A suggested area to perform this study is the Gower peninsula in South Wales. The specific location of this study was Port Eynon bay, Kilvrough.
The Gower Peninsula
Once the study area is chosen it is important to make certain preparations before starting any measurements:
- For every shore to be studied, obtain 8 random co-ordinates (i.e. when drawing a graph one obtains X, Y co-ordinates to plot points). Make a results table with the following information: Type of shore, Limpet, Width, Height, Orientation. It is recommended to do measurements on at least 40 limpets (A larger number of limpets will give a better of the diameter/height ratio distribution of the study area. It will also give a better representation of the mean in the population, therefore allowing for better results in a statistical test. The same number of limpets should be studied on each shore)
E.g. Sheltered shore
- Group members must be well dressed for the conditions of the shore; cold weather, rain, winds etc. Wellington boots are highly recommended for protection from water and the sharp rocks.
The following is what each person of the group should do at their corresponding shores.
- Once the middle shore region is reached:
With the interval marked string, set up a study area of 10 X 10 meters in the same way as one would draw co-ordinate axes for graphs (perpendicular lines). This is important in order to follow an orderly method of work, and to apply the use of the random co-ordinates.
- Use the co-ordinates produced before to identify the area where the quadrat will be placed (using the markings on the string as values for the co-ordinates) Place one of the corners of the quadrat on this point. (if there is a large rock pool where the co-ordinates indicate, produce new random co-ordinates until an adequate area is found)
- Identify the limpets that are closest to the four corners of the quadrat and the one closest to the middle (see picture 3). Use the vernier callipers to measure their height and diameter.* Remember to be accurate to 0.05 millimetres (Be careful with the metal rod sticking out of the calliper!)
- Now take the second set of coordinates and repeat step 5. Do this for all sets of coordinates.
- Reunite all the data with group members in order to perform a statistical test.
*If time allows measure the orientation of each limpet with the compass. This may give additional information on the nature of how limpets find their homes.
The statistical analysis method used for the data attached was a confidence level test. (The method and the results obtained are attached) This method was used due to the fact that it is an ideal test for relatively large samples. The test can prove to be very useful by giving an idea of the statistical significance when comparing the ratios of three shores together. The results obtained here can be used to see if the hypothesis previously stated is correct:
- Null hypothesis: the diameter/height ratio of limpets does not differ significantly between the shores they inhabit. i.e. This ratio could be explained by other factors outside the scope of this investigation.
After performing the confidence level test, it can be concluded graphically (graphs attached) that there is indeed a significant statistical difference between the mean ratios of limpets on the exposed and sheltered shores. One can therefore reject the null hypothesis of no relationship between ratio and shore type.
The confidence limit graph shows that there is no overlap between the confidence limits of the sheltered shore and the exposed shore. The confidence limits of exposed shores and semi-exposed shores do not overlap either; again this shows a statistical significance in the difference. The confidence limits of semi-exposed and sheltered overlap almost completely; we cannot therefore conclude that the slight difference between the mean ratios of these two shores is of any statistical significance.
The frequency histogram gives us an idea on how the diameter/height ratio in limpets is distributed over the sample taken. Each type of shore has an evident pattern. The exposed shore bars tell us that a greater proportion of limpets in the sample have a low ratio; the modal groups being (0.2-0.3) and (0.3-0.4). The most common ratio for the semi-exposed shore is tending towards higher values (0.6-0.7). For the sheltered shore one can identify a similar distribution, but the modal group is in the middle of the range (0.5-0.6), giving what resembles a normal distribution.
The patterns displayed in these results can be explained in various ways. The limpets must deal with different factors in order to ensure survival; wave action, strong winds that make the environment dry, predators, food availability etc. Sometimes they will cope with strong tides and winds by finding a microhabitat under a rock or using the edge of a rock pool as a shield. In reality, not all limpets will be able to find the ideal microhabitat. To deal with harsher conditions, the only way to cope would be to increase their strength and reduce their exposure. The latter is the reason why we found the mean ratio on the exposed shore to be reduced; a larger foot and a lower height (0.366 being the mean). The frequency histogram is representative of this as it shows that most ratios tend towards the low values (the ranges 0.2-0.3, and 0.3-0.4 with 13 each) The low ratios will make the limpets more streamline in the water, which will prevent them from being washed away from their rock scar when moving around finding food.
It will also help them in the low tide; preventing them from being washed away by breaking waves that do not have to be dealt with when the tide is high.
Their large diameter means their foot is bigger (a large foot can produce a pressure of 5.3 kg cm/2, changing the diameter will change the pressure value significantly), therefore increased suction maximises adhesion to the rock and prevents them from losing humidity in strong winds. The limpets that do not have these characteristics will find it very difficult to survive unless they have a home that is sheltered permanently (which is why one finds tall limpets occasionally on the exposed shore, 1 limpet in the range 0.7-0.8, none in the range 0.8-0.9). They will die out and the limpets with the ideal characteristics will survive. It can be said that natural selection is at work on this shore. Selection pressure, taking the form of environmental conditions, has pushed the distribution of the population (in this case we have a sample) towards lower diameter/height ratios. The number of limpets with higher ratios will remain existent, but very low. Limpets with very low ratios (only a couple of limpets very near to 0.2 were found) would also be existent, but low in number. A normal distribution for this shore would look like this:
The sheltered shore sustains limpets with a higher ratio (increased tallness, reduced diameter, 0.589 being the mean). This suggests that the limpets do not need to compromise their tallness and the size of the foot does not have to be large to survive. When looking at these results from an evolutionary point of view, it can be said that if the environment changes drastically (i.e. tide increases due to a natural occurrence; global warming causing poles to melt, a meteor crashes in the area etc.) selection pressure could push the distribution towards lower ratios. Having an intermediate ratio would therefore be beneficial from an evolutionary point of view. Limpets with high ratios would quickly disappear, but the ones tending towards the intermediate/lower ratios would survive and these characteristics would remain in the population.
The results obtained at the semi-exposed shore are very similar to the ones obtained at the sheltered. From the frequency histogram we can see that the range is similar, and the modal group is higher (0.6-0.7), the mean is a bit lower (0.552). The similarity can be due to the fact that limpets apply strong pressure to the rock no matter where they live, and tallness will only be compromised when faced with increased harsh conditions (as on the exposed shore). The conditions here will not be harsh enough to reduce the tall limpet population and therefore the distribution will be very similar to the sheltered shore. One could have expected a greater difference between the means of these two shores as opposed to almost identical results, this possible “anomaly” could be down to sample size which will be discussed further on.
After having performed this study it is possible to identify certain aspects of the method that make the data collection more difficult. At the sheltered shore there was a very high population of barnacles that covered most of the limpets; this wasn’t the case at the other shores. Even if the limpets were uncovered, identifying them would be an arduous task; it is very hard to identify the species by just their shells unless one has a wide biological experience on the shore. The only fool proof method to identify them would be to detach them from their scar. They do not reattach readily, they would die and the ecological nature of this study would be lost. The barnacles made it impossible to identify the species of limpet, sometimes the diameter/height ratio depends on the species. We can not be completely sure if the difference in ratios on the shores is down to species diversity.
Callipers are accurate devices, but sometimes it is difficult to read the decimal values of the length and height correctly. Limpets that lived in crevices were quite hard to measure with the callipers due to physical restrictions; there is room for error in these measurements. This happened on the sheltered shore more than the other two; the distributions obtained could be different from the ones in reality; the conclusions reached could therefore be imprecise.
Orientation of limpets is difficult to take into account because one has to assume where the front of the limpet lies without being able to actually know. Therefore no concrete conclusions could be drawn from the data obtained.
For the semi-exposed shore only 20 samples were collected as opposed to 40 as on the other shores. This makes it difficult to say the results from this shore are a representation of the whole population. It is quite probable that the ratios are distributed in a different way when looking at a more representative sample.
Apart from the latter, the results seem to make sense. The samples don’t produce perfect normal distributions or patterns, but this is just due to the fact that we are not counting the whole population (the trend observed in the whole population would be clearer). A larger sample will always be an improvement someone can make if they have the time.
All these factors combined can push the results towards inaccuracy. The confidence limits might change and the difference in means might prove to lose statistical significance. The confidence limits obtained at the exposed shore are well separated from the ones obtained at the other two shores. I think there is enough space between the two means to be confident that the trends obtained are not down to chance; they do have a statistical and scientific importance.
Picture 4
If we divide the length of the shore in 12 parts, the middle shore region will be found in the shaded area; the area bounded by 5/12 down the shore and 8/12 down the shore. Limpet measuring should start at mean tide level time, when the water is exactly 6/12 down the shore.
Bibliography:
Cremona, Julian. A field Atlas of the Shore
Little, Colin and Kitching. J.A The Biology of Rocky Shores
Hawkins, Stephen J. The Rocky Shores
Sheltered, Semi-Exposed, and Exposed are measures of the exposure of the shore facing the tide
A flat area of rock shaped by the years of presence of the limpet
Mean tide level: Time of day when the tide is covering half the tidal range. The time at which this happens will vary depending on the day. In our study mean tide level was at 12:03
