Discussion

__Discussion__

For our experiment we have looked at the relationships between 4 high and low altitude sites and have compared 2 phenotypic traits; the colour of the shell (whereby a snail can be separated into two groups, brown and pink or yellow) and by the number of bands the snail had on its shell (either 0-3 or 4-5). By grouping these traits we are more likely to see a significant difference between sites. Our first step was to determine whether there was a significant difference between the shell colour of the high altitude sites grouped together and the low altitude sites grouped together by performing two separate chi squared (χ2) tests, with the null hypothesis being ‘there is no significant difference between the phenotypes at this altitude’. At the high and low altitudes the χ2 test produced the values of 3.36 and 4.63 respectively. We did this test at the 5% significance level at had 3 degrees of freedom making our critical value 7.82. As you can see both our values were less than our critical value allowing us to accept H0, and concluding that there was no significant difference. We followed this by combining these two χ2 tests producing one, making our new null hypothesis: ‘there is no significant difference between the shell colour phenotype in the high altitude compared with the shell colour phenotype of the low altitude’. This produced a χ2 value of 7.99; the critical value at a 5% significance level with 9 degrees of freedom is 12.59, therefore again we can accept H0 and assume that there is no significant difference in shell colour amongst altitude in our 8 sites, thus me moved on to looking at the relationship between number of bands and altitude. We conducted this experiment slightly differently, by grouping sites 1-2 and 3-4 in both altitudes, as values of some sites had less than 5 snails reducing the accuracy of the measure. The first null hypothesis we tested was: ‘there is no significant difference in the band number phenotype amongst the high altitude populations’, with Yates correction the χ2 value was 10.10, the degrees of freedom is 1, which means using a 5% significance level the critical value is 3.48. As 10.10<3.48 we can reject H0, therefore showing there is a significant difference in the number of bands in the high altitude populations. We repeated this procedure with the low altitude populations and again could reject H0 due to a χ2 value of 7.32 showing there is also a significant difference between low altitude populations. Our final step was to group the high altitude sites and test them against each low level site individually, this produced 4 χ2 tests each with a 5% significance level and 1 degree of freedom giving a critical value of 3.84, when compared to the grouped high altitude sites the low altitude sites 1, 2, 3 and 4 produced the χ2 values 6.1, 1.68, 2.11 and 0.01 respectively, this shows that only site 1 shows a significant difference against the high altitude sites. Due to all 4 high altitude sites and all 4 low level altitude sites showing significant difference amongst them we cannot conclude that the difference was due to altitude. Therefore our experiment shows that the polymorphisms in //Cepaea nemoralis// is more likely to be due to random processes i.e. genetic drift than natural selection.