190 Ostenfeid: Distribution of Sea-Grasses. 



whole group migrated from the Indopacific to the Caribl>ean Sea 

 at a time when there was a connection between them through the 

 Isthmus of Panama. Two of these species have reached the Ber- 

 mudas, but elsewhere they are not found outside the Caribbean 

 region, the Atlantic Ocean being very poor in sea-grasses. 



7. The genus Posidonia has at the' present time one species 

 along the south coast of Australia, and the other in the Mediter- 

 ranean. It is supposed that they represent the last remnants of 

 a genus whose home was somewhere in the Indian region, and 

 that it w^as driven away towards the north and the south. 



8. The other species of the Mediterranean group, Cymodocea 

 jiodosa, also migrated from the south-east into the Mediterranean, 

 while the two Zosteras of this sea came from the north. 



9. The genus Phi/llospadix is restricted to the North Pacific. 

 Its morphological characters indicate its derivation from Zostera. 



10. The narrow-leaved, small Zosteras (Z. nana, and the three 

 Australian species) are supposed to be the older type of the genus, 

 as they still have the " retinaculum " (the scale attached to the 

 flower). When taken together, their distribution is " bipolar," 

 with a few outposts in the Tropics. 



11. It is supposed that the genus Zonfera originated in a warm 

 sea and migrated towards the north and the south. 



12. The younger type, Zostera marina, is yet an old species. 

 It is supposed that it also originated in a warm sea (perhaps in 

 the Indopacific region), wandered northwards, and in one manner 

 or another came from the Pacific into the Atlantic, or vice versa. 



13. With the exception of Phyllospadix, which originated in the 

 North-Pacific, and arose from Zostera, all the genera of sea-grasses 

 are supposed to have arisen in the Tropics, where the home of most 

 of them still is, Zostera marina being the only species which extends 

 into the Arctic Sea. 



14. The distribution of the species still requii-es investigation. 

 This applias especially to the three Australian Zosteras, which are 

 little known, both systematically and geographically. 



