24 CURCDLIONID.^. 



development, and it is interesting to note that this has not led 

 to the reappearance of true legs, but that a quite different mecha- 

 nism has been evolved to enable tlie larva to maintain its position 

 and move about the surface of the plant. This is effected by 

 means of a viscous secretion which covers the whole body in a 

 very thin layer and within which the larva progresses by uii- 

 dulatory movements, often aided by the presence of setigerous 

 tubercles on the lower surface. A number of different observers 

 have definitely stated that this secretion, or a similar one used 

 by mauy genera for the lining of their cocoons, is emitted from 

 a retractile dorsal tubercle on the last segment just above the 

 anus ; but recently T. Knab (Proc. Ent. Soc. Washington, 1915, 

 p. 154) has contended that this statement is inaccurate and 

 that the substance is emitted from the anus itself. His view is 

 in agreement with the earlier observations made by Jj'abre on 

 Larinus maculosus, Schh., and L. itrsus, Y. (Souvenirs Entom. vii, 

 pp. 68, 8l) and on Gionus similis, Miill. (op. cit. x, p. 84). Eabre 

 considered that the secretion was derived from the Malpighian 

 tubes, and Knab agrees that this is probably the principal source, 

 but that the rectal glands also contribute to it, and possibly like- 

 wise tlie siirfa,ce glands distributed over the body. When these 

 external-feeding lai-vse are full-grown, they construct from this 

 secretion cocoons which are attached to the food-plant. This 

 work is done entirely with the mouth, the viscous material being 

 drawn out into silk-like threads. In the case of Hyj^era the 

 cocoons are elegant net-work structures of a white, yellow or 

 green colour, through which the pupa can be seen ; while those 

 of Gionus are bladder-like, and in some instances present a very 

 striking resemblance to the seed-capsules of the plants on which 

 they occur. 



The movements of coleopterous pupae are generally very limited, 

 but tliose of some species of Hypera are capable of twisting 

 themselves round and round in their cocoons with extraordinary 

 rapidity when disturbed. This is perhaps done for the purpose 

 of disconcerting a possible enemy, and is comparable with the 

 habit of certain spiders of rapidly vibrating their webs in similar 

 circumstances. The j)upa3 of some species oi Lixus are capable of 

 moving up and down the hollow stems in Avhich they lie by means 

 of their dorsal bristles. In this connexion it may be mentioned 

 that one of the so-called " jumpiiig seeds," which frequentl_y 

 attract attention, is due to the movements of a Curculionid larva, 

 Nanophijes tamnricis, Gyl., which lives in the seeds of a European 

 species of tamarisk. 



The external-feeding larvse of an i^merican weevil, Gcelor/aster 

 liiuratus, Dietz, are recorded by Knab {I.e. p. 155) as covering 

 themsehes \\ith their own excreta for purposes of concealment, 

 as is done by the larvae of the asparagus beetles (Grioceris). 

 Some species of the semi-aquatic genus Phytohius also show a 

 tendency in this direction. 



