Sponge-fauna of Norway. 447 



about the level of the first and second vesicles of the incur- 

 rent canal system, the connective-tissue corpuscles have under- 

 gone a remarkable internal change (PL XVII. fig. 18). 

 Within the granular protoplasm a smooth shining globule 

 makes its appearance ; it is colourless, transparent, homoge- 

 neous, and liighlj refringent. In some corpuscles only one 

 such body is present; in others several, lying in close contact 

 with flattened apposed faces. The number in different groups 

 does not follow any regular series, such as 1, 2, 4, 8, &c., but 

 any number may occur from 1 to 8, and perhaps more : nor 

 are the granules of a group all of the same size ; there may 

 be one large and several smaller ones of various degrees of 

 minuteness. Sometimes they lie in immediate contact with 

 the protoplasm, more often separated from it, lying in a 

 vacuolated space. We are able fortunately to determine the 

 stage in which they earliest appear, by finding them in evi- 

 dently very young corpuscles, distinguished by the large 

 quantity of their finely granular protoplasm, which takes a 

 specially deep stain with reagents. From this starting-point 

 we can readily trace their history as they are followed deeper 

 into the interior of the sponge. In corpuscles a stage older 

 than the preceding we find the protoplasm becoming less 

 granular, staining much less deeply with carmine, and dimi- 

 nishing likewise in quantity, so that it forms a mere spherical 

 or oval shell around the granules, but still retaining 

 its outward radiating processes (PI. XVII. fig. 19) ; these, 

 however, in the next stage also disappear, and the corpuscle 

 becomes simply a mere oval or spherical sac, filled with the 

 products of its metamorphosis or secretion, amidst which the 

 nucleus lies concealed (PI. XVII. figs. 26, 45, 46). The 

 shining granules next begin to diminish in number and 

 size, and at length finally disappear, leaving as an effete resi- 

 duum the investing sacs, which, lined by a small quantity of 

 protoplasm produced sometimes into branched processes and 

 showing the now reexposed nucleus, contribute largely to the 

 histological elements of the gelatinous tissue (PI. XVII. 

 figs. 31, 44). 



Tiie manner in which the fat-like granules make their ap- 

 pearance and their subsequent history seem to point to their 

 being food-reserves of some kind ; but of what kind in parti- 

 cular, one cannot safely even conjecture. They stain deeply 

 with carmine, turn brown, and not blue, with iodine (i.), do 

 not dissolve in ether or chloroform (ii.), nor in boiling water 

 (iii.), nor in strong sulphuric acid (iv.) ; strong acids, indeed, 

 like nitric and sulphuric, seem to have no action upon them 

 in the cold, even after prolonged treatment ; iodine does not 



31* 



