Obersteiner, The Structure of the Nervous System. 75 



and Lenhossek, Verb. d. anat. Gesellsch., 1891) teach that the 

 ependyma cells, as the epithelium in question is called, have a 

 very considerable share in the formation of a supporting frame- 

 work, which extends through the central nervous system from 

 the central ventricle to the surface clothed with the pia mater.' 



The supporting substance, in which 1 include everything 

 which is neither nerve substance nor vessels, is certainly not to 

 be regarded as a single kind of tissue. It is at least probable, 

 though not generally accepted, that real connective tissue par- 

 ticipates in the formation of this supporting substance, especially 

 of the larger septa. Much of that which has heretofore been 

 described as neuroglia — a finely granular mass, or intercellular 

 substance, which fills the larger and smaller interstices between 

 the other elements of the central nervous system — proves, under 

 improved methods of investigation,, to be a felt, or net- 

 work, of the finest nerve fibres. Here belongs also Leydig's 

 dotted substance in the lower animals. Hut evtn the familiar 

 spider cells of the central nervous system, Deiters' cells, which 

 are an essential ingredient of supporting tissue, are not so far 

 removed from the nervous elements as has been heretofore sup 

 posed. Even by the earlier investigators it it was assumed that 

 the glia cells also take their origin from the ectoderm ; but it 

 was reserved for Ramon y Cajal (numerous works in recent 

 years) and Lenhossek accurately to describe them as arising, 

 like ganglion cells, from the epithelium of the medullary tube. 

 Consequently the distinction between ganglion cells and glia 

 cells must be dropped ; and one really finds structure in the 

 central nervous system which show intermediate stages. In this 

 connection I mention the so-called " granules," which are met 

 with in the bulbus olfactorius, in the retina, but especially 

 in larger numbers in the granular layer of the cerebellum. Re- 

 cent investigations, especially of Ramon y K ajal (Inlernat. 

 Monatsschr., 1890) and Kolliker (Zeitschr. f. wiss. Zool., 1890), 

 have shown that most of the granules v.\ the granular layer are 

 of nervous nature and are in connection with nerve fibres. On 

 the other hand their nuclei are .stained deep blue by haematoxy- 



Ullustrations of the structure here referred to may be seen on p. 

 45 and Plate VII, Fig. 9 and Plate X, of this number. [Tr.] 



