298 FIRST PRINCIPLES OP NATURAL CLASSIFICATION. 



this transportation occasioned by our ignorance of some 

 unknown law of variation in analogies, resulting from the 

 peculiar situation of the groups compared, in the higher 

 circles to which they belong ? Let us first try to verify 

 the former supposition, and see what results would attend 

 such a disposition of the analogies. 



Circle of the Tenuirostres. Circle of the Insessores. Circle of the Class Aves. 



Trochilidze. Tenuirostres. GRALLATORES. 



Cinnyridae. Fissirostres. NATATORES. 



Meliphagidaa. Scansores. RASORES. 



Paradisidas. Conirpstres. INSESSORES. 



Promeropidas. Denttrostres. RAPTORES. 



(362.) It might be perfectly easy to show analogies 

 (whether true or false is not now the question) between 

 the TrochilidcB and the Tenuirostres, the Meliphagides 

 and the Scansores, the Paradisidee and the Coni- 

 rostres ; but when we proceeded to the others, and en- 

 deavoured to make out what possible resemblance, even 

 the most remote, can be found between the Cinnyridce 

 and the Fissirostres, the Promeropidce and the Denti- 

 rostres, we must give up the whole theory of stating 

 analogies on these principles ; for only three out of five 

 wear the least appearance of truth, and even these violate 

 many other conclusions of a more certain nature. The 

 Melliphagidce, for instance, are related to the Scansores 

 by affinity and not by analogy., because they form the 

 immediate passage, or point of connection, between the 

 Tenuirostres and the climbing birds. 



(363.) Nor shall we get over this difficulty by stating 

 the question under another form, viz., by preserving 

 the natural series of the insessorial and the primary 

 circles, but transposing that of the Trochilidte, thus : 



PW./M< r"w/e Transposed Circle of True Circle of True Circle of the 

 Primary Circles. fhg TrocMid(B v thc Insessores . Orders. 



1. Typical. Trochilidae. Conirostres. INSESSORES. 



2. Sub-typical Cinnyridas. Dentirostres. RAPTORES. 



(" Promeropidae. Fissirostres. NATATORES. 



3. Aberrant. -< Paradisidze. Tenuirostres. GRALLATORES. 



(. Meliphagidaa. Scansores. RASORES. 



But here our difficulties, so far from being diminished, 

 are increased : in the first place, we make absolute 

 affinity subordinate to the purpose of preserving an ap- 

 pearance of regularity in our analogical series; for it can 



