96 Mr Carpenter on Unity of Function 



rudiments of wings may usually be traced in the caterpillar, 

 assuming the form of lamina? of mucous tissue, and permeated by 

 tracheae;* and during the chrysalis state, their development 

 proceeds gradually towards the form which they ultimately as- 

 sume. Now, as the tracheae permeate not only the wings but 

 the whole bodies of insects, it is evident that this circumstance 

 of itself assists us little ; the development of the wings of some 

 aquatic insects, however, affords us more valuable testimony. 

 " As long as the insect dwells in the water, its rudimental 

 wings are true water gills ; but so soon as it has quitted the 

 water, they transform themselves into air gills; for, in both 

 cases, fluids circulate in their vessels, which doubtlessly receive 

 oxygen from the air." -f* 



To enter into all the arguments by which this position might 

 be supported, and to refute the objections which may be urged 

 against it, would lead me too far away from my present object ; 

 but I may observe, that it is only by taking an extensive view 

 of comparative structure that we can have any hope of arriving 

 at accurate results ; and great care is necessary to dismiss from 

 our minds all prejudice in favour of any particular mode of oi- 

 ganization as a standard or type of the rest. Let us suppose 

 an entomologist to form his views of the structure of animals in 

 general from that of the articulata ; he would expect to find 

 the wing of a bat or bird constructed on the model of that of 

 an insect ; and yet he would not be acting more absurdly in 

 maintaining that this organ is an appendage to the respiratory 

 system in vertebrated animals (especially considering its re- 

 markable connection with this system in birds) than many en- 

 tomologists in being led by their previous acquaintance with 

 other types of structure, to consider the wing of an insect as a 

 modification of its leg. 



In speaking of the separation existing between different 

 groups of organized beings, it is to be recollected that the mi- 

 nor variations from a particular type, whether that be of a 

 class, order, genus or species, are frequently of such a charac- 

 ter as to approximate some of its divisions to neighbouring 

 groups ; and that, sometimes, the minor or secondary character 

 may become so predominant as to leave us in doubt to which 



" Burmeister's Entomology, p. 426. t Ibid. 442. 



