270 Ernst Martert. 



system is more a sort of feeling and guess-work than the 

 result of deep research. This is especially the case with the 

 Passeres. Here it is where work is most required, but it is 

 possible that anatomical studies will rather lead to negative 

 results than to positive ones, inasmuch as they may very likely 

 show that some of oui' so-called families are not separable 

 at all. Great care is necessary in 'reaching conclusions, and 

 here, too, the examination of many species and specimens is 

 necessary. There is, in my opinion, more individual varia- 

 tion than many people imagine, and two doubtful genera are 

 not good genera, because two species, one out of each — say, 

 Turdus musicus and Menda merlila — are distinguishable by 

 their skeletons. All the species of the genus should (if pos- 

 sible) be studied, and especially the doubtful ones. 



But I must conclude. Let me say in a few words that 

 much work is still to be done in all branches of Ornithology, 

 but that this work must be done more thoroughly, more con- 

 scientiously ; that it must be based on a wider basis than 

 usual ; that deductions should be made from as much material 

 as possible, and that we should most of all increase and 

 shape our knowledge of the various species and geograjmical 

 forms of birds which forms the most important basis of all 

 our work. A great help and step to this end is the con- 

 scientious study of entire genera or families, while the 

 scattered and often hasty descriptions of single new species 

 and subspecies are very often only a bother to one who 

 reviews, studies, and finally knows a certain genus or family. 

 On the whole, however, we can look confidently, hopefully 

 and well pleased on the present state of Ornithology, and we 

 only hope that the interest in it may continue to grow, and 

 that our work will increase in depth as far as it lays in our 

 power. 



