ORDERACTINOIDEA. 73 



fined to the side which faces obliquely outward and downward ; and 

 it is by this means that the horizontal growth is carried on. 



e. Again, the budding polyps are sometimes confined to two oppo- 

 site sides of a branch, and pinnate forms result — that is, branches, 

 with two opposite rows of branchlets, as shown in figure 32, repre- 

 senting part of a branch of the Gorgonia setosa. In this species, 

 there are one or two rows of minute polyps on one side of a branch, 

 and one on the other ; the branch elongates by a succession of buds, 

 the new buds opening at the very apex. Branchlets — or pinnules, 

 as they are called — bud from both sides, and from either of the 

 rows, on the side which has two, but from only one at a time. 

 There is usually an interval of five or six polyps on each side be- 

 tween the pinnules, and owing to this they are mostly about one-third 

 of an inch apart. The buds are sometimes alternately from the two 

 rows, but often continue in one for some distance, and then change to 

 the other, or alternate again. Owing to this want of perfect uni- 

 formity, and sometimes a spiral twist in the stem, the pinnules are 

 somewhat irregular, or a little zigzag in position. The pinnules 

 elongate by apical budding to a certain length, without any increase 

 in diameter ; but they sometimes give out lateral pinnules below, and 

 thus commence to become branches. In the change of a pinnule to 

 a branch, one or two from among the lower polyps begin to bud : the 

 growing pinnule goes on elongating, and shortly, on the other side of 

 the same, another polyp, or pair of polyps, buds, and originates a second 

 branchlet; and then, when lengthened at apex a little farther, another 

 starts on the opposite side, each new budding-point being at a nearly 

 uniform distance from the apex. In this manner, the lengthening 

 pinnule becomes a pinnate branch. 



f. The positions of branches, as well as their size, are strikingly 

 alike in different specimens of the same species. The angle which 

 the polyps make with the axis of the stem, is the angle with which 

 the new branch begins. This angle varies little in the same species; 

 it is sometimes quite small, and the branchlets are then nearly erect 

 and crowded together ; but sixty degrees is the more common angle, 

 and in some instances it is ninety degrees, or the branch is even 

 reversed a little. The branches, when spreading, usually curve 

 upward as they elongate themselves, and sometimes become vertical, 

 an eflfect which appears to proceed in part from the influence of light; 

 that is, the propensity of the polyp to grow upward towards the light. 

 The horizontal Madrepores (plates 32, 33) follow the same principle, 



19 



