Browne — Flora and Fauna of Valencia Harbour, Ireland. 695 



It will be seen from the list of species that most of the Medusae 

 belong to one or other of the two orders — Anthomednsae and Lepto- 

 medusse. These orders contain nearly all the species usually found 

 in our seas. Though only a few Medusae have been connected with 

 Hydroids, yet probably an alternation of generations will be ultimately 

 proved to exist in all the species belonging to the two orders. 

 Tp to the present time no evidence has been found of the direct 

 development of a Medusa fi'om a Medusa in either the Anthomedusse 

 or Leptomedusae. 



The MedusK belonging to the other orders, the Trachomedusge 

 and the IS'arcomedusae, may be regarded as visitors to our shores. 

 Their natural habitat is the open ocean, and their development is 

 probably direct, as we have no evidence of the existence of any 

 Hydroid form belonging to them. 



Our Medusae may, therefore, be divided into two groups : — {a) 

 Littoral (Anthomedusse and Leptomedusae ; (h) Oceanic (Trachomedusae 

 and I^arcomedusae). Certain species of the Leptomedusae may be 

 regarded as visitors to British waters, as they have been very rarely 

 recorded (such an one is Octorchis) ; but they may be common elsewhere, 

 and probably belong to Hydi'oids inhabiting grounds far away fi'om 

 our shores. 



The geographical position of Valencia Harbour makes it difficult 

 there to distinguish the regular inhabitants of the coast from the 

 oceanic species and the casual visitors ; but, by comparing them with 

 Medusae found at places further from the Atlantic, an approximately 

 correct estimate can be obtained. For this comparison I select two 

 places : — Plymouth, and Port Erin, in the Isle of Man, where the 

 Medusoid fauna is fairly well known to me. 



The Yale'N'cia Medtts^ coitpahed with Species eoitni) at PLYiiouiH 

 ANT) Poet Eeen. 



All the Anthomedusae found at Valencia have, with one exception,, 

 been taken at or near Plymouth. This exception is Margelispyramidata, 

 which has only been recorded from the west coast of Scotland. Since 

 this species almost certainly comes from a Hydroid belonging to the 

 genus Bou(/amviUia, or a closely allied genus, its distribution must 

 mainly depend upon the distribution of that Hydroid, which may Kve 

 in fairly deep water ojS the Atlantic coasts of Ireland and Scotland. 



I have taken at Port Eiin nearly all the Valencia Anthomedusas ;. 

 and, from the known distribution of the missing species, they, with 



