38 PROCEEDINGS OF THE MALACOLOGICAL SOCIETY. 



39 out of 2,994, or 1'3 per cent. There is no evidence that large 

 or small individuals are more likely to lose their apical whorls 

 than those of moderate size, and the error introduced in this 

 way may be neglected. 



§ 4. With respect to the first question, whether shells from 

 individual trees close to one another show differences in size, there 

 are data for six trees in area A, three trees in area B, and six trees 

 in area D. The figures for these fifteen lots are given in table I, 

 and the result of the appropriate calculations in table II. From 

 these it appears that the shells from closely adjacent trees are 

 demonstrably different in size in two instances only, Ab being 

 definitely shorter than Ac, and Da broader than Df . With six trees 

 in area A there are fifteen comparisons and fifteen possible differences, 

 in area B three, in area D fifteen, in all thirty-three, or, if we take 

 altitudes and diameter separately, sixty-six. Of these two only are 

 present. This negative result throws no light on the question as 

 to how far a familial assembly of CI. bidentata ranges ; it might mean 

 that the range is greater than the area served by a single tree, or 

 that families living near one another are not distinguishable in 

 size with the available data. 



Table II. — SHOWING THE SIGNIFICANT DIFFERENCE IN 

 ALTITUDE { + ) AND DIAMETER (o) FOR FIFTEEN 

 SEPARATE TREES IN THREE AREAS. 



Aa Ab Ac Ad Ae Af Ba Bh Be Da Db Dc Del De Df 



§ 5. That the shells in the different areas are largely differentiated 

 in size is shown pretty plainly if individual trees in one area are 

 compared with individual trees in another. Taking table II as 

 a whole, there are 105 possible comparisons; eighteen differ in 

 altitude only, one in diameter only, and twenty-seven in both 

 altitude and diameter, forty-six in all. The comparison is, however, 

 best made in the simple form of taking each area as a whole, as in 

 tables III and IV. The shells from each area differ in altitude 

 from those in each of the other areas, with the exception that 

 C and E are not differentiated. Nine of the ten possible differences 



