iQii.] N. AnnandaIvE: Sponges and Vcrnidid Molluscs. 49 



Spiroglyphus are closely allied groups of species which many 

 authorities recognize only as subgenera. 



The sponges associated with these shells must be regarded as 

 varieties of a species originally described by Carter from the Gulf 

 of Manaar under the name Discodennia sccpirellifera and now as- 

 signed to the genus Racodiscitla . Those that help to form the 

 Siliquaria-masses are hereafter described as the types of the new 

 variety siliquariae of the species, those associated with the Spiro- 

 glyphus as the types of the new variety spiroglyphi. The two 

 varieties differ not only in colour and compactness but also as 

 regards certain details of spiculation. 



The Spiroglyphus-masses must have when fresh a very gorgeous 

 appearance, the sponge being red or orange, the shells of a deli- 

 cate pink and the body of the molluscs yellow. The largest of 

 the specimens (pi. viii, fig. 2), of which a considerable number 

 were obtained, measures about 47-4 cm. by 38-6 cm. by 18 cm. (the 

 last measurement being that of the depth), weighs (dry) nearly 

 22 lb. and has a regular oval or wreath-shaped form (pi. viii, fig. 2). 

 The smallest, which is very irregular in outline, measures 17-5 cm. 

 ^y I4'5 cm by y^ cm. In the centre of each mass there is usually 

 either a concavity at the base or an empty space extending from 

 the base through to the upper surface. This concavity probably 

 represents that formerly occupied by the object to which the 

 Spiroglyphus shells originally attached themselves on quitting their 

 active larval life — in one specimen this object remains in the form 

 of an oyster-shell, in another in that of the shell of a gastropod ; 

 for there is evidence that molluscan shells when not occupied by a 

 living animal dissolve rapidly in the Bay of Bengal, even in 

 shallow water. -'Dead" shells (to use a conchologist's phrase) 

 from the Bay are usually very " dead," their surface being much 

 corroded ; and immediately north of the area on which the 

 Spiroglyphus-masses occur, the bottom of the sea is coated with a 

 recent conglomerate formed of partially dissolved shells consoli- 

 dated with sand-grains into a layer of stony hardness. Most of the 

 Spiroglyphus-masses seem from their regular outline to be complete 

 in themselves, but some of the smaller specimens appear to have 

 once formed parts of wreath-shaped masses which have been broken 

 by some accident and have repaired the actual fracture in the 

 course of growth The collection offers no proof of the exist- 

 ence of masses larger than the larger one of which the measure- 

 ments are given. 



The Siliquaria-masses are not so numerous as the Spirogly- 

 phus-m.asses and must have been much less conspicuous objects 

 when fresh. None of our specimens appear to be complete, for all 

 are irregular in form and show evidence of being merely fragments 

 of larger masses. The largest fragments obtained are about half 

 the size of the largest Spiroglyphus-masses. While the Spiroglyphus 

 shells radiate more or less distinctly from a common centre, run 

 mainly in a horizontal direction and have a very open and irre- 

 gular spiral, those of Siliquaria cochlearis are coiled in almost 



