VARIATION AND ZOOGEOGRAPHY OF LIGUTJS IN FLORIDA. 455 



6 foot line, extending with little interruption from Grassy and Duck Keys to 

 the mainland. Long Key, which may belong to this group of keys, is "a long 

 strip with vegetation hardly suited to Liguus. Drymams was found on the low 



bushes" (C. B. Moore). 



Grassy Key (or Big Grassy Key) . The hybrid race roseatus X castaneozonatus 

 was taken. Both forms are very similar to those of Chokoloskee Key, drawn in 

 ftes 14 14c. Mr. J. S. Raybon, who collected these shells for Mr. Moore in 

 1904 reports dead shells in abundance, but only three living ones found after 

 a long search, although the hammock appears very favorable. Two specimens 

 of crenatus are in the collection of the Academy, collected about twenty years 

 ago by Mr. Joseph Willcox. They are like the Duck Key form. No crenatus 

 were among the few Grassy Key shells taken by Mr. Raybon. 



West Crawl Key, searched by Mr. Raybon in 1904, afforded only quantities 

 of long-dead fragments, one of which shows the bands of castaneozonatus. It 

 is evidently the same race as that on Grassy Key. 



Duck Key. A few fragments, the last whorl only, were taken by Mr. Raybon 

 in 1904, the freshest one not unlike the form of crenatus from Grassy Key and 

 Middle Cape Sable, shown in PI. XXXVIII, fig. 16d. As the other fragments are 

 too old to show color I leave the form among those of adjacent keys, since it 

 would probably prove to be a hybrid race if adequate material was available. 



Key Vaca. Dead shells of the same roseatus X castaneozonatus hybrid race 

 were taken in 1904 by Mr. Raybon on Key Vaca in an old field about three miles 

 west of the middle of the key 



Key Vaca (PL XXXVII, figs. 9, 9a-d). About midway of Key Vaca Mr 



Raybon (1904) found eleven living Liguus. The shells are thin, long-spired 

 composed of 7 to 8 whorls. The columella is of the thin crenatus type. Color- 

 ation as in pale crenatus (white or xanthistic) in some shells, but in others pecu 



liarly marbled with deep brown or black 



ground. The marbled 



form, which may be called marmoratus, is probably a new mutation of crenatus, 

 or of crenatus X castaneozonatus, which now forms a considerable element in the 

 race. The color forms represented are as follows: 



1. Plate XXXVII, fig. 9c. White, with a few olive-yellow or olive-green lines 

 on the last half whorl, four shells. Another is similar except for its very pale 



ochre tint. 



2. Plate XXXVII, fig. 96. Similar but strongly xanthistic, also with olive- 

 green lines, two shells. . 



3. Pale ochre, brightest near the lip, with the very tip of the apex, and streaks 

 on the parietal wall roseate, an area around the columella purple; one shell. 



4. Plate XXXVII, figs. 9, 9a, 9d. Embryonic and first neanic whorls white, 

 spirally striate ; on or at the end of the fourth whorl small brown spots in spiral 

 series appear above and below the sutural white borders. About a whorl jurtner 

 on the spots of the lower series become confluent into a blackish-chestnut band, 

 from which wide black or purple-brown flames rise to the spots along the suture 



* 



