380 DE. G. ELLIOT SMITH ON 



the eighth nerve on the size of the lobus flocculi ; but, from a careful comparison of a 

 large series of mammals, these tAvo hypothetical factors impress me as being the most 

 potent inflaences which determine the size and complexity of the cerebellum. 



There is one other point regarding the cerebellum which does not directly concern 

 us in this memoir, but may be mentioned en passant. The growth of the cerebelluui is 

 almost purely an increase in the superficial extent of the cortex, for, so far as we are 

 aw-are, the cerebellar cortex does not appreciably advance in the richness of its histo- 

 logical elaboration in the Mammalia. In other words, the resultant nerve-fibres do not 

 become appreciably more abundant per unit of surface area. Therefore the problem of 

 the mutual adaptation of the rapidly increasing cortical sheet, which is growing in terms 

 of the square, to the proportionate increase of tlie white or medullary core, which from 

 the nature of the case must grow in terms of the cube, becomes much more purely 

 mathematical than is the case in the cerebral cortex. Consequently the surface of the 

 cerebellum becomes exceedingly comidex in order that the surface of the medullary 

 matter may approximately be equal to that of the cortical matter. In a large mammal 

 the cerebellum is much larger, and correspondingly more complex, than it is in a smaller 

 mammal of the same family. AVhile all parts of the cerebellum share in the increase, 

 its effects become especially noteworthy in the region which I have called " area 

 crescens." Among tlie factors which cause the rapid expansion of the area crescens, 

 liowever, the prllial influence takes a predominant share. A glance at the cerebellum in 

 the Primates conclusively demonstrates this. 



In Chlamydophoriis we find a small and simple cerebellum which is constructed 

 according to the same plan that ])revails in many members of the Rodeutia and 

 Chiroj)tera, and in most Insectivores and Marsupials. There is the most remarkable 

 apparent resemblance between the general shape of the organ in Notoryctes and 

 ClilamydopliorKs. But we find a much higher degree of complexity in the Edentate than 

 in the Marsupial, A\hich is all tlie more impressive when Ave consider the similar modes 

 of life of these two animals, and their not dissimilar bodily dimensions. Even in the 

 much larger Marsuj)ial Perameles the cerebellum is not so richly folded as it is in the 

 small Ch lamydophorns. 



In Xenurus we find a cerebellum conforming to the same type, and quite as highly 

 elaborated as that of the Eabbit, and in tlie larger Armadillos (such as Basypus) we find 

 the same kind of progressive uncovering of the anterior lobe as in certain Eodents *. 

 This demonstrates the gradual transition from the simpler type of flattened cerebellum 

 to the more complex type found in all the other families of Edentata, as w^ell as in the 

 Ungulata and CarniA^ora. 



This plumper type of cerebellum with an exposed anterior lobe may possibly represent 

 a higher type of cerebellum. We find that it is gradually attained in the Armadillos as 

 the animal increases in size, i. e. as the somatic factor increases. It seems not improb- 

 able, however, that some other factor, such as the size and degree of complexity of the 

 pallium, enters into this change. Por in the Kangaroo (MacrojJiis), in which the somatic 



* Vide Parsons's figures, op. cit., Proc. Zool. Soc. London, 1894, p. 687. 



