A bell curve is symmetrical, which is to say that on both sides of the mean the shape of the curve is precisely the same. Each side is a mirror image of the other side. Unimodal means that there is only one mode for the set -- that is, only one number occurs most frequently.
Bell curves are used to describe probabilities and as such the area of the curve is equal to 1 percent of the values. That is to say that each iteration along the axis beneath the curve -- if aggregated -- will equal 1.00. This allows any population that can be described by the bell curve to be interpretable.
The mean and standard deviation of a bell curve remain the same. The mean of the curve -- or normal distribution -- is always zero and the standard deviation is always in base 10 from the mean. Traditionally, increments of 1.0 in standard deviations are used to represent data and deviation from the mean score of the set.
Within the bell curve are a set of fixed proportions based upon standard deviations from the mean. Within one standard deviation from the mean, 68 percent of the population is included. Two standard deviations represent 95 percent of the population, or data set. And within three standard deviations, 99.7 percent of the population is accounted for.