upon lepidopterous lai'vce, cOc. 



293 



anything which suggested a deep furrow between them 

 would destroy the protective resemblance. At one point 

 only in the body of a Geometer larva is there any 

 necessity for maintaining apparent continuity with the 

 food -plant, and at this point only in the larvas which I 

 have examined fleshy tubercles like those described 

 above (in Catocala, &c), are developed. In fig. 1 the 



Fig. 1. 



posterior part of the body of the larva of Rumia 

 cratcegata is shown (x 9 diameters), and the fleshy 

 processes are seen to exactly correspond to that part of 

 the body which would otherwise cause a dark shadow in 

 the deep cleft between itself and the branch. The effect 

 of the fleshy processes is diagrammatically shown in 

 fig. 2, in which («) represents a section across two 

 branches immediately below the point at which they 

 have divided, (b) represents a section across the larval 

 body between the two posterior pairs of claspers, 

 showing the formation of a deep cleft, while (c) shows 



