460 AXIS-CYLINDER AND DENDRITE STAINS. 



long) should be put for two to five days into the hardening mixture, and 

 for one to two into silver nitrate. 



Epidermis of Lumbricus. Three to six days in the mixture, and two 

 in the silver, or double impregnation if necessary. 



Nervous system of Helix (glia- cells). The above mixture for eight to 

 ten days, then silver of 0-75 to 1 per cent. 



As a general rule, the younger the subject the shorter should be the 

 hardening. If it has been too short, sections will have a brownish-red 

 opaque aspect, with precipitates, and irregular impregnation of cells 

 and fibres. If it has been too long, the ground will be yellow, without 

 precipitates, but with no impregnated elements, or hardly any. 



This process has the advantage of great rapidity, and of sureness 

 and delicacy of results, and it is the one that has found most 

 favour with other workers. But for the methodical study of any 

 given part of the nervous system GOLGI himself prefers the 

 following : 



883. GOLGI' s Bichromate and Nitrate of Silver Method. MIXED 

 PROCESS. Fresh pieces of tissues are put for periods varying from 

 two to twenty-five or thirty days into the usual bichromate solution 

 ( 881). Every two or three or four days some of them are passed 

 into the osmio-bichromate mixture of the rapid process, hardened 

 therein for from three or four to eight or ten days, and finally 

 impregnated with silver nitrate and subsequently treated exactly 

 as by the rapid process. 



The reasons for which Golgi prefers this process are : The certainty 

 of obtaining samples of the reaction in many stages of intensity, 

 if a sufficient number of pieces of tissues have been used for the 

 purpose. The advantage of having at one's disposal a considerable 

 time some twenty-five days during which the tissues are in a 

 suitable state for taking the silver. The possibility of greatly 

 hastening the process whenever desired by simply bringing all the 

 pieces over at once into the osmic mixture. Lastly, a still greater 

 delicacy of results, particularly noticeable in the staining of axons 

 and their collaterals. 



884. Theory of Impregnation. It was once held that the reaction 

 depends on the formation in the tissues of a precipitate of some salt 

 of silver. And Kallius has put forward the suggestion that this 

 precipitate may consist of a protein-silver-chromate combination. 

 But this seems to B. Lee incorrect (see 1913 Ed.). In agreement 

 with v. Lenhossek, he finds that the colouration is not due to a 

 visible precipitate, but is a true stain accompanied, particularly in 

 unsuccessful impregnations, by precipitates which not only do not 



