1907.] 



XATLRAL SCIKN'CES OF PIIILADKLPHIA. 



239 



the cause of this chan<2;e may be unknown the chansje itself is unques- 

 tionable, and in the probable phylogen}- of this particular series of 

 forms it precedes the modification of the later shell stages. 



Probable Phylogenetic Relations. 



The Lower Claiborne races of Athleta petrosa show not a little racial 

 variation especially in the smooth stage. They are nevertheless 

 typical examples of the main stock. The later Claiborne race of 

 Alabama probably ow^es its origin to some such assemblage as the 

 first strain from the Lower Claiborne of Moseley Ferry. The second 

 strain from INIoseley Ferrj^ — that is the one which shows such marked 

 Jackson tendencies — is a probable ancestor which, through acceleration 

 and the multiplication of plaits, passed into the Montgomery and 

 Jackson races of Jackson time. 



Fig. 6. Fig. 7. Fig. S. 



Fig. 6. — Athleta peti-osa Con. St. Maurice, Louisiana. Immature indi\-idual ex- 

 hibiting the sculpture of the cancellated stage. Lengtli = 10 mm. 



Fig. 7. — Athleta clayi. St. Maurice, Louisiana. Immature individual exhibiting 

 the sculpture of the cancellated stage. Length = 12 mm. 



Fig. S. — Athleta sayana Con. Claiborne, Alabama. Mature indi\idual showing 

 in the sculpture of its later whorls marked departures from the A. petroso 

 main stock. The secondary cancellation is shown in the last whorl. Length 

 = 48 mm. 



Athleta clayi is almost identical with the examples of A. petrosa 

 main stock in everything except its smooth stage. In its smooth stage 

 it approaches Athleta sayana of the later Claiborne beds of Alabama. 

 We are reasonably safe in concluding that A. sayana was derived from 

 the A. petrosa main stock through the Lower Claiborne A. clayi. In 



