620 Trow . — The Karyology of Sctprolegnia. 
results, so far as concerns the nuclei of the zoospores and 
mycelium, were satisfactory, but the study of the sexual 
organs could not be carried out properly by this method. The 
staining was effected on the slide in all later experiments, and 
the advantages of this method compensate for its laborious- 
ness : the stain comes into direct contact with the protoplasm 
and does not have to pass through more or less impermeable 
membranes. 
The sections were fixed to the slide with a mixture of 
white of egg and glycerine, freed from paraffin and stained 
with Schneider’s acetic carmine, gentian-violet and eosin, 
or Kleinenberg’s haematoxylin, and all three methods gave 
excellent results. I have used the two former very extensively ; 
the latter has furnished me occasionally with beautiful pre- 
parations. The best haematoxylin-stained preparations were 
obtained by using a very dilute solution, allowing it to act for 
twenty-four to forty-eight hours, and then partly decolorizing 
with acid alcohol, alum solution, or best of all by a saturated 
solution of picric acid. [See Zimmermann (’92).] Schneider’s 
acetic carmine was diluted so that the solution contained 
from two per cent, to five per cent of acetic acid, and the 
objects remained in the stain for from twelve to twenty-four 
hours. 
For a long time I failed entirely to apply Gramm’s method 
of double staining with gentian-violet and eosin, although 
I had been using it successfully for twelve months in the 
laboratory for a great variety of purposes. In the case of 
material fixed by absolute alcohol, I believe it is impossible to 
secure good results, as the gentian-violet taken up by the 
nucleus is almost instantaneously taken out again by the 
absolute alcohol used in dehydration. With material fixed 
by mercuric chloride, the same difficulty presents itself, but 
not to such an extent as to make it impossible to surmount. 
When successful, the preparations are so excellent that for 
most purposes they are far superior to those produced by any 
other method. The difficulty of carrying out this method 
satisfactorily is great, and at first only a small percentage of 
