One element which is usually found unpalatable to modern thought is the miraculous ; and one way of dealing with it has been simply to suppress the miraculous features.1

The presence of miracle does not of itself invalidate a legend. The story that a certain arahat attended an assembly may be true, even if we are told that he passed through the air on his way thither. To the chronicler this feature was miraculous, but at the same time quite normal for an arahat. When however we are told that Buddha paid three visits to Ceylon, we get no nearer to historical fact by suppressing the circumstance that he went through the air. The presence of miracle has in fact little to do with the question whether some historical basis underlies a legend. Normal circumstances are quite as likely to be invented as miracles. A much more important means of testing a legend is to compare the different forms in which it appears. It may have been elaborated, or an elaborate legend may have been rationalised. Additional incidents may be inserted in awkward places, or quite contradictory accounts of the same circumstance may be recorded.