GROWTH. 155 



muscular power that is large at the outset, and increases as 

 the cubes of its dimensions: its dense medium supports it. 

 The exceptional continuance of growth observed in creatures 

 so circumstanced, is therefore perfectly explicable. 



46a. If we go back upon the conclusions set forth in the 

 preceding section, we find that from some of them may be 

 drawn instructive corollaries respecting the limiting sizes of 

 creatures inhabiting different media. More especially I refer 

 to those varying proportions between mass and stress from 

 which, as we have seen, there results, along with increasing 

 size, a diminishing power of mechanical self-support : a rela 

 tion illustrated in its simplest form by the contrast between 

 a dew-drop, which can retain its spheroidal form, and the 

 spread-out mass of water which results when many dew-drops 

 run together. The largest bird that flies (the argument 

 excludes birds which do not fly) is the Condor, which reaches 

 a weight of from 30 to 40 Ibs. Why does there not exist a bird 

 of the size of an elephant? Supposing its habits to be 

 carnivorous, it would have many advantages in obtaining 

 prey: mammals would be at its mercy. Evidently the 

 reason is one which has been pointed out the reason that 

 while the weight to be raised and kept in the air by a bird 

 increases as the cubes of its dimensions, the ability of its 

 bones and muscles to resist the strains which flight neces 

 sitates, increases only as the squares of the dimensions. 

 Though, could the muscles withstand any tensile strain they 

 were subject to, the power like the weight might increase 

 with the cubes, yet since the texture of muscle is such that 

 beyond a certain strain it tears, it results that there is soon 

 reached a size at which flight becomes impossible: the struc 

 tures must give way. In a preceding paragraph the limit to 

 the size of flying creatures was ascribed to the greater 

 physiological cost of the energy required; but it seems 

 probable that the mechanical obstacle here pointed out has 

 a larger share in determining the limit. 



