The Polychaete genus Brachiomma sp. is a highly invasive genus of fan worm that poses a threat to uninvaded ecological systems. As a result, developing a method to limit the distribution of this genus may be critical in protecting many habitats. One such avenue to do so is the through development of a biological pesticide based off of the presence of predation. This is because Brachiomma sp. respond to predation by retracting their feeding appendage, a brachial crown back into their calcareous tubule. As a result, when a predator is detected, a significant trade-off occurs, remain at risk of predation or reduce nutrient intake. Reducing nutrient intake leads to a decrease in adult fecundity, size and longevity meaning this is a possible way to limit the distribution of this genus. The experiment performed looked at the time taken by Brachiomma sp. to evert the brachial crown when in the presence of damaged conspecific alarm cues and Fenneropenaeus murguiensis kairomones. Results suggest that chemical cues do not significantly affect the time taken to evert the brachial crown. As a result, a biological pesticide cannot be developed from either of these chemical cues. 

Key Words: Chemical cues, predator kairomones, conspecific alarm cues, brachial crown

Kingdom: Animalia
Phylum: Annelida
Class: Polychaeta Order: Sabellida Family: Sabllidae Genus: Brachiomma

In ecological communities, predators are critical in determining the structure of the system and also influences the diversification and evolution of other species (Nosil and Crespi, 2006). In a community, predators typically follow the population growth of their prey species as more prey means more resources to support a larger population, the inverse is true also (Stenseth et al., 1997). At the level of the individual, a prey's behaviour, morphology and physiology often changes in response to predation (Preisser et al., 2005). Overall, the early detection of predatory cues and subsequent antipredatory response is crucial in diminishing the prospect of being predated upon. 

In order for a prey's antipredator behaviour to be triggered, detection is required but the predator does not necessarily need to be physically present for this to be achieved. Specifically, chemical cues that indicates the presence of a predator can be in the form of kairomones directly released from the predator, alarm cues emitted by damaged prey items and dietary cues that are expressed in the form of digested prey or faeces excreted by the predator (Wormington & Juliano, 2014). Based off information acquired via chemical cues, the prey item can detect the presence of a predator and subsequently elicit an antipredatory response (Relyea, 2004). Once the presence of alarm cues are detected via chemoreceptors, the prey item typically activates a predetermined behaviour that does not need to be learned. As a result, only when it is required is the antipredator behaviour initiated. Mosquito larvae, snails and newts have all been observed responding to alarm cues as well as numerous other animal groups (Kesavaraju & Juliano, 2010).

Currently, there are three levels of responses to predator detection, each with different metabolic costs. In order from highest to lowest, these are: Permanent changes in morphology, reduced resource acquisition and behavioural changes including increased vigilance and evasion (Roux et al., 2013). At an individual level, it is crucial that risk of predation accurately reflects the prey items behavioural response. This is because an individual's ability to grow and subsequently reproduce is hindered by long periods of antipredator behaviour (Constanzo et al., 2011). This has been demonstrated by Relyea in 2004 where tadpoles responded differently to the presence of predators depending on the likelihood of predation.

The response of prey species to chemical cues is observable in sedentary and sessile species of marine invertebrates, particularly Brachiomma sp. (a genus of fan worm). Due to the confined nature of their environment, increased vigilance and evasion are not options to avoid predators. Furthermore, there are no current changes associated with morphology in response to predation. As a result, in order to reduce the risk of predation Brachiomma sp. must respond to the chemical cues through changes in behaviour. Protection from predation comes in the form of a solid tubular structure that surrounds the body and is typically composed of mucous (Dill & Fraser, 1997). Additionally it also acts as a mechanism to attach the organism to the substrate. When predator detection occurs, Brachiomma sp. will usually withdraw their feeding structure, a brachial crown back into the tube so that no part of their body is exposed (Dill & Fraser, 1997). The issue with this is that whilst the crown is retracted, the organism is unable to feed and so a significant metabolic trade-off occurs. This is critical as it impedes growth and reproduction further highlighting the importance of having a finely tuned assessment of the overall risk of predation. 

Understanding the effect of predatory cues of Brachiomma sp. has potential implications on controlling aggressive invasive species. Threatened Brachiomma sp. that initiate antipredator behaviour inhibits nutrient uptake which leads to an increased development time resulting in the reduced fecundity, size and longevity of adults. Using this information, predatory cues could be beneficial in controlling the distribution of invasive members of Brachiomma sp. It has been suggested by Op de Beeck that a biological pesticide combined with a synthetic natural stressor (predatory chemical cues) would result in lower dosages of biological pesticides required, a reduced chance of resistance developing and a diminished chance of it impacting other species. 

Currently little literature exists on using a biological pesticide/synthetic predator kairomone to limit the distribution of Brachiomma sp, particularly within an Australian context. However, a native predatory species is ideal given that the chemosensory cells of Brachiomma sp. (along with many other species) has evolved to determine the differences between species. A non-native predator has not evolved alongside with Brachiomma sp. and as a result Brachiomma sp. would not be able to recognise the kairomones the predator releases. With this in mind, Fenneropenaeus merguiensis (colloquially known as the white banana prawn) is a likely candidate to develop a synthetic kairomone from given that the species lives in sympatry with Brachiomma sp. and are known predators of several members of the genus. Additionally, damaged conspecifics could also be used to develop a biological pesticide from as they release alarm cues when predated on. 

Aims: To determine the length of time it takes to evert the brachial crown in Brachiomma sp. when exposed to predatory kairomones from F. merguiensis and alarm cues released by damaged conspecifics. 

Research Question: How do predatory kairomones from F. merguiensis and alarm cues released by damaged conspecifics influence the length of time it takes to ever the brachial crown in Brachioma sp.

Hypothesis: Both kairomones from F. merguiensis and the alarm cues from damaged Brachiomma sp. will increase the period of time for the brachial crown to evert. 

Experimental Design
Brachiomma sp. were collected from Manly Boat Harbour via settlement plates. These were then transported to the University of Queensland for experimentation. The control was in the form of distilled water. F. merguiensis kairomones were gathered by collecting water from a bucket that had contained F. merguiensis. Conspecific alarm cues were gathered by collecting water from a vial that contained mechanically damaged Brachiomma sp. A Petri dish was then filled with 10mL of seawater whilst a single Brachiomma sp. specimen was removed from the settlement plate and placed in the dish. The Petri dish was then left for 5 minutes to decrease the stress levels of the animal and so that the brachial crown had time to evert. 2mL of the control was then added to the dish at which point the brachial crown retracts. The amount of time it took for the brachial crown to evert again was recorded with a maximum time limit of 300 seconds. In order to remain consistent, the brachial crown was defined as everted when it is fully extended which can be seen in figure 1. This was repeated a further 14 times. This was then repeated for both treatments so that there were 15 replicates for each and 45 specimens were experimented on in total. 

Statistical Methods
In regards to the statistical analysis, a histogram with treatment on the x-axis and time (s) taken to evert brachial crown on the y-axis was generated. Additionally, a one-way ANOVA was performed to determine if there were significant effects of the treatments on the time taken to evert the brachial crown. 

9 data points had to be rejected, 4 from the control, 2 from the damaged conspecifics alarm cues and 3 from the  F. merguiensis kairomones. This is because they either a) did not retract their brachial crown to begin with or b) did not evert their brachial crown within the 300 second time limit. 

As seen in Figure 1, there were noticeable differences between all three treatments. Specifically, the average time taken to evert the brachial crown when the F. merguiensis kairomones was the treatment was 61.7122 seconds longer than the control test which is depicted in figure 2. However, this difference was found not to be significant (F_1,21=4.2222, P=0.0525) which can be seen in figure 5. Additionally, it was found that overall the treatment does not have a significant effect on the time taken to evert the brachial crown (F_2,33=2.0208, P=0.1486) as shown in figure 3. It was found there was no significant difference between the conspecific alarm cues and F. merguiensis kairomones (F_1,23=0.421, P=0.5229) (figure 6). Similarily, there was no significant difference between the control and the conspecific alarm cues (F_1,22=2.02, P=0.1693) as shown in figure 4. 

Analysis of Results
The results of the study indicate that damaged conspecific alarm cues and F. merguiensis kairomones do not have a significant effect on the time taken to evert the brachial crown in Brachiomma sp. This means that it is unlikely that an effective biological pesticide can be developed from these chemical cues. However, this experiment has demonstrated that these animals have the ability to respond to chemical compounds. Schaum et al. has shown that the sedentary polychaete species Hediste diversicolor possesses strong chemosensory abilities and given the this species has a relatively similar lifestyle to Brachiomma sp. it is likely that Brachiomma sp. may also possess similar abilities. Currently, not much is known about the mechanism or location of the chemosensory abilities of Brachiomma sp. or H. diversicolor but other species of benthic marine polychaetes have been examined and comparisons can be drawn. It has been found via confocal laser scanning microscopy that the feeding palps contain many sensory cells of which some are chemoreceptors (Lindsay et al., 2004). As a result, it is fairly feasible to assume that Brachiomma sp. also have chemoreceptors located in the same place.

What these results also indicate is that the Brachiomma sp. responds to the presence of predation in the same way irrespective of the type of chemical cue. For this genus, their response is to increase the amount of time the brachial crown is retracted for. Additionally, Brachiomma sp. responds to mechanical stimulus and changes in light intensity in a similar manner as they both perceive these as evidence of predation (Bok & Nilsson, 2016). In regards to light intensity, the genus has developed photoreceptors in the compound eyes which are located on the branchial crown, specifically the radioles (Bok & Nilsson, 2016). The compound eyes may not have the greatest visual capabilities but are adept at detecting shadows and thus can determine the presence of a predator (Bok & Nilsson, 2016). 

Although the difference may not be significant, there is still a noticeable difference between the damaged conspecific alarm cues and kairomones from F. merguiensis. One possible explanation for this is that the chemoreceptors are more sensitive to predatory kairomones than they are to conspecific alarm cues. Additionally, the predatory kairomones could be viewed as greater evidence of predation than conspecific alarm cues. This could be due to the fact that predatory kairomones indicates the direct presence of a predator whereas conspecific alarm cues may not directly be involved with predation. Currently, no literature exists on the mechanisms behind why conspecific alarm cues and predator kairomones could be perceived as different levels of predation risk. 

The study clearly shows the behavioural changes in terms of risk of predation versus optimising feeding time. The changes in the time taken to evert the brachial crown demonstrates the judgements and adjustments they are able to make when tracking food availability in the environment and predatorial cues. Subsequently, this has implications on fecundity, growth and respiration and the overall growth of a population. As a result, this can still be used to limit distribution of this invasive genus although it does not necessarily have to be through biological pesticides, rather mechanical stimuli or changes in light intensity are also viable options.

Limitations and Future Recommendations 
One aspect that could be improved upon is the conditions that the specimens were held in when performing the experiments. Clearly a laboratory environment is not optimal given that it is not close to a natural environment especially considering that they are a sessile adults and had to be removed from the place they settled. This probably lead to the animals being under significant stress which could have lead to skewed results. Additionally, another area that could be improved upon is how the treatment is added to the Petrie dish. The treatments were added by putting a pipette in the water and subsequently squeezing which did cause ripples in some cases. This means that the specimens may have been responding to the mechanical stimuli or the changes in light intensity rather than the chemicals in the water. Another issue with the methods was the total number of replicates used, 45 is typically not enough to conduct thorough research. The final issue with the methods was that some of the animals may have not had enough time to adjust as well as possible to the unfamiliar conditions. Again, this would have lead to elevated levels of stress and skewed results. 

In regards to future recommendations, it may be useful to test different concentrations of damaged conspecific alarm cues and F. merguiensis kairomones as these may prolong the amount of time it takes to evert the brachial crown. Additionally, using different predatory kairomones to determine whether they have a different effect on Brachiomma sp. compared to F. merguiensis may be useful in examining whether there are other suitable candidates to develop a synthetic kairomone from. Furthermore, seeing how individual species of Brachiomma sp. respond to damaged conspecific alarm cues and F. merguiensis kairomones would expand our knowledge. The final piece research I recommend for the future is heating the treated water above 80 degrees Celsius and subsequently treating Brachiomma sp. with the water once cooled. This is to determine whether the alarm cues and kairomones are protein based. 

Overall, the results indicate that different chemical cues do not have a significant effect on the length of time it takes to evert the brachial crown. However, it has been demonstrated in this experiment that members of this genus do have the ability to respond to chemical predatory cues and so could be further researched to develop a biological pesticide from. Light intensity and mechanical stimulus are also other avenues that could be used to limit the distribution of this highly invasive genus. 

I would like to acknowledge the School of Biological Sciences at the University of Queensland for providing all necessary equipment to complete the research. Additionally, I would like to thank professor Bernie Degnan and associate professor Sandy Degnan for their guidance in experimental design and overall assistance. Finally I would like to thank the tutors for assistance in completing the experiment and keeping the specimens alive whilst not being experimented on.

Bok, M.J. & Nilsson, D. 2016. Fan worm eyes. Current Biology Magazine, 26, pp. 907-908.
Constanzo, K.S., Muturi, E.J. & Alto, B.W. 2011. Trait-mediated effects of predation across life history stages in container mosquitos. Ecological Entomology, 36, pp. 605-615

Dill, L.M. & Fraser, A.H. 1997. The worm re-turns: hiding behaviour of a tube-dwelling marine polychaete, Serpula vermicularis. Behavioural Ecology, 8, pp. 186-193.

Kesavaraju, B. & Juliano, S.A. 2010. Nature of Predation Risk Cues in Container Systems: Mosquito Responses to Solid Residues from Predation. Annals of the Entomological Society of America, 103, pp. 1038-1045.

Lindsay, S.M., Riordan, T.J. & Forest, D. 2004. Identification and Activity-Dependent Labeling of Peripheral Sensory Structures on a Spionid Polychaete. Biological Bulletin, 206, pp. 65-77.

Nosil, P. & Crespi, B.J. 2006. Experimental Evidence that Predation Promotes Divergence in Adaptive
Radiation. Proceedings of the National Academy of Sciences of the United States of America, 103,
pp. 9090-9095.

Op De Beeck, L., Janssens, L. & Stoks, R. 2016. Synthetic predator cues impair immune function and make the biological pesticide Bti more lethal for vector mosquitoes. Ecological Applications: A Publication of the Ecological Society of America, 26, pp. 355-366. 

Preisser, E.L., Bolnick, D.I. & Benard, M.F. 2005. Scared to Death? The Effects of Intimidation and Consumption in Predator-Prey Interactions. Ecology, 86, pp. 501-509. 

Relyea, R.A. 2004. Fine-Tuned Phenotypes: Tadpole Plasticity Under 16 Combinations of Predators and Competitors, Ecology, 85, pp. 172-179.

Roux O., Diabate, A. & Simard, F. 2013. Larvae of cryptic species of Anopheles gambiae respond
differently to cues of predation risk. Freshwater Biology. 58. pp. 1178-1189.

Schaum, C.E., Batty, R. & Last, K.S. 2013. Smelling Danger - Alarm Cue Responses in the Polychaete Nereis (Hediste) diversicolor (Muller, 1776) to Potential Fish Predation. PLOS One. 8, pp. 1-11. 

Stenseth, N.C., Falck, W., BjØrnstad, O.N. & Krebs, C.J. 1997. Population regulation in snowshoe hare and Canadian lynx: Asymmetric configurations between hare and lynx. Proceedings of the National Academy of Sciences of the United States of America, 94, pp. 5147-5152.

Wormington, J.D. & Juliano, S.A. 2014. Hunger-dependent and sex-specific antipredator behaviour of
larvae of a size-dimorphic mosquito. Ecological Entomology, 39, pp. 548-555.