346 



MEDUS/E OF THE WORLD. 



Wherever a rudimentary or newly arisen tentacle lies on the bell-margin, it will always, normally, 

 be found to lie just in front of a newly arisen sense-organ, and just after a larger tentacle, i. e., one 

 of an earlier cycle." 



There is thus a cyclic symmetry in the order of appearance of the marginal sense-organs 

 with reference to the tentacles. 



New tentacles develop 4 at a time 90° apart so that they occupy identical positions in the 

 four quadrants. Very often, however, the 4 new tentacles do not appear simultaneously; 

 but first 2 appear 180° apart and then 2 more 90 from the first pair. 



Fig. 197. — Sequence of the development of the lithocysts and tentacles in Gonionemus mur- 

 bachii. A, youngest stage; D, mature state. After Perkins, in Proc. Acad. Nat. Sci., 

 Philadelphia. The sequence of the lithocysts is shown in Arabic numerals, I being the 

 oldest (first formed) and 10 the youngest (last formed). 



The order of appearance of the tentacles and the lithocysts will be made clear through an 

 inspection of the text-figure 197 reproduced from Perkins, 1903, Proc. Acad. Nat. Sci. 

 Philadelphia, for 1902. The lithocysts finally become nearly as numerous as the tentacles. 



Miss Ida Hyde, 1902, has made an interesting study of the nervous system of G. murhachii. 

 She finds that there are three marginal nerve-rings, one below and two above the attachment 

 of the velum; and there are also nerve chains along the radial-canals and a nervous network 

 over the subumbrella. The ganglion-cells are bipolar and multipolar. 



The variations of the medusa have been studied by Hargitt, 1901. Among about 2,000 

 specimens he found a range of from 2 to 6 radial-canals and cases of forked and trifid ten- 

 tacles were observed. The forking may arise at the base or from the shaft of the tentacles. 

 Hargitt's smallest medusa was 2 mm. wide and had 29 tentacles, while his largest was 19 mm. 

 wide and had 68 tentacles. One 15 to 16 mm. wide had 72 tentacles. About 5 per cent of the 

 medusae have other than 4 radial-canals. 



The regeneration of Gonionemus has been studied by Hargitt and also by Morgan. 

 Hargitt carried out an interesting series of experiments in grafting in the medusa and he 

 finds that they show a well-marked polar orientation. No successful aboral grafting of medusae 

 could be made, but the bell-margins of different individuals would readily unite. 



When the bell-margins of two individuals were united by their bell-edges, the margins of 

 each having been cut away, they soon coalesced everywhere around the rim of each, excepting 



