190 PLATYSTERA. 



species agreeing most perfectly in their structure, 

 except in a single point, i. e. the tail of one being 

 very long, and of the other very short. In a family, 

 or a sub-family, such a variation may be expected, 

 because every ornithologist knows that these two 

 forms of tail are always to be seen in good sized 

 groups; the flycatchers, for instance, and in the 

 thrushes and warblers : but such a decided differ- 

 ence has always been considered sufficient to con- 

 stitute a generic character, and in former years, we, 

 no less than the generality of mpdern systematists, 

 carried this belief so far as to make it often the 

 only character of a new genus. In this we have all 

 much erred, not, indeed, in making the distinctions, 

 but in giving them a higher value than they really 

 possess, a value, in short, which analysis would 

 have shewn they were not entitled to. Greater expe- 

 rience, however, has solved this difficulty. In the 

 first place, the long-tailed or rasorial form, very ge- 

 nerally passes into the short- tailed or natatorial; 

 and this, not by graduating steps, but as it were 

 abruptly ; that is to say, the difference of the tail 

 alone marks the passage. The crow and the mag- 

 pie, the blue and the long-tailed tit, are among the 

 most familiar instances of this sudden change ; not 

 more striking, indeed, than those we have here 

 exhibited in the sub-genera Todus and Platystera. 

 Now, there are two modes of accounting for this ; 

 the first is, that Nature preserves her distinction of 

 types even in the series of the species comprising a 

 sub-genus, when that sub-genus is full and circular; 



