508 GILBERT C. BOURNE. 



fuuctioii. I have, however, observed iu some Madreporaria 

 traces of a coiled filament within the ovoid body, which leads 

 me to the conclusion that they are degenerate nematocysts. I 

 have not been able to discover any traces of a coiled filament 

 in Alcyonium, Gorgonia, or Ileliopora, but the ovoid bodies 

 have in other respects the characters of developing nematocysts, 

 dnd it is not unreasonable to conclude that the sclcroblasts in 

 the course of their migration from the ectoderm have carried 

 with them some of the most characteristic constituents of that 

 layer in the shape of nematocysts, which, being functionless, 

 have undergone degeneration. 



One of the smallest sclerites which I was able to discover is 

 shown in fig. 1, scl. It consists of a single minute crystal, 

 •01 mm. long by '0025 mm. broad, lying in a distinct vacuole 

 in the centre of the scleroblast. Adjoining it is another 

 scleroblast, in which there is a minute nodule, some '005 nam. 

 in its greatest diameter, in addition to a relatively large and 

 already irregular spicule. 



Figs. 2 and 3 represent larger sclerites enclosed in the cells 

 which have given rise to them. Fig. 4< shows what are ap- 

 parently two cells containing sclerites fusing together, but it 

 is probable that this is the result of division of the nucleus of 

 the scleroblast without corresponding division of the cytoplasm, 

 and the two sclerites which are contained in the binuclear cell, 

 after reaching a certain size and form, are becoming joined 

 together. Fig. 5 shows the normal condition of a young spicule 

 of Alcy onium ; the sclerite has the characteristic shape, some- 

 what resembling a caudal vertebra, and is enclosed in a proto- 

 plasmic sheath containing two nuclei. In older and more 

 complicated spicules I have counted three or four nuclei, and 

 where two are present they are generally at the ends of the 

 spicule, and not at the sides as in fig. 5. Though the exist- 

 ence of two, three, or four nuclei suggests the presence of as 

 many cells, I do not think that more than one cell, i. e. more 

 than one continuous body of cytoplasm, is concerned in the 

 formation of a single spicule in Alcyonium digitatum. 

 The scleroblasts, as is seen in fig. 1, are often coenocytes con- 



