440 VARIATION AND ZOOGEOGRAPHY OF LIGXJUS IN FLORIDA 



2. Liguus fasti atus roseatus (PI. XXXVIII, figs. 11-13, 18-19a) 



pure 



the 



occupies keys between those of pure crenatus and those of the hybrid forms 

 east (fig. 3) and in western Florida it ranges north of the hybrid form (fig. 5 ( page 

 442). It occurs only on keys which from their positions must have remained 

 connected with the mainland after the separation of those occupied by L. crenatus 

 These keys must in their turn have been isolated in the gradual disintegration of 

 the land prior to the isolation of those occupied by the chestnut-zoned race. 11 This 

 is so obvious, both in several colonies in the west and in the area occupied by seven 

 colonies north of Key Largo, that we cannot escape the conclusion that roseatus 

 migrated into its present localities after the migration of crenatus and before that 

 castaneozonatus . That the position of the colonies of roseatus relative to the 



of 



other forms is fortuitous seems so improbable as to be negligible. 



In roseatus, as in crenatus, there has been conspicuous local differentiation. 

 In view of the prevalence of similar forms over far greater areas where the 

 colonies are not isolated, it can hardly be doubted that isolation has been instru- 

 mental in the evolution of these incipient races. As noted in the descriptions of 

 the colonies, they are characterized by varying degrees of intensification of the 

 several factors involved in the roseatus organization, and not by new characters. 



3. The hybrid colonies occupy areas more central than the pure races of 

 crenatus and roseatus, or than L. solidus. On the west coast, at Chokoloskee 

 and Rabbit Key, the hybrid segregates into roseatus (PL XXXVIII, figs. 14, 14a, 

 146) and castaneozonatus (PL XXXVIII, figs. 14c, 14d, 14e) . Apparently crenatm 

 did not penetrate so far north on this coast. The pure colonies of roseatus lie 

 farther north. 



In the south and east, Middle Cape Sable, Grassy and Vaca Keys, the main- 

 land bordering Biscayne Bay and Key Largo (see fig. 6, page 442, and fig. 3, 

 page 441), the hybrid race segregates into crenatus, roseatus and castaneozonatus. 

 The isolation of these hybrid colonies from one another is conspicuously less 

 wide than in the case of the pure races, though the rapid disintegration of the 

 periphery of the peninsula has already brought about some separation. Key 

 Largo apparently formed a peninsula of the mainland up to a very recen ^P erl0< ? 

 The Arch Creek colony is now separated from those south of Miami 



There 



Sable 



probably almost or quite continuous hammock from Miami to Detroit. How 

 far this great colony or series of colonies extends in the direction of Cape 

 is unknown. 13 The colonies on Grassy and Vaca Keys are the most isolated, du 

 great shoals lie in the interval, and it seems reasonable to suppose that 

 extensive tract of land formerly connected these south-central keys wi ^ 

 mainland, since the castaneozonatus and roseatus elements of their Liguus 

 cannot be derived from the keys either east or west. In this connection it y 



n A single colony of roseatus, that at South Cape Sable, is in an area where we wot i e*P ^ 

 •iH 1-opo Tt ; a ««* „«i;ir«i,, +V.O+ i,w»oi onnrUfiftno \t lrnown would indicate an early iso 



hybrid race, 

 colony. 



understand that south of D 



; and the east coast I am told 



glade 



