Mammals typically have seven cervical, or neck, vertebrae (left) and 12 thoracic, or chest, vertebrae
Imagine a long-necked animal.
Most people will, I suspect, picture a giraffe. Other likely candidates are swans or long-necked dinosaurs or plesiosaurs; many vertebrates have evolved long, relatively flexible necks, the better to reach food that is otherwise out of reach, or is more easily captured with a mobile head on a flexible stalk.
Now picture the anatomical features that produce that long neck. That's often more difficult if you haven't poked around in comparative anatomy, but basically, the neck has a core structure of bony vertebrae stacked like spools along its length, each one separated from the other by a joint. This series of joints is what gives the neck its flexibility and, as you might guess, the more numerous the joints, the more flexible the structure would be. One simple question to ask is how many vertebrae are present in the neck of these various animals? It's easy to count; just tally up the vertebrae from the base of the skull to the first vertebrae that bear ribs. The ones that lack ribs are the cervical (a fancy word for "neck") vertebrae, and the first ones that have ribs are the thoracic ("chest") vertebrae. Easy, but we get one surprising result.
Plesiosaurs, those aquatic reptiles of the Mesozoic, are impressive: Some species had about 40 cervical vertebrae. Modern birds also have representatives with numerous cervical vertebrae, up to 25 in swans. Diplodocids, the characteristic long-necked herbivorous dinosaurs, had 12 to 13.
Giraffes have seven.
Even stranger, imagine any short-necked mammal—a dolphin, a mole, a cow, a human being—and you get exactly the same answer: Typically, they all have just seven vertebrae, no matter how many millimeters or how many meters long their neck might be. (There are some obscure exceptions: Manatees and two-toed sloths have six cervical vertebrae, and three-toed sloths have nine. Otherwise, seven is the rule for mammals.)
What a curious limitation! Other vertebrate lineages vary the number of cervical vertebrae more freely, presumably as a response to selection for greater length or flexibility, but mammals are rigidly locked in to having just seven, even in situations where increasing numbers might well be adaptive. It's also a specific constraint on only the cervical vertebrae of mammals. You, for instance, probably have 12 thoracic vertebrae, while your dog would have 13, and your horse would have 18. The number of thoracic vertebrae in mammals ranges between nine and 23, but almost always the number of cervical vertebrae is seven.
The question that has probably already popped into your head is "Why?"—what force is it that could possibly be so tightly constraining this detail of morphology in not just our species, but in every species of the order mammalia?
One possible explanation is that it is an example of the "you can't get there from here" problem. There are some changes that are so complex and require so many steps that we do not expect to be able to see any population evolve the new feature in a single or even a few generations—we don't expect to ever see a mammal born with a fly's compound eyes, for instance, because there are too many differences in too many genes to generate any kind of detectable shift toward the fly pattern in any reasonable period of time (evolution, of course, can work over eminentlyunreasonable periods of time). Perhaps the number of cervical vertebrae is regulated in such a complex way that mutations that change the number are absurdly unlikely, and so we see no variations—and since natural selection works only on existing variation, it can get no traction in selecting for fewer or more neck bones in mammalian populations.
This doesn't seem to be the case, for several good reasons. Vertebrae number in general seems to be fairly flexible: Witness the different numbers of cervical vertebrae in birds, dinosaurs, and plesiosaurs, for instance, or the variation in thoracic vertebrae in mammals. It doesn't look as if the developmental mechanisms that count out vertebrae are that complicated or fixed (and I'll be saying more about that specific subject in a future column). The trump card, though, is the observational evidence: Individual people are walking around right now with fewer than seven cervical vertebrae. The presence of ribs on the seventh vertebra (making it by definition a thoracic vertebra) occurs in about 1 percent of the human population. The individual variation is present in mammalian populations, yet somehow neither selection nor drift leads to species with variant vertebrae number.
Another explanation is that selection actively culls out individuals with variant cervical vertebrae number, and the evidence is growing that mammals, including humans, are fiercely selected for precisely seven cervical vertebrae. That may sound unlikely to you—there probably aren't any death certificates out there that specify "too few neck bones" as the cause of death—but that's because the deaths are reaped in the population of developing fetuses or by indirect means.
Evolutionary development biologist, Frietson Galis and others examined fetuses that had been spontaneously aborted, or medically aborted due to detected fetal abnormalities, and discovered that an astonishing 55 percent of them had ribs on the seventh vertebra. Recall that adults have this same pattern with a frequency of about 1 percent. It's clear that having six instead of seven cervical vertebrae is associated with a severely deleterious effect on viability. An embryo that forms ribs on the seventh vertebra has only a 20 percent chance of surviving to see its first birthday, relative to an embryo with no such deviation. It's not simply that different numbers of vertebrae are inconsistent with life in a direct way; individuals with such a defect tend to also have many other abnormalities, as if an early perturbation in pattern formation leads to a growing, fatal avalanche of developmental errors.
Even if the fetus makes it to term and is born, there are other dangers. Children with cervical ribs have been found to have a 120-fold greater chance of developing certain cancers over children with seven normal cervical vertebrae.
It's highly unlikely that any benefit from an incremental change in the number of neck bones could compensate for the associated 80 percent mortality rate before one year of age and a much greater chance of cancer! Mammals appear to have evolved an as yet unidentified linkage between the genes that pattern the vertebral column and genes that protect against cancer, or a deeper integration of cervical
patterning genes and other patterning genes such that disruption of one yields wider and intolerable changes in the overall integrity of the developing organism. Those linkages may confer other benefits on the animal, but they also reduce morphological flexibility—or perhaps, to put a more positive spin on it, confer greater reliability and robustness on the process of development.
We're used to thinking of selection operating as a consequence of competition between individuals: The slower gazelle is eaten by the lion, the lions susceptible to a virus succumb to disease, the finches best able to forage on thick-shelled seeds survive a drought. There is also a kind of internal selection going on in development, selecting for embryos that successfully assemble a coherent body plan, with their parts neatly integrated into mutually supporting relationships. Before an organism as a whole can compete in the world outside, its cells and tissues and organs have to learn to dance together, to be a coordinated unit, and the formation of a cohesive individual is the first and possibly most important target of selection in multicellular organisms.