductive structures instead of body plans, a single family of plants might include trees and non-trees. For example, every plant with a pea-like flower was put into the family Fabaceae, regardless of whether the species is a tree or a small herbaceous plant.

Note, however, that the “body plan” system for categorizing animals and the “reproductive structure” system for categorizing plants are both based on morphology — that is, the shape of the organism or of key parts of that organism.

After Darwin, as biologists came to accept evolution as the driving force behind the diversity of life, they saw that our system of biological classification provides a rough approximation of actual evolutionary relationships. It turned out that the Linnaean methods for categorizing plants and animals were fortunate choices. For example, we now know that the reproductive structures of plants are powerful clues regarding their evolutionary relationships. Biologists came to feel that any ongoing updates to the taxonomic system ought to reflect our latest understanding of phylogeny (evolutionary relationships). In addition to morphology, crucial data in the 19th and 20th century came from two other lines of inquiry:

• Historical geology (especially the analysis of fossils and the surrounding strata)
• Developmental biology (the way that various multicellular organisms grow from one-celled zygotes into complex multicellular structures)

By the 1970s another important tool was biochemical analysis — looking for similarities in the natural compounds found in related species, along with similarities in the biochemical pathways employed by the organisms. In recent years we have entered an era in which we can directly compare the genomes (the complete DNA) of two organisms — providing us with the best tools yet for discerning the evolutionary relationships between species. These tools now allow scientists to revise our biological classification system at an unprecedented pace.

As it turns out, all living things share certain genetic similarities with all other living things. For example, humans and earthworms may seem quite different, but we share a lot of genes. Humans share even more genes with fish, even more with lizards, even more with wolves, and even more with chimpanzees. So this gives us a good idea of how related we are to each of these other creatures. Also, it’s not just a matter of how many genes are shared. The shared genes exhibit slight differences between species, and the details of these differences provide great insight into evolutionary relationships. Even within a single species (such as humans), the DNA varies from one individual to another, and therefore comparing DNA is helpful for determining the degree of relatedness. This is especially true when comparing the entire genome, including the parts that don’t correspond to any genes.

What About Foxes?

Let’s now return to our discussion of foxes, to consider how we might distinguish a fox from all other animals.

People living in northern Europe, as Linnaeus did, tended to encounter just one species of fox (the Red Fox) and one species of wolf. The only similar animals in the vicinity were domesticated dogs. This made it rather simple for the average person to distinguish between foxes, wolves, and dogs. But if you consider the entire world, then there are dozens of similar species to consider. Worldwide, there are about 20 species that include “fox” as part of the common English name, including the Red Fox, Gray Fox, Arctic Fox, Kit Fox, and so on. There are several species of animals called wolves, several species called jackals, and several species called dogs. (These wild “dogs” are quite distinct from domesticated dogs, which are descended from wolves.) Altogether, there are around 35 species in this family of animals, which is called Canidae (or informally, the “Dog Family”).

The biological classification of Canidae started out as a formalization of commonsense concepts. The family was divided into several genera, such as Vulpes for foxes and Canis for wolves. As additional canid species from around the world received scientific scrutiny, some were placed in Vulpes or Canis, but many were placed in other genera. Currently Canidae is divided into 13 genera, several of which contain only one living species each:

The grouping of the 35 living species of Canidae into 13 genera provides a rough idea of how the species are related to one another. However, the fine evolutionary detail is completely missing. Contrary to the implications of this diagram, these 13 genera are not all equally related — some pairs are much more closely related than other pairs. The same issue exists within the genus Vulpes or any other genus that has more than two species — some pairs of species in the genus are quite closely related, while other pairs are less related.

To illustrate the evolutionary relationships within any group of living things, scientists now use a clade diagram. Such diagrams are a cornerstone of the new science of cladistics. For example, the diagram below shows how the 13 genera of Canidae are related to one another, according to a recent interpretation of the genetic data:

As you examine this diagram, you can trace a path from the box that says Canidae to any box that represents a genus. Each fork in the path represents the last common ancestor for two or more genera — in order words, it represents a point where a single ancient species diverged into two species. Note that the 13 genera did not split off from each other all at once. It would be more accurate to picture them splitting off one by one. You can see that the genus most closely related to Canis is Cuon, and the genus most closely related to Vulpes is Nyctereutes. But what else does this diagram tell us? And if this is a clade diagram, then what exactly is a clade?

To define a clade, you choose a single species that lived sometime in the past. That species, plus all the species that descended from it — including both living and extinct species — together form a clade. For example, all 12 living species in the genus Vulpes are descended from a single ancestral species of fox that lived about 9 million years ago, and all known descendants of that ancestor are included in this genus. Therefore the genus Vulpes is a clade. All 35 living species in the family Canidae are descended from an ancestral species of canid that lived about 30 million years ago, and all known descendants of that ancestor are included in Canidae. Therefore the family Canidae is also a clade.

However, a clade doesn’t have to correspond to a genus, a family, or any other of the 7 levels in the traditional system of biological classification. For example, in the diagram above, a clade named Vulpini consists of the three fox-like genera shown in purple: Vulpes, Nyctereutes, and Otocyon. Contained within that clade is a smaller clade consisting of just Vulpes and Nyctereutes. In fact, every horizontal black line in the diagram corresponds to a clade.

Of course, this only works if each of the genera listed above is a genuine clade. The past twenty years have seen many changes in how the 35 canid species are grouped into genera, in part to ensure that each genus does indeed correspond to a clade. Biologists now take it for granted that all genera and all families in our standard classification system ought to reflect actual clades. Whenever new research shows that a currently defined genus or family is not a clade, then taxonomists soon revise that genus or family — provided that enough genetic data is available to do so.

The clade diagram below is adapted from a 2012 paper (and is therefore just slightly out of date). It illustrates the evolutionary relationships between 11 of the 12 living species in the genus Vulpes:

As you can see, the genus Vulpes includes many smaller clades. For example, the Arctic Fox, Kit Fox, and Swift Fox together form a clade. The last common ancestor of these three species lived around 4 million years ago, much more recently than the last common ancestor of the entire genus. Note that every species is a member of countless different clades. Pick any ancestor of that species, and you have just defined a clade that includes your lucky species. Pick a different ancestor — and you have just defined a second clade that is either a subset or a superset of your first clade.

So far in this discussion, the diagrams have only shown extant (living) species. But a clade also includes all the extinct descendants of the ancestral species. It can be difficult to identify and properly categorize all the extinct species in a family, but Canidae is known to include well over 100 extinct species, many of which are in genera that are also extinct. Extant genera can also contain extinct species. For example, the genus Nyctereutes contains only one living species (the Raccoon Dog), but scientists are aware of 7 extinct species in the same genus.

Is There a Scientific Definition for “Fox”?

After all this discussion, we still don’t have a precise definition of what a fox is. Any dictionary will define the word “fox” — but that definition is unlikely to be useful for distinguishing a fox from other types of canids. One way out of this dilemma is to say that any creature that has “fox” in its common name must be a fox — but then you have issues such as the Flying Fox, which is actually a bat, not a fox.

In fact, there is no precise scientific definition for the word “fox” — but science will get you partway there. At the very minimum, the 12 species in the genus Vulpes are certainly all foxes. These are sometimes called the “true” foxes, because they share a genus with the Red Fox, the original fox of the Linnaean classification system. If you include the 6 species in the genus Lycalopex, the South American foxes, then you’ve got 18 species of fox — although they don’t form a clade. But this still omits several other canids that are popularly called foxes, such as the Gray Fox (Urocyon cinereoargenteus). Note that the most recent common ancestor of the Red Fox and the Gray Fox is ancestral to all living species in Canidae, including wolves and jackals.

Because foxes don’t form a clade, we can’t use evolutionary relationships to define the word “fox”, unless we limit our definition to the “true” foxes in the genus Vulpes. The alternative is to use morphological criteria. For example, we might say that a fox is any canid that has a narrow snout, a bushy tail, and whose adult height at the shoulder is less than a certain maximum. However, any such definition is likely to have exceptions — foxes that fail to meet the criteria, or other canids that meet the criteria. The upshot is that it is difficult to provide an unambiguous scientific definition that distinguishes foxes from all other canids. This doesn’t mean that you shouldn’t use the word “fox” — it just means that two reasonable people might disagree as to whether a particular species is covered by the term.

So feel free to argue that a Gray Fox is indeed a fox, or to argue that a Gray Fox is not a “true” fox. There are valid lines of thought behind both arguments. We use a lot of words that have fuzzy boundaries, and yet many of these words are still useful. The word “tree” is quite useful, even though there is a somewhat fuzzy boundary between trees and shrubs — and neither term is of much use for grouping plants according to their evolutionary relationships. As for me, I will continue to call a Gray Fox a type of fox — because to me it is indeed a fox!

For additional reading on topics related to evolution, see the chapter “Survival of the Fittest” in my book The Stickler’s Guide to Science in the Age of Misinformation.