Select the search type
  • Site
  • Web

Student Project

Epifaunal Associations of Pyura praeputialis

Robyn Davies 2020


Pyura praeputialis, a solitary ascidian native to Eastern Australia and Invasive in Chile has been shown to engineer ecosystems by forming dense beds in the intertidal zone in wave-exposed rocky shores. The macro-invertebrate epifaunal communities found on the tunics of these ascidians at two sites in South-East Queensland; Kirra Beach and Point Arkwright were recorded and compared. The findings suggest that there is variation between the communities found at these sites, with more filter feeders found at the site with lower wave energy and higher water temperature. Comparisons with epifaunal communities in the species’ invasive range show similarities in a high proportion of species diversity of moreactive feeders than filter feeders at both sites. However, there was much greater differences in how phyla contribute to diversity between Australian sitesthan sites in Chile. The diversity of species found in small samples of these communities highlights the importance of P. praeputialis for maintaining biodiversity on rocky shores in southern Queensland. Further investigating whether species found in these communities are found elsewhere is informative for making management decisions about how potential P. praeputialis declines could affect biodiversity.


Studying ecosystems dominated by a single species can be useful for understanding the complex and diversity of species interactions that can exist in communities. The species which dominate such ecosystems can sometimes be described as ecosystem engineers. Jones, Lawton et al. (1994) describes an ecosystem engineer as a species which controls the availability of resources to other species by causing changes in biotic or abiotic materials, thereby creating habitat. He provides many examples of ecosystem engineers, from elephants to ants, which change the structure of the environment (Jones, Lawton et al. 1994). A number of solitary ascidian species belonging to the genus Pyura (Phylum: Chordata, Order: Ascidiacea) have been described as model ecosystem engineering organisms.


Pyura praeputialis (Heller, 1878) is a large solitary species native to the intertidal rocky shores of eastern Australia. In its adult, sessile phase, it grows up to 18 cm tall, attached to hard substrates such as rock in areas of strong wave action (Endean, Kenny et al. 1956; Fairweather 1991). Monteiro, Chapman et al. (2002) proposed that the species can exist in either clumped or sparse distributions. In some areas, the species forms Pyura beds, which exhibit the highest biomass density of any intertidal organism in the literature (Rius, Teske et al. 2017).


In the past, Pyura in Australia have been identified as Pyura stolonifera a species originally described in South Africa (Dalby 1996; Davie 2011; Fairweather 1991; Monteiro, Chapman et al. 2002). Recently, DNA sequencing has suggested that P. stolonifera is a species complex including five distinct species (Rius and Teske 2013; Teske, Rius et al. 2011). P. praeputialis is distributed from Double Island Point, Queensland, to the southern coast of Victoria, where it’s range overlaps with Pyura doppelgangera (Endean, Kenny et al. 1956; Rius and Teske 2013). These developments present new questions about the ecology of P. praeputialis in the context of being a separate species to South African species. Furthermore, Castilla, Collins et al. (2002) used molecular techniques to identify a population of Pyura in a 60-70km area in the Bay of Antofagasta, Chile as Pyura praeputialis recently introduced from Australia .


In areas where P. praeputialis has a high biomass and highly clumped distribution, the species can be considered an ecosystem engineer due to its control of the use of substrate by other marine intertidal organisms. In Chile, P. praeputialis was found to constitute 97% of total biomass in the intertidal zone(Ortiz, Campos et al. 2013). Extensive research has therefore been conducted into the ecosystem engineering effects of such a dominant invasive species (Castilla, Manríquez et al. 2014; Cerda and Castilla 2001; Ortiz, Campos et al. 2013; Pacheco and Andrade 2020; Rius, Teske et al. 2017).


P. praeputialis matrices have been shown to have higher macroinvertebrate species diversity than surrounding areas without Pyura (Castilla, Guinez et al. 2004). Proposed mechanisms by which P. praeputialis provides habitat for a higher diversity of species includes protection from predation and desiccation due to higher structural complexity, facilitation by enhanced ecosystem functions and increased recruitment due to higher spatial and temporal variation (Castilla, Guinez et al. 2004). Studies in the native range of P. praeputialis have also recorded associated macro-invertebrate community assemblages (Dalby 1996; Monteiro, Chapman et al. 2002).


This study aims to record the macroinvertebrate communities near the northern limit to the range of P. praeputialis at two sites; Kirra Beach and the more northern Point Arkwright. In particular, Endean, Kenny et al. (1956) proposes a north-south geographic boundary, where richness of temperate species begins declining around the sunshine coast and is replaced by richness of tropical species increases rapidly towards Double Island Point. Whether there is an effect of changing environmental conditions across this gradient on both the growth of P. praeputialis and the epifaunal species is interesting, to measure the impacts of environment on these communities. Furthermore, comparing these communities to those found in recently invaded sites in Chile may provide insight into whether ecosystem engineers have evolutionary effects on the ecosystems associated with them (Jones, Lawton et al. 1994).

Materials and Methods


15 Pyura praeputialis individuals were collected from Kirra Beach and Point Arkwright headland, Queensland (Figure 1) during the ebb tide on April 29th and May 18th 2020 respectively. Both sites were located on rocky headlands near sandy beaches (Figures 2-5). Individuals were selected so that the sample was approximately representative of the size distribution at each site. Ascidians were removed from rocks in the mid intertidal zone using a hammer and chisel and stored in seawater for up to a day before data collection (Dalby 1996).

Data Recording

The epifaunal community composition of each sample was measured by photographing each novel species and counting all individuals found in each sample of P. praeputialis. Each species was identified to the lowest taxonomic level possible using resources including, but not limited to: Davie (2011); (Glasby 2003) and iNaturalist (Cerda and Castilla 2001). The classification “Polychaeta” was used despite its paraphyletic status due to it’s prevalence in the resources used. The length and diameter of each ascidian was recorded when the siphon was retracted (Dalby 1996). Each ascidian was dissected along the mid-sagittal, cutting first between the siphons to sever the ganglia (Figure 6)(Dalby 1996). The length and diameter of the internal cavity of the tunicate was also measured. To measure maturity, gonad index of each individual was scored using the key adapted from Dalby (1996) (1 = gonads absent or gametes absent, 2 = gonads small relative to body and gametes present, 3 = gonads large relative to body and gametes abundant). The abundance and species identity of symbionts inside the branchial sac were recorded.

Environmental Conditions

To compare the environmental conditions at each site, data was retrieved from 2 Queensland Government Waverider buoys near the sites. The Mooloolaba wave monitoring site (26° 33.960' S, 153° 10.870' E) was chosen to estimate Point Arkwright conditions and Tweed Heads wave monitoring site (28° 10.655' S, 153° 34.594' E) to estimate conditions at Kirra Beach. The maximum wave height per 27 minutes from 2000 to 2004 from these locations was used to estimate the wave energy in the area. Sea surface temperature readings at these sites between 2018 and 2020 were extracted to test whether the more northern site could be described as more tropical than the southern site.


Statistical Analysis

All data processing was performed in Excel and RStudio. To determine whether site had any effect on the growth rate of P. praeputialis, the effect of site on gonad index was modelled using an ANCOVA with cavity area, calculated from cavity length and width, as a covariate. Differences in mean external size, calculated from external width and diameter, and cavity area between sites were tested using Welch Two Sample t-tests. Logistic binomial generalised linear models were used to estimate the effect of site on the abundance of species and individuals of different feeding modes. Logistic The effect of site on lifestyle modes at a species and individual organismal level was estimated with logistic multinomial generalised linear models. Sea surface temperature and maximum wave height data extracted from wave rider buoys offshore from collection sites were compared in a Welch’s two-sample t-test.

To compare between the sites within the native range of P. praeputialis sampled in this study and sites in the invasive range in Chile sampled by Cerda and Castilla (2001), species presence data was adapted into the same format as data from the former. Lifestyle and feeding mode of species were also determined using online resources. An NMDS was performed to test for differences in within phyla diversity between sites.

Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6


Size and Maturity of Individuals

While gonad index did have a significant effect on cavity area (ANCOVA: F1,28=102.19, p-value = 7.57×10-11), the effect of site was not significant (ANCOVA: F1,28=3.086, p-value = 0.0899) (Figure 7). Data also suggested no effect of site on cavity area (t = 1.4208, df = 24.982, p-value = 0.1677) or external area (t = 1.7945, df = 23.169, p-value = 0.08579).         

Community composition

23 and 31 epifaunal species were identified on the 15 ascidians sampled at Kirra Beach and Point Arkwright respectively. 216 organisms were recorded at Kirra Beach and 164 at Point Arkwright. There were Chitons and Anemones on P. praeputialis at Kirra beach and more sponges and colonial ascidians at Point Arkwright. 

Comparison of Kirra Beach and Point Arkwright

Individual organisms at Kirra Beach were more likely to be filter feeding organisms than at Point Arkwright (GLM: Binomial logistic regression, z=-6.586 p-value = 4.51×10 -11). Species present at Kirra beach also tended to be more likely to be filter feeding (GLM: Binomial logistic regression, z = -2.113, p-value = 0.03458).

Abundance of errant individuals was significantly lower than sedentary and sessile individuals at Kirra Beach (GLM: Multinomial logistic regression, errant vs sedentary: z =3.766307, p-value = 0.00017, errant vs sessile: z =2.680, p-value = 0.007352). There were no significant differences in the species diversity within lifestyle modes between sites.

Mean maximum wave height was significantly higher in Tweed Heads (2.1) than Mooloolaba (1.9) (t = 48.678, df = 110610, p-value < 2.2e-16) (Figure 10). Mean sea surface Temperature was higher in Mooloolaba (24.1) than Tweed Heads (23.5) (t = 25.788, df = 42969, p-value < 2.2e-16)(Figure 11).


Comparison of Australia and Chile

At Australian sites, Annelida contributed the most to species diversity, particularly at Kirra beach, whereas in Chilean sites, there was the most diversity within Mollusca. The NMDS revealed that there was much greater difference between the two Australian sites than between sites in Chile (Figure 12).

There was little difference in the proportion of species using filter feeding compared to predation/scavenging between sites, with more active modes of feeding employed by more than 70% of species at all locations (Figure 13). There was much more evident differentiation in diversity within feeding modes between sites (Figure 14). In particular Australian sites tended to host less errant species.

Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12
Figure 13
Figure 14


As in Pyura beds distributed throughout the southern hemisphere, P. praeputialis beds at the two study sites in south east Queensland have both been shown to host a diverse macroinvertebrate epifaunal community (Cerda and Castilla 2001; Dalby 1996; Davis, Walls et al. 2018; Monteiro, Chapman et al. 2002; Ortiz, Campos et al. 2013; Rius, Teske et al. 2017). There were qualitative differences in community composition between the samples at sites, with only 13 of 55 species recorded shared between the sites. However, lack of replication within sites does not allow quantitative conclusions to be drawn about how similar these sites are.


Characterisation of Epifaunal Community

Fewer species were recorded during this study than similar studies in Chile and Australia. Monteiro, Chapman et al. (2002) identified 45 species associated with Pyura beds around Sydney, while Cerda and Castilla (2001) recorded 74, 67 and 79 species at 3 sites in the Bay of Antofagasta, Chile. It is likely many species were missed due to sampling method, with many individuals falling off during extraction of the ascidian, and many being missed during data collection due to being small, or concealed within the algae or in burrows in the tunic. Both the aforementioned studies had a very high diversity of small molluscs which was not the case in this study. This prompts further investigation into whether these species were lost in sampling or absent in the Northern end of the Australian distribution of P. praeputialis. Further, of the individuals recorded, it is likely some similar species were grouped together. Therefore, it should not be concluded that these sites had lower species richness than sites in other studies. Improving sampling techniques so that the epifauna growing on each ascidian can be identified would allow a species abundance curve to be used to estimate the true species abundance found at the site. Preserving epifauna and identifying them using a dissecting microscope would also reduce the likelihood of grouping similar species. An interesting finding was a domination by tubicolous “polychaetes” similar that found in Northern New Zealand where Pyura doppelgangera is invading rocky shores and competing with green lipped mussels (Davis, Walls et al. 2018). It is likely the tunic of P. praeputialis provides an ideal substrate for burrowing and tube-building worms, due to its’ rough and penetrable nature. This is an ideal example of Pyura’s role as an ecosystem engineer by providing suitable settlement habitat (Castilla, Guinez et al. 2004).

Differences Between Australian Sites

Differences between environmental conditions (Figures 10-11) suggest regional differences are present between the sampling locations. A limitation to this analysis is that the only data available for wave height for both sites was from 2000 to 2004, which is not directly relevant to the communities found in 2020, although this study assumes that a period of 4 years is enough to extrapolate overall climatic spatial trends. Furthermore, the differences in both variables are a matter of 20 cm and less than a degree of temperature, and it is likely that smaller scale differences in site, such as aspect of the coastline and rugosity would have a greater impact than these regional differences. Measurements of size and gonad index suggests there is no obvious stunting of growth caused by differences in environmental conditions, despite the Point Arkwright site beginning to approach the northern limit of Australian P. praeputialis populations (Endean, Kenny et al. 1956). The correlation between gonad index and size shown in Figure 7 suggests that size is a good estimator of the age and reproductive output of individuals. In the context of this study, no differences in the sizes of individuals means there is no difference in the area for settlement of epifaunal species per ascidian. This also suggests that there are no differences in reproductive output and population-age structure between sites.


Observed differences in dominant feeding and lifestyle strategies between sites of could be due to regional differences such as different wave energy and sea surface temperature, or due to site specific differences, such as direction the shore faces and structure of the rock substrate. This large-scale spatial variation in the environment may lead to high diversity, not only at a scale of tunicates, but between rocky outcrops. Robinson, Finelli et al. (2013) suggests that sites with higher wave energy may have more active feeders because passive feeders tend to lose their prey more often, despite encountering more food items due to higher flow rate. The higher proportion of filter feeding inviduals at Kirra Beach, which has higher regional wave heights agrees with this hypothesis. Further measurements of flow rates on P. praeputialis beds would be useful to investigate this finding. Studying sites from Double Island Point, south along the temperature gradient would also be interesting to determine whether presence of more tropical or temperate species in these ecosystems is affected by regional sea surface temperatures.


Native vs Invasive Sites

Non-metric multidimensional scaling found greater differences between Australian sites due to differences in contribution of different phyla to overall species richness than between sites in Chile (Figure 12). These differences could be caused by species in Australia being adapted to this niche. By engineering this habitat, it is possible this species has had evolutionary effects on other organisms in Australia as proposed by (Jones, Lawton et al. 1994). Castilla, Lagos et al. (2004) found that, by providing protection from desiccation at low tide, P. praeputialis facilitated subtidal organisms to inhabit intertidal regions. It would be interesting to investigate whether this is the case in Australia by sampling along the tidal zone at Pyura dominated sites. If this is not the case in Australian Pyura beds, and there are species specifically adapted to inhabiting Pyura dominated intertidal beds, this has implications for the management of bait collection of P. praeputialis. Particularly because strong harvesting pressure has been shown to reduce P. praeputialis populations, therefore any species endemic to these ecosystems are also at risk of decline (Castilla, Manríquez et al. 2014).


I’m grateful to Bernie and Sandie Degnan for their expertise and passion for sharing the wonders of marine invertebrates. I’d also like to thank Maximiliaan Koebrugge and Gurion Ang for assistance collecting and processing samples.


 Invertebrates of the Coral Sea. Available from: Accessed [26-05-20]


Castilla, J.C., Collins, A.G., Meyer, C.P., Guiñez, R., and Lindberg, D.R. (2002) Recent introduction of the dominant tunicate, Pyura praeputialis (Urochordata, Pyuridae) to Antofagasta, Chile. Molecular Ecology 11(8), 1579-1584.


Castilla, J.C., Guinez, R., Caro, A.U., and Ortiz, V. (2004) Invasion of a rocky intertidal shore by the tunicate Pyura praeputialis in the Bay of Antofagasta, Chile. Proceedings of the National Academy of Sciences of the United States of America 101(23), 8517-8524. [In English]


Castilla, J.C., Lagos, N.A., and Cerda, M. (2004) Marine ecosystem engineering by the alien ascidian Pyura praeputialis on a mid-intertidal rocky shore. Marine Ecology Progress Series 268, 119-130.


Castilla, J.C., Manríquez, P.H., Delgado, A., Ortiz, V., Jara, M.E., and Varas, M. (2014) Rocky Intertidal Zonation Pattern in Antofagasta, Chile: Invasive Species and Shellfish Gathering. PLOS ONE 9(10), e110301.


Cerda, M., and Castilla, J.C. (2001) Diversidad y biomasa de macro-invertebrados en matrices intermareales del tunicado Pyura praeputialis (Heller, 1878) en la Bahía de Antofagasta, Chile. Revista chilena de historia natural 74, 841-853.


Dalby, J.E. (1996) Nemertean, copepod, and amphipod symbionts of the dimorphic ascidian Pyura stolonifera near Melbourne, Australia: specificities to host morphs, and factors affecting prevalences. Marine Biology 126(2), 231-243.


Davie, P. (2011) 'Wild guide to Moreton Bay and adjacent coasts.' 2nd ed. edn. (Queensland Museum: South Brisbane, Qld.)


Davis, A.R., Walls, K., and Jeffs, A. (2018) Biotic consequences of a shift in invertebrate ecosystem engineers: Invasion of New Zealand rocky shores by a zone-forming ascidian. Marine Ecology-an Evolutionary Perspective 39(3), 10. [In English]


Department of Environment and Science, Queensland Government. (2018a) Coastal Data System - Waves (Mooloolaba).


Department of Environment and Science, Queensland Government. (2018b) Coastal Data System - Waves (Tweed Heads). 


Department of Environment and Science, Queensland Government. (2020a) Coastal Data System - Waves (Mooloolaba). 


Department of Environment and Science,Queensland Government. (2020b) Coastal Data System - Waves (Tweed Heads). 


Endean, R., Kenny, R., and Stephenson, W. (1956) The Ecology and Distribution of Intertidal Organisms on the Rocky Shores of the Queensland Mainland. Marine and Freshwater Research 7(1), 88-146.


Fairweather, P.G. (1991) A conceptual framework for ecological studies of coastal resources: An example of a tunicate collected for bait on Australian Seashores. Ocean and Shoreline Management 15(2), 125-142.


Glasby, C.F., K. (2003) POLiKEY. 2 edn. 


Google Earth (2015) Image Landsat / Copernicus. 

iNaturalist. Available from Accessed [26/05/20].

Jones, C.G., Lawton, J.H., and Shachak, M. (1994) Organisms As Ecosystem Engineers. Oikos 69(3), 373-386. [In English]


Monteiro, S.M., Chapman, M.G., and Underwood, A.J. (2002) Patches of the ascidian Pyura stolonifera (Heller, 1878): structure of habitat and associated intertidal assemblages. Journal of Experimental Marine Biology and Ecology 270(2), 171-189.


Ortiz, M., Campos, L., Berrios, F., Rodriguez, F., Hermosillo, B., and Gonzalez, J. (2013) Network properties and keystoneness assessment in different intertidal communities dominated by two ecosystem engineer species (SE Pacific coast): A comparative analysis. Ecological Modelling 250, 307-318.


Pacheco, A.S., and Andrade, D.G. (2020) Decline of a non-native ecosystem engineer and its replacement with a native on rocky shores: effects on the diversity and structure of benthic communities. Marine Biodiversity 50(1), 2.


Rius, M., and Teske, P.R. (2013) Cryptic diversity in coastal Australasia: a morphological and mitonuclear genetic analysis of habitat-forming sibling species. Zoological Journal of the Linnean Society 168(3), 597-611. [In English]


Rius, M., Teske, P.R., Manriquez, P.H., Suarez-Jimenez, R., McQuaid, C.D., and Castilla, J.C. (2017) Ecological dominance along rocky shores with a focus on intertidal ascidians. In Oceanography and Marine Biology: An Annual Review, Vol 55. Vol. 55. (Eds. SJ Hawkins, AJ Evans, AC Dale, LB Firth, DJ Hughes and IP Smith) pp. 55-85. (Crc Press-Taylor & Francis Group: Boca Raton)


Robinson, H.E., Finelli, C.M., and Koehl, M.A.R. (2013) Interactions Between Benthic Predators and Zooplanktonic Prey are Affected by Turbulent Waves. Integrative and Comparative Biology 53(5), 810-820. [In English]


Teske, P.R., Rius, M., McQuaid, C.D., Styan, C.A., Piggott, M.P., Benhissoune, S., Fuentes-Grünewald, C., Walls, K., Page, M., Attard, C.R.M., Cooke, G.M., McClusky, C.F., Banks, S.C., Barker, N.P., and Beheregaray, L.B. (2011) "Nested" cryptic diversity in a widespread marine ecosystem engineer: a challenge for detecting biological invasions. BMC Evolutionary Biology 11(1), 176.