Wednesday, June 4, 2014

Jerry Coyne Doubts That Symbionts Cause Speciation. Let's Discuss.

Any new or unexpected concept/data will be confronted with criticisms and doubts. But over roughly the 20 years that I have been studying speciation by symbiosis (TREE 2012 review), I found that the most common criticism is the least formidable. It concerns the fact that symbionts have not been shown to be a "cause of speciation", and it has just been used again by evolutionary biologist Jerry Coyne. It is time to clear this up once and for all to generate some unity around speciation by genetics and symbiosis. I have been deeply influenced by Coyne's seminal work on speciation genetics. I admire his research, yet he has not engaged me in discussions on speciation by symbiosis despite my attempts to share/discuss our work with him. I welcome the opportunity to debate him on this issue.

From Quanta Magazine, "There are no studies as of yet that prove definitively that the microbiome causes speciation," Coyne states.

Note: I previously addressed this concern in a 2003 book chapter on Symbiosis And The Origin of Species.

The criticism, while true, fails to distinguish speciation by symbiosis from speciation by nuclear genes. Rarely do speciation workers know whether the particular isolating barrier that they are investigating is or was a cause of speciation. Speciation genes found in the nucleus make a big splash in evolutionary biology, and rightfully so, but often they are found in such old species pairs (i.e., Drosophila melanogaster and D. simulans diverged millions of years ago) that researchers can not claim they are a cause of speciation as well. Read that twice.

The concern by Coyne and others is a more general and practical problem to the study of speciation, as the formation of species takes a long time. Whether or not microbial symbionts prove to be a causative agent now is not necessarily an indicator of whether symbionts have played a role in speciation in the past or will play a role in the future. The same holds true for speciation genes in the nucleus. Most of the time, the best that biologists can do is discover the factors that cause reproductive isolation and thus the entities (genes or microbes) that can cause speciation, rather than cause speciation. 

Most reproductive isolation likely evolves in allopatry as an incidental byproduct of divergence. Simply demonstrating the genetic or microbial basis of speciation may be the closest we get to this issue. In the Quanta Magazine, I think Bradford Harris gets it right when it comes to why we see the inequality in judging speciation studies.


  1. This actually seems like a more experimentally tractable problem than most, in the field of speciation. There are two parallel arenas of divergence--in the host genes (from drift and selection), and in the microbiota. What happens if you hold host genetic background constant, and modify the microbiota? Conversely, what happens if you choose a few recently-diverging species pairs, cure them of their microbes, and re-infect with a shared community? And what happens to the progeny several generations down the line, if the original microbiota is reintroduced? This is a great opportunity to introduce experimentation where descriptive studies and inference have dominated. But I'd also like to see large-scale surveys of the microbiota within host species pairs across their range(s)...there's an analogy to the delineation of subspecies here: only with a big survey of organisms in the wild can you get at whether variation is partitioned meaningfully between OTUs; in the case of the microbiota, differences that appear important in the lab might dissolve with increasing sample size and a real ecological context. How about sampling wild populations of Nasonia? Or...say...foliar endophytes of woody plants in the Amazon basin? ;-)