308 EVENINGS AT THE MICROSCOPE. 



their wayward traverses across each other, and intertwin- 

 ing their long thread-like tentacles, we will select one or 

 two for examination, as samples of their kindred. And 

 first let me separate this active little Beroe (Cydipjjepomi- 

 formis), which I dip out with a tea-spoon and transfer to 

 this other glass jar, that we may watch its form and 

 movements unaffected by the presence of its companions. 



We see, then, a little ball, almost perfectly globular, 

 except that a tiny wart marks one pole, of the size of a 

 small marble, and apparently turned out of pure glass, or 

 ice, or jelly — according to your fancy, — perfect transpa- 

 rency and colourlessness being its characteristics ; so much 

 that it is not always easy to catch sight of the little crea- 

 ture, except when we allow the light to fall on the jar in a 

 particular direction. From two opposite sides of the globe 

 proceed two threads of great length and extreme tenuity, 

 which display the most lively and varied movements. 



These filaments shall occupy us for a few moments. 

 We trace them to their origin, and find that they proceed 

 each from the interior of a lengthened chamber, on each 

 of two opposite sides of the animal. Suddenly, on the 

 slightest touch of some foreign object, one of the threads 

 is contracted to a point and concealed within its chamber, 

 but is presently darted forth again. When the lovely 

 globe chooses to remain still, the threads hang downward, 

 gradually lengthening more and more, till their ex- 

 tremities lie along the bottom of the jar extended to a 

 length of six inches from the chamber. Then we see 

 that this delicate thread is not simple, but is furnished 

 along one side, throughout its length, at regular distances, 

 with a row of secondary filaments, which project at right 

 angles from the main thread. 



These secondary filaments constitute an important ele- 

 ment in the charm which invests this brilliant little crea- 

 ture. They are about fifty in number on each thread, and 

 some of them are half an inch long, when fully extended, 



