26 THE HUMANIZING OF THE BRUTE. 



if below, the length of the side o a would either not 

 correspond to that of the outer edge a g or at least 

 the oblique position of the funnel would make it im- 

 possible to wind the other half of the leaf around it. 

 The margin, therefore, must be a curved line. But 

 this curved line again would be either convex or con- 

 cave; or partly convex, partly concave. Of these pos- 

 sible cases, the first two would be impractical; for the 

 merely convex margin has all the disadvantages of a 

 straight one, and, besides, would make the poor 

 beetle do superfluous work in rolling up a part of the 

 leaf (0 b a c) that is of no use in the formation of the 

 funnel. A funnel with a merely concave margin 

 would have too many windings closely packed at the 

 top, and thus overtax the strength of the builder; 

 and the funnel, which is subsequently to serve also 

 as food for the larvae, would perhaps have dwindled 

 down too much in size. 



There remains only the concavo-convex line 

 of section, in which again either the convex or 

 the concave part might be longer. And here, as Was- 

 mann justly remarks, "the technical ingenuity of our 

 architect shows itself in its brightest light. ' ' For what 

 would be the result if the convex part were longer? 

 We need only cut such a funnel from paper and see. 

 First of all, it is not pointed enough. Besides, the 

 part h b is not in the spiral of the point, as it should 

 be, but along the vertical axis, and thereby the curve 

 o cb d will no longer have the required length. And 

 the part c a b d would most unsuitably protrude above 

 the funnel's apex. At any rate, the funnel would be 



