420 CHAPTER XXXIV. 



The impregnation is a partial one, by which is meant that of 

 all the elements, whether nervous or not, that are present 

 in a preparation, only a limited number are coloured. 

 That is one of the great advantages of the method. For if 

 all the elements present were coloured equally, you would 

 not be able to see the wood for the trees, for you would hardly 

 be able to follow any one element for more than a very short 

 distance. But Golgi's method selects from among the ele- 

 ments present a small number which it stains with a great 

 intensity and very completely that is to say, throughout 

 a great length, so that they are both very clearly separated 

 from those elements that have remained uncoloured, and also 

 can be followed out for a great distance. 



Axis-cylinders are generally impregnated only so long as 

 they are not medullated. In the adult the method stains 

 nerve-cells and their processes, so far as these are not 

 myelinated ; but if it be wished to impregnate the axis- 

 cylinders of the cerebro-spinal axis the method is best 

 applied to embryos or new-born animals at a time when the 

 fibres have not become surrounded by their sheath of 

 my elm. 



There is no other method which will allow cell -processes 

 to be followed out for such great distances. But the method 

 does not demonstrate at the same time the histological detail 

 of other tissues that may be present in the preparations, and 

 all cytological detail is lost. It is par excellence a special 

 method. 



Nervous tissue is not the only thing that is impregnated 

 in these preparations ; neuroglia, connective tissue, fibrils, 

 etc., are impregnated, and the method has been applied with 

 success to the study of such things as bile-capillaries, gland- 

 ducts, and the like. Both on account of this character, and 

 on account of the capriciousness with which the impregnation 

 takes hold of only certain elements of the preparations, care 

 must be exercised in the interpretation of the images obtained. 

 A further source of possible error is found in the fact that 

 the method frequently gives precipitation-forms of the silver 

 salt that simulate dendrites and other structures (see FRIED- 

 LAENDER in Zeit. iciss. Mik., xii, 1895, p. 168, and the plate 

 in the following number.) 



The method has been applied with success to the tissues 



