176 . [January, 



separate and sometimes stalked ; let me quote two families in which 

 they are never separate. Of the GEcophorklce I have myself examined 

 800 species, and probably about 8000 specimens ; of the Gelechiadce 

 about 500 species, and 3000 specimens ; the whole at present classed 

 in about 150 genera : in this entire number I have not found even a 

 single specimen in w r hich these veins are separate. And it should be 

 observed that in so large a number there might well have been two or 

 three exceptional instances, of the nature of monstrosities, which 

 would yet not have invalidated the importance of the character. Here, 

 then, is absolute fixity of a particular neural structure, extending 

 throughout a very great number of species, and therefore applicable 

 as a family character ; yet I should never for a moment suppose that 

 this mass of evidence afforded the slightest indication as to the per- 

 manence or value of the same character in Stathmopoda, for instance. 



The questions on the elementary principles of classification in the 

 latter part of Lord Walsingham's paper are exceedingly simple, and 

 I have fully explained them elsewhere ; but as I suppose they are 

 meant to be answered, and as I probably do the majority of British 

 Lepidopterists no injustice in imagining that any information whatever 

 on neural structure is new to them, I will give a brief exposition. 



(1). Where two forms of structure occur in the same species, are we to infer 

 progressive development ? No : we are to infer variability only ; variability is a 

 necessary condition of development, but in no way involves it. If, by continued 

 observation, we found that one form was becoming scarcer and the other more 

 numerous, we might infer development. 



(2). When development takes place, which is the older or more ancestral form ? 

 This may sometimes be indeterminable ; but it is often possible to decide by a con- 

 sideration of allied forms, and a recognition of the principles (a) that a lost organ 

 cannot re-appear, (b) that a rudimentary organ is rarely re-developed, (c) that a new 

 organ is never spontaneously evolved out of nothing, but is a modification of some- 

 thing previously existing. To the third of these principles an exception must be 

 made in the case of monstrosities ; but if monstrosities are ever regularly reproduced 

 under natural circumstances, the probability of its having happened in any given 

 case is so small, that it may be practically neglected. 



(3). Does the multiplication, or the diminution of veins constitute improve- 

 ment ? According to the third principle, multiplication of veins can never take 

 place. 



(4). When the apical vein is forked, does it result from the junction of two 

 original veins, or the branching of one ? It follows from the third principle that a 

 simple vein cannot emit a branch. This may also be very simply proved by induc- 

 tion ; for it will be found in all cases that when a vein is branched, the simple 'vein 

 which should come next to it is missing. Thus, in the case of Cerostoma-, it is only 

 necessary to count the veins on the wing margin to discover that the number is the 



