1872.] . ■ 81 



pale yellow is changed into red {Chri/sopn). The hypodermal colors 

 can be changed even by a voluntary act of the animal, and the new 

 colors disappear again (Cassida). The hypodermal colours are the 

 only ones on which the animal has any influence either involuntarily 

 by the action of the nutritive fluid or voluntarily. The epidermal 

 cells are placed entirely outside of any influences of the animal, when 

 once established. It will perhaps be possible to prove that the so- 

 called mimetic colors are all hypodermal colors. 



The hypodermal colors seem to be produced by a photographic 

 process (I know no better expression), the epidermal colors by a 

 chemical process of combustion or oxidation. Would it be possible to 

 prove that by a photographic process even the colors of the surround- 

 ing world could be transmitted, a great step towards an undertaking 

 of the phenomena would be given. The fact of course is very proba- 

 ble, at least in some instances. 



In observing the mimicry, the pattern of an insect must be 

 clearly separated from the color. In fact the pattern is not the 

 product of an accidental circumstance, but apparently the product 

 of a certain law, or rather the consequence of certain actions or 

 events in the interior of the animal and in its development. The 

 proof is very easily afforded by the regularity of the pattern in a 

 genus, or a family of insects. If studied carefully and comparatively 

 the pattern in a genus is the same or is only more or less elaborated. 

 The number of such families is so exceedingly great that some example 

 wall readily occur to every one. 



Moreover, a certain and constant pattern can be found for the 

 head, a different pattern for the segments of the thorax, and a 

 different pattern for the segments of the abdomen. This pattern is 

 in the different segments of the abdomen {Hymenopiera, Dipfera, 

 Neuroptera, Orthoptera) always the same, only more or less elaborated, 

 and less finished in the first and last segments. In some way, the 

 same is true for the thoracic segments. 



In some few instances, I was able to observe how the pattern is 

 produced. In the Odonnta (Dragon flies), at the moment of trans- 

 formation, the thorax is transparent, and shows no colors at all. At 

 this time, the muscles are without importance, and in process of forma- 

 tion. The thoracic muscles, as is well known, are, in the Odonata, very 

 powerful, and also very extraordinary as regards the shape of their 

 .tendons. Just along outside the muscles are dark lines more or less 

 well finished, and resulting from the action of the mia spoi; Jlhi^ 

 irritafio, {hi affjuxns. I believe that it would not be un at there is a 



