238 GENERAL BIOLOGY 



Study 32. An attempt at interpreting a possible phytogeny, 



Materials needed : The completed drawings from study 

 30, with homologies fully determined and verified. 



Construct a genealogic tree for each of the three series, 

 that shall show a possible genetic relationship (based only 

 upon the data furnished by the venation of the figures). 

 Assume that the wing of figure 138 is primitive. Pick out 

 the form most like it to go near the foot of each tree. Single 

 out in each series the different ways in which the type has 

 been modified, and make as many principal branches as 

 there are different kinds of divergence. Pick out the most 

 specialized forms for the tips of the longest branches. 

 Arrange the others in position in accordance with their 

 degrees of divergence, and let the branching and the length 

 of the twigs represent this. Derive no form directly from 

 any other that is in any respect more specialized. Compare 

 all wings in each series together with respect to each charac- 

 ter, the divergence of the tips of the subcosta, the fusion of 

 the tips of the first fork of the media, etc. Remember that 

 each species is the end of Its own special line of development, 

 and place each at the end of a twig. 



The record of this study will consist in three genealogic 

 trees (which may be combined into one) , drawn without 

 any superfluous branches and with all the forms figured 

 (including Tipula, drawn) located thereon. 



It need, perhaps, be stated concerning genealogic trees, 

 that they generally err in being more explicit than the 

 known facts warrant. The figure of a tree does not present 

 a good likeness of evolution as it lies before us at the present 

 time, because the branches of the tree are conjoined in per- 

 fectly definite relations. Lines of development are in fact 

 traceable backward only a little way, and are then lost in ob- 

 scurity. The liverwort shown in figure 147 presents a truer 



