parb 1] OF THE UPPER CRETACEOUS IN HUNGART. Ill 



bigger creatures. This dwarfing is very similar to that observable 

 as dating from Pliocene times on various Mediterranean islands, 

 where only the comparatively large mammals decrease in size, 

 while the originally rather small tortoises and originally small 

 mammals retain their normal size, or seem even to become bigger. 1 

 Insular isolation, therefore, seems to act differently according to 

 the size of the animals that are isolated. Apart from all this, 

 we can discern some of the various reasons that led to the ex- 

 termination of this fauna. One factor was assuredly the invasion 

 of Asiatic mammals, which probably disturbed the helpless Dino- 

 saurs of the terra firma, such as Strutliiosaurus and 3£egalosaurus, 

 in much the same manner as the recent placental mammals of 

 Australia disturb the Marsupials of that country. The possi- 

 bility of such an invasion becomes quite evident if one studies 

 Karpinsky's maps, reproduced in Dacque's book (14). 



Another factor which probably led to the extermination of the 

 herbivorous Dinosaurs inhabiting the lakes and swamps was the 

 drying-up of the country, and lack of marsh- vegetation. 



The flora of the Danian and Senonian rocks of South-Eastern 

 Europe is different from the Eocene flora. In Danian times the 

 forests of Eastern Europe consisted, as shown by the works of 

 L: de Launay (15), Zeiller (79), Staub (75), and Tuzson (76), of 

 Cunningliamites, Damarites, Ternstromia, Gredneria, Sassafras, 

 Picas, Tree-Eerns, and Araliaceas : in their shade grew ferns, such 

 as Asplenium, Gleiclienia, and Pecopteris. In the forest-glades 

 lofty palms arose, such as Sabal and Jurania; while on the 

 marshy grounds near the lakes, Pandanites trees grew on a ground 

 covered with Arundo. All this indicates a moist and warm climate. 

 In contrast to this, the Hungarian Lower Eocene contains already 

 great numbers of walnut-trees (77), and therefore trees that regu- 

 larly shed their leaves. This shows, that on the Eocene continent, 

 perhaps because of lack of moisture, the temperature had sunk lower. 



It is a well-known fact that the higher Dicotyledonous flora 

 appears in the Cretaceous Period, and thus it seems that the change 

 in the flora precedes the change in the animals that feed upon it. 

 Curious as this may appear, there is a parallel in earlier ages, for 

 at the end of the Palaeozoic Era the appearance of the Glosso- 

 pteris flora in the Southern Hemisphere and its spread north- 

 wards precede in a similar manner the extinction of the herbivorous 

 Palaeozoic reptiles Pareiasaarits, Dueynoclonts, and Dinocephalia. 3 



The explanation is quite simple, for the flora is directly affected 

 by every change of climate ; while the animals for the greater part 



1 D. M. A. Bate, ' On a small Collection of Vertebrate Remains from 

 the Har Dalam Cavern, Malta, &c.' Proc. Zool. Soc. 1916, p. 421 ; ' On 

 Remains of a Gigantic Land Tortoise from the Pleistocene of Menorca ' Geol. 

 Mag. 1914, p. 100 ; ' Preliminary Note on the discovery of a Pigmy Elephant 

 in the Pleistocene of Cyprus ' Proc. Roy. Soc. vol. lxxi (1903) p. 498 ; and 

 ' Further Note on the Remains of Elephas Cypriotes from a Cave-Deposit in 

 Cyprus ' Phil. Trans. Roy. Soc. ser. B, vol. cxcvii (1905) p. 347. 



2 See fig. 325, p. 430, in Potonie's ' Lehrbuch der Paliiobotanik ' 2nd ed. by 

 Gothan. Berlin, 1922. 



