Select the search type
  • Site
  • Web

Student Project

Selectivity in the filter feeding behaviour in Australian feather-duster worms (Family: Sabellidae)

Harriet Allen 2016


Feather-duster worms are key members of the benthic marine community, filtering the water to remove and consume small planktonic particles. Considering their potential use as bio-filters and bioremediators, and the growing concern over the abundance of marine micro-plastics in the marine environment understanding the feeding behaviour of the organisms near the base of the food chain is important. This study provided sabellid worms from Moreton Bay, Australia, with three different sized algal solutions, and analysed the difference in optical density after one day, and therefore how many cells had been consumed. The largest algae type showed no significant difference to the controls, suggesting that the feather-duster worm either did not, or could not consume this algae. The two smaller algal treatments showed a significant difference to the controls, with the smallest algae having the largest decrease in optical density and cell count. Understanding the reasons for these differences will require further study and even gut-contents analysis, but there are two main hypotheses. Firstly, the smaller algae may be the optimal size for the feather-duster worm and so the highest and most efficient feeding rate was used in order to feed as much as possible while the environment was good. Secondly, it may be the case that the sabellid worm had to feed at a higher rate in order to make up for the reduced nutrient load in the significantly smaller algae. 


Filter feeding organisms are key members of benthic communities. Filter feeders often have fine sieve-like appendages which capture small particles in the water for consumption, which in turn prevents a build up of nutrients and sediment in the water. This can allow for increased water clarity, aiding organisms that photosynthesise to obtain energy and can prevent build ups of bacteria or harmful sediments (Gili & Coma, 1998). Sponges are one such group key for water filtration in the marine environment, heralded as having particularly high filtration rates, with some able to filter up to 6L h-1 of seawater (Stabili et al., 2006; Gerrodette & Flechsig, 1979). As well as being important for maintaining water quality, many sedentary filter feeders are selective about what they ingest, this preferential feeding behaviour has been observed in bivalves and sponges (Shumway et al., 1985; Stabili et al., 2006). This selectivity may be because the energetic benefits would be outweighed by the cost of handling and processing such large particles whereas very small particles simply will not get caught by the filter feeding apparatus. From this it would not be unreasonable to suggest that there may be an impact of the particle size upon the rate of consumption, perhaps due to energy gain or energy expenditure to digest such items, and that there will probably be an optimal cell size for ease of digestion and maximum nutrient gain. This is reflected in many optimal foraging behaviour models (Schoener, 1971; Werner & Hall, 1974).

However, sponges and bivalves are not the only filter feeders in the marine benthos; polychaetes, such as those from the order Sabellida, can also have key roles in maintaining water quality and reducing particle loads in the surrounding environment, despite sometimes being seen as pest species (Davies et al., 1989). Feather-duster worms (Family: Sabellidae) are sedentary, filter feeding, marine polychaetes, consuming suspended planktonic particles, such as bacteria, algae, detritus and other microscopic organisms, in feather like appendages which they extend out of the burrow when not under threat (Rouse, 2004; Tamaru, Ako & Baker, 2011). Feather-duster worms are cryptic, living in burrows they either secrete or create from sediment, depending on the species (Rouse, 2004). Their feather like appendages are part of a radiolar or branchial crown which is covered in cilia. These cilia create currents and water flow to allow them to sift the particles from the water. The cilia also carries the caught food down the radioles towards the gut (Rouse, 2004; Lewis, 1968; Nicol, 1930). Feather-duster worms have been found to have a very efficient filtration system, (Licciano, Stabili & Giangrande, 2005) and capture selectivity has been observed in these marine worms. It has been found that feather-duster worms can sort the particles passing through their branchial crown, distinguishing between the size of the particles ingesting small particles and rejecting those that are too large (Lewis, 1968; Licciano, Stabili & Giangrande, 2005; Licciano et al., 2007). Small sabellids have been found to selectively consume particles of 1-7µm, and occasionally larger particles (Lewis, 1968). As previously stated, it may be the case that the size of the worm limits the particle size consumed, either through the cost of digestion compared to the energy gain, or just the ability to actually catch and consume extremely large or extremely small particles. These factors may well affect the rate of cell consumption for an individual.

Feather-duster worms are an important marine invertebrate group as they are major components of bio-fouling communities (Tovar‐Hernández, Méndez & Villalobos‐Guerrero, 2009). Biofouling communities have a large impact upon human utilisation of the oceans. Preventing biofouling is a large economic, labour intensive and time consuming burden, particularly in aquaculture, costing many industries billions of dollars every year in losses and control (Dürr & Watson, 2010). Because of this feather-duster worms are often seen as pests due to their biofouling capabilities and their capability to rapidly form dense aggregations, as well as the ease with which they can be translocated and compete with native species (Currie, McArthur & Cohen, 2000). However, feather-duster worms are also crucial members of reef communities, forming a large proportion of the biomass, helping to build the reef itself and being part of the diet for many other reef species (Coleman & Williams, 2002). As well as this, filter feeding feather-duster worms may be useful to the marine environment as important bio-filters and may be of use in bioremediation (Licciano, Stabili & Giangrande, 2005; Licciano et al., 2007). It may be possible to use sabellids as bioindicators for the health of the water, as the concentrations of waste particles might help us detect pollutants and contaminants that are at low concentrations (Licciano et al., 2007). One useful application for sabellids would be filtering sewage and urban or agricultural run-off (Licciano et al., 2007). However, different species of feather-duster worm would cope with these stressors differently and so only the more robust species would be useful in this application. Furthermore, considering the growing concern over the abundance of marine micro-plastics (Cole et al., 2011; Wright, Thompson & Galloway, 2013), it would be interesting to see if particle size has an effect on the feeding rates, as well as the largest particles simply being mechanically excluded, as this could have implications for plastic ingestion and potential build up in the wider ecosystem. This would also affect the suitability of sabellid worms for use as bio-filters if pollution would reduce their filtering ability. Understanding the clearance capabilities of these animals and their target particle size will help in not only understanding their ecology and feeding behaviour, but also their potential usefulness as bio-remediators and biofilters as well as the potential impacts of marine pollution. 

This study will focus specifically on sabellids from Moreton Bay in South Eastern Queensland from the genus Branchiomma (WoRMS, 2016; Figure 1). These feather-duster worms were abundant on the Moreton Bay encrusting plates and tended to have a dark green or brown tube formed by sediment and mucus. I will investigate whether these feather-duster worms exhibit a difference in feeding behaviour and the rate of algal consumption when exposed to algae of different sizes. The three different algal solutions used in this experiment will be Nannochloropsis sp., Thalassiosira sp. and T-Iso sp. aquarium feed solutions (Figure 2). Nannochloropsis sp. and T-Iso sp. have been used in sabellid studies before, finding that worms fed with T-Iso sp. tended to have the higher survival and growth rate (Tamaru, Ako & Baker, 2011), so I took these solutions to be suitable feed for the sabellids used in this study.

I predict that the larger algae will be consumed at a lower rate than the two smaller algal types, as it is outside of the observed particle size consumed in other sabellid studies (Lewis, 1968). I also hypothesise that the smallest sized algae, Nannochloropsis sp., will have the most cells consumed. This may be because the cells are a smaller size and so to get enough nutrients a higher quantity will need to be consumed, or perhaps simply because there are more cells per ml in the solution.

Figure 1
Figure 2

Materials and Methods

To investigate whether the size of the particle affects the feeding behaviour of the feather-duster worms, worms were provided with one of three different algal treatments and left for a set period of time to filter feed in that treatment. The difference in the colour of the solution, recorded as optical density using a spectrophotometer, was recorded at the start and end of the experiment along with the mass and cell count at the start of each run. 

The feather-duster worms for this experiment were removed along with as much of their casing as possible using forceps from encrusting plates. These encrusting plates were collected from Moreton Bay, Southeast Queensland. After removal each feather-duster worm was weighed (wet mass) and placed into individual wells and one of the three algal treatments applied. For each well I recorded the starting optical density of the treatment using a spectrophotometer as well as the ending optical density, starting cell count and, from this, I calculated the change in cells and therefore the rate of feeding was calculated for each individual. Each feather-duster worm was left for 27 hours to feed.

The controls in this study were individual wells with the algal treatment applied, but no feather-duster worms. This allowed a comparison between how much the optical density changed with and without feeding from the feather-duster worms. These were compared to see if there was a significant difference, which, if there was, would suggest that the feeding was the cause of the increased change and therefore the differences between treatments could be analysed for differences 

Algal treatments 

In this study three different sized algae were used, to see if there would be a difference in the feeding behaviour of feather-duster worms as a result of the size of the food particles. The three algal solutions used were concentrations of Nannochloropsis sp., Thalassiosira sp. and T-Iso sp., standard ‘Instant Algae®’ marine microalgae aquarium feed solutions provided by the University of Queensland (Figure 2). Each algal solution was diluted down to a suitable feeding concentration, but enough to be able to observe a visible change in colour or algal density due to the feeding. Algae was kept refrigerated between experiments. 

The three algal solutions were diluted down to 5% of the original bottled concentration with filtered seawater. I then randomly assigned and applied 50µl of each of the algal dilutions to the feather-duster worms placed in individual wells of 12 well dishes in a diluted form (Figure 3). Overall there were 40 feather-duster worms used, 13 in each of the Nannochloropsis sp. and Thalassiosira sp. treatments and 14 in the T-Iso sp. treatment. There were also 4 controls for each treatment. Each of the well containing the feather-duster worms already contained 200µl of filtered sea water, making the experimental concentration 1%. Filtered seawater was used in this study to make sure that only the only food available in the wells was the algae,no other plankton or sediment upon which the feather-duster worms could have fed.

The well plates were kept at room temperature in a laboratory over a period of 27 hours, kept under normal circadian rhythms and with lid to prevent evaporation of the water, which could concentrate the algal solution making it seem like less had been consumed by the feather-duster worms 

Optical Density and Cell Counts 

To understand whether algal size has any effect on feather-duster worm feeding capabilities I recorded the change in algal concentration. To do this the optical density of the each of the treatments were be recorded at the start and end of the experiment for each well. A 1% dilution from the main stock for each algae was placed into a spectrophotometer and the optical density at 550nm recorded at the start of each run of the experiment. A wavelength of 550nm was used as 550nm is where chlorophyll has low absorbance, and optical density measures the density of the sample, not how well pigments absorb the light (Griffiths et al., 2011). 550nm is a typical wavelength to use when looking at algal optical density (Myers et al., 2013 ). However, for the difference in colour to actually mean anything the optical density had to be related to cell density. Using a 0.5% solution of the usual algae feed stock a 10µl sample was used to count the cells using a hemocytometer. I then scaled the results up to fit the 1% dilution. The resulting number of cells per ml could then be related to the recorded starting optical density. Then the difference in optical density allowed me to calculate the proportion of cells that the feather-duster worms had ingested in that specific time frame, and by extension rate of consumption as a factor of body size or mass to be calculated.

Data Analysis

The data were analysed using R-studio (Version 0.98.1062). ANOVAs and linear models were used to compare the three algal treatments, with Tukey HSD post hoc tests and t-tests for clarifying any significant differences that were found between them.

Figure 3


There was a significant difference between the three treatments provided to the Branchiomma sp. feather-duster worms and the controls (ANOVA: F2,40=9.74, P<0.001), with no significant difference found in the change in optical density between  the three controls themselves. There was a significant difference between the three treatments with the feather-duster worms (ANOVA: F2,40=8.613, P<0.001; Figure 4): The largest sized algae, Thalassiosira sp., had a mean change in optical density of 0.074±0.011nm, which was not significantly different to the control. T-Iso sp. had a mean change of 0.285±0.060nm, which was significantly different from the control (T-test; t14=2.17, P=0.048). The smallest algae, Nannochloropsis sp., changed by 0.328±0.049nm, which was again significantly different from the control for this treatment (T-test; t13=3.40, P=0.005). Tukey HSD post hoc tests showed that there was a significant difference between the Thalassiosira sp. and the two smaller algal treatments (p<0.05), but no difference in the mean change in optical density between the two smaller algae, Nannochloropsis sp. and T-Iso (p=0.783).

The number of cells consumed in each of the three feather-duster worm treatments was significantly different between the three treatments (ANOVA: F2,40=46.43, P<0.001; Figure 5), with Tukey HSD post hoc tests showing a significant difference between all three (P<0.05). There were more cells consumed in the Nannochloropsis sp. treatment (mean=300±32million cells), than the T-Iso sp. treatments (mean=88±20million cells), with the largest cells, Thalassiosira sp., having the fewest number consumed (mean=6±0.8million cells).

The change in optical density when compared to the change observed in the controls ([change in treatment]-[change in control]) showed a significant different between the three treatments, (ANOVA: F2,40=9.74, P<0.001; Figure 6) with the difference shown in the Thalassiosira sp. treatment being significantly different from the two smaller cell types (Tukey HSD; P<0.005) but there was no significant difference found between the two smaller cell types. This analysis allows insight as to how much more than the controls the feather-duster worms changed the cell quantities within their wells, with the fan worms fed T-Iso sp. and  Nannochloropsis sp. consuming significantly more than the larger cell type. However when the changes seen in the controls were accounted for, there was then a greater change in optical density in the T-Iso sp. treatment (mean=0.217±0.038nm) than the Nannochloropsis sp. treatments (mean-0.195±0.040nm)(difference=0.022nm). However this difference is not statistically significantly different.  

Figure 4
Figure 5
Figure 6


Statistical analysis showed that there was no significant difference in the optical density, and by extension the cell count, between the controls and the fan worms for the largest algae, Thalassiosira sp., suggesting that none, or very little of the Thalassiosira cells were consumed by the feather duster worms, supporting my hypotheses. Between the two smaller algae, the only significant difference was found in the number of cells which had been consumed, with Nannochloropsis sp. having a decrease in cell count 3x that of T-Iso sp.. This may be a signal that this was the algal size nearest to optimum algal size, as the rate of consumption was highest. However it may also be that, due to the small size of Nannochloropsis, the small cells do not contain enough nutrients and so feeding rates need to be increased in order to meet the sabellid’s energy demands. Indeed, other studies found that T-Iso sp. resulted in larger feather duster worms than Nannochlropsis, due either its higher energy content or fatty acid contents, suggesting that it is the better nutrient(Tamaru, Ako & Baker, 2011). Additionally to this, when the controlled were accounted for in the change in optical density in this study, it did appear as if there was more of a change in the T-Iso sp., which would support this papers findings. However since their study looked at different feather-duster worms it may be the case that the sabellids in this study require a different suite of nutrients in order to survive, or maybe they are less able to process the larger T-Iso cells. 

From this study there is no way of knowing which of these two hypotheses fit the results, I can only say that there was a difference in the feeding on these three algal solutions, with Thalassiosira sp. consumed the least and Nannochloropsis sp. the most. Future studies could look at feeding preference in the same worm when presented with different algae in the same solution, however this would require a gut contents analysis afterwards. It would also be interesting to study the nutrient gain from each of these algae, much like Tamaru, Ako & Baker’s study (2011), and to see whether those algae that provided less essential minerals, or a lesser energy gain, were consumed at a higher rate as compensation or whether they were consumed at a lesser rate as they were inefficient sources. If a food source was inefficient it may be better to eat just enough to survive and hope a new and more beneficial food source comes along, however as the risk of death increases the organism is more likely to stop being specialist and switch to a generalist feeding pattern, this is called prey switching or dietary plasticity. It would also be interesting to track the optical density over time and see how the feeding rate changes. It may be that as time goes on the less beneficial algae will be consumed at an increasing rate as it is needed to survive whereas before it was more to tide the feather-duster worm over until something better came along. 

The worms used in this study were most likely stressed, which could be seen by them ejecting themselves from their tubes. If stressed they may behave differently in their feeding behaviour. One good future study would be to allow them to acclimate to the new environments and then apply the treatments. However this would be methodologically complicated as they would need plankton in order to survive, but the experiment needs the worms to be exposed to only the filtered sea water and the algae. 

One major difficulty was finding a balance between providing enough food so that there would be a visible colour change, but not giving the feather-duster worms too much algae, and leaving the experiment for the correct amount of time so that the feather-duster worms do not run out of oxygen. This is especially a problem since feather-duster worms have evolved to feed in low concentrations of nutrients and the concentrations in this study were fairly high (Tamaru, Ako & Baker, 2011). It would be interesting to see if different sized algae affect the worms differently at different concentrations and whether size of the algae would actually affect the feeding rate at near-natural concentrations.

As well as simply studying the feeding behaviour of Branchiomma sp. and the restraints of cell size, this study also suggests that feather duster worms do filter feed enough to make a significant difference in cell concentrations in a fairly concentrated solution of algae in under 48 hours, meaning that they may well be a suitable candidate for use as bio-filters and bio-regulators in polluted environments. 

The largest algae used in this study was 7-10µm in size, and this was not significantly consumed by the feather duster worms. Most studies report microplastics being found at sizes of 300µm or larger, in which case small filter feeding animals such as these Branchiomma sp. would be safe, preventing early accumulation in the food chain (Hidalgo-Ruz et al., 2012). However, microplastics have been recorded as sizes as small as 1.6-3µm, which would be within the range at which this species feeds (Hidalgo-Ruz et al., 2012; Wright, Thompson & Galloway, 2013). Furthermore, there is now a growing concern over nanoplastics in the marine environment, which filter feeders such as Branchiomma sp. may well be susceptible to (Cole et al., 2011).

In conclusion this study has found that the size of the algae presented to a sabellid worm does affect the rate at which the algae is consumed. However the rate of consumption may be affected by factors such as nutrient or energy content as well as the size of the particle itself. Understanding the feeding behaviour of these animals will allow us to further understand their ecology. This increased understanding of sabellid feeding systems and ecology may aid us in finding methods to control them as biofouling pests, but may also aid us in identifying those that will be of use to polluted marine areas as bio-remediators and biofilters. We will also be able to identify which species may be at risk from marine microplastics, and by extension which species might be at risk from plastic accumulation. However, all of these applications are very theoretical and at present not enough is known about this particular Branchiomma sp. to state anything with certainty, but we do know that the particle size appears to affect the consumption rate of algae with the upper limit appearing to be around 7-10µm. 


A huge thank you to Bernie and Sandy Degnan and to all the tutors from the BIOL3211 Marine Invertebrates course at the University of Queensland for all their invaluable help, tuition and guidance. Also, thank you to the University of Queensland and to all the bioscience laboratory staff for helping make sure that the lab sessions ran as smoothly as possible. 


Cole, M., Lindeque, P., Halsband, C. & Galloway, T.S. (2011). Microplastics as contaminants in the marine environment: A review. Marine Pollution Bulletin, 62, 2588-2597. 

Coleman, F.C. & Williams, S.L. (2002). Overexploiting marine ecosystem engineers:potential consequences for biodiversity. TRENDS in Ecology & Evolution, 17(1), 40-44.

Currie, D.R., McArthur, M.A. & Cohen, B.F. (2000). Reproduction and distribution of the invasive European fanworm Sabella spallanzanii (Polychaeta: Sabellidae) in Port Phillip Bay, Victoria, Australia. Marine Biology, 136(4), 645-656.

Davies, B.R., Stuart, V. & de Villiers, M. (1989). The filtration activity of a serpulid polychaete population (ficopomatus enigmaticus (fauvel)) and its effects on water quality in a coastal marina. Estuarine, Coastal and Shelf Science, 29(6), 613-620. 

Dürr, S. & Watson, D.I. (2010) ‘Biofouling and antifouling in aquaculture’ in (ed.) Dürr, S. & Thomason, J.C. Biofouling. Blackwell Publishing Ltd: Singapore.

Gerrodette, T. & Flechsig, A.O. (1979). Sediment-induced reduction in the pumping rate of the tropical sponge Verongia lacunosa. Marine Biology, 55(2), 103-110. 

Gili, J.M. & Coma, R. (198). Benthic suspension feeders: their paramount role in littoral marine food webs. Trends in Ecology and Evolution, 13(8), 316-321.

Griffiths, M.J., Garcin, C., van Hille, R.P. & Harrison, S.T.L. (2011). Interference by pigment in the estimation of microalgal biomass concentration by optical density. Journal of Microbiological Methods, 85(2), 119-123. 

Hidalgo-Ruz, V., Gutow, L., Thompson, R.C. & Thiel, M. (2012). Microplastics in the Marine Environment: A Review of the Methods Used for Identification and Quantification. Environmental science and Technology, 46, 3060−3075.

Lewis, D.B. (1968). Feeding and tube building in the Fabriciinae (Annelida: Polychaeta). Proceedings of the Linnean Society London, 179(1), 37-49.

Licciano, M., Stabili, L. & Giangrande, A. (2005). Clearance rates of Sabella spallanzanii and Branchiomma luctuosum (Annelida: Polychaeta) on a pure culture of Vibro alginolyticus. Water Research, 39, 4375-4384.

Licciano, M., Stabili, L., Giangrande, A. & Cavallo, R.A. (2007). Bacterial accumulation by Branchiomma luctuosum (Annelida: Polychaeta): a tool for biomonitoring marine systems and restoring polluted waters. Marine Environmental Research, 63, 291-302.

Myers, J., Curtis, B. & Curtis, W. (2013). Improving accuracy of cell and chromophore concentration measurements using optical density. Bmc Biophysics, 6(4). 

Nicol, E.A.T. (1930). The feeding mechanism, formation of the tube, and physiology of digestion in Sabella pavonina. Transactions of the Royal Society of Edinburgh, 56, 537-598.

Rouse, G.W. (2004). ‘Annelida: Polychaeta’ in (Ed.) Mule, C.M. & Yong, H.S. Freshwater Invertebrates of the Malaysian Region. Academy of Sciences Malaysia, Kuala Lumpur. 

Schoener, T.W. (1971). Theory of feeding strategies. The Annual Review of Ecology and Systematics, 2, 369-404.

Shumway, S.E., Cucci, T.L., Newell, R.C. & Yentsch, C.M. (1985). Particle selection, ingestion, and absorption in filter-feeding bivalves. Journal of Experimental Marine Biology and Ecology, 91(1), 77-92. 

Stabili, L., Licciano, M., Giangrande, A., Longo, C., Mercurio, M., Marzano, C. & Corriero, G. (2006). Filtering activity of Spongia officinalis var. adriatica (schmidt) (porifera, demospongiae) on bacterioplankton: Implications for bioremediation of polluted seawater. Water Research, 40(16), 3083-3090. 

Tamaru, C.S., Ako, H. & Baker, A. (2011). Growth and survival of juvenile feather duster worms, Sabellastarte spectabilis, fed live and preserved algae. Journal of the World Aquaculture Society, 42(1), 12-23.

Tovar‐Hernández, M.A., Méndez, N. & Villalobos‐Guerrero, T.F. (2009). Fouling polychaete worms from the Southern Gulf of California: Sabellidae and Serpulidae. Systematics and Biodiversity, 7:3, 319-336.

Werner, E.E. & Hall, D.J. (1974). Optimal Foraging and the Size Selection of Prey by the Bluegill Sunfish (Lepomis Macrochirus). Ecology, 55(5), 1042-1052. 

World Register of Marine Species (2016). “WoRMS Taxon Details. Genus: Branchiomma”. Available from: [accessed May 31 2016]. 

Wright, S.L., Thompson, R.C. & Galloway, T.S. (2013). The physical impacts of microplastics on marine organisms: A review. Environmental Pollution, 178,483-492.