FORMS OF SPICULES. 73 



used^ the stalactitic crystals were found turned into rounded 

 rods, bulging at many points into beads and variously bent, 

 twisted, and interwoveu, so as to bear some resemblance to 

 the form in which mineral matter is deposited in the skeletons 

 of some of the Echinodermata. Other phosphates (phosphate 

 of calcium) were found in irregulai", elongated, curved, and 

 branching masses." Calcium carbonate is deposited in albu- 

 men in the form of small spheres covered with curved and 

 pointed spines ; calcium oxalate in gelatin forms mulberry 

 masses, spheres, dumb-bells, feathered octahedra with curved 

 pinnae, etc., and uric acid is still more protean. The curious 

 ball- and cone-shaped deposits which corrosive sublimate 

 forms in balsam are known to everbody. And, as Bowman, 

 Vogelsang, and others have shown, some of the forms assumed 

 in colloidal media by common salt, santonin, salol, pyrogallol, 

 antipyrin and other substances are marvellous in their com- 

 plexity and beauty. And it must not be forgotten that all 

 these forms are determined by colloidal media alone. As 

 stated above, crystalloid media have an immense influence 

 in producing the most varied complex aggregate crystals, 

 and if, in addition to these, there are present, as is the 

 case in living organisms, colloidal media of highly complex 

 and variable constitution, the possibility of the develop- 

 ment of elaborate crystallomorph forms is obvious. It is 

 true that these crystallomorphs are all composed of so- 

 called crystalloid matter, whereas the greater number of 

 spicules are composed of colloid matter (mainly opal), and, 

 as I have before stated, I am not aware of any experimental 

 or other evidence that colloids are capable of assuming 

 crystallomorphic forms. However, it is well known that 

 typical colloids like egg- and serum-albumen and certain 

 globulin proteids are easily capable of ordinary crystallisation 

 and in consequence we have some reason to suppose that 

 colloid crystallomorphs are at least possible. Applying our 

 present scanty knowledge of crystallomorphs to the subject 

 of spicules, the researches of Gautier (7) and others have 

 made known to us that the protoplasms, so to speak, even of 



