PROTOZOA AND SPONGES. 443 



this slice. Those take a curious form, exactly that of 

 the pins Avhich we use on our dressing tables ; each 

 consisting of a cylindrical slender rod, pointed at one 

 end, and at the other surmounted by a globular head, 

 the whole formed of glass— ^/?i^ glass literall}'. Yen 

 Bee them bristling all round the edge of the section, 

 being etuck into the surface of the sponge, exactly as 

 pins are loosely stuck into a pin-cushion. The heads 

 and points, too, project into the cavities ; more, how- 

 ever, than they did during life, for you must make al- 

 lowance for the shrinking of the soft parts ; and thus 

 you perceive how the whoie structure is permeated by 

 these glass}' pins, which seem to be entangled together 

 quite at random without rule or arrangement. And yet 

 there is an arrangement discernible here ; for the canals 

 are formed by the manner in which tnese are grouped ; 

 and this is seen much more clearly, in the case of the 

 three-rayed needles of lime in the GrantiGB. Mr. Bow- 

 erbank has shown that in G. com.pressa the sul.)stance is 

 divided into very regular chambers in a double series, 

 separated by a diaphragm, whose axis is at right an- 

 gles to the axis of the sponge ; and that these chambers 

 are defined by walls made up of the three-rayed needles 

 ill their ir.utual interlacement. 



