The Evolution of a Short Back

One of the issues raised by the recent Sarmiento comments is that of the Miocene apes and the evolution of a short back. All extant apes possess a “short back,” by which we mean a reduction in the lumbar spine combined with an upward elongation of the blades of the pelvis.  This back is a nice, stable structure that allows large-bodied suspensory primates to withstand bending moments without breaking their backs.

Adolph Schultz's drawings of backs. Note the extreme shortening of the back in Pan, achieved by reducing the number of lumbar vertebrae, extending the ilia toward the head, and "capturing" the lumbars between the ilia. Hylobates shows some of these adaptations as well.

Old World Monkeys all have 7 cervical (neck) vertebrae and 13 thoracic vertebrae. In the lumbar region, the number is either 6 or 7, and then they possess a sacrum made of a few vertebral segments, plus a tail.

Apes kept this formula for a while (Proconsul had 6 or 7 lumbars 23-17 MYA, and so did Nacholapithecus at around 15 MYA). Oreopithecus had 5 lumbars, and moved one of them down to the sacrum so that it had 6 sacral vertebrae.

An interesting thing about vertebral formula is that it is actually fairly polymorphic.  The typical human formula is 7-12-5-5, but I know someone with a 6th lumbar, and it’s fairly common for people to have 6 sacral segments. Some people have 29 vertebral segments, while others have 30. Pretty neat. The other apes have this kind of variation, as well.

The most common formula for gibbons is 7-13-5-5 (total of 30) (Hylobates lar) or 7-13-4-5 (29) (H. syndactylus).  Orangs come in at 7-12-4-5 (28); Gorillas are at 7-13-4-5 (29), Common chimps (Pan troglodytes) have as their most common formula 7-13-4-6 (30), as do bonobos (Pan paniscus).  Humans, again, usually have 7-12-5-5 (29), and Australopithecus had 7-12-6-4 (29)

It gets interesting when you look at the differences in variation for chimps vs. bonobos.  After 7-13-4-6, the most common counts for chimps are 7-13-4-5 and 7-13-3-6 (Total number of vertebrae is 29).  Bonobos have 7-13-4-7, or 7-14-4-6 (Total # 31).  Bonobos have at least one more vertebra than chimps and gorillas a significant amount of the time.  In fact, if you pick a random bonobo out of a drawer and compare it with a random chimp, the probability that you’ll get a bonobo with more vertebrae is 87.7%!

So how do we interpret that number?  Well it certainly calls into question the idea that chimps and gorillas share their lumbar formula with a common ancestor.  If anyone should be similar, it’s the two species of chimpanzee, right?.  But the two chimps are more different from each other than common chimps and gorillas!  David Pilbeam (2004) calculated a “Similarity index” based on vertebral counts and found that Chimps and Gorillas have a SI of .86.  But the two chimps have an SI of only .39.  It’s possible that bonobos have reverted back to a more ancestral form, but could there be a selective advantage for having a longer back?

There are other differences between chimps and gorillas, too.  We’ve been talking about the shortening of the back, but another important aspect to stabilization of the back is a lengthening of the the blades of the pelvis called the ilia.  In gorillas, the ilium is extended so that its anterior border extends to the topd of the last lumbar vertebra.  In chimps, the border of the ilium extends further, so that the last two lumbar vertebrae are “entraped” by the ilium.  Orangutans share this feature with chimps. This may seem trivial, but it demonstrates that animals with more active locomotor regimes can converge on the same degree of entrapment.

It’s looking like there’s a lot of homoplasy going on here.  Superficially, it might appear that all extant apes have the same short back, but if we tease apart different aspects of that back, we find a much more complicated picture. Parallelisms abound, but the taxa in which we find them don’t make sense if we try to reduce them to common ancestry.

It’s a lot of potential homoplasy- but is it an unbelievable amount? Consider that in the spider monkey, the lumbar region is also shortened and entrapped by the ilia.

McCollum (2009) brings up a good point as well:  If humans evolved from a short-backed ancestor, they would not have been able to lordose whatsoever.  A short back is remarkably specialized for suspensory locomotion and isnt’ flexible at all.  When short-backed primates try to walk bipedally, the have to adopt the bent-knee, bent-hip gait.  On the other hand, when you train a long-backed macaque to walk bipedally as part of a kabuki drama, they develop lordosis and a bicondylar angle. Bipedal locomotion and suspension are different enterprises entirely.

Proconsul, our basal ape, had a vertebral formula of 7-13-6/7-4/5, for a total number of 30-31.  It had a flexible back and a short sacrum, just like humans do now.  If we accept that this body plan was retained until the LCA between chimps and humans, it means that the extant apes evolved short backs independently, but since they all have interesting little differences, I don’t think that this is too much of a stretch. We all started out with a Proconsul-like body, and there are only so many things you can do with a body like that.

This is especially true when you start to select for the flexible shoulders required of careful climbing and clambering, as we already see in Morotopithecus 20 mya.  The flexible shoulders that we see in extant and extinct apes are achieved by moving the shoulder sockets so that they point laterally, which requires restructuring of the back.  As a result, a mass of muscles and tendons which help support the back- the errector spinae- is reduced.  If our ape friends already committed to that reduction in muscle mass and back support because they needed a flexible shoulder, they need to compensate for that reduction in support in another way. It’s no surprise that we see convergences like those in the shortened back evolve several times in the hominoids.  Large-bodied, weak-backed apes who are experimenting with suspension require that support, and there is strong selective pressure for it.

If we don’t accept that the last ancestor we share with chimps had a long back, then we would have evolved a short back, reverted back to a long back, and only then would our backs have been free to lordose.  We’d have had a long pelvis, and then we would’ve had to have evolved a shorter pelvis than any other primate, licketty split.  We would have evolved lumbar entrapment, followed by lumbar liberation.

Not entirely impossible, but it seems like a lot of reversions. The beauty of it is that we don’t even need to look at the fossil record.  Even if we are just focusing on the comparative anatomy of living apes, we can detect subtle differences between them, and shared retentions between humans and monkeys.

McCollum MA, Rosenman BA, Suwa G, Meindl RS, & Lovejoy CO (2010). The vertebral formula of the last common ancestor of African apes and humans. Journal of experimental zoology. Part B, Molecular and developmental evolution, 314 (2), 123-34 PMID: 19688850

Pilbeam, D. (2004). The anthropoid postcranial axial skeleton: Comments on development, variation, and evolution Journal of Experimental Zoology, 302B (3), 241-267 DOI: 10.1002/jez.b.22

Preuschoft, H., Hayama, S., & Günther, M. (1988). Curvature of the Lumbar Spine as a Consequence of Mechanical Necessities in Japanese Macaques Trained for Bipedalism Folia Primatologica, 50 (1-2), 42-58 DOI: 10.1159/000156333

NAKATSUKASA, M., KUNIMATSU, Y., NAKANO, Y., & ISHIDA, H. (2007). Vertebral morphology of Nacholapithecus kerioi based on KNM-BG 35250 Journal of Human Evolution, 52 (4), 347-369 DOI: 10.1016/j.jhevol.2006.08.008

Advertisement