In many cases, these two statements mean the same thing. However, they can also be quite different.
Testing a hypothesis consists of first saying what you believe will occur with some phenomenon, then developing some kind of test for this phenomenon, and then determining whether or not the phenomenon actually occurred. In many cases, testing of a hypothesis need not involve any kind of statistical test. I am reminded of this quote by the physicist Ernest Rutherford - If your experiment needs statistics, you ought to have done a better experiment.
That being said, testing of hypotheses normally does use some kind of statistical tool.
In contrast, testing of significance is a purely statistical concept. In essence, one has two hypotheses - the null hypothesis, which states that there is no difference between your two (or more) collections of data. The alternative hypothesis is that there is a difference between your two samples that did not occur by chance.
Based on the design of your study, you then compare the two (or more) samples using a statistical test, which gives you a number, which you then compare to a reference distribution (like the normal, t, or F distributions) and if this test statistic exceeds a critical value, you reject the null hypothesis and conclude that there is a difference between the two (or more) samples. This criterion is normally that the probability of the difference occurring by chance is less than one in twenty (p<0.05), though others are sometimes used.