306 CONTRIBUTIONS TO THE NATURAL HISTORY 



appear as a flat disk entirely spread out, the marginal tentacles spreading also in the same 

 plane. Or the disk may be swollen slightly into a flat hemisphere, with a uniform curve 

 from the margin to its summit, the tentacles hanging down at right angles all around 

 (Plate VII. Fig. 10, 11). Or the prominent disk may be more swollen in the centre 

 than towards the margin, which may spread, while the central part of the disk is con- 

 siderably raised. Again, the whole disk may be contracted so much as to form a 

 regular rounded hemisphere (Plate VII. Fig. 1), with hanging marginal threads ; or the 

 margin may be still more contracted, and give the whole body an almost spherical shape, 

 when the lower opening is perhaps reduced to half the diameter of the globular body. 



Again, the muscles may contract in such a manner, as to give a quadrangular form to 

 ' the disk with rounded angles, and the upper portion may be more or less raised (Plate 

 VII. Fig. 8). In such a position, the margins are inflected inwards between the arms of 

 the cross, and the fringes form also a more or less quadrangular fringe, and the inner 

 seam is bent over between the arms of the cross, and covers them more or less. 

 If, in such a position, the animal contracts still further, and the radiating muscular 

 bundles between the arms be particularly active, the quadrangular shape of the outline 

 may be still more brought out, and it may even assume the form of an emarginate four- 

 rayed star, when the inner seam covers almost entirely the arms of the cross, and shuts 

 completely the lower surface. In that case the marginal fringes are bent inwards, and, 

 the lobes in which they are inserted form, as it were, four valves, moving up and down, 

 and alternately opening and shutting the lower cavity. Finally, the contraction of the 

 radiating muscles may be such as to divide the fringed margin into more or less numer- 

 ous lobes (Plate VII. Fig. 7), which may form an undulating circle, or more or less prom- 

 inent lobes almost as completely isolated as if they were separated by fissures. 



In order fully to understand the structure of the central cross of the lower surface, 

 it is necessary to be acquainted with the morphology of the mouth of other Medusae, 

 and to have traced its successive changes in some of the higher Medusae, from its earliest 

 condition up to the time when it assumes there its most complicated structure. 



In the young Aurelia, the mouth is a mere quadrangular opening, with prominent 

 margins. These margins, however, soon project ; the angles become more prominent, 

 and then elongate ; and this elongated portion at length widens and is enlarged into flat 

 tentacles, frequently called by zoologists the arms of the mouth. In the perfect adult 

 Aurelia, the mouth is a simple cross-fissure, with prolonged thin, membranous margins. 

 The margins of each arm of such a cross are laid flat against each other, and constitute 

 compressed channels opening into the mouth. 



Now let us for a moment suppose that such a mouth, instead of being confined to a 



