358 



The cause of this disparity between my conclusions and 

 those of the Countess von Linden may, as far as I can see, be 

 found in two circumstances : firstly, Dr. von Linden seems not 

 to have paid sufficient attention to the younger stages in 

 the wing-colour development ; and secondly, she was handicapped 

 by her conviction of the correctness of Eimer's hypothesis 

 about the primitive nature of longitudinal striping in different 

 classes of animals. 



Returning after so many years to the study of this interesting 

 subject, I have extended my investigations to the families of 

 Pieridce and Pafilionidœ-, and paid more attention to the ex- 

 ternal appearance of the pupal sheath. Taking the point of 

 view indicated by Poulton about the real nature of the chrysalis, 

 we may by its help get to a series of developmental stages of 

 the wings, at least of the forewings. With regard to the 

 succession in time of their appearance, they are the following : 



1. The uncoloured wing-rudiments in the full-grown cater- 

 pillar. 



2. The wing-sheaths of the newly formed pupa. 



3. The wing-sheaths of the fully coloured pupa. 



4. The uncoloured wing-rudiments within the young pupa. 



5. The newly coloured wings within the pupa, showing the 

 so-called primitive pattern. 



6. The different stages of the definite pattern. 



This succession in time need not, however, necessarily 

 correspond with the real morphological sequence of patterns 

 on the wings. 



On the contrary, the colour-pattern of the wing-sheath of 

 the chrysalis undoubtedly represents a farther advanced degree 

 of development than the primitive coloration of the wing- 

 rudiments within this sheath. We are thus, by studying the 

 colour-pattern, led to the same conclusion as Poulton arrived 

 at for the wing-size and form, and (using his own words) "for 

 the venation of the wing-sheath in comparison with the course 

 of the main tracheœ of the pupal wing, which will ultimately 

 be enclosed as important elements in the imaginai veins, and 

 which at this time possess an arrangement different from that 

 which they will then assume." This conclusion may be thus 

 formulated : that the ornamentation of the pupal wing-sheath 



