On the Occurrence of Gelatinous' Spicules. 



317 



had to be picked out one by one with needles in order to make preparations 

 suitable for more minute examination, but sufficient of the sponge-tissue 

 remained to enable me to make the following observations. 



When mounted in a drop of absolute alcohol the colloscleres are perfectly 

 distinct and may sometimes be seen to be enclosed each in a thin-walled 

 vesicle (fig. 2). When a drop of water is run in under the cover-glass of such 

 a preparation the colloscleres suddenly swell up to several times their 

 previous volume and become extremely transparent (fig. 2a ; fig. 3, b, c). 

 The swelling takes place chiefly on the convex side, which becomes more 

 strongly convex, while the line of demarcation between it and the surrounding 

 water becomes almost invisible. The contour of the concave side, on the 

 other hand, is quite bright and distinct, and usually distinctly double. 

 I conclude from this that the substance of which the collosclere is composed 

 is denser and less absorbent of water on the concave than on the convex side. 

 I also conclude that in life these spicules, if spicules they can be termed, 

 exist in the sponge in the swollen, gelatinous condition, and that the con- 

 tracted state first observed is v due simply to the withdrawal of water by 

 means of alcohol. It seems remarkable that they should retain their power 

 of absorbing water and swelling up even after more than twenty years in 

 Canada balsam. 



In a very strong cold solution of caustic potash the colloscleres, already 

 swollen in water, swell up further and then dissolve. 



When a 5 per cent, solution of hydrochloric acid is run in under the cover- 

 glass they swell up and become invisible in the surrounding tissues, but there 

 is no effervescence and I do not think they really dissolve. At any rate, the 

 contour of the concave surface may remain distinctly visible even after 

 prolonged action of strong hydrochloric acid (fig. 3, V), and a preparation 

 stained with paracarmine and mounted in Canada balsam after treatment 

 with dilute hydrochloric acid shows at least one collosclere quite distinctly. 



The colloscleres stain quite readily with paracarmine, but with a solution 

 of iodine in potassium iodide I obtained only negative results. When 

 examined with the polariscope they exhibit no optical activity either when 

 contracted in alcohol or when swollen in water. From these observations 

 I think it may be concluded with a reasonable degree of certainty that the 

 colloscleres are composed of colloidal silica containing a much higher per- 

 centage of water than the ordinary siliceous spicules of the same sponge. 



Teased preparations, stained with paracarmine and mounted in Canada 

 balsam after removal of the sand-grains, throw a great deal of light on the 

 origin of the colloscleres. They are found to be associated in the mesogloea 

 with large spherical cells, which, for reasons which will appear directly, 



