after the Development of Swi,rm-larv(e. 341 



duction first perish, so that a successive dying-ofF is observed. 

 Hence it happens that we find sponges containing larvee or 

 gemmules the exterior of which may appear quite healthy, 

 while the inner parts are already in course of breaking down 

 by the development of the germ-materials, or ah-eady quite 

 destroyed. Gotte further explains those cases in which, in 

 spring and summer, we find perfectly developed gemmules, 

 together with ova or semen, by the supposition that in them 

 the germ-formation taking place in the preceding autumn was 

 prematurely interrupted, and the sponge liybernated with the 

 gemmules sticking in its soft parts. 



At a former meeting (21st December, 1886) I noticed the 

 freshwater sponges living in the Spree and in the Tegelsee, 

 and stated that EjpTiydatia fluviatilis living in the Tegelsee 

 never forms gemmules there, but is perennial. This sponge 

 is therefore a favourable object for deciding the question as to 

 the duration of the life of this species and testing the correct- 

 ness of the opinion of Laurent and Gotte of the death of the 

 ii^ongiTlce through sexual reproduction. 



During the last three winters I have kept large and small 

 («. e. down to 1 centim.) specimens of the above-mentioned 

 species from October to March, May and June, in aquaria, 

 some standing in heated, others in unwarmed rooms. A part 

 of them, and especially all the larger specimens, have always 

 perished in course of time. In many of the smaller ones, 

 however, the original size became considerably diminished, as 

 in the case of the larger specimens ; here also the outer mem- 

 brane became closely applied to the sponge-body or disap- 

 peared altogether ; here also the points of the bundles of 

 spicules projected more or less ; but the efferent canal only 

 disappeared in a part of them. The others almost constantly 

 showed an osculum upon each sponge ; the presence of an 

 external membrane closely applied to it was ascertained under 

 the microscope, and the current of water flowing tln-ough the 

 Spongilla was demonstrated. These little sponges lived 

 longest, but they also perished in May and June. 



The dying oft" of the Spongillce observed in the aquarium 

 appears, however, to occur by no means so frequently in 

 nature. Every one who has kept freshwater sponges in 

 aquaria knows how difficult it is to keep even small specimens 

 alive for a few months. On the other hand this is easily 

 done (see Lieberkiihn and Gotte), as I have also observed, 

 with young Bjjongilloe reared from larvie. It was only after 

 many attempts that I succeeded in realizing, at least approxi- 

 mately, the conditions which are necessary for the prosperity 

 of larger specimens. My failures in past years in attempting 



