Stephens — The Fresh-Boater Sponyes of Ireland. 233 



slightly above the surface. The .specimens are of all sizes, up to about 

 20 mm. in diameter. Those from the rivers tend to be rather larger, but 

 this is usually the ease with sponges taken in such a habitat (see p. '11'.)). In 

 this conne.\iou it may be mentioned that Dr. Aunandale (5) describes and 

 figures specimens of the Himalayan race of E. flvvialilis, which were growing 

 on stones in the form of fiat, circular films. A few of the Irish specimens 

 tend to be thicker, and are like little rounded cushions, while a number 

 of examples taken from the River Erne early in the year form small, smooth, 

 rounded masses growing on water-plants, and are rather soft to the touch. 



The oscula are small, but are rendered more conspicuous by the well- 

 marked, branching, sub-dermal canals which radiate from each osculuni. 



The sponges are bright green in colour when exposed to the light, and 

 greyish- white when shaded from it. In the latter case they resemble to a 

 remarkable degree the lake-form of lleteromeycnia Byderi, but, unlike that 

 species, they flourish on the limestone. The specimens are nearly always 

 densely crowded with enibrj'os. 



With regard to the structure of the skeleton, the main fibres run 

 vertically upwards from the base to the upper surface of the sponge, branching 

 once or twice in their course. They consist of multiserially arranged 

 megascleres, bound together by a small amount of spongin. The tips of the 

 terminal bundles of spicules project very slightly above the surface of the 

 sponge. The main fibres are united by single spicules, or by bundles of two 

 or more spicules lying at right angles to them. 



In specimens from various points along the shores of Lough Erne and of 

 Lough Gill, from Lough Feeagh, County Mayo, and Lough Derg, County 

 Tipperary, the megascleres are rather short, fairly thick, or sometimes very 

 thick, abruptly pointed oxea, which are microspined to the very tips. A few 

 among them are smooth. In some examples many of the oxea have a central 

 swelling; in others only a few possess it. In the sponges from Oakport 

 Lough, County Eoscommon, and Ballyscanlan Lough, Co. Waterford, for 

 example, the spicules are very similar to the foregoing in shape and size, but 

 are smooth. The spicules from the latter locality have often a central 

 swelling of the shaft, and are particularly like those described and figured 

 by Miiller (27) for a sponge which he regarded as a probable variety of 

 U.Jluviatilis, and which will be referred to more fully later on. 



Gemmule-spicules have been very carefully searched for in many prepara- 

 tions made from these sponges, but without success, except for one malformed 

 amphidisc found in one of the Ballyscanlan Lough specimens. 



The size of the megascleres varies somewhat from one specimen to 

 another; they are, on the whole, between O'lS-0'27 mm. in length. In some 



