115 



diately, from the two initial cells before described, in two cells 

 usually smaller than the other polypiferous cells, and of a different 

 form, and I have not yet been successful in seeing polypes in them. 

 One of this primary pair, moreover, differs from every other cell in the 

 branch, in its supporting a cup, and it is at this point that we first 

 notice the appearance of the latter very curious organ. There is no 

 communication, however, between the cup and the cell. The first 

 two polypiferous cells being formed in this way, the series of which 

 each of them is the commencement is thus continued. The primary 

 cell which bears the cup is continuous by a contracted portion with 

 the cell immediately above it (see fig. 3, d), which gives off from 

 its superior and internal or axial angle a narrow tubular prolonga- 

 tion, which above the next pair of cells expands into the cell on the 

 opposite side of the axis, and which constitutes one of the pair 

 of cells next but one above ; unless, as is frequently the case, the 

 branch bifurcates immediately above the second pair of cells, in 

 which case, the tubular prolongation of the cell is continued into the 

 inferior or single cell at the base of one of the branches of the fork. 

 The other primary cell, or that which forms the pair with the cup- 

 bearing cell, sends up a prolongation from its superior and internal 

 angle, which above the level of the first regular pair of cells expands 

 into the alternate cell on the same side as itself, or one of those con- 

 stituting the second pair of cells. The continuation of the series 

 along the branch takes place in the same way, by each cell giving off 

 a tubular prolongation from its superior and internal angle, which, 

 after the first bifurcation, is always placed on the anterior aspect of 

 the branch, and gives off or expands into a polypiferous cell at each 

 alternate pair, and always at the same side of the axis in the same 

 internode, and alternately on the other side in each fresh internode 

 into which the series enters. We have thus from the primary pair of 

 cells traced the origin of two series, each of which constitutes half of 

 the two first branches into which the polypidom divides, and to each 

 of which series one of the first pair of regular cells belongs, and one of 

 the second pair, if there be two pairs before the first bifurcation. Sup- 

 posing this to be the case, as in fig. 6, it will be seen that the pri- 

 mary cells are marked a and a", the second pair of cells or the first 

 regular pair b and b", and the third pair c and c". Now, of these 

 four regular cells, two only are derived from the primary pair, viz. b 

 from a, and c" from a". It remains to account for the origin of b" and 

 c. This will be seen, looking at fig. 6, to take place thus : — Above 

 and behind the primary cell (a") is placed a cup, which forms, as it 



