22 THE NAUTILUS. 



Fig. 16. Form b. Key West, A. N. S. 104029. 



Fig. 17. Form b. Babia Honda Key. A. N. S. 105521. 



Fig. 18. Form d. Bahia Honda Key. A. N. S. 105521. 



Fig. 19. Form b. Bahia Honda Key. A. N. 8. 105521. 



Fig. 20, 21. Form c. Key West. A. N. S. 104029. 



Fig. 22. Form d. Key West. A. N. S. 104029. 



Fig. 23. Form c. New Braunfels Texas. A. N. S. 68888. 



Fig. 24, 26, 27. Form c. Summerland Key. A. N. S. 102742. 



Fig. 25. Forme. Bahia Honda Key. A. N. S. 105521 . 



(Key West, figs. 1, 3, 5, 6, 7, 8, 12, 13, 15, 16, 20, 21, 22. 

 Bahia Honda Key, figs. 9, 10, 14, 17, 18, 19, 25. Summerland 

 Key, figs. 4, 24, 26, 27. New Braunfels, Texas, figs. 2, 11, 23). 



( To be concluded^). 



TJNIO (LAMPSILIS?) PURPURIATUS SAY. 



BT L. S. FRIKRSON. 



Though Mr. Say describes very well his Unio purpuriatus, its 

 identity seems to have been nearly lost, That it may again be 

 known to students, it is now figured, plate III, figs. 4 S 5, and his de- 

 scription repeated. " Transversely oval, slightly oblong, in some 

 specimens with a little tendency towards ovate ; dirty yellowish or 

 fuscous, obscurely radiate with blackish green ; beaks behind [be- 

 fore] the middle, hardly raised ; anterior [posterior] edge rounded, 

 or rather composed of two nearly rectilinear lines ; posterior [an- 

 terior] margin rounded ; hinge margin regularly rounded ; cavity of 

 the hinge membranes (behind [before] the beaks) narrow but very ob- 

 vious ; basal margin arquated, a little compressed in the middle and 

 sometimes almost contracted in that part ; within purple, margin 

 livid, anterior [posterior'] subroargin iridescent; primary teeth 

 nearly direct, rather thick, striated ; lateral teeth hardly extending 

 beyond the sinus of the hinge margin. Length [height] one inch 

 and three-fifths ; breadth [length] three inches and one-tenth. Con- 

 vexity over nine-tenths of an inch. 



REMARKS. Conrad's identification of this species with U. medell- 

 inus Lea no doubt has confused the situation. Crosse and Fischer 

 doubted this identification of Conrad's, but they erred in thinking 

 that the name was badly spelled and meant to be purpuratus, preoc- 



