168 I. IJIMA : HEXACTINELLIDA. I. 



to eacli other, than at other places which face a wider inter- 

 space (see PI. V, fig. 43). 



The variously sized, flat patches of archseocytes may lie so 

 close against the chamber-wall, apparently closing up the proso- 

 pylar meshes in such spots, that in surface views (fig. 37) and 

 often also in optical sections of the wall, they appear to constitute 

 a part of it. But such is not really the case, as can be distinctly 

 made out in many places in specially good sections (fig. 43). 

 The flagellated raembrana reticularis runs uninterruptedly 

 alongside the groups. 



I have not observed karyokinetic figures in the archfeocytes. 

 Nevertheless, many of these are apparently undergoing multi- 

 plication, especially in the immediate neighborhood of chambers. 

 This is indicated by the fact that in some exceptional cases two 

 or even three nuclei have been found inclosed in the same cell- 

 body, and also by the great diversity (from only two cells up- 

 ward) in the size of the archajocyte-groups. 



The growth of the latter may take place not only by cell- 

 division of their cells but also by fusion of originally separate 

 grouj)s. For, not infrequently small groups consisting of as yet 

 a very small number of cells lie so near to one another that, 

 should they continue to grow by cell-multiplication, they must 

 soon come into contact. There also exist somewhat larger groups 

 whose irregular shape strongly suggests their origin by such a 

 union of two or more originally separate groups. These facts 

 make me believe that the arch(eocyte groups or masses are not 

 necessarily to be considered as derived each from a single mother- 

 cell. 



As already indicated, the archfeocyte-groups, generally flat 

 and small, attain a very considerable size in the deeper parts of 



