82 BULLETIN OF THE 



to tweuty-two days because of differences of temperature. Hoffman 

 ('81) gives the following table for herring: — 



Temperature 3° to 5° C. hatched in 40 days. 



u 70 „ go u a 15 a 



The best results are obtained by keeping an even temperature, as cold 

 as possible without, of course, going below the freezing point. Dead eggs 

 turn opaque, and should be immediately removed from the good ones, 

 since their presence is injurious. 



In an Appendix is given a description of a simple hatching apparatus 

 which I constructed and found to be satisfactory. It embodies all the 

 points essential in caring for fish eggs, from the youngest stages to 

 the time of hatching, and is adapted for use in the laboratory and in 

 experimentation. 



Killing. — Fixing. — Staining. 



Every student of fish embi-yology has some favorite method of killing, 

 fixing, and staining, Henneguy recommends that fish eggs be killed in 

 1% osmic acid, in which the eggs are left for a few minutes, and then 

 transferred to Muller's fluid for three or four days. AVhile in Milller's 

 fluid the egg membrane is cut with fine scissoi's and the yolk is allowed 

 to dissolve out. In using this method I found the first obstacle to be due 

 to the fact that the yolk is dissolved out in such a manner as to be sub- 

 sequently precipitated around the eggs, which are thus left imbedded in 

 the precipitated mass. This difficulty can be overcome, however, by 

 cutting the egg membrane in such a way as to catch an air bubble within 

 the egg shell and thus to float the egg in the fluid. Care must be taken, 

 however, that tlie egg floats in such a position as to keep the entire 

 embryo submerged. Other and more serious obstacles to Henneguy's 

 method are found in the difficulty of staining well after its employment, 

 and in the dissolving out or injury to the parablast, which is of impor- 

 tance in the question here considered. 



For killing young stages Hoffman ('82) used osmic acid of from ^^% 

 to ^%, and for older ones either 5% bichromate of potassium or picro- 

 sulphuric mixture. 



Rider ('84) objects to the use of chromic acid followed by alcohols, 

 being of the opinion that it coagulates the deutoplasm so as to distort 

 the embryo in a confusing way. 



I was not very successful with Hoffman's osmic and bichromate methods. 



