11 



pigment has taken place here and this accounts for the narrow yellow ring again 

 surrounding this black. For the black pigment, which there as elsewhere in 

 the wings, had first covered the original colour, has, under the said stress, 

 thence pressed towards the centre, where consequently, it has increased most 

 and has thus produced the intense black colour but at the same time it has 

 retired from the place where it occurred first and in this manner has caused 

 the original colour — now faded to yellow — which formerly it covered, to 

 reappear as a narrow ring. 



Conceived in this fashion the colour arrangement of this ocellus is easily 

 understood without having recourse to chemical action to which Bateson draws 

 attention. Providing, of course, one is acquainted with the phenomenon of 

 colour evolution and has been taught to interpret the distribution of the black 

 pigment over the original, or subsequently faded, pigmental colours in a manner, 

 which in fact reminds us of the flowing forward of a liquid, as the expression 

 of the stress of this evolution. 



Since this is the case with this ocellus it seems reasonable, as regards other 

 true ocelli, to explain the cause of their origin in a similar manner even though 

 the course of the process of colour evolution in every species, or at least in 

 every genus, may not be so clearly traceable and adventitions circumstances, 

 such as the appearance of interference colours, may impede its investigation. 



Other ocelli exist which must also be classed with the true ones but which, 

 in part at least, have presumably originated in a different way. I refer to 

 those occurring in many caterpillars of the Sphingidae, and which have played 

 an important role in the fantastic theory of mimicry. As far as I have been 

 able to find out they sometimes start from a colour spot on each segment, 

 the remnants of the longitudinal lateral stripes occurring in earlier stages of 

 development and which subsequently either disappear entirely or at least strongly 

 diminish ; these remnants appear to assume a round shape during the growth 

 of the caterpillar and in this manner form large round ocelli on such articu- 

 lations which increase inordinately in size. The course of this development, 

 however, requires further and special investigation. 



The ocelli, first referred to, are of great importance for a correct judgment 

 on the question of the alleged seasonal forms. For it is asserted that these 

 markings only occur in the wet season while in the dry-season forms of the 

 same species they are absent, and that consequently the forms of the dry and 

 the wet season can be distinguished by the absence or presence of the ocelli. 

 It is true this assertion is based upon a certain amount of correct observation ; 

 but this has become the subject of much wild speculation so that in the end 

 much misconception has arisen. Much stress has frequently been laid upon 



