1873.] COLOURING IN INSECTS. 155 



limits, the species presenting a corresponding range of variation in 

 its disguising characters. 



In illustration of this class I may mention the " leaf-butterflies " 

 (Kallima paralekta and K. inachis), as also some of the true " leaf- 

 insects" (Phasmida). The objects here imitated are leaves in 

 various stages of decay ; and as these are of variable hues, the insects 

 present varieties of corresponding shades of colour. 



III. Cases in which the imitated object undergoes a change of 

 character once during the lifetime of the individual, whose disguise 

 changes in correspondence. 



In illustration of this class I will name the larvae of two of our 

 native Geometrse, viz. Geometra papilionaria and Acidalia dege- 

 nerative, the former of which is dull purplish before hibernation, 

 but becomes green in the spring ; while the latter, from the period 

 of its emergence from the egg in July to the middle of September, 

 is of a greenish-brown colour, but changes to a rusty brown at this 

 period preparatory to hibernation *. 



It is obvious that a change of habit in the species entailing a new 

 relationship with the environment is equivalent to a change of cha- 

 racter in the environment itself. In this class therefore I include 

 such larvae as those of Thecla betulce, the genera Smerinthus and 

 Sphinx, Macaria alternate, &c, which are green when feeding on 

 their respective food-plants, but become brown previous to pupation, 

 at which period the insects require to crawl over the ground to find 

 a suitable burying-place. 



IV. Cases in which the colour of the imitated object undergoes 

 periodic change, the protective characters of the species changing in 

 correspondence. 



This class includes those cases in which, like the Alpine Hare, 

 the Ermine, and the Ptarmigan, the disguising characters change 

 with the season. 



In concluding this classification I would call attention to the 

 manner in which the characters of the imitated object and the dis- 

 guising characters of the species vary together. The truth thus 

 brought to light I hope to fully investigate in a future paper. 



Variable protective colouring. 



I have already had occasion to mention that there existed a cer- 

 tain number of cases that could not be included in any class of the 

 above arrangement ; it is to these residual instances that I would 

 wish to direct the attention of naturalists in the present paper, as 

 their study offers a wide and interesting field for observation. The 

 particular class of cases that I propose to include under the term 

 " variable protective colouring " will be best understood from the 

 following examples which I have collected from various sources. 



First, with reference to insects in the larval state. Fabricius 



* See the recently published description of this larva by Mr. William Buckler, 

 of Emsworth, 'Entom. Monthly Mag.' Oct. 1872, p. 115. A somewhat similar 

 change of colour is recorded in the larva of Gnophos obscurata by Mr. Hellins, 

 I. c. June 1871, p. 20. 



