376 STUDIES IN EVOLUTION 



In a second group, (h~) long form^ the shell is elongate- 

 spatulate, and proportionately deeper than the normal, but, 

 with a single exception, individuals have not been observed 

 to exceed a length of 10 mm, and a width of 7 mm. This 

 variation is so persistent that it appears to be well founded 

 genetically, and not merely an occasional occurrence. Trac- 

 ing backward from the mature shell to the earlier stages of 

 development, both this and the normal form are found merging 

 into each other ; hence both have had a similar starting-point. 

 The long form, however, reaches maturity of development at 

 a very early age, and never approaches the size or proportions 

 of the normal adult. A tendency to obesity is especially 

 noticeable in the group (I ), the majority of such individuals 

 being: below the normal fuU-cjrowth. 



A single adult example shows traces of broad, rounded 

 plications on each side of tlie fold and sinus, a singular 

 condition in a species uniformly non-plicate. 



Incipient Form (Plate XXI, figures 4, 4 a, 10, 10 a). — Shell 

 2.5 mm. in length by 1.75 mm. in width; elongate-oval; 

 beak elevated, straight, acute. Pedicle-aperture of the ven- 

 tral valve very broad, triangular, extending to, but not 

 encroaching upon the apex. Dorsal beak full, rounded, but 

 inconspicuous. Valves convex just below the beaks, becom- 

 ing depressed toward the anterior margin. The shell is 

 proportionally much narrower than the adult form. 



The starting-point of this series is precisely the same form 

 of shell as that taken for the incipient stage in the species 

 Meristina rectirostris. Under the discussion of that species 

 reference has been made to the im})ossibility of sejiarating 

 these two species, in their earlier stages, and the imj^ression 

 of the specific characters may be regarded as of subsequent 

 development. 



Developmental Variations. 



The surface characters being unvariable, the important 

 changes in development are confined, as far as observable, to 

 the pedicle-aperture and deltidial plates. As already observed, 

 the beak is incurved so early in the history of the individual 



