AI.KINS : GROWTH AND VARIATION OF UNIO PICTORUM. 



229 



ft 



Now, the width : length and thickness : length ratio curves were 

 plotted (figs. I and 2 respectively) ; the data are given in Tables I 

 and II below ; the distance between successive abscissae was o"o5 

 over a range of 0*50 (from 2 "05 to 2 '5 5), and 0*02 over a range of 

 o*22 (from o"54 to 076) respectively. The width : length curve in 



Table I. 



Width/Length ... 2-05 



210 



2-15 



2'20 



2-25 



2-30 



2-35 2-40 



2-45 



2'5o 



2-55 



No. of Individuals | 2 



14 



13 



39 



35 



52 



50 31 



13 



7 



4 



Table II. 



Thickness/Length 



o"54 



•56 



•58 



•60 



•62 



•64 



•66 



•68 



•70 



72 



•74 



•76 



No. of Individuals 



I 



4 



13 



22 



39 



48 



41 



38 



26 



9 



8 



I 





particular shews some irregularities, but these are of little importance 

 in view of the probability of the occurrence of errors of measurement, 

 which it is impossible entirely to eliminate. 



There can be little doubt, so far as these characters go, that the 

 group of shells measured forms a single biological entity, without any 

 tendency towards a differentiation into two or more forms. Neither 

 is any such tendency shown in other features, less susceptible of 

 direct measurement, such as thickness of shell, position and character 

 of the teeth, and so on. We may, therefore, proceed with confidence 

 io enquire into the manner of growth of the species. 



The length has been taken as the standard because it appears to 

 give the best measure of the degree of growth attained by the shell. 

 All shells of the same length were collected together, and the width 

 and thickness of each were used to determine the maximum and 

 minimum and the mean width and thickness corresponding to each 

 particular length. The results are shewn in Table III, and are 

 plotted in figs. 3 and 4. Mean values are shewn as crosses, and the 

 upper and lower limits by dotted and continuous irregular lines. The 

 straight lines shew the width or thickness to be expected if the mean 

 value of each corresponded throughout the series with the most fre- 

 quently occurring width : length or thickness : length ratio, as shewn 

 in figs. I and 2 respectively. Below a length of 23 mm. the number 

 of specimens is too low to give trustworthy data, and the curves below 

 this length are too irregular to be of much value ; the same is true 

 above a length of 35 mm, 



o 



