Reproduction, larval dispersal and population genetics of the sponge Rhopaloeides odorabile

Reproductive strategies and larval dispersal behaviours are key factors in the distribution patterns of marine taxa. Furthermore, our understanding of population distributions, and the processes that influence and maintain them is fundamental to the conservation and management of marine benthic environments. Whilst reproductive and larval dispersal processes are well documented for a range of sessile marine invertebrates our understanding is predominantly related to corals. For other dominant sessile invertebrates, such as sponges, these processes are less well understood, even though they form an important component of the benthos. Sponges are found from polar to tropical seas, in both marine and freshwater environments, with an estimated 15, 000 species worldwide. Their ecological roles are equally diverse with functional roles implicated in reef consolidation, and bio-erosion, habitat refuges, and benthic-pelagic coupling processes. Despite their presence and important functional roles, our understanding of processes that influence populations are still unclear. This study contributes to this understanding by quantitatively examining the reproductive processes, larval dispersal behaviours and population genetic structure of Rhopaloeides odorabile (Dictyoceratida; Demospongiae), a common sponge species of the Great Barrier Reef (GBR). Sexual reproductive processes of Rhopaloeides odorabile were quantified using histological sections of samples from Pelorus Island collected over two reproductive seasons. Further sampling was undertaken to compare reproduction of populations from different locations including Pelorus Island (inner shelf reef), Rib Reef (mid shelf reef), and Pith Reef (outer shelf reef). Rhopaloeides odorabile is viviparous and gonochoristic. Commencement and cessation of gametogenesis coincides with rising and falling sea surface temperatures (≈ 20-28oC). Spermatogenesis occurs from September until December. Females initiate oogenesis in October with the asynchronous development of oocytes, embryos and larvae occurring within brood chambers. A larval release period of 5-6 weeks occurs during January and February. There is a clear gradient of female reproductive effort across the shelf reefs with reproductive output increasing with increasing distance from the coastline. Female reproductive effort for R. odorabile varied significantly across the shelf reefs on two levels. Firstly, sponges from the outer and mid reefs have a reproductive output, per unit volume, of up to 15 times more than inner reef sponges. Secondly, in conjunction (and co-related) to the reproductive output is investment into the size of oocytes. Mean oocyte diameters increased in size from inner to mid and outer reefs. Inner reef sponges recorded mean oocyte sizes of 156 μm (±9.4) compared to mean sizes of 203μm (±8.7) and 195μm (±8.9) for mid, and outer reef sponges respectively. Therefore, both increased egg size and egg numbers contribute to a much larger reproductive output. Reproductive output for inner shelf sponges may be compromised from exposure to terrigenous run-off resulting in sub-optimal habitat conditions. The poor motility of many sessile marine invertebrate larvae makes active dispersal and habitat selection challenging. Manipulative experiments were undertaken using larvae from R. odorabile to quantify behaviours associated with vertical migration, phototaxis, swimming ability and settlement responses to cues associated with light, settlement surface topography, coral rubble and biofilms. Following an afternoon release larvae are cued by light and migrate vertically to the surface for 6 - 18 hours. From 24 hours larvae move from the surface to the bottom, maintaining this position for up to 54 hours before settling or dying. Larvae do not display gregarious settlement patterns or a preference for settlement surface topographies but do preferentially settle to light exposed surfaces rather than shaded sites. Settlement to individual treatments of biofilms or coral rubble was higher when compared to controls with no cue. In addition, higher larval settlement to treatments containing non-sterile coral rubble compared to those containing sterile coral rubble demonstrates settlement is almost certainly a response to the biofilm, rather than the physicality associated with coral rubble. However, the transition from initial settlement to metamorphosis was higher when treatments comprised a combination of biofilm and coral rubble substrate suggesting a combination of cues is responsible for optimum recruitment. Therefore, vertical migration to surface waters is likely to facilitate passive dispersal over medium to large distances via wind driven surface currents. Larvae have the ability to swim and explore the substratum on their return to the benthos. The importance of settling to light exposed, biofilmed substrata facilitates settlement to microhabitats favourable to continued survival and growth (recruitment). Population genetic results from allozyme and mtDNA data of R. odorabile support genetic differentiation at both the broadest (i.e. 10's kms), and smallest spatial scales (1 km) examined in this study. In addition, levels of genetic connectedness are also observed across the same spatial scale comparisons, providing an overall pattern of chaotic genetic structure. Whilst extensive larval dispersal would explain genetic homogeneity over broad spatial scales, fine scale population subdivision would also suggest larval dispersal is restricted between some of the sites examined. The behavioural strategies of vertical migration, coupled with pre-competency periods of up to 54 hours for R. odorabile larvae, provide a mechanism for passive dispersal via current systems throughout the sampled sites in this study. However, recognising that the GBR comprises a patchy mosaic of reef systems, many associated with self-entrained circular or eddy current systems, is of critical importance in explaining patterns of genetic variation, and therefore larval dispersal patterns. It is proposed that the degree of larval dispersal is subject to reef specific hydrodynamics, and current patterns at the time of larval release, explaining the genetic patchiness observed in this species. Varying intensity of current regimes at spawning times will lead to localised recruitment at some sites or the occurrence of random or occasional dispersal over extensive distances, thereby explaining genetic homogeneity between more distant sites. Although R. odorabile broods larvae, larval dispersal as inferred from population genetic data, and coupled with larval vertical migration behaviour, suggests larval dispersal can occur across moderate spatial scales beyond natal reefs. Overall, the variable patterns of reproductive output together with seasonality demonstrated to date for sponges, make generalisations of trends difficult, and illuminate the complexity in reproductive ecology for Porifera. Importantly, recognising the significance of reproductive processes for benthic invertebrates, will contribute to our understanding of the roles they play in the recruitment and maintenance of populations. This is particularly relevant for sponges, which are a major part of many benthic environments, but attract a comparatively minor focus of research effort to manage and conserve reefs