Some species of moths spin a very slight silken tissue for their cocoons, being apparently intended more to retain them from falling than to afford protection from other accidents. The gipsy-moth (_Hypogymna dispar_), rare in most parts of Britain, is one of these. It selects for its retreat a crack in the bark of the tree upon which it feeds, and over this spins only a few straggling threads. We found last summer (1829), in the hole of an elm-tree in the Park at Brussels, a group of half a dozen of these, that did not seem to have spun any covering at all, but trusted to a curtain of moss (_Hypna_) which margined the entrance. (J. R.) In a species nearly allied to this, the yellow-tussock (_Dasychira pudibunda_, Stephens), the cocoon, one of which we have now before us, is of a pretty close texture, and interwoven with the long hairs of the caterpillar itself (see figure _b_, page 17), which it plucks out piecemeal during the process of building,--as is also done by the vapourer (_Orgyia antiqua_, Hubner), and many others. [Illustration: Cocoon of Arctia villica.] [Illustration: Net-work cocoon.] These are additional instances of the remarks we formerly made, that caterpillars which spin a slight web are transformed into perfect insects in a much shorter period than those which spin more substantial ones. Thus the cream-spot tiger (_Arctia villica_, Stephens) lies in chrysalis only three weeks, and therefore does not require a strong web. It is figured above, along with another, which is still slighter, though more ingeniously woven, being regularly meshed like net-work. A very prettily-netted cocoon is constructed by the grub of a very small grey weevil (_Hypera rumicis_), which is not uncommon in July, on the seed spikes of docks (_Rumices_). This cocoon is globular, and not larger than a garden pea, though it appears to be very large in proportion to the pupa of the insect, reminding us not a little of the carved ivory balls from China. The meshes of the net-work are also large, but the materials are strong and of a waxy consistence. Upon remarking that no netting was ever spun over the part of the plant to which the cocoon was attached, we endeavoured to make them spin cocoons perfectly globular by detaching them when nearly finished; but though we tried four or five in this way, we could not make them add a single mesh after removal, all of them making their escape through the opening, and refusing to re-enter in order to complete their structure. (J. R.)