The Evolution of the Kidney 5 1 



which excessive quantities of water could be absorbed, and 

 the ostracoderms and early fishes had to compensate for this 

 excessive influx by increasing the excretion of this substance. 

 Their battle against fresh water was only half won. Evolu- 

 tion frequently works by adapting old things to new uses, 

 and it seems that no better way could be devised to get the 

 surplus of water out of the body than to have the heart pump 

 it out; and the easiest way to do this was to prepare a filtering 

 device by bringing the pre-existing arteries into close juxta- 

 position with the pre-existing coelomic tubules, to form the 

 coelomate glomerulus which, as a lobulated tuft of capillaries, 

 still hangs free in the pericardial cavity of some of the lower 

 vertebrates. Later a direct connection was effected between 

 arteries and tubules outside the coelomic cavity, to form the 

 typical glomerulus as found in mesonephros and metanephros 

 of the higher animals. But in many recent fishes and Am- 

 phibia, the mesonephric tubules still retain their ancient con- 

 nection with the body cavity. The essential point is that the 

 renal glomerulus was evolved independently of, and long 

 after the evolution of the renal tubule. And it will be re- 

 called that in the ontogenetic development of the human 

 embryo the glomerulus is not brought into conjunction and 

 connected with the tubule of the metanephros until some 

 time after this tubule has been formed; it is possible that this 

 interval between the development of the tubule and the 

 glomerulus is an ontogenetic recapitulation of the phyloge- 

 netic interval which separated their evolution some four or 

 five hundred million years ago. 



But the very nature of a high-pressure filtration system 

 permits not only water to be pumped out of the body, but 

 also most of the osmotically active constituents of the plasma, 

 which means all the valuable constituents except the proteins 



