Discussion

On the first two days that I recorded data on Batillaria minima, there was no visible input from the ocean in the tidal pool where I placed the rocks.  However, on the third day, when I visited the site in the morning and the afternoon, there was regular spray from the ocean.  This caused a difference in temperature from the previous day by 2° C (table 1.1).  It also caused sedimentation to build up on the rocks and wave action to move the rocks (thankfully they were well marked).

Concentration of snails around the rocks differed each day, but were present around all rocks on all days.  On the third day, patterns of aggregation were not uniform enough to quantify the concentrations.  Because of this, the estimation used, and the uncertainty, no statistical analysis was conducted on this information. Instead, percent cover under and above the rocks were used as the indicies of comparison.

 On all three days most snails were found aggregating around or underneath the rocks, usually in a circular pattern.  Because the tide pool had few other visible species, the fairly regular movement patterns agree with a finding of Underwood and Chapman (1994) that Littorina unifasciata, another intertidal gastropod, moved faster and more directly in simple habitats than in complex ones (2).  

Rock 3 was the only rock that consistently did not have a circular pattern when removed, perhaps due to its curved shape.  An ANOVA determined that there was a significant difference between the mean percent cover above the rocks and the mean percent cover under the rocks (tables 2.3, 3.3, 4.3). P-values for the three days were 0.0155, 0.0046, and 0.0139 respectively, which are all significant because they are less than .05, which indicates that it is unlikely that these results could be attributed to random distributions.  This shows that the conditions, above or below, caused the distributions. Greater numbers may have been found under the rocks than above because there was more muddy sediment under the rock, which is the preferred habitat of of B. minima (7). On day two, the hottest day with the most direct sunlight, the difference between the two conditions was the greatest; suggesting that position on or under the rock could be the result of higher temperatures (tables 2.3, 3.3, 4.3).  However, the study by Wiser et al (1981) shows that B. minima can withstand high temperatures (11).  This suggests that perhaps it was sunlight, which was more intense day 2, rather than heat, that caused the density of aggregations to increase.  This is also consistent with a study that Chapman and Underwood (1996) did on Littorina unifasciata, which found that the temperature of the snails did not affect their tendency to aggregate (3).

T-tests were used to compare the difference in percent cover above each rocks (table1.4). Although no significant difference was found between the percent cover above the rocks, because no p-values were below .05, many were close to .05 which suggests the possibility of a difference. This was particularly noticeable for rock 2.  All p-values calculated comparing rock 2 to the other rocks were all below .1 (p= 0.086595, 0.085143, 0.06346, 0.066435, table 1.4).  This was not true for any other rock. Rock 2 was also the darkest rock, suggesting, again, that the aggregation of B. minima  was an attempt to regulate or increase their body temperature.  The darker color might also have been favored as a defense mechanism because it is close to the dark color of many of the snails, possibly creating a camouflage effect.  These results could be due in part to organic matter of some type or source already present on the rock, because I had previously found B. minima aggregated around a food source.  If this experiment were to be replicated, I would suggest that the rocks somehow be sanitized or perhaps man made material could be used instead of rocks to eliminate confounding variables.

Single factor ANOVA were used to compare the percent cover and above the rocks between the days (tables 1.5 and 1.6) and the percent cover and above the rocks between the rocks (tables 1.7 and 1.8).  These tests found that there was a significant difference under the rocks between the days (p=0.011528, table 1.5) and percent cover above between the rocks (p=0.030711, table 1.8).  This means the averages were different under the rocks between the days, due to differences of the days (weather, temperature).  The difference between the rocks indicates that differences in the rocks were also causing different intensities and patterns of aggregation.

T-test were also used to compare percent cover under the rocks (table 1.3).  These results seemed more uniform and no pattern was evident.  There was, however, a significant difference between the percent cover under rock 1 and rock 4 (p=0.010656, table 1.3).  The physical characteristics of these two rocks show no obvious reasons for this difference.

The third day was the only day I was able to visit the site at two different times, in the morning and the afternoon (the usual time).  If temperature and water input do play a role in distribution, it would be useful to visit the site at several times during the day to account for temperature and water input differences (low tide verses high tide).

If this experiment was more standardized, the results may have been clearer. Beck (2000) found that the manipulation of gastropod habitat structure can confound results (1). I found this held true for B. minima. Because I was manipulating the rocks based on multiple characteristics, it was often difficult to interpret results, because one or more of the characteristics (size, shape, color) could have been causing the snails to aggregate at different densities.