Table of Content
HomeIntro & Methods
Nahant
Results
At first glance it appears that there is a trend for Hermit crabs to choose Littorina littorea shells over the other species present in the intertidal zone (figure 1). After putting my data set through a Chi-Square test I got 4.3 as my test statistic which is not higher than the critical value of 5.9 at a p-value=.05 (figure 2). There was also a large distribution of Littorina littorea in my sample sites (figure 3)
At first glance it appears that there is a trend for Hermit crabs to choose Littorina littorea shells over the other species present in the intertidal zone (figure 1). After putting my data set through a Chi-Square test I got 4.3 as my test statistic which is not higher than the critical value of 5.9 at a p-value=.05 (figure 2). There was also a large distribution of Littorina littorea in my sample sites (figure 3)
Discussion
From these results it appears as though there is no trend for Pagurus longicarpus to choose on shell over another. Because my Chi-Square value was not higher than the critical value I can't reject the null hypothesis that there is no trend for shell preference. One reason for this could be that in the wild Pagurus relies on what it can find and may not have the treat of getting to choose what shell it wants to be in. It has to pick a shell that is just the right size so that it can fit in it, but at the same time has to have some room for it to grow in. If littorea shells or saxitillus shells are the only thing that it can get its claws on then it probably doesnt have a choice in the matter introducing an element of chaos that could cover up any trends. Brendan Clancy's project from 2006 shows that in the lab a P. longicarpus has tendencies to pick one shell over another depending on its size. Some problems that my data set has are its small sample size collected over a short period of time. If the sample size was around 30 or so crabs a trend might have popped up instead of hiding behind random variation. Also a lab part to this experiment would have allowed me to bipass or account for randomness and given me some better data.
Figure 1: Observed values for P. longicarpus
| Observed | Expected | |
| 10 | 6 | |
| 5 | 6 | |
| 3 | 6 | |
| 18 | 18 | Total |
| Tide Pool 1 | Tide Pool 2 | Tide Pool 3 | Tide Pool 4 | |
| L. littorea | 16 | 23 | 26 | 42 |
| L. saxitillus | 11 | 5 | 7 | 15 |
| L. obtusata | 4 | 1 | 0 | 5 |
| Nucella | 3 | 6 | 0 | 0 |