( 449 ) 



have a far greater number of well-marked bands, which have nothing to do witli 

 " Rieselzeichunng." 



Bat of much greater importance than the number of bands is the question, 

 whether the ancestral pattern did really consist, as maintained by Eimer, of con- 

 tinnons bands. If this jjoint were demonstrated in Elmer's books by convincing 

 evidence, if he had shown that the banded forms of a group of allied species were 

 the phyletically older, the spotted and streaked forms the phyletically youuger ones 

 in all groups of Lepidoptera, nay, even only in Butterflies, this result would be 

 worthy of the highest comment, and far outweigh all the mistakes in the special 

 classification of the species. 



I find five arguments brought forward in support of that contention, namely: — 



(1) The streaked, spotted, and unicolorous wings are derivable from the banded 

 wing. — -Yes ; but exactly as the presence of spots is explainable by assuming that 

 bands were broken up into spots, the jiresence of bands can be explained by assuming 

 that spots had fused to bands ; and the same can be said of the development of spots 

 from streaks, and of streaks from spots. The question is, have we to conclude that 

 tlie line of development was from bands to spots to streaks, as Eimer maintains for 

 Lepidoptera, or from streaks to spots to bands, as Escherich says of Beetles ? or was 

 the spotted wing the original from which the banded wing developed in one, the 

 streaked wing in another direction ? All three possibilities would equally well 

 exjjlain that there is a connection between the banded, spotted, streaked wings of 

 different species. 



(2) The series of allied forms put together in each group, says Eimer, 

 demonstrate the road Evolution has taken in evolving one from the other, and give 

 as strong evidence for the bands being the ancestral pattern, as the facts of 

 Palaeontology furnish evidence for other conclusions in Evolution. — That the 

 phyletic connection of the forms of Papilios as accepted by Eimer is to a large 

 extent erroneous we have shown above ; but let us assume that in Arth'ddunq the 

 roads Evolution had taken were demonstrated, only for the sake of argument. If we 

 thus know that there is a connection from one species to the other in a group of near 

 relatives, the series of forms .representing the road Evolution has taken, we have a 

 road that leads both ways, from bands to spots and from spots to bands, and the 

 jiroof of there being such a road does not provide us with the knowledge of the 

 direction in which Evolution has traversed it, does not give an answer to the question, 

 which stejis in the mutation of the pattern are the youuger, which the phyletically 

 older ones, and hence there is no justification for a comparison with the facts of 

 Palaeontology that do give an answer to that question. 



(3) It has been shown in Artbildung, says Eimer, how minute characters 

 appear in single individuals, increase in other examples, become more fixed, and 

 appear as the characters of varieties and species, developing further in allied species, 

 and thus form a connection between series of species, and such mutations have been 

 demonstrated from the banded to the unicolorous wing, so that consecjuently the 

 bands must represent the ancestral pattern. — 1 have not found an instance in 

 Artbildung I. where it is shown that a banded wing develops into a .sijotted wing 

 and then becomes unicolorous. The banded forewing of Cosmodesmus becomes uni- 

 colorous by obliteration and fusion of bands, and by a sudden change of the ground- 

 colour into black, as demonstrated in Artbilduny I. AVhere do the spots come in ? 

 Further, the same series of species which demonstrate, according to Artbildung, 

 the progressive development of certain characters A, demonstrate also, according to 



