( 90 ) 



is to say, minus the terminal nasal cartilages — was 96 mm. Tlie brain is therefore 

 rather less than one-third of the length of sknll. These measnrements are not far 

 ofFtliosc exhibited by some of the early and small-brained Ungnlates of the Eoeeno. 

 They are, indeed, almost exactly those of Tillotkcriuin fodiens as reproduced in the 

 text-books, and not far off those of Conjphodon, while in other cases the brain is 

 between one-third and one-fourth of the length of the skull. The proportions 

 contrast markedly with those oi Erinaceus, which has a brain (jnite half the Imgth 

 of the sknll. Moreover, the size of the hemispheres of Centctes is unusually small as 

 compared with the rest of the brain, another feature in which this lowly insectivorous 

 creature comes near to some of the early eutherian mammals. Seen laterally 

 (PI., fig. 2) the latter are seen to be ou a level with the rest of the brain ; there 

 is no strongly ])rojectiug temjwral lobe projecting downwards below the level 

 of the rest of the hemispheres, and of the brain generally ; this is again a feature 

 that is worthy of note in connection with the comparisons here made. 



As may be seen from the drawing, which represents the upper aspect 

 of the brain (fig. 1), the two hemispheres, closely ajiplied to each other in the 

 middle line, diverge from each other posteriorly ; one-third of their total length 

 is thus deilected, for the median part of the hemispheres which are in contact 

 measure only 10 mm. along the line of contact, while the total length of each 

 hemis]ihere measured in a straight line is 15 mm. From this there results a 

 complex ex]iosure of the corjiora ipiadrigemiua. It is stated by Huxley * that in 

 ICrinaccus the hemispheres sometimes " hardly cover the corpora quadrigemina." 

 I interpolate the word " sometimes," since Dr. Uobsou figures a brain of that 

 animal in which the optic lobes most clearly and entirely covered by the cerebrum. f 

 The same is the case with Tupnia, whose brain has been figured by Garrod : % tl'e 

 latter observes that even the cerebellum is just covered at its anterior border by 

 the hemispheres, and the optic lobes are therefore entirely concealed ; but then 

 Tiipaia, with its many Lemur-like characteristics, is plainly a more specialised 

 insectivore than is Centetes. I believe that there is no mammal whose corjiora 

 ijuadrigemina are so fully exposed as they are in this Madagascar insectivore 

 Centetes. The corpora quadrigemina are largely exposed in some other genera, 

 in Rbjnchocyon, Petrodromus and Macroscelides ; § and also in rodents, particularly 

 in Coelngeni/s paca ; but here, as in other genera of rodents, at least part of the 

 four convexities are overlapjied by the otherwise divergent hemispheres. || In 

 Centefrs not one scrap of the coi'pora quadrigemina are hidden when the brain is 

 viewed from aliove ; and I may add that my brain, though preserved in alcohol, 

 has retained the appearances that it presented when in the skull, and in a fresh 

 condition. The cerebral hemispheres of Centetes are not absolutely smooth, as 

 are those of TtipaUi, according to Garrod. There is a rounded furrow on the 

 outer part of each hemisjjhere, of which the concavity is turned towards the 

 middle line of the cerebrum. This rudimentary fissure — it is indeed hardly more 

 than a faint de{)ression — is curiously like a similar furrow which is figured and 

 described in the brain of the common Hedgehog. There is, moreover, a distinct, 



* A Manual af the Anatomy of Vertvhrated Animalu, p. 447. Loinlon ; Churchill, 1871. 



f On the othnr hand T.cche figures a br.iin (_Svensk Vctennlt. Ak. JIaiidl. 21, No. 11, PI. IV., li.,'. J'.l) 

 in whicli the corpora are partly expo.scit. 



X " .Votes on the Visceral .Vnatomy of the Tupaia o£ Burmah (Tupaia bclangeri)," P.X.S. 18711 p. 'M\. 



% Peters, lieinv luich Mosmmhlqnr, 1858, PI. XXIV., figs. 10, 12, 13. 



II lieddard, " On the Cerebral Convolutions of the Cerebral Hemispheres in certain Uudonte," Ihhl. 

 1832, p. .-,96. 



