Revise and rewrite: Revision of final paper draft using the peer review worksheet
Please check the “Peer view worksheet” and see if there is any missing points or thing you need to change and add on the “EEMB 138 paper”. Correct it for the final version of this paper. Hope you can help me out with a better paper. I also attached the material that you may need. Thanks.
Revise and rewrite: Revision of final paper draft using the peer review worksheet Please check the “Peer view worksheet” and see if there is any missing points or thing you need to change and add on t
Ideal Free Distribution: Dispersal of Goldfish ( Carassius auratus auratus ) in Response to Patch Quality Your Name Here University of California, Santa Barbara Abstract In this study, I investigated the presence of ideal free distri bution in the goldfish Carassius auratus auratus . I hypothesize d that the fish would distribute themselves in the manner predicted by ideal free theory so that there would be equal ratios of fish to food in each patch . By providing patches of food that dif fer ed in size, I was able to observe how the fish dispersed as a resul t of the differences in profitability between food patches . I found that when I did not provide food, the fish were clumped; when the patches were of equal size, the fish distributed the mselves randomly between patches; and when one patch was larger, the fish distributed themselves in an ideal free manner by clumping on the side with more food. This dispersal pattern should help maximize the fitness of each individua l by maximizing resour ce intake, a close correlate of direct fitness. Introduction Animal distribution depends largely on the dis tribution of food. An animal will forage in the location (patch) which will offer it the maximum benefit (Milinski 1987). When patches of food diffe r in profitability, animals should distribute themselves between patches so that each individual receives equal benefit . Under these conditions, there should be an equal proportion of food to animals in e ach patch (Milinski 1987, 1988). This situation resu lts in an evol utionarily stable strategy ; if an individual does not distribute it self in this ideal manner , it will incur a cost by not obtaining the optimal amount of food and will be selected against (Milinski 1988). This type of distribution is known as idea l free distribution (IFD). Ideal free theory can be used to predict the dispersal of animals between patches. IFD depends on several assumptions of the behavior of the animals involved. The only factor in the environment to which the animals react sh ould be the food, so if they conform to IFD, they should not school in response to predation risk. In order for true ideal free distribution to occur, each individual within a patch must have the same food intake . T herefore, there can be no competition or resource guarding ; these behaviors would result in differing resource intakes between individuals and hence a different distribution . Ideal free distribution has been repo rted in a number of animals , such as three -spined sticklebacks ( Gasterosteus aculea tus ) (Milinski 1987), ducks ( Anas platyrhynchos ) (Milinski 1987), flamingoes ( Phoenicopterus ruper ) (Arengo and Baldassarre 1995), zooplankton (Daphnia hyaline x galeata ) (Lampert et al. 2003), and tadpoles ( Rana temporalis ) (Veeranagoudar et al. 2004). Fo od distribution has affected all these animals such that they conform to IFD to maximize their resource intake, but there are other factors in the environment that affect their distributions. Differences in competitive abilities may alter the ratios of ind ividuals between patches (Milinski 1987, 1988) , and predation risk can cause animals to school in order to protect themselves from predators (Ryer and Olla 1998) . In this study, I examined the effect of differences in patch quality on the distribution of goldfish ( Carassius auratus auratus ), a species of freshwater fish that originated in Asia and has been domesticated for more than 300 years. We tested whether or not goldfish conform to IFD by providing different amounts of food on either side of a tank. I predict ed that 1) when no food was provided for the fish, they would distribute themselves equally relative to the two sides of the tank and evenly relative to each other (in the absence of food, there is no resource around which they clump themselves ); 2) when even amounts of food were provided on each side of the tank, the fish would distribute themselves as in the first treatment (distributed evenly between patches) ; and 3) when we provide d three times as much food on one side of the tank than on the other side, there would be, on average, three times as many fish on the “high -food” side than on the “low -food” side. Methods We tested our fish in 35.5 L (20.125 x 10.375 x 12.125 in) tanks that had a filter on one side only . Each fish tank (N=21) containe d 10 fish. We drew a line down the center of the tank s to divide them into two sides (patches) . When food was provided for the fish, it was dropped through the hole s in the screen on the tank top . T wo experimenters dropped food simultaneously into each pat ch every thirty seconds . Observations for each treatment were taken every twenty seconds for five minutes, and five minutes were taken between treatments to allow the fish to re – establish their distributions and to consume food (if there was any in the tan k). The fish on both the filter and non -filter side were counted to determine which side had more fish. The fish were observed during three treatments: no food, even food, and uneven food. No Food Treatment : In this treatment, no food was provided; instead , we observed the fish to find what their behavior was in the absence of food. We recorded the number of fish on each side (filter or non -filter) and then determined the D -value (abs(# fish on filter side – # fish on non – filter side)) for each observation. The D -value determines where the fish are in relation to each other. We performed a (2 -tailed) binomial test on the mean values for each tank to determine whether or not the fish were exhibiting a side bias and whether or not they were clumped (determined by the D -value) . Even Food Treatment : In this treatment, e ach side of the tank received two small pellets of food every thirty seconds. The same data were recorded for this treatment as were for the no -food treatment. The same tests were also performed, a nd we added a (2-tailed ) t-test to determine the effect that the addition of even amounts of food had on the clumping of the fish. Uneven Food Treatment : Food was provided in this treatment, but six pellets were given on one side, while only two were given to the other side. After five minutes of feeding and observations, we switched which side received more food to test for a side bias. All statistical tests performed on the even -food treatment data were performed for the data collected from this treatment . Results No Food Treatment : There were 21 tanks of fish subjected to this treatment; on average, there were more fish on the filter side of the tank in nine tanks, and there were more fish on the non – filter side of the tank in 12 tanks (Figure 1) . This result was not significantly different from random (Two -tailed Binomial test, n=21, r=9, r crit =5, p>0.05) . The mean D -val ue for this treatment was 3.522. This value was greater than what would be expected if the fish were distributed randomly relative to eac h other , so the fish were clumped in this treatment (Two – tailed D -test, D obs =3.522, D crit =3.46, D obs > Dcrit ). Even Food Treatment : Out of the 21 tanks, 12 had more fish on the filter side, and nine had more fish on the non -filter side (Figure 2) . This res ult was not significantly different from random (Two -tailed Binomial test, n=21, r=9, r crit =5, p>0.05). The mean D -value was 3.113, which lies within the values that would be expected if the fish were distributed randomly relative to each other , which mean s that the fish were neither evenly distributed nor clumped (D obs =3.113, Dcrit =3.46, D obs < Dcrit ). The difference in the D -values between the no -food treatment and the even -food treatment was not significant (Two -tailed T -test, df=40, t=1.039, t crit =2.021 , t Dcrit ). Here, the difference between the D -value for the even -food treatment and the D -value for the uneven -food treatment was significant (Two -tailed T-test, df=40, t=4.163, t crit =2.021, t>t crit ). Discussion Our results suggest that goldfish may conform to ideal free distribution . While the fish in this study did not conform per fectly to IFD in all treatments, they began to follow that dispersal pattern by the end of the experiment. Under the no -food treatment, the fish did not show a bias towards either side of the tank. They were clumped, which is not what I expected, given th e even lack of food on each side of the tank. It is possible that the fish were schooling to defend themselves from predators. If the fish clump together, there is less of a chance that each particular individual wil l be the eaten if a predator attacks the schoo l (the dilution effect) . Without food present, there would be no incentive for the fish to leave the protection of the school (Ryer and Olla 1998). In the even -food treatment, the fish did not exhibit a bias towards either side of the tank (i.e., ei ther patch). They distributed themselves randomly with respect to each other. While the fish were no longer clumped, they were not distributed evenly as I predicted. The goldfish did not conform to IFD in this treatment; however, they were less clumped tha n they were in the fi rst treatment, so it is possible that our results represent a trend towards ideal free distribution. Under the uneven -food treatment, the fish showed a bias towards the high -food side of the tank in every trial. They were clumped on t he high -food side, which means that they altered their dispersal in response to the uneven distribution of food. Here, the goldfish did conform to ideal free distribution, meeting my prediction and supporting my hypothesis that goldfish dispersal can confo rm to IFD. This study examines the distribution of fish under different treatments of food dispersal. The results of our study support the ideal free theory because the fish had begun to conform to IFD by the end of the experiment. These fish would have b een distributing themselves in this manner because doing so would maximize their resource intake . Maximizing resource intake would likely increase their lifetime fitness by allowing them to survive to reproduce and to put the maximum amount of energy and r esources into their offspring. The goldfish did not always conform to this ideal distribution; when predation was most likely their primary concern (due to the absence of food throughout the tank ), the fish schooled to protect themselves. However, when foo d became available, the fish began to distribute themselves in an ideal free manner to increase their food intake . The behavior of these fish (and other animals which conform to IFD) is adaptive because maximizing resource intake should help maximize lifet ime fitness. Literature Cited Arengo, F. and G.A. Baldassarre. 2002. Patch choice and foraging behavior of nonbreeding American Flamingos in Yucatan, Mexico . Condor 104 (2):252 -257. Lampert, W. et al. 2003. Trade -offs in the vertical distributi on of zooplankton: ideal free distribution with costs? Proceedings of the Royal Society of London B – Biological Sciences 270(1516) : 765 -773. Milinski, M. 1987. Competition for non -depleting resources: the ideal free distribution in sticklebacks. In: Fo raging Behavior (A.C. Kamil et al., eds. ), pp. 363 -388. New York: Plenum Press. Milinski, M. 1988. Games Fish Play: Making Decisions as a Social Forager. Trends in Ecology & Evolution 3(12): 325 -330. Ryer, C.H. and B.L. Olla. 1998. Shifting the balance b etween foraging and predator avoidance: the importance of food distribution for a schooling pelagic forager. Envi ronmental Biology of Fishes 52 (4):467 -475. Veeranagoudar, D.K. et al. 2004. Foraging Behavior in tadpoles of the bronze frog Rana temporalis : Experimental evidence for the ideal free distribution. Journal of Biosciences 29 (2): 201 -207. Tables and Figures Figure 1. Side of tank with majority (more than 5) of fish : Filter side (9) and Non -filter side (12). Standard Error bars re present 1 SE ±. Figure 2 . Side of tank with majority of fish: Filter side (12) and Non -filter side (9). Standard Error bars represent 1 SE ±. Side of Tank with Majority of Fish — No Food Treatment 0 5 10 15 20 Filter Side Non-Filter Side Side of Ta nk w ith Ma jority of Fish Number of Tanks Side of Tank with Majority of Fish — Even Food Treatment 0 5 10 15 20 Filter Side Non-Filter Side Side of Ta nk w ith Ma jority of Fish Number of Tanks Figure 3. Side of tank with majority of fish: High -food side (21) and low -food side (0). Standard Error b ars represent 1 SE ±. Side of Tank with Majority of Fish — Uneven Food Treatment 0 5 10 15 20 25 High Food Side Low Food Side Side of Ta nk w ith Ma jority of Fish Number of Tanks