160 PLATE LXII. 



tions from the apex of the sponge, rather exceeding ten 

 lines in diameter, had probably been not less than four 

 inches in height. 



The habit in this species of discharging the fecal 

 streams by means of large cloacge is quite in accord- 

 ance with that of the nearly allied British species D. 

 fruticosa. 



This species is one of the very few sponges in which 

 a section at right angles to the surface in the dried 

 state does not give a good illustration of its structure ; 

 a portion of the longitudinal run of the skeleton affords 

 a much more correct idea of its anatomical characters, 

 and these are very much more developed in the thick- 

 ened base and apex of the bulb of the sponge, and in 

 the stout prehensile and cloacal organs. 



Sections of the dried mass of interstitial membranes 

 and sarcode that had subsided within the outer crust 

 of the sponge during its drying afforded but very httle 

 information regarding its mode of disposition and 

 structure ; but from one of the large fragments sent to 

 me by Mr. Peach I obtained sections at right angles to 

 the surface of the sponge, which presented every indi- 

 cation of the tissues having retained their natural 

 characters and positions. 



There was every indication of the membranes having 

 been arranged in lines radiating from the centre to the 

 circumference of the sponge, the greater portion of 

 the n imerous tension spicula assuming that direction, 

 with comparatively a few others dispersed among them 

 at various angles. There were also innumerable gem- 

 mules in a fine state of preservation evenly dispersed 

 on the membranes. These organs with a linear power 

 of 700 are distinctly seen to be filled with minute 

 granules. The average diameter of the gemmulo was 

 TsVo inch. Amidst these were also an abundance of 

 minute, simple, and contort, bihaniate retentive spicula, 

 exceedingly minute, and very liable to escape observa- 

 tion amidst the sarcode unless examined with a powei" 

 of about 1000 linear. The leng-th of one of the 



