548 NEUROGLIA AND SENSE ORGANS 



though a product of differentiation of the protoplasmic portions 

 of the astrocytes, never become entirely independent of the latter. 

 These fibres appear to correspond to those stainable by the 

 methods described in §§ 1083 et seq. The astrocytes prevail in 

 the white matter of the central nervous system, and form the bulk 

 oi the fibrous neuroglia (§ 1082). 



The gold chloride and sublimate method leaves unstained a third 

 category of elements, the existence of which was at first recognised 

 by Cajal by means of this negative character, but they were sub- 

 sequently studied by him in superficial sections of pieces stained 

 by his uranium nitrate method and other cytological methods. 

 The cells belonging to the category now considered appear in 

 uranium nitrate preparations as roundish elements, but, as a 

 matter of fact, they also are provided with a changing number 

 of variously-ramified protoplasmic processes (see § 1082). As 

 Cajal was not able to come to any definite conclusion in regard 

 to their nature, he proposed to term them the " third element,''^ 

 i.e., a category of cells which though non-nervous in character, 

 do not plainly form part either of the connective tissue (blood- 

 vessels, pial septa) or of the neuroglia, this term being, in Cajal's 

 opinion, reserved for those elements which are genetically derived 

 from an evolution of the ependymal epithelium. 



1094. AcHucARRo's Tannin Method and Del Rio-Hortega's 

 Modifications. The methods described in this paragraph can be 

 considered as the direct outcome of various efforts at modifying 

 the Bielschowsky method for sections (§ 1018) in such a way as 

 to obtain a neuroglia stain. As a matter of fact, they all stain 

 both neuroglia cells (astrocytes) and connective tissue elements. 

 In other words, they are not elective, and may be used for the 

 study of reticular tissue in non-nervous organs, as well as of other 

 histological details in nervous and non-nervous tissues. 



Perusini's modification of Bielschowsky's method {Neurol. 

 CentrbL, xxix, 1910, p. 1256) should be first remembered. Pieces 

 of fresh material were fixed in Weigert's formalin-copper acetate- 

 chromium fluoride mixture for neuroglia stain (§ 1083), cut by 

 the freezing method, and stained as by Bielschowsky's method 

 for sections, without pyridine treatment. Achucarro did the 

 same, except for silvering by Ramon y Cajal's reduced silver 

 process. 



AcHUCARRo's tannin method (Bol. Sac. Espan. Biol., Madrid, 19H, 

 p. 139) consisted in putting sections made from frozen formol material 

 into a cold-saturated solution of tannin and warming this until vapour 

 arose. Without waiting for the tannin to become cool again, the 

 sections were, one by one, quickly rinsed in water and put to stain for 

 about ten minutes into three successive glass dishes, each containing 

 10 c c. of distilled water and 6 to 8 drops of Bielschowsky's ammoniacal 

 silver nitrate-and-oxide bath, prepared beforehand, as described in 

 § 1018. As soon as they turned dark yellow, they were transferred into 



