46 ECHINOIDEA. I. 



SO that there can be no doiibt tliat it is a genuine ophicephalous pedicellaria. It is a highly curious 

 faet that each of these three kinds of pedicellariæ, two of which show a very perfect development, are 

 only foiind in a single genus, vvhile none of the other Echinothurids seem to have a corresponding 

 form of pedicellariæ. 



The tridentate pedicellariæ are very richly developed in the Echinothurids. Most frequently 

 their form is simjjle; the valves are leaf-shaped, and the blade is more or less filled by a net of meshes 

 which may be very spinous. In another common form the edges of the blade are involuted, so that 

 only the point of the blade is somewhat widened; in this form the blade is commonly strongly bent, 

 so that the valves are widely separated, and only join with their points when the pedicellaria is closed. 

 Both these forms may be found in the same species; and in a group of species, A. variitin and the 

 species most nearly allied to it, even three different kinds of tridentate pedicellariæ are found, viz. 

 besides the two mentioned forms a short, broad one with coarsely serrate edge (PI. VIII, Figs. 4, 27). 

 A peculiar short and broad form is found in Pli. luculcntunr, it recalls to some degree an ophice- 

 phalous pedicellaria, but as it has no indication of an are, there can scarcely be any question of inter- 

 preting it as any thing else than a form of the tridentate pedicellariæ. The tridentate pedicellariæ 

 may be very large, especially those with involuted edge; these have commonly a verv short neck. 



The triphyllous pedicellariæ (PI. XII, PI. XIII. Fig. 23) are very well developed in the Echino- 

 thurids; peculiar to these in comparison with the triphyllous pedicellariæ of the Echinids is the faet 

 that the upper edge of the apophysis spreads over the lower part of the blade, and continues up along 

 its sides; in some, for instauce Pli. placenta, this cover-plate is not niuch developed, in most species 

 it is highly developed, and covers a great part of the blade. Generally there are then some large 

 holes in the median line, and some smaller holes around; the part continuing upward along the lateral 

 edges of the blade, is most frequently without holes. The upper edge of the blade is generally finely 

 serrate. The holes in the blade are always placed in rather regular curves from the raiddle obliquely 

 upward on either side. — The peculiar bottle-shaped, two-valved pedicellaria, figured by Agassiz from 

 Phormosoma tcnuc (Chall. Echinoidea. PI. XLIV. Fig. 21) is presumably an abnormal form. I have 

 examiued a couple of the type specimens, but have only found the connuou, three-valved form. 

 Agassiz (Chall. Echinoidea. p. 84) thinks that this bottleshaped pedicellaria is only a modification of 

 the ordinary type of pedicellariæ, in which the terminal edge becomes raised to form a spoon-shaped 

 valve>. This is absolutely wrong; one form is a triphyllous pedicellaria, the other a tridentate one. 



The stalk of the pedicellariæ in by far the greatest number of Echinothurids is thiu, irregularly 

 perforated, not distiuctly tube-shaped (PI. XIV. Fig. 31). In the large tridentate pedicellariæ, as in A. 

 varium, also the stalk is somewhat coarser; the stalk of the ophicephalous pedicellariæ of Tromikosoma 

 is a rather thick tube. In Pli. asterias the construction of the stalk is quite exceptional among the 

 Echinothurids; it consists of some long, very thin calcareous threads, only united at the ends of the 

 stalk, at most connected in the intervening part by quite few transverse ridges. 



Also the iuner anatomicai strncture seems to yield good systematic characters. Thus Bell 

 (Catalogue p. 142) mentions as a chief difference between the genera P/ioriiiosoiiia and Ast/iriwsoiiia that 

 the latter has highly developed v longitudinal muscles > dividing the body-cavity iuto chambers, while 

 such muscles are wanting in Phori/iosoma. — To this, however, is to be remarked that the specimens 



