INTRODUCTION. 5 



It may be seen from this table, that there are four modes 

 at least in which the parts of an insect may vary ; and as 

 the great difficulty of making use of an organ in arrange- 

 ment must depend on the irregularity and confusion which 

 arise from the interference with each other of these several 

 modes of variation, it follows that those parts are the best 

 calculated to serve as the basis of classification which vary 

 in the least number of different ways. Hence we are led to 

 conclude, that the mentnm, oculi, and maxilla are the parts 

 of an insect which are the most to be attended to in our 

 endeavours to arrive at a natural system. But the first 

 and last of these are very much preferable, for the attain- 

 ment of our object, to the eye ; inasmuch as the variation 

 of the form of this organ in different species is by no means 

 so easily seized. 



If however I have laid peculiar stress on the modifi- 

 cations of the maxillae % I wish to be understood as by 

 no means undervaluing those characters which may be 

 drawn from the more obvious parts of an insect. Such 

 characters, indeed, once that the chain or order of na- 

 ture is discovered and established, are the most useful, 

 because the ordinary observer can by their means arrive 

 at the same conclusions with the anatomist, without 

 giving up the time and attention requisite for the dis- 

 section of the parts of the mouth. At the same time we 

 ought to be very careful in the use of artificial characters ; 

 and to recollect, that in natural history we have always 

 good reason for suspecting methods. Indeed, the interests 

 of science and that love of truth which every scientific 



a " Maxillam constantissimam irwenimus, vix in congeneribus aberrat." 

 Phil. Ent. p. 93. — " Maxillae et labium ejusdem speciei tunc constautissima 

 semper simillima." Ibid. p. 94. 



