ABT. 20 FLOEIDA TREE SNAILS — SIMPSON 13 



passage of dry-land plants and to a considerable extent the migra- 

 tation of snails. Liguus could only cross this great marsh during a 

 time when the sea was driven in over it by hurricanes and then on 

 floating timber. 



During the time of the second elevation there was probably an 

 almost continuous hammock along much of the south shore of the 

 mainland; in fact a considerable amount still remains. The Liguu^s 

 castaneozonatus, vacaensis, Uneoldius, 7'oseatus, and niarmoratus 

 crossed over from the keys and became establislied on the south and 

 southwest parts of the mainland. By and by the sea began to seri- 

 ously gnaw away at the great coral island. High tides occur on 

 one side at different times from what they do on the other, and I 

 have seen at low tide on one side of one of these coral keys many 

 streams of water passing through, from a tiny trickle to those of the 

 size of small rivers. Soon the carbon dioxide in the water ate the 

 rock away until the weakened roof fell and later an open channel 

 formed Avhich cut the island in two. This action, aided by the fury 

 of the sea when driven in b}^ storms, has been continued until this 

 once greatly elongated island has been cut into more than 30 islets 

 and islands, each high enough to bear upland hammock, besides a vast 

 number that are only clothed with mangroves. Of course when a 

 ke}' was cut in two it checked the passage of the Lu/mis from one 

 island to the other, and we can form some idea of the time of arrival 

 of the snails by their distribution. The subspecies matecunibiensis 

 did not arrive until Upper Matecumbe was cut off from the keys to 

 the north and south of it, for we have never found it on any other 

 of these islands. But it came in time to cross the land bridge, for we 

 find it one of the most abundant of the tree snails of the Flamingo 

 region. Oxi/styla -flaridensis^ which inhabits the keys from Largo 

 southAvestward, crossed this bridge and is very abundant in the Cape 

 Sable area ; but O. reses, which inhabits the keys from Key West to 

 Vaca, apparently has not passed any farther up them nor crossed to 

 the mainland. A beautiful form, suberenatvs, was common in the 

 center of the upper chain from probably Largo to Grassy Keys. The 

 subspecies eUiottensis has its metropolis on the island for which it is 

 named but has been found on Old Rhodes and possibly upper Largo. 

 It w^as a late arrival and probably was kept from an extensive 

 distribution because of the cross-channels. 



Four subspecies of Ligims Avhich inhabit almost the entire rano© 

 of the Upper Keys — co.stameozoimtus, roseatus, ■mannorafuf^. arid 

 luteus — are also found on the mainland on the rocky southeastern 

 ridge, although a tract of Everglades and a series of shallow sounds 

 separate the two regions. During time of hurricanes when the sea is 

 driven against southeast Florida with tremendous force the entire 

 chain of islands may be overflowed and water driven in until it covers 



