THE HISTOLOGICAL STRUCTURE OF THE GILLS, ETC. 93 
shell. Ortmann has used-this method when a microscopical ° 
examination was made. 
The fixatives used were Petrunkevitch, corrosive sublimate 
with glacial acetic acid, Perenyi, Cox-Golgi, and Zenker. 
Of these the best results were obtained with the first three, 
Zenker’s fluid was entirely unsatisfactory and the Cox-Golgi 
was almost as bad. Both showed a decided tendency to distort 
the gill, and were very hard to.remove in washing. Only one 
gill prepared in each of these two fixatives was sectioned and 
the results were so unsatisfactory that the sections were dis- 
carded without an attempt being made to study them. The 
Petrunkevitch fluid was found to be the most satisfactory in 
every way. - The gills were washed in tap water for twenty- 
four hours, and in the case of the corrosive acetic until the 
iodine test showed that excess fixative had been removed. The 
material fixed in Zenker’s fluid was washed in semi-darkness 
in an unsuccessful attempt to avoid precipitation. 
The Heidenhain, Ehrlich and Delafield haematoxylins were 
found to be the most satisfactory stains. All gave very good 
results and there was little to choose from among them. 
Mayer’s haem-alum was also used but usually the stain of the 
necleus with it was too faint. 
The material was stained in bulk for twenty-four hours, 
being taken from 70% alcohol into the stain; it was then 
taken back into 70% alcohol to avoid any distortion due to 
possible diffusion currents and then carried through the alco- 
hols into xylol, remaining twenty-four hours in each. After 
the material had been in xylol twelve hours, melted paraffin 
was slowly: added until the limit of solubility of the paraffin 
was reached. In this way the paraffin was slowly infiltrated 
into the tissue at the same time that clearing was taking 
place. The vials containing the gills were then placed on a 
warm plate until most of the xylol was evaporated, and then 
the tissues were transferred to the melted paraffin of the im- 
bedding oven and imbedded in the usual way. By this method 
of paraffin infiltration ali sudden changes of temperature were 
avoided and the concentration of the paraffin in the tissues 
very gradually increased as the xylol evaporated. 
Serial sections were used throughout the investigation, the 
