FOOD OF INSECTS. 423 



vour even small birds; but this is wholly denied by 

 Langsdorf, who declares that it eats only insects a . This 

 species, as well as another tropical one, Thomisus ve- 

 natorius, the European Cteniza cementaria, and many 

 others, construct in the ground very singular cylindrical 

 cavities, and therein carry and devour their prey. These, 

 being rather the habitations of insects than snares, I 

 shall describe in a subsequent letter. Lycosa saccata, 

 the species whose affection for its young I have before 

 detailed, and not a few others of the same family, com- 

 mon in this country, in like manner seize their prey 

 openly, and when caught carry it to little inartificial ca- 

 vities under stones. Dolomedes Jimbriatus hunts along 

 the margins of pools ; and Lycosa piratica and its con- 

 geners not only chase their prey in the same situation, 

 but, venturing to skate upon the surface of the water 



itself, 



" . . . . bathe unwet their oily forms, and dwell 

 With feet repulsive on the dimpling well." 



The Rev. R. Sheppard has often noticed in the fen 

 ditches of Norfolk a very large spider which actually 

 forms a raft for the purpose of obtaining its prey with 

 more facility. Keeping its station upon a ball of weeds 

 about three inches in diameter, probably held together 

 by slight silken cords, it is wafted along the surface of 

 the water upon this floating island, which it quits the 

 moment it sees a drowning insect, not, as you may con- 

 ceive, for the sake of applying to it the process of the 

 Humane Society, but of hastening its exit by a more 

 speedy engine of destruction. The booty thus seized it 

 devours at leisure upon its raft, under which it retires 

 when alarmed by any danger. 



8 JSemerkungen auf einer lleise um die Welt. i. 63. 



