nensis in the Norfolk stone-bed, resting on the old chalk land-surface at Horstead, which specimen was not preserved, but of which he obtained two molars from the owner of the pit in which the bones were found.
The relative abundance of the remains of Mastodon in the Norfolk and Suffolk beds is important. In the Norfolk stone-bed Mastodon arvernensis is undeniably very much more abundant than in the Suffolk bone-bed. This was sufficiently evident before the Suffolk area had been so largely worked for the phosphatic nodules. Now that the Suffolk bed has been carefully sifted and turned over for so many acres, there is danger of overestimating the abundance of its mammalian fauna, as compared with that from Norfolk. Teeth of terrestrial mammalia are of the extremest rarity in the Suffolk bed ; and it is only because of the high price offered for them, and the constant operations of the " coprolite-diggers," that so many of them have been found. Comparing equal areas of exploration, the molars of Mastodon arvernensis are very much more abundant in Norfolk than Suffolk. In the Norfolk stone-bed the local collectors find that Mastodon is about twice as abundant as Elephas meridionalis, whilst Equus and species of Cervus are more abundant than either.
A fact of importance with regard to the occurrence of Mastodon arvernensis in the Suffolk bone-bed is, that molars of this animal have been obtained from the bed with the soft bony fangs adherent, whilst in the case of nearly all other associated mammalia the enamel crowns only are found.
It is important to notice that the bones of terrestrial mammals are almost unknown in the Suffolk bone-bed, whilst they are abundant in the Norfolk bed. It is also necessary to bear in mind that though the Suffolk bone-bed occurs below the Coralline Crag as well as below the Red Crag, yet it has not been proved that the mammals common to the Norfolk and Suffolk beds, viz. Mastodon arvernensis, certain species of Cervus and Equus, are found in the bed when below the Coralline Crag. A Mastodon tooth which I have seen from that situation is not Mastodon arvernensis, but belongs to the Trilophodont species to be described below.
These, being the facts of the case, it is well simply to state the hypotheses by which Ave may account for the occurrence of Mastodon arvernensis, — in Norfolk on the one hand, associated with Elephas meridionalis ; in Suffolk on the other hand, in most intimate connexion with Upper-Miocene mammals.
1. Seeing the very fragmentary nature of the remains found in both the Suffolk and Norfolk bone-beds, we might suppose that the absence of Elephas in the Suffolk bed, and of Miocene forms in the Norfolk bed, is due to imperfect knowledge of the contents of the beds, which may constitute but one fauna. This is negatived by the improbability of such an association of forms to constitute a fauna as we should then get, and by the fact that the Suffolk bed at least has been remarkably well searched.
2. Another explanation might be found in regarding Mastodon arvernensis as an annectant form, one which lived first with a Mio-
