As global climate change and human activities have altered the balance of marine ecosystems, there is great concern that the intrinsic capacity of reef-building corals to adapt and survive would be overwhelmed by the profound environmental change. However, beside adapting to environmental changes with their multiple approaches to nutrition and flexible reproductive patterns, corals might also form species complexes by species differentiated in morphology, genetics, and environmental adaptability, so as to response to environmental changes at the metapopulation level. The broadcast spawning reef-building coral Galaxea fascicularis is the representative species in the South China Sea. According to the morphology of MpM nematocysts and the indels in the non-coding region of the mitochondrial genome, Galaxea fascicularis was differentiated into two species (S and H) with different characteristics such as growth and heat resistance. Overall, the ratio of S/H is close in Xisha Islands and Zhongsha Atoll, but in Xisha Islands populations (XA, n=84), H was dominant (82.1%), while in Zhongsha Atoll populations (MA, n=55), S was the majority (78.2%). Ten microsatellite markers were used to analyze the genetic structure of 5 XA and 11 MA populations. The results showed that: allele richness, observed and expected heterozygosity all revealed a moderate genetic diversity. Overall, the genetic diversity of S populations was lower than that of H populations, and that of XA was lower than that of MA.  



    AMOVA analysis showed that the genetic differentiation (F_ST) between S and H populations was large (>0.25) both within and between regions, indicating a reproductive isolation between them. In contrast, the F_ST between XA-S and MA-S was only 0.069, which was significantly smaller than the F_ST between XA-H and MA-H (0.108). Accordingly, the Neighbor-Joining phylogenetic tree and the principal coordinate analysis (PCoA) showed that all the S populations converged into one branch, while H populations were divided into two branches (XA-H and MA-H), indicating that H populations may be geographically isolated between Xisha Islands and Zhongsha Atoll. The STRUCTURE results supported that H populations were divided into two sub-groups, MA-H and XA-H. MA-H had almost no genetic structure and had little gene flow with XA-H, while XA-H showed gene flow from MA-H. As for S populations, they were not divided into subgroups by region, and the gene flow was strong (3.37) between XA-S and MA-S.



    Considering the difference in depth between XA (0-20 m) and MA (20-30 m), H has the advantage of heat resistance so that it prevailed in the relatively high temperature environment of XA, and forms a unique subgroup (XA-H). In MA-H, the heterozygote excess (Ho/He, 0.62±0.05/ 0.55±0.13) probably implies that MA-H was subjected to balancing selection. The coral reef in Zhongsha Atoll is relatively deep, and the heat pressure might be lower than that of Xisha Islands, so S (relatively intolerant to heat stress) could become the dominant population in Zhongsha Atoll because its advantages of growth and asexual reproduction, which supported the recruitment and genetic structure of XA-S through the larvae’ gene flow. In summary, the members of Galaxea fascicularis species complex showed a flexible ability to co-adapt to environmental changes with their complementary ecological characteristics in the central South China Sea.