upon leyidopterous larva, &c. 291 



worked out the arrangement of the primitive tubercles 

 very completely, and I only mention my own observa- 

 tions in connection with the subject as an entirely 

 independent confirmation of his conclusions. I think 

 it is extremely probable that the white spots occurring 

 in pairs upon the dorsal surface of the mature larvae of 

 Cluerocampa elpenor and C. porcellus (figured by Weis- 

 mann) represent one of the two pairs (probably the 

 posterior) of the dorsal primitive tubercles. If this be 

 the case they will be found, with the other pair, at 

 earlier stages of the ontogeny, and will be found to 

 assume the appearance of tubercles. Although the 

 traces of the primitive tubercles merely take the form 

 of white patches in Dr. Hickson's larva, yet there 

 is no other Sphinx larva hitherto described in which 

 the position of so large a part of the primitive 

 characters are distinctly indicated in the later stages. 

 The ontogeny of such a larva is probably extremely 

 uniform. 



3. FUETHER OBSERVATIONS ON THE PROTECTIVE ATTITUDE 



of Geometer larva : — a. The spiral or irregular position 

 assumed by young larvce resting on leaves.- — I am now able 

 to add observations of two species, in addition to those 

 previously described. The young brown larvae of Selenia 

 lunaria twist themselves into an irregular spiral when 

 seated on the leaves of their food-plant, and this attitude 

 is sometimes assumed by the mature larvae when resting 

 in such a position. The young brown larvae of Rumia 

 cratcegata also have the same habit, generally sitting on 

 the edge of a leaf close to the piece which they have 

 eaten out of it. In this position the larva suggests 

 most strongly the appearance of a small part of the 

 leaf which has been injured in some way, and has curled 

 up and turned brown, but still remains adherent by one 

 end to the uninjured part of the leaf. 



ft. The supporting thread sometimes dispensed with in 

 the twig-like attitude. — Although the almost invisible 

 thread is invariably made use of in ail the most perfect 

 and elaborate resemblances to twigs, I have observed 

 that in an effective and well-concealed form of this 

 protective appearance the larva (in the case of A. betu- 

 laria) is supported in another way, viz., by holding a 

 branch, leaf-stalk, or leaf in the clasp of its thoracic 



