462 PROCEEDINGS OP THE ACADEMY OF [1881. 



releasing the statoblasts on the first approach of winter, these are 

 provided Avith either a single row or a more complicated series of 

 marginal tentacular hooks, by wliich they become matted together, 

 entangled with roots, stems, etc., or held to rough places on planks 

 or stones. 



The same relation to the pei-manency of their skeleton structure 

 we find existing amongst these genera and species of fresh-water 

 sponges. The statosplieres of nearly all species are provided 

 with some arrangement for protection and retention. These vary 

 greatly in kind and degree, inversely according to the protection 

 afforded them by the surrounding skeleton. Perhaps the lowest 

 in the series in this regard is Meyenia Leidyi. This is a thin 

 encrusting sponge ; the skeleton spiculje stout and firmly matted 

 together ; maintaining the position and form of the mass through- 

 out the year. The statosplieres are formed in the autumn in the 

 lowest parts of the sponge, within special capsules formed by 

 interlacing spiculai. It is hardly possible these should wash away, 

 and accordingly we find no means provided peculiar to themselves 

 for detaining them. Their armor consists of a closely laid series 

 of birotulate spiculse with entire margins, excellent as a shield 

 but hopelessl}^ useless as a means of retention. On the other 

 hand no apparent means of diffusion are provided, and as a con- 

 sequence the species seems to be extremely local ; none having 

 been noticed except in the stream where the first specimen was 

 gathered, and within a few yards of the probable spot. 



Spongilla fragiliH of Leidy,when seen during the summer-time, 

 nearly resembles in form the above-mentioned species ; its skeleton 

 structure, however, is much more fragile and is frequently detached 

 and washed awa}', leaving a uniform series of statoblasts standing 

 side by side, with no special coating of spicules for each, as in 

 most other species, but grouped and held together by a common 

 coating of cellular or granular matter, covered by and imbedding 

 a great number of cylindrical spined spicules. A variety of this 

 is often observed (whether it differs specifically in other respects 

 he could not be certain) in which the statospheres are segregated 

 into groups of four or more, spherically enclosed in a similar 

 coating, thus appearing like one large seed. While the statoblasts 

 of the former arrangement retain their positions during the winter 

 and germinate there in the spring, it may be that this is a character 

 assumed for diftusive propagation. 



In Spongilla lacustris and similar branching sponges, the appar- 

 ently conflicting ends of retention and diffusion are attained in a 

 different way. The " seed " are formed in the interstices of both 

 the sessile and the branching portions. In the former they are 

 retained during the winter, partially by the agency of recurved 

 spines upon the acerates projecting from the seed coat; while the 

 fragile branches soon break off and float their contained stato- 

 spheres to distant parts. 



The massive sessile character of many sponges, repeated through 



