176 



THE AMERICAN NATURALIST [Vol. XL 



methods was altogether successful. The cocaine — a 2% solu- 

 tion in 50% alcohol added drop by drop —produced extension of 

 gill and cephalic tentacles. In specimens killed after this treat- 

 ment, the subradular organ was likely to be extruded. Chlore- 

 tone — crystals gradually added to sea water — produced at first 

 a general extension but a larger dose brought about contraction. 

 This agent is especially useful for narcotizing parts — as gills or 

 tentacles — which it is desired to study while still alive. Exten- 

 sion of mantle tentacles is best obtained by killing in Gilson's 

 fluid, extension of the mantle in general by this method or by 

 allowing the animal to die in stale sea water, or, as recommended 

 by Fisher, in fresh water. 



The most satisfactory killing agents I have found to be picro- 

 sulphuric acid, chrom-alcohol (equal parts of 70% alcohol and 

 'tV % chromic acid), and corrosive sublimate with 5 to 20 % of acetic 

 acid. A weak — 5% — aqueous solution of sublimate preserves 

 the external cilia better than a stronger one; this solution also I 

 have employed with success for material in which it was desired 

 to demonstrate mucus. The various osmic acid solutions — vom 

 Rath's, Hermann's, Flemming's — have no marked superiority 

 except for demonstrating certain glands as noted in the section on 

 the integument. Picro-sulphuric acid has the advantage when 

 used for very small specimens, of decalcifying the shell while leav- 

 ing it in situ; in specimens killed in corrosive acetic the shell 

 parts from the animal, in consequence I suppose of the pressure 

 due to the more rapid evolution of gas brought about by the 

 larger proportional amount of acid. An acid killing agent, as 

 Bernard has pointed out, is desirable because it at once coagulates 

 the mucus and thus renders the goblet cells more conspicuous. 



For purposes of dissection it is desirable to have some specimens 

 killed in formalin. The comparative transparency produced by 

 this agent as well as its slight swelling action are often of advan- 

 tage; it has the further good quality that it preserves at least for 

 some time the color of the nephridial epithelium. A 2%, solution 

 is most satisfactory and it is of course desirable that the solution if 

 acid should be neutralized. Opaque specimens for dissection are 

 best killed in Gilson's fluid. It is perhaps unnecessary to ad<l 

 that details can often be made out better in specimens in which 



