224 rROCEEurxGS of the academy of [-^pril, 



their general anatomy (lack of lateral branches), not npon the 

 fact that in the one case they are nnicellnlar and in the other mul- 

 ticellnlar. In other words, in considering the homologies of organs 

 one shonld not look foT cell nuclei. The grosser differentiations of 

 an organ are of the first importance in phyletic classification. A 

 compound eye of an Arachnid is essentially a single lens in con- 

 junction with retinal cells arranged in retiuulse; and we characterize 

 it so in comparison with the compound eyes of insects or crusta- 

 ceans, and need not define it further in terms of finer structure of 

 rhabdomes of nerve endings. The fingers on the hand of a urodele 

 may be compared with those of a mammal, because of their essen- 

 tial similarity and corresponding relations to the fore-limb, though 

 in the one the skeletal rays are cartilaginous and in the other 

 osseous. The method should be in general exactly that by which 

 an anatomist examines any part — first, a careful examination of 

 the whole with the naked eye, then of a part with a low degree of 

 magnification, then of smaller parts with greater magnification — 

 but with this difference, that the more minute the examination 

 becomes the less general value the parts have for the purposes of 

 phylogeny. 



To summarize, all characteristics of an organ have value for the 

 phylogenist, except the more superficial ones of color and consist- 

 ency. Those characteristics which most completely represent the 

 organ as a whole are the ones of most importance, and minute 

 details of structure may lead to erroneous conclusions with regard 

 to homologies. Degree of conservatism is also a criterion of the 

 value of a characteristic. Broader homologies must be based upon 

 the relations to the rest of the organization, and upon the grosser 

 differentiations. 



VII. Comparison of the Methods of Embryology and 

 Comparative Anatomy in Determining Homologies. 



The organism is to be classified at the end of its development, 

 for reasons given in section III. But so far there has been no 

 discussion as to whether for determining homologies the organism 

 sliould be considered at this stage alone, or whether in previous 

 stages also. The question is, is the determination of homologies to be 

 made on the data of comparative anatomy or of comparative embry- 

 ology, or upon both ? In other words, are homologies to be drawn 



