A pre-zygotic barrier to hybridization in two con-generic species of scleractinian corals

Hybridization is often cited as a potential source of evolutionary novelty in the order Scleractinia. While hybrid embryos can be produced in vitro, it has been difficult to identify adult hybrids in the wild. Here, we tested the potential for hybridization between two closely related species in the family Fungiidae. We mixed approximately 5000 eggs of Ctenactis echinata with sperm from C. crassa. No hybrid embryos were produced. This observation adds to a growing body of evidence for pre-zygotic barriers to hybridization in corals and challenges the claim that hybridization is a major source of evolutionary novelty in the order.

Hybridization is a controversial topic in coral reef ecology 1,2 . While small numbers of hybrid embryos can be produced in a few species in vitro 3 , the evidence for hybrids in the field is often equivocal because the genetic techniques used for corals cannot distinguish between hybridization and incomplete lineage sorting 4 . In fact, only one of the over 1300 species in the order is generally accepted to be unequivocally of hybrid origin: Acropora prolifera 1,5 . Nonetheless, hybridization is often invoked as a source of evolutionary novelty in the order Scleractinia 6,7 .
Here, we report an incidental observation on the potential for hybridization between two closely related scleractinian corals species in the family Fungiidae, Ctenactis echinata and C. crassa. These species are sympatric, often dominating large multi-specific assemblages of fungiid corals throughout the central Indo-Pacific 8 . These species can generally be distinguished on the basis of the density of septa and the shape of septal dentitions, however, in Okinawa, these features are very similar and the most useful diagnostic character is a strong arch in the corallum of C. crassa ( Figure 1A and B) 9 . Both species are gonochoric, that is each colony is either male or female, and reproduce by broadcast spawning, releasing gametes into the water column for fertilization 8 ( Figure 1C and D). At our study site on Sesoko Island (26°38'13.00"N; 127°51'56.24"E), Okinawa, Japan, spawning occurs following the full moons from July to August 8 . Furthermore, both species release gametes at the same time 8 and consequently there is the potential for hybridization. In the days before the predicted date of spawning in July 2013, we collected four colonies of C. echinata and six colonies of C. crassa, to produce larvae for other experiments.
While the species are relatively easy to identify, determining the sex of each individual prior to spawning is impossible without destructive sampling to expose the gametes. Consequently, we placed each individual in a separate 20 L bucket containing sea water in the open air at approximately 20:00 h in order to sex each individual once gametes had been released. On the night of 27 July between 22:30 and 23:30 h three C. echinata and five C. crassa spawned revealing that the three spawning C. echinata were female, while four C. crassa were females and one was a male. The size of the eggs of each species at the time of release was distinct with a range in maximum diameter of 244-266 μm in C. echinata and 133-155 μm in C. crassa. In contrast to earlier work on C. echinata 10 , we saw no symbiotic algae in the eggs of either species. We collected approximately 5000 eggs from the three C. echinata females and mixed them with sperm from the C. crassa male. The viability of the C. crassa sperm was tested by mixing it with C. crassa eggs, however, we could not quantify the viability of the C. echinata eggs because no C. echinata sperm was available on the evening of the experiment. Nonetheless, eggs from these colonies of C. echinata did produce viable larvae for use in later experiments. Approximately 100 eggs were observed under a stereo-dissecting microscope for cleavage, indicating fertilization, every 2 to 6 h over the next 24 h. At no point did we observe cleavage in the cross between species indicating that no hybrid embryos were produced and none of the approximately 5000 eggs remained intact after 24 h. In contrast, over 90% of C. crassa eggs in the positive control were fertilized within 2 h. We conclude that despite synchrony in the time of gamete release between these two closely related sympatric species there appears to be strong pre-zygotic mechanism to avoid hybridization. While our observations are preliminary and in only one direction (i.e. we did not cross C. echinata males with C. crassa

Amendments from Version 1
In response to feedback from the reviewers, we have updated our Observation Article as follows. In response to comments by Yossi Loya, we now note in the text that we did not quantifiy the viability of the Ctenactis echinata eggs by crossing them with C. echinata sperm because no sperm was available at the time. We have also corrected one citation. In response to a comment by Bernie Degnan, suggesting we had only looked at 100s of eggs, we now state that none of the approximately 5000 C. echinata eggs exposed to C. crassa sperm were fertilized. In response to email correspondence with Bert Hoeksema, we have changed the name of the species in Figure 1 C & D from Fungia repanda to F. fungites. We have also provided more detail on how the Ctenactis species were identified and have corrected the relevant citation.

REVISE D
females) we predict that hybridization between these species is unlikely. This observation adds to a growing body of evidence indicating strong pre-zygotic barriers to hybridization in many scleractinian corals [11][12][13] .
Author contributions AHB, VRC & JF conceived the study and performed the experiment. All authors contributed to writing the manuscript.

Competing interests
No competing interests were disclosed.
authors recognise the limitations of this observation -only small numbers of eggs (100's) were observed and only eggs from were available to be fertilized (i.e. they did not cross males C. echinata C. echinata with females) -but rightly point out that this study provides further evidence that hybridization is C. crassa not as widespread amongst scleractinian corals as often portrayed in the literature. However, the analysis of reciprocal crosses and a larger numbers of eggs is necessary before it can be said with some confidence that these congeners can not hybridize.
I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
No competing interests were disclosed. Data and Conclusions: The authors note that they did not cross males with C. echinata C. crassa females; however they also did not test the positive control of crossing males with C. echinata C. females (all their experimental specimen were females). Nevertheless, this echinata C. echinata does not diminish their prediction that hybridization between these species is unlikely. The paper does not diminish their prediction that hybridization between these species is unlikely. The paper contributes further information to the controversial topic of potential hybridization and breeding incompatibilities within the mating systems of broadcast spawning reef corals.
I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
No competing interests were disclosed. Thank you for your comments. We have corrected the typos you identified, changed the reference as requested and added a sentence to clarify the controls that were used to test for gamete viability. Images 1 C & D are presented as an example of the spawning behavior of fungiids. none Competing Interests: