446 



NA TURE 



\_Sept. 4, 1884 



that the marine coast-beds and the associated plant-beds are in 

 Africa much newer than the inland sandstone formation, but it 

 is not impossible that the relations may really be the same as in 

 India, and that the Stormberg beds of the inland formation may 

 be the equivalents of the Upper Jurassic or even the Cretaceous 

 marine beds on the coast. "Ihe discovery of plants identical with 

 those of the Jurassic (probably Upper Jurassic) beds of Queens- 

 land in the Stormberg series may of course be taken for what it 

 is worth ; it is of quite as much importance in indicating the age 

 of the rocks as the occurrence of dicynodont reptiles in the 

 Permian of Russia and in the lower Godwanas of India. 



Altogether there is quite sufficient probability that the upper 

 Karoo or Stormberg beds are of later age than Triassic to justify 

 the protest which I made last year against a skull being described 

 from these beds as that of a "Triassic " mammal (Q. J. G. S. 

 xl. p. 146). The practice, so common amongst palaeontologists, 

 of positively asserting as a known fact the geological age of 

 organisms from beds of which the geological position is not 

 clearly determined is very much to be deprecated. 



I have called attention to the occurrence of boulders in the 

 Talchir beds in India, the Ecca beds of South Africa, and the 

 Bacchus Marsh sandstones and Ilawkesbury beds of Australia. 

 The idea has occurred quite independently to several different 

 observers that each of these remarkable formations affords 

 evidence of glacial action ; and although, in the case of India 

 especially, the geographical position of the boulder-bed within 

 the tropics seemed for a long time to render the notion of ice 

 action too improbable to be accepted, further evidence has so far 

 confirmed the view as to cause it to be generally received. Even 

 before the Australian boulder-deposits had been observed, it 

 was suggested that the Talchir beds and Ecca conglomerate 

 might be contemporaneous (Q.J. G. S. xxxi. p. 528), and that 

 the evidence in favour of a Glacial epoch having left its traces in 

 the Permian beds of England (Q. J. G. S. xi. p. 185) might 

 possibly indicate that the Indian and South African boulder-beds 

 are of the same geological epoch. The discovery of two similar 

 deposits in Australia adds to the probability that all may have 

 resulted from the same cause and may record contemporaneous 

 phenomena. It would be very unwise to insist too much on the 

 coincidence. 



It would be easy to call attention to further examples of dis- 

 crepancies in paL-eontological evidence, but I should weary you 

 and nothing would be attained by going through instance after 

 instance of deposits in distant parts of the world, the age of 

 which has been solely determined by the examination of a few 

 fossil forms of land and fresh-water animals and plants. I have, 

 therefore, only taken a few with the details of which I have had 

 occasion to become acquainted. In some of the most important 

 cases I have mentioned, such as those of the Pikermi and Sivvalik 

 faunas, the Cutch (Umia bed.) flora, and that in the lower Coal- 

 Measures of Australia, the conflict is between the evidence of the 

 marine and terrestrial organisms. Manifestly one or the other 

 of these leads to erroneous conclusions. 



The general opinion of geologists is in favour of accepting the 

 evidence of marine organisms. The reason is not far to seek. 

 So far as I am aware no case is known where such an anomaly 

 as that displayed in the Gondwanas of India has been detected 

 amongst marine formations of which the sequence was un- 

 questioned. In the Gondwanas we have a Rhtetic flora overlying 

 a Jurassic flora, and a Triassic fauna above both. In Australia 

 we find a Jurassic flora associated with a Carboniferous marine 

 fauna, and overlaid by a Permian fresh-water fauna. The only 

 similar case amongst marine strata is that of the well-known 

 colonies of the late M. Barrande in Bohemia, and in this instance 

 the intercalation of strata containing later forms amongst beds 

 with older types is disputed, whilst the difference in age between 

 the faunas represented is not to be compared to that between 

 Triassic and Jurassic. 



There is, however, another and an even stronger reason for 

 accepting the evidence of marine instead of that afforded by 

 terrestrial and fresh-water animals and plants. If we compare 

 the distribution of the two at the present day, we shall find a 

 very striking difference, and it is possible that this difference 

 may afford a clue to the conditions that prevailed in past 

 times. 



Wanderers into what they fancy unexplored tracts in pale- 

 ontology are very likely to find Prof. Huxley's footprints on 

 the path they are following. I have had occasion to turn to a 

 paper of his on Hyperodopedon (Q. J. G. S. xxv. p. 150), that 



very curious reptile already mentioned, of which the remains 

 occur both in Great Britain and in India, and I find the following 

 remarks, which appear so exactly to express a portion of the 

 view to which I wish to call your attention, that I trust I may 

 be excused for quoting them. Prof. Huxley writes : — 



" It does not appear to me that there is any necessary relation 

 between the fauna of a given land and that of the seas of its 

 shores. The land-fauna; of Britain and Japan are wonderfully 

 similar ; their marine fauna; are in several ways different. 

 Identical marine shells are collected on the Mozambique coast 

 and in the easternmost islands of the Pacific ; whilst the fauna; 

 of the lands which lie within the same range of longitude are 

 extraordinarily different. What now happens geographically 

 to provinces in space is good evidence as to what, in former 

 times, may have happened to provinces in time ; and an essen- 

 tially identical land-fauna may have been contemporary with 

 several successive marine fauna;. 



" At present our knowledge of the terrestrial fauna; of past 

 epochs is so slight that no practical difficulty arises from using, as 

 we do, sea-reckoning for land-time. But I think it highly 

 probable that sooner or later the inhabitants of the land will be 

 found to have a history of their own." 



When these words were written more than twenty-four years 

 ago, scarcely one of the geological details to which 1 have called 

 your attention was known. I need not point out how wonderful 

 a commentary such details have afforded to Prof. Huxley's 

 views. 



I have no desire to quote authority. I fear that in the facts I 

 have been laying before you my quotations of the most authori- 

 tative writers have been made less for the purpose of showing 

 reverence than of expressing scepticism. My reason for calling 

 attention to Prof. Huxley's views is different. I entirely agree 

 with them ; but there is, I think, something to be added to 

 them. There is, I believe, an additional distinction between 

 land and marine faunas that requires notice, and this distinction 

 is one of very great importance and interest. It appears to me 

 that at the present day the difference between the land-faunas of 

 different parts of the world is so vastly greater than that between 

 the marine faunas that, if both were found fossilised, whilst there 

 would be but little difficulty in recognising different marine 

 deposits as of like age from their organic remains, terrestrial and 

 fresh-water beds would in all probability be referred to widely 

 differing epochs, and that some would be more probably classed 

 with those of a past period than with others of the present 

 time. 



I had proposed to enter at some length into this subject, and 

 to attempt a sketch of the present state of our knowledge con- 

 cerning the distribution of terrestrial and marine faunas and 

 floras. But I found that it was impossible to do justice to the 

 question without making this address far longer than is desirable, 

 ami I have already taken up more time than I ought to have 

 done. I can therefore only treat the subjects very briefly. 



As you are doubtless aware, the most important work upon 

 the distribution of terrestrial animals yet published is that of 

 Mr. Wallace. He divides the earth's surface into six regions — 

 Palnearctic, Ethiopian, Oriental, Australian, Neotropical, and 

 Nearctic. Some naturalists, with whom I am disposed to 

 agree, consider Madagascar and the adjacent islands a seventh 

 region, and it is possible that one or two other additions might 

 be made. 



These regions are essentially founded on the distribution ol 

 Vertebrata, especially mammals and birds, and the following 

 table, taken from Wallace's lists, shows the percentage of peculiar 

 families of Vertebrata and peculiar genera of Mammalia in each 

 region, Mammalia being selected as being more characteristic 

 than birds, and better known than reptiles, amphibians, or 

 fishes : — ■ 



Australian . 

 Neo rpopicAL 



Nearctic 



amals and reptiles are too few in number to 



