390 OHAPTRE XXXII. 



it is waslied out for from six to twelve hours (not more in any case) in 

 absolute alcohol, cleared in clove oil imbedded in paraffin, and cut. It 

 is said that if sections are mounted by the vrater method the stain will 

 be extracted ; but why not make the water acid, § 291 ? 



The usual histological stains either leave them colourless 

 <ir stain the surrounding plasma more strongly than the fibrils 

 themselves. There may thus be produced a " negative " 

 image of fibrils which does not really show the true neuro- 

 fibrils, and being taken for them may prove a source of error. 

 The following methods are such as have been recommended 

 as giving true stains of the fibrils. 



767. Neurofibrils: Silver Methods. — Those most used are 

 Ramon y Oajai/s and Birlschowsky's. The essential differ- 

 ence between the two is that Ramon employed a single 

 impregnation bath — of nitrate of silver; whilst Bielschowsky 

 employs two — one of nitrate and one of silver oxide dissolved 

 in ammonia. 



Fixing agents should in general be neitfral or alkaline; for 

 acids or oxidants divert the impregnation from the fibrils to 

 other elements, chiefly tigroid and nuclear elements. 



Sections should be thin — not more than 15 // to 20 /t. 



Toning with gold has the effect of intensifying the stain 

 in the fibi'ils and lightening it in surrounding elements, thus 

 giving enhanced contrast. It also serves to favour the pre- 

 Rervation of the stain. 



The object of fixing with hyposulphite is to remove from 

 the tissues any unreduced silver salts which might cause a 

 loss of contrast by darkening the ground of the preparations. 

 With well reduced preparations, such as Ramon^s are in 

 general, it is not necessary. 



Bielschowsky's methods have the advantage of being- 

 applicable to larger specimens than Ram6n'.s, for they give a 

 more uniform impregnation through the whole thickness of 

 the objects (especially with the central nervous system of 

 man) ; whilst Ramon's (especially his formula la) only gives 

 the desired results in a — sometimes very thin — layer between 

 an overstained outer and an understained inner one. 



Bielschowsky's are applicable to xery old formol material. 

 Bayon [Die liisfologisclien Uvtersuclinngs-llrfliodeii dcs Nerv(n- 

 systrms, p. l.'j?) has succeeded with material four years 



