1907.] XATIHAL SCIIONCES OF IMU LADKLI'lllA. 237 



In these ten Lower Claiborne races it is evident that the early stages 

 vary more than do the later stages. The end of the cancellated stage 

 comes very constantly toward the close of whorl 8. The end of this 

 stage is not as distinct as its beginning, and it is true that the table 

 exaggerates slightly the sharpness of this termination of the cancellated 

 stage. Though the change from the cancellated to the spiny stage is 

 gradual it is usually accomplished within half a whorl, and this trans- 

 formation takes place with great regularity during the latter part of 

 whorl 8. To sum \\p, the table may slightly exaggerate the sharp- 

 ness of the change, but it in no way exaggerates the fact that the 

 end of the cancellated stage occupies in this series of forms a much 

 more definite place in the ontogeny than does its beginning. 



Table II shows that in the Athlda petrosa races of Jackson time the 

 beginning of the cancellated stage is on the whole earlier than in the 

 forms of the Lower Claiborne, In some cases this acceleration is 

 quite marked, as is seen when we compare, for example, the Jackson or 

 Montgomery races with the ^It. Lebanon race of Table I. The Lower 

 Claiborne St. Maurice race of Athleta petrosa. seems to have attained 

 about the same degree of acceleration as the later races of Table II. 

 The Lower Claiborne ^loseley Ferry race, which exhibits the tendencies 

 of the later Jackson time races, has been introduced into Table II as a 

 probable ancestral f(rm. 



In Table III the Lower Claiborne race of Athleta petrosa from St. 

 Maurice and ihe Athleta clayi, also of St. Maurice, are compared. Ath- 

 leta sayana from the later Claiborne beds is shown as a probable 

 descendant of ^4. clayi. 



Taking all three tables into consideration, it is evident that the 

 smooth stage is by far the most variable stage in Athleta petrosa and 

 its derivatives, .4. clayi and A. sayana. Still another remarkable fact 

 is brought out by these tables when we examine the changes of the 

 smooth stage in relation with the changes in the later stages; It is 

 only in Athleta sayana. a form where the smooth stage has been restricted 

 to a little over one whorl, that we find any marked modification of the 

 characters of the spiny stage. Again, it is only in .4. sayana that we 

 find the spiny stage beginning at such an early point in the ontogeny 

 as the sixth whorl. 



In Athleta clayi and .4. sayana the smooth first whorl is much larger 

 than in any of the assemblages of .4. petrosa. The apex of -4. sayana 

 is larger than the apex of A. clayi, and in addition the later stages of 

 A. sayana are markedly different from those of A. petrosa. On the 

 other hand, the later shell characters of .4. clayi are almost identical 



