The rock band theory of morphology  (2016.07.22 [Fri.])

 I’ve been thinking about the “identity” of music groups . Take the Beatles—John Lennon, Paul McCartney, George Harrison, and Ringo Starr. If you asked most people about who was the indispensable member, they would probably say that they couldn’t conceive of a Beatles reunion after Lennon’s death. That much seems clear. But whether that is generally true for other bands is a different question.

 The Eagles, for example, got back together and even after the death this year of core member Glenn Frey, the three band mates Don Henley, Joe Walsh and Timothy Schmit are still touring. Don Felder could also probably be persuaded to rejoin, as could the second bassist, Randy Meissner, and maybe even Bernie Leadon, the banjo player on the first two albums. But if Don Henley were (perish the thought) to die, that would mean no vocalist for classics like “Hotel California” and “Desperado,” which I believe would also be the end of the Eagles as a band. (For me, that may have already happened once they lost Felder’s guitar.)

 The Doobie Brothers provide an even more extreme case in point. Of the founding members, Tom Johnston, Patrick Simmons, John Hartman and Dave Shogren, Simmons is the only one to have stayed with the band throughout its long career. Now, we can again hear Tom Johnston’s vocals and guitar but from 1979 to 1982 Patrick Simmons was the only original member still in the group, and they even disbanded for a time around 1990. But no one would say that the Doobies were all about Simmons; rather, it seems the core musicians changed over time. After Tom Johnston fell sick in Phase I, Jeff “Skunk” Baxter and Michael McDonald stepped up and the band began to sound like Steely Dan, a massive change in the group’s style. (And when I say fundamental, I mean it was as if Kyary Pamyu Pamyu were to join the Doobies from tomorrow.) So while the Doobies have persisted in name, the membership and indeed the group’s musical signature have moved on. The only constant has been change.

 Seen in that light, the irreplaceability of the members of the Beatles may be the exception to the rule. (In fact even the Beatles had some changes in members when they were just starting off.) Cases like this in which the members are identified 100% with the band are rare, with maybe the only comparable instances being bands like Loggins and Messina or Simon and Garfunkel, which are named after their members. (I suppose the Everly Brothers and The Carpenters fit this pattern as well.)

 So, why all the musical background? Well, it strikes me that there are some real parallels between the identity of morphological structures we see in organ homologies throughout animal evolution and that of rock bands over time. Take body segments. The parasegments in insect embryos are controlled by segment polarity and pair-rule genes and segment structural patterns, but that doesn’t mean that these genes, which were first studied in Drosophila, play the same developmental roles in every insect. What’s more, in some arthropods (such as spiders) Notch-Delta signaling has functions that entirely absent in insects. Within the Arthropoda, the insects are something an outlier, and flies are stranger still. That said, flies did not acquire a completely unique form of body segmentation, and it is accurate to understand their segmentation in the fly as a homologous continuation of the broader arthropod routine. (And with the knowledge that Notch-Delta functions in vertebrate body segmentation as well, some recent thinking has it that segmentation is a homologous feature across all bilaterians.)  The upshot of this is that, while morphological homologies may be conserved, the underlying developmental programs (i.e., substructures) can change. Thus, oftentimes the genes that give rise to homologous structures are not themselves homologous. In evo-devo this is known as “developmental system drift.”

 This occurs because, over generations, mutations appear in the genome and in the resultant developmental programs, but at the same time the phenotype exerts string selection pressure on what morphological patterns can emerge. It’s as if evolution tells us, “Change what you like, just don’t mess with the segmentation pattern.”

 Of course in some cases even the phenotype may change within the limits of the system’s functionality. For example, in North America there is a locust that vocalizes using its jaw, due to changes in wing morphology that made it unable to emit chirps in the manner of its ancestors, by rubbing the hind limbs against the wing. But even now that it calls using its jaw, its back leg still pumps up and down against the empty air. In its ancestors, this motion was part of a courtship behavior that produced sound and apparently was successful in attracting mates, as evidenced by the emergence in later generations of jawed locusts, which continue to emit calls as part of their inherited reproductive behavioral system. The song remains the same, only the choice of instruments has changed.

 So for homologies in organ or behavior, there is no simple link to their ontogenic program, selection logic, or genetic sources. Instead, all are layers in what can only be understood collectively as a “continuous module” (akin to a band identity), in which the constituent elements nonetheless enjoy relative freedom. But among these there are also those genes, structures and morphological patterns with an obstinate identity that persists across generations. Perhaps it would be appropriate to think of the extraocular muscles (six muscles that control the movement of the eye) as a Beatles-like system in our own bodies. In which case, from now on I’ll be calling my m. rectus lateralis “Paul.”