10 THE CRASPEDOTE MEDUSA OLINDIAS AND SOME OF ITS NATURAL ALLIES. 



And one is very apt to interpret such a different behavior towards stains as proving 

 different origin. This is the fundamental principle of double staining, but if applied 

 too liberally it is very apt to lead one into unwarranted conclusions. The researches 

 of Fischer ('99) have taught us that this principle must be checked and rechecked 

 constantly by other considerations to afford justifiable and reliable conclusions. The 

 origin of a structure can be regarded as definitely proved only by being traced to its 

 first beginning. 



The cells that are destined to form the otolith gradually undergo histolysis; 

 the cytoplasm becomes more and more attenuated, and the nuclei enlarge and become 

 clearer and vesicular; finally they disintegrate and disappear, leaving in their stead 

 a sparsely granular substance, which at first stains tolerably well, and in which small 

 blocks of chromatin and some irregular fibres can be observed. At this stage the 

 stalk is very distinct and consists of a deeply staining, finely granular protoplasmic 

 mass containing several nuclei (Fig. 19). The otolitic substance gradually becomes 

 more refringent, at the same time losing its affinity for stains, and finally comes to 

 consist of numerous concentric layers. The otolith is surrounded by a nucleated 

 protoplasmic investment to the last. In the stage shown in Figure 19, the lining 

 epithelium of the otocyst is separated from the ectoderm only an exceedingly thin 

 layer of mesogloea; but in older stages the entire organ is pushed far into the 

 jelly and has usually no cellular connection with the ectoderm. Exceptionally, 

 however, such a connection persists through life. The otocysts are throughout life 

 more or less closely applied to the endoderm of the circular canal, from which they 

 are frequently separated by a veiy thin layer of jelly. 



The colors * of the medusa are as follows : for the tips of the exumbrellar tentacles 

 a beautifully transparent, shining lilac, for the next adjoining portion a shining sma- 

 ragdine-green, and for the peronia and the basal portion of those exumbrellar tentacles 

 that arise some way up the umbrella ivory-black, the latter thinning out and passing 

 into the green portion or separated from it by a short colorless stretch; for the radial 

 and circular canals a deep scarlet, and for the centripetal canals a lighter scarlet. 

 There is also a small smaragdine-green triangular area at each corner of the base of 

 the manubrium, on either side of which is a lilac area of the same color as that of 

 the tips of the exumbrellar tentacles ; there is in addition one or two somewhat irregu- 

 lar longitudinal streaks of lilac along the middle of each side of the manubrium. The 

 tips of some of the filiform velar tentacles are occasionally just tinged with green 

 and lilac. The gonads are egg-yellow. In small specimens the more axial portion of 

 the manubrium is egg-yellow, but in larger ones it is almost entirely colorless. 



* The names of the colors used in this description are in accordance with the system proposed by Saccardo ('94). 



