Handbuch der Systematischen Ornithologie 31 



and especially of migratory animals, is made up of annual 

 cycles, as is the life of plants, which have their fixed and 

 determinate seasons for flowering and fruiting. This is the 

 key to the impulse of the spring migration, of which the fall 

 migration is but the necessary complement, inasmuch as in 

 most instances the winter conditions of the breeding grounds 

 of most species are prohibitive of their continued residence 

 therein throughout the year. " {" Auk, ' ' XXV, 1908, pp. 332- 

 333.) These facts and conclusions are so correct and final 

 that no other theory is necessary. As the ' ' Auk ' ' is numbered 

 among the journals used by Dr. Reichenow, we fail to under- 

 stand why he overlooked these investigations of ]Mr. Chapman. 

 In fact, I do not believe that he overlooked them, but it seems 

 impossible to convince any of the European savants that any- 

 thing good can come out of America. It is high time that a 

 good many of them should have their eyes opened to the fact 

 that the "uneducated Americans" are doing a goodly piece of 

 the world's scientific work, but from personal experience I can 

 say that they die hard. Several pages are devoted to the 

 fossil birds and the classification of birds according to Fiir- 

 bringer is quoted in full. The writer then proceeds to give 

 his own system. He says that a system based upon the inter- 

 nal organs has a high value, but that the internal organs are 

 just as much subject to changes as the external parts through 

 the conditions of living, food and motion. The author says 

 that such genealogical rows as Fiirbringer's have a high 

 value to give further investigations the right direction, but 

 can not ser\e as systems which have the practical value to 

 give a clear perspective of the masses of forms so as to learn 

 to know the manifold forms. For this there is needed a 

 "logical system" based on a few apparent characteristics. 

 The genealogical representation, which should teach how the 

 various forms have developed out of one another, presupposes 

 the knowledge of the separate individual forms, while the 

 system should first teach us the knowledge of these forms. 

 In a practical system the principal point is to limit the coordi- 

 nate groups as much as possible in regard to number, and 

 rather to create subordinate categories and in a logical way to 



