﻿April, 1857.] 



ELLIOTT SOCIETY. 



181 



5 JIB- ORDER — EXOS TOM A TA. 



Bell generally shallow, seldom deep. Ocelli rarely present ; 

 marginal capsules, always present, in the earlier stages of growth, 

 and generally throughout life. Development sometimes by alter- 

 nate generation sometimes by a direct metamorphosis, during 

 which multiplication by gemmation or fission may take place. 

 In either case the cavity of the disk or bell appears to be never a 

 closed sack, but always open, being gradually formed by the 

 growth outward and downward of a fold from the base of the pro- 

 boscidiform projection which becomes the digestive trunk. The 

 latter is thus uncovered and free during growth. In the numer- 

 ous cases of alternate generation, the young medusa-buds appear 

 to be almost universally protected as are the gemmiparous polyps 

 by a horny case, which is an expansion of the hard rind of the 

 polyp stems. In the case of direct metamorphosis, the free lar- 

 vae, sometimes seek protection in the bell-cavities of other medusae, 

 imitating as parasites the normal condition of the young Tubula- 

 ria and like Endostomata. They appear also to derive their nour- 

 ishment entirely from the medusae upon which they fix them- 

 selves. 



Remark. — The absence of ocelli is almost universal among these 

 medusae, such organs being found only in Thaumantias (Gegenb.) 

 and a few allied genera. On the other hand, the remarkable 

 marginal sense-capsules, (which, were either overlooked or con- 

 founded with ocelli, and incipient tentacula, by Forbes,) are un- 

 mistakably present in every genus well enough known to make 

 its position in the group certain, with the single exception of 

 Thaumantias, in which if representatives of these organs exist, 

 they have not yet been pointed out. Ocelli and marginal cap- 

 sules ("the greater ocelli" Agass,) exist together in the genus 

 Tiaropsis as defined by its celebrated author. These capsules 

 are certainly, in a general sense, as appendages of the disk-mar- 

 gin homologous with the tentacula, but they constitute a second 

 and very distinct class of appendages. A tentaculum never be- 

 comes a capsule nor a capsule a tentaculum, but so far as my obser- 

 vations extend they are distinguished from each other in form, 

 even in their earliest stages of growth. Besides this, the capsules 

 are developed in a different position, i. c, in a circle within or 

 below the circle of tentacula, and these two circles do not coalesce 

 until comparatively late in life, and even then their mode of dis- 

 tribution, especially in the higher genera, preserves an evident 



