TRANSACTIONS OF SECTION C. 647 



in the Invertebrata amoimts only to 6 out of 88, or about seven per cent., while in 

 the Vertebrates it is not less than 12 out of 40, or 30 per cent. These figures imply 

 that the amount of ordinal change in the fossil Vertebrates stands to that in the 

 Tnvertebrata in the ratio of 30 to 7. This disproportion becomes still more marked 

 when we take into account that the former had less time for variation than the 

 latter, which had the start by the Cambrian and Ordovicean Periods. It follows 

 also that as a whole they have changed faster. 



The distribution of the extinct orders in the animal kingdom, taken along with 

 their distribution in the rocks, proves further that some types have varied more than 

 others, and at various places iu the geological record. In the Protozoa. Porifera, and 

 Vermes there are no extinct orders ; among the Ccelenterates one : the E,ugosa ; iu the 

 Echinodermata three: Cystideans, Edriasterida, and Blastoidea; in the Arthropoda 

 two : the Trilobita and Eurypterida. All these, with the solitary exception of the 

 obscure order Rugosa, are found only in the Primary rocks. Among the Pisces 

 there is none ; in the Amphibia one ; the Labyrinthodouts ranging from the Car- 

 boniferous to the Triassic Age. Among the Keptilia there are at least six of 

 Secondary age: Plesiosauria, Ichthyosauria, Dicynodontia, Pterosauria, Therio- 

 dontia, Deinosauria ; in the Aves two : the Saururse and Odontornithes, also 

 Secondary. In the Mammalia the Amblypoda, Tillodontia, Condylarthra, and 

 Toxodontia represent the extinct orders — the three first Early Tertiary, and the last 

 Pleistocene. It is clear therefore that, while the maximum amount of ordinal 

 variation is presented by the Secondary Reptiha and Aves, all the extinct orders 

 in the Tertiary are Mammalian. 



If we turn from the extinct orders to the extinct species, it will also be found 

 that the maximum amount of variation is presented by the plants, and all the 

 animals, excepting the Mammalia, in the Primary and Secondary Periods. 



The general impression left upon my mind by these facts is that, while all the rest 

 of the animal kingdom had ceased to present important modifications at the close of 

 the Secondary Period, the Mammalia, which presented no great changes in the 

 Secondary rocks, were, to quote a happy phrase of Professor Gaudry, ' en pleine 

 evolution ' in the Tertiary Age. And when, further, the singular perfection of the 

 record allows us to trace the successive and gradual modifications of the Mammalian 

 types from the Eocene to the close of the Pleistocene Age, it is obvious that they 

 can be used to mark subdivisions of the Tertiary Period, in the same way as the 

 reigns of kings are used to mark periods in human history. In my opinion 

 they mark the geological horizon with greater precision than the remains of the 

 lower members of the animal kingdom, and in cases such as that of Pikermi, 

 where typical Meiocene forms, such as Deinotheria, are found in a stratum above 

 an assemblage of marine shells of Pleiocene age, it seems to me that the Mammalia 

 are of greater value in classification than the Mollusca, some of the species of which 

 have been living from the Eocene down to the present day. 



Yet another important principle must be noted. The fossils are to be viewed 

 in relation to those forms now living iu their respective geographical regions. The 

 depths of the ocean have been where they are now since the earliest geological 

 times, although continual geographical changes have been going on at their 

 margins. In other words, geographical pi'ovinces must have existed even in the 

 earlier geological periods, although there is reason to believe that they did not 

 differ so much from each other as at the present day. It follows from this that 

 the only just standard for comparison in dealing with the fossils, and especially of 

 the later rocks, is that which is offered by the fauna and flora of the geographical 

 province in which they are found. The non-recognition of this principle has led to 

 serious confusion. The fauna, for example, of the Upper Sivalik Formation has 

 been very generally viewed from the European standpoint and placed in the 

 Meiocene, while, judged by the standpoint of India, it is really Pleiocene. A 

 similar confusion has followed from taking tlie Meiocene flora of Switzerland as a 

 standard for the Tertiary flora of the whole of the Northern Hemisphere. 



It now remains for us to see how these principles may be applied to the co- 

 ordination of Tertiary strata in various parts of the world. In 1880 I proposed a 

 classification of the European Tertiaries, in which, apart from the special cha- 



