MUSEUM OF COMPARATIVE ZOOLOGY. 255 



clinging to the bottom by the upper (aboral) surface of the bell is also found 

 in a genus widely separated in our classification from Cassiopea. We noticed 

 last summer a young Cyanea (C.arctica), which was kept in the aquarium at 

 Newport, fasten itself in the same way, and adopt the same sluggish movement 

 of the bell-margin which is so characteristic of Cassiopea. It is not impossible, 

 although as yet not supported by observation, that the "aboral papillae " (Bull. 

 Mus. Comp. Zool., VIII. 8, p. 669, PI. VII. fig. 1) of the young Cyanea may 

 serve to anchor the young medusa in this posture.* 



The bell of C. fronalosa is flat and disk-shaped, in larger specimens with a 

 diameter of a foot or a foot and a half. When seen from the aboral pole | lig. 

 10), two regions can be distinguished on the surface. Of these the central part 

 has a circular form, and a slightly concave surface. It is bounded by the 

 circumference of a circle, whose diameter is about three fourths that of the 

 whole disk, and whose circumference limits that rigid (fig. 7, u) portion of 

 the medusa bell by which it is attached to the bottom. 



The most marked feature in the structure of the central portion of the bell 

 is the possession of sixteen radial stripes (e), which can best be seen from the 

 aboral surface. These stripes are simple thickenings of the bell walls, and 

 are most clearly defined near the periphery of the central region already 

 described. At that point they assume a pyriform shape, while nearer tin- 

 centre of the disk they become narrower until they disappear. A periphery 

 drawn through the ends of these bodies, most distant from the middle of the 

 bell, bounds the rigid portion of the umbrella and divides the central part from 

 the flexible margin. The outline of the stomach cavity (s) can be easily made 

 out through the aboral bell walls, in which, at this point, there are traces of the 

 radial stripes. If the substance of the bell be cut in such a way as to make 

 a cross section of a pyriform body, it will be found that it has a milky-white 

 color, while adjoining parts of the bell are brown and green. Its tissue is also 

 more compact than that of the rest of the bell. They seem to impart a greater 

 rigidity to the bell walls, and not to be simply superficial coloration as some- 

 times supposed. 



The marginal portion of the bell arches upward in the natural position of the 

 medusa, and is very flexible. It is much thinner than the central part, and is 

 almost wholly without radial markings. 



The bell rim is destitute of tentacles. It has, however, marginal bodies 

 which distantly represent these structures in tentaculated Discophora. These 

 structures are of two kinds (v I, o 1). They assume either the form i 'I' serrations 

 {rl) placed peripherally (velar lappets), or take a crescentic shape (ocular lap- 

 pets), (pi). In the latter case they are arranged iii pairs and mark the position 

 on the bell margin of the sense bodies. There are sixteen pairs of these struc- 

 tures (ol) having the more rounded shape, and each pair marks the position 



* In the genus CepJua we find structures homologous to the aboral papflll 

 Burning the form of small excrescences. Haeckel, op. cit., \\ 674, PI. XXXVI. 

 figs. 3, 5, 6. 



