OF DENISON UNIVERSITY. 77 



pero or cap of the tuber is covered by the cerebrum. The lobe is 

 much larger, for example, than the precrucial lobe of the hemisphere. 

 As seen from above, the outline is ovoid with the larger end caudad 

 and the longer axis passing obliquely caudo-ventrad. The ventral as- 

 pect is acutely ovate or subtriangular with the apex mesi-caudad. In 

 its natural position, the infero-cephalic surface is curved and the 

 greater portion is covered by the fibres of origin of the olfactory nerve. 



The pero (as this term is used by Wilder) consists of a cap or bus- 

 kin-like body which entirely clothes the lobe. Its substance extends 

 nearly to the ventricle, which latter is, however, clothed with a larger 

 or smaller mass of white matter with a gray envelope differing totally 

 from that which pertains to the pero proper. Passing entad, we en- 

 counter successively the following layers, i. The fibre layer, which 

 consists of apparently irregularly disposed fibre clusters intercrossing 

 confusedly. Among the fibres are numerous blood-vessels and the 

 usual inoblasts and leucocytes accompany the fibres. 2. The glom- 

 erulary layer is of the usual structure ; fibres from the cells next to be 

 described pass into the dense ball of neuroglia and emerged clothed 

 with their myelin as distinctive olfactory fibres. It is suggested that 

 the glomerule is simply a locus devoted to the manufacture of the 

 myelin sheaths of the nerves. In this case the only active agents in 

 the process must be the large number of wandering cells or Deiter's 

 corpuscles which congregate about the glomerules in large numbers. 

 If the Deiter's cells have this function here it would be easy to pred- 

 icate a similar function of the same cells elsewhere. There are many 

 reasons for such an assumption in other parts of the brain. The pres- 

 ence of exactly similar cells in the tracts of medullated fascicles else- 

 where is suggestive. (See also FlechsigV works on the development 

 of the sheaths.) The same explanation may be given for the exactly 

 similar cells in the laminated granular layer. 3. The external neu- 

 roglia layer is in this case not distinct from the layer next entad, as 

 the ganglionic layer is dispersed throughout them both. In this re- 

 spect the opossum differs from any other mammal examined. The 

 ganglion cells are perhaps most abundant near the external boundary 

 of the laminated granular layer, but are irregularly distributed to the 

 very base of the glomerulary layer. The ganglion cells are of pyra- 

 midal form and large size with a (generally single) strong apex process 

 which can be traced as a large non-medullated fibre to the glomerulary 

 layer. From the base several processes extend latero-entad or laterad 



