43 



especially interesting to the Conchologist, because by collecting 

 snoh cases he may occasionally procure species of fresh-water 

 shells that he has searched for in vain ; the Caddis-worm is a 

 better collector than he ! It has occurred to me that the cases 

 formed of seeds may play a part in the distribution of water 

 plants hitherto unsuspected. Caddis-worms no doubt form part 

 of the food of many aquatic birds, as also of those of the snipe 

 family, &c. ; it wonld be impossible for these to pick the larvae 

 cat of their cases, and these latter are commonly swallowed with 

 the inmates ; the hard seeds and other substances would probably 

 be passed in an undigested condition by the birds, when perhaps 

 many miles from the locality where they were swallowed ; hence 

 plants may be thus distributed. This suggestion requires 

 working up from actual observation, but at present I think there 

 is some force in it. A very common foi-m of these tubular cases 

 is that in which stoney fragments or sand is exclusively used ; 

 these are usually curved, and taper a good deal to the tail-end, 

 resembling shells of the genus Dentalium in form. Such cases 

 differing only in size, amount of curvature, coarseness or fineness 

 of materials, &c., are constructed by a vast number of different 

 species and even of genera. Very rarely {Setodes, t&c.) the case is 

 formed wholly of hardened silky secretion, with no covering of 

 extraneous matters. One of the most wonderful of British 

 forms of case is that constructed by the genus Molanna. It 

 may be found in parts of the Eavensbourne, where the bottom is 

 sandy, and in ponds on Hayes Common, &c. These cases are 

 wholly made of fine sand ; internally there is the usual tube, 

 but the sides are broadly expanded, so that the form is long-oval, 

 the upper side slightly convex, the lower slightly concave ; the 

 larva lives on the sandy bottom, from which the case is scarcely 

 distinguishable unless the inmate move, and an additional pecu- 

 liarity is that the upper or convex side is produced at the mouth- 

 end far beyond the lower side, so as to form a covering for the 

 larva when feeding at the sandy bottom. A still more remarkable 

 instance occurs in a small case from Ceylon, in which the actual 

 case is merely an ordinary straight tube, but to the mouth-end is 

 affixed a circular shield, concave beneath, and equally protecting 

 the inmate when its anterior segments are extended in search of 

 food. Some cases {Goera, dr.) are nearly oblong in shape, con- 

 sisting of a tube formed of fine gravel, to the sides of which are 



