On the connective tissues and body cavities of the Nemerteans. 7 



1) Delafield's baematoxyliu (diluted with an equal part of water) 

 20 minutes, followed by a cone, aqueous sol. of eosin, 5 minutes, 



2) Ehrlich's baematoxyliu (undiluted) 1 — 3 bours, then eosin (as 

 above) 5 minutes. 3) The Eiirlich-Biondi stain (orange G + methyl 

 green + acid fucbsin), staining 3 hours with the solution made from 

 the precipitated powder, manufactured by Grübler; this is the only 

 stain I have found for the parenchym cells. 4) Indigo-boraxcarmine (after 

 NoRRis & Shakespeare) used in aqueous solution, for about 48 hours. 

 These four stains produce clear differentiations of the tissues, and 

 are to follow fixation in corrosive sublimate. After fixation in Her- 

 mann's fluid, I stain 50 bours in a cone, solution of saff'ranin in 70 "/„ 

 alcohol, then 1 hour in cone, aqueous solution of gentian violet, and 

 lastly 1 — 2 minutes in a cone, aqueous solution of orange G (this 

 being a modification of Flemming's methods, in : Arch. mikr. Anat., 

 1891; Z. wiss. Zool., 1891, and particularly adapted to the study of 

 nuclear structures). After Hermann's fluid, the Ehrlich-Biondi stain 

 (as above) may also be used, from 12 to 24 hours. For material 

 fixed with Flemming's fluid, the following stains are most advisable: 



1) Delafield's haematoxylin followed by eosin, as given above; 



2) Grenacher's alumcarmine (alcoholic solution, to prevent decay, as 

 recommended by v. Mährenthal), 24 hours; 3) Mayer's cone, solu- 

 tion of cochineal in 70 7o alcohol, though this is less serviceable than 

 1 and 2. 



I have found no methods of staining in to to, which are satis- 

 factory for cytological details. 



I have also experimented with the following stains, which are not 

 to be recommended : Victoria blue in aqueous sol., alone, and followed 

 by eosin. Acid fucbsin in aqueous and alcoholic sol., also in com- 

 bination with Delafield's or Ehrlich's haematoxylin. Dahlia in 

 cone, aqueous sol., with the addition of acetic acid. Brasilin (aque- 

 ous and alcoholic solutions), alone, and in combination with eosin. Saf- 

 franin (cone. sol. in 70 *^/o alcohol), also followed by Delafield's 

 haematoxylin. Methylen blue (in cone, aqueous and alcoholic sol.), 

 10 — 15 minutes, followed by acid fucbsin, 15 minutes, or eosin, 

 5 minutes; a fair double stain for gland cells is produced by the 

 combination of methylen blue with brasilin (Schaudinn's method); 

 methylen blue does not stain at all, on material fixed with chromic 

 acid, but always to some extent on corrosive sublimate preparations. 



It is remarkable that corrosive sublimate is a poor reagent for 

 the preservation of Amphiporus virescens, while it is the best for the 



