292 BIOLOGY: GENERAL AND MEDICAL 



ing great progress through the labors of Lamarck, who 

 founded the science of invertebrate paleontology, and 

 Cuvier who founded that of vertebrate paleontology, 

 and it was discovered that the "fossil remains" afforded 

 an insight into the nature and structure of creatures 

 that had long ago inhabited the earth, but became dis- 

 placed by now existing forms. Fossil animals, there- 

 fore, came to require a place in the genealogical tree, or 

 system of classification. 



Should it ever become possible to become acquainted 

 with all of the animals that have lived as well as all of 

 those that now live, and to place them in correct and 

 orderly position with reference to one another, the ar- 

 rangement would show the exact genealogy of every 

 group and its complete family tree. 



Such a family tree would also show that the different 

 groups of animals that we now know have not, as the 

 early systematists imagined, developed one into the other, 

 but that they are simply branches of the same great 

 family tree so that to get from one to the other one 

 would be obliged to descend one branch to some common 

 intermediate, or even go back to the main trunk and 

 ascend again in order to reach the new branch. This 

 is a most fundamental conception. The various animals 

 we now know are the newest buds and sprouts upon the 

 apical twigs of boughs, branches, and limbs of the tree 

 of life that has been growing and spreading ever since 

 life first made its appearance. 



The ambition of every systematist of the present time 

 is to so arrange the known living and extinct organisms 

 as to make them find their proper places in the evolution- 

 ary sequence. This is quite a different matter from that 

 of arranging them in the order of development one into 

 the other. 



How nearly we are in a position to complete the gene- 

 alogical tree may be judged by a hasty comparison of 

 the known living and fossil forms. Allowing a liberal 

 margin for error, it may be surmised that there are known 



