282 THE entojiolo(;ist's kecukd. 



the connections and distinctions between different species increases ? 

 Siirely it is not illogical to treat the nomenclature of species in the 

 same way as that of genera, tribes and families, nor unscientific to 

 consider knowledge of the phylogeny and distribution of species of 

 more importance than cabinet labels. 



To this point I shall have to recur. Meanwhile, to make my 

 position clearer, and to answer the objections made to it, let me come 

 down to details. I heartily rejoice that it is in the power of so 

 insignificant a person as myself to supply so eminent a naturalist as 

 Mr. Prout with much of the information which he expresses himself 

 in the April number of the Jvnt. hecurd as l)eing anxious to acquire. 

 "Is it conceivable," he asks, " that any thoughtful entomologist has 

 really imagined a type form exists in nature apart from the subjectivity 

 of the individual student?" It is, of course, open to question whether 

 I am " thoughtful " or even an " entomologist," as neither of these 

 terms is defined (my critic, I notice, very rightly hungers for definitions) 

 in the paper in question ; but certainly, I not only imagine, but am 

 entirely confident, that a type form does exist in nature, probably in 

 every case, in the perfectly objective sense in which I use the word. 

 As to what this sense is, Mr. Prout has accused me of placing myself 

 " in a hopelessly unscientific position by not even attempting approxi- 

 mately to define " my " conception." I am very sorry to be obliged, 

 in self-defence, to point out that I hare explained on page 8 of my 

 Introduction, that I hold that the word " type " should be restricted 

 to the " most generally distributed form." I further explained that 

 I did not consider this to be ideal, as what Mr. Prout so excellently 

 calls the " phylogenetic type " is scientifically preferable ; but as all 

 the probabilities are obviously in favour of the most widely distributed 

 form being the most ancestral, and therefore the "phylogenetic type," 

 as well as the most numerous, and therefore the "numerical type," 

 and consequently also the " average type," it seems to me that my 

 original definition of the " type " as " the most generally distributed 

 form " was less wide of the ideal than I had myself imagined. With 

 regard to applying tests to wing-markings only, it would, of course, be 

 both illogical and unscientific to do so, but in most cases more 

 fundamental differences would involve specific distinction. Incidentally 

 my definition supplies an answer to another of Mr. Prout's criticisms. 

 Taking the " type " to be the " most generally distributed form," I do 

 not consider it to be in any way " a miracle of good luck " that in the 

 vast majority of cases the species have " happened to be first described 

 from specimens conforming to" my "ideas of a type"; on the 

 contrary, I should regard it as a miracle of ill-luck, not to say a 

 monument of premature definition, if the first described forms had 

 not been almost always in my sense typical. Further, if such a 

 miracle of ill-luck /lad occurred, I should certainly have l^een " logical 

 enough to inundate " him " with changes " without the slightest 

 compunction, a contingency which it is happily unnecessary to con- 

 template. Perhaps, by the way, Mr. Prout would take pity on my 

 stupidity and explain what precise meaning the expression " type of 

 a nmiic'' is intended to convey; I have puzzled my head over it for 

 some time and cannot understand it; it evidently means sDiiiethiwi, 

 but the something eludes me. Another point on which Mr. Prout 

 much desires enlightenment is how a stable nomenclature can possibly 



