OP ARTS AND SCIENCES. 251 



by monographers about certain groups or families, but that is all. I 

 believe that a more detailed study of pattern and of the diflerent 

 patterns which are to be found in different groups, and perhaps the 

 development of the law according to which the pattern is changed in 

 diiFerent groups, would advance us nearer to the knowledge of its 

 nature and origin. What I know about it is only the first step iu this 

 direction. 



The pattern is not the product of an accidental circumstance, but 

 apparently the consequence of certain events or actions in the interior 

 of the insect, mostly at the time of its development. The proof is 

 easily afforded by the regularity of the pattern in the same genus or 

 in the same family. If studied carefully and comparatively, the pattern 

 of such a genus is the same for all species, but for some of them more 

 or less elaborated. The number of such genera, and even of such 

 families having the same pattern, is so large, that some will be readily 

 remembered by every naturalist. In some families a peculiar and 

 constant pattern can be observed for the head, a different one for the 

 thorax and its limbs, and again another one for the segments of the 

 abdomen. The latter is on the different segments (Hyraenoptera, 

 Diptera, Neuroptera, Pseudoneuroptera) mostly the same, but more or 

 less elaborated, and less finished in the first and last segments. To a 

 certain extent the same can be said about the segments of the thorax. 



Weismann has studied carefully the origin of the pattern of cater- 

 pillars of the Sphingidaj. The caterpillars were chosen by him 

 because sexual selection is thus excluded. When just hatched they 

 are mostly colorless, but after a few hours they almost always become 

 uniformly green.* The pattern a2:)pears generally after the first 

 moult in the form of longitudinal bands or lines. There is a dorsal 

 line, and one ou each side a subdorsal one, and lower down one along 

 the stigmata. In Cha^rocampa the subdorsal line is changed on 

 two segments of the thorax into large eye-spots, and disajjpears 

 entirely on all the following segments. In Deilephila the same line 

 forms a ring-spot on the last segment, which in some sj^ecies is re- 

 peated in former segments. In Smerinthus oblique lines cross the 

 subdorsal line on each segment, and these oblique lines receive colored 

 borders in Sphinx and Acherontia. The caterpillars of Cha^rocampa 

 and Deilephila change later the green color to a darker one. 



* The young larva of Deilephila euphorbice is entirely black on leaving the 

 egg (Sclnvarz), or pale and becoming black after half an hour ("Weismann). 



