To come next to the Red Crag, the reader will observe that a percentage of 25 is given of shells unknown as living, and this increases to 31 in the antecedent Coralline Crag. But the gap between these two stages of our Pliocene deposits is really wider than these numbers would indicate, for several reasons. In the first place, the Coralline Crag is more strictly the product of a single period, the Red Crag, as we have seen, consisting of separate and independent patches, slightly varying in age, of which the newest is probably not much anterior to the Norwich Crag. Secondly, there was a great change of conditions, both as to the depth of the sea and climate, between the periods of the Coralline and Red Crag, causing the fauna in each to differ far more widely than would appear from the above numerical results. The value of the analysis given in the above table of the shells of the Red and Coralline Crags is in no small degree enhanced by the fact that they were all either collected by Mr. Wood himself, or obtained by him direct from their discoverers, so that he was enabled in each case to test their authenticity, and as far as possible to avoid those errors which arise from confounding together shells belonging to the sea of a newer deposit, and those washed into it from a formation of older date. The danger of this confusion may be conceived when we remember that the number of species rejected from the Red Crag as derivative by Mr. Wood is no less than 25. Some geologists have held that on the same grounds it is necessary to exclude as spurious some of the species found in the Norwich Crag proper; but Mr. Wood does not entertain this view, believing that the spurious shells which have sometimes found their way into the lists of this crag have been introduced by want of care from strata of Red Crag.