196 



INTRODUCTION. 



of deciduous trees ; 'in such cases she partly cuts through the 

 stem, so that the fruit or shoot falls at about the time that 

 the larva is full grown aud ready to undergo its further trans- 

 formations, which take place underground. 



A large number of Curculionid larva? change to pupae under the 

 earth, but this is by no means always the case ; the species 

 belonging to the very large and universally distributed genus 

 Apion, for instance, as far as is known, undergo all their trans- 

 formations in the flower-heads, seed-vessels, pods, leaf-stems or 

 stalks in which they were originally hatched. The Cossonin2e 

 are wood-feeders, and are of interest in that two- thirds of the 

 described species belong to insular faunas ; this was especially 

 brought out by Mr. Wollaston who described one genus Mycro- 

 ceylobius, containing thirteen species, all peculiar to St. Helena, 

 and considered them to be the archaic remnants of an ancient 

 fauna in that detached island. 



A few genera, e. g., Bagous, Eubrychius, Litodactylus, etc., are 

 subaquatic, and the species of the two latter genera swim rapidly 

 with their hind legs like a frog or a Dytiscid. 



Although the members of the family, however different, are, 

 almost without exception, easily identified as belonging to it, yet 

 there is an enormous diversity of form and sculpture. Perhaps 

 some of the most extraordinary forms are found among the 

 Attelabinje and their allies, some of which are armed with long- 

 thorns or spines at the sides. 



Certain genera possess an extraordinarily hard integument which 

 is calculated to protect them effectuall)' against enemies ; this is 

 notably the case with the genus Brachycerus. I have made experi- 

 ments with dry and hollow specimens of a comparatively small 

 species of this or a closely allied genus, and have found that when 

 laid upon a board with another board above they would bear 

 a weight of more than a quarter of a hundredweight without 

 giving or breaking ; on ordinary ground and in a living state they 

 would, of course, bear much more. 



The clothing of the upper surface varies a great deal. The 

 surface is often quite bare, but it is the rule rather than the 

 exception for certain parts of the upper and under surface, if not 

 all, to be covered with scales of varied shape and proportions ; 

 some of these are very brilliant, and render their possessors very 

 striking objects ; larger or smaller hairs or setae are often 

 present. 



We have already alluded to the fact of the jaws being situated 

 at the end of the rostrum ; as a rule their motion is horizontal, 

 but in the case of Bcdaninus it is vertical. The character is so 

 peculiar that it might with reason be held to constitute the 

 Balanininje a separate division or subfamily. 



The difficulty of making any satisfactory arrangement of the 

 Cukculionidvi; may be gathered from the fact that Lacordaire 

 considers it to consist of no less than eighty-one tribes, most of 

 which are divided into groups of varying number. In his- 



