146 Records of the Indian Museum. [Vol. X, 



The genus only differs from the subgenus Stratospongilla of the 

 genus Spongilla in being devoid, apparently in all circumstances, 

 of true microscleres. The skeleton of most species ^ resembles 

 that of Baikalospongia, except that the reticulation is never quite 

 so dense and the sponge is therefore even more fragile. B. inter- 

 media is, however, a connecting link in this respect. In all the 

 species of Nudospongilla I have examined both subdermal cavities 

 can be traced, but in one Syrian form {N . aster) the disposition of 

 the oscula somewhat resembles that characteristic of both Lubo- 

 mirskia and Baikalospongia. 



The genus M etschnikowia was described by Grimm from the 

 Caspian Sea. His paper, which is apparently in Russian and was 

 published in 1876 or 1877, is not available to me. Dybowski 

 {1880); pp. 52-59) has redescribed three species, as well as re- 

 defining the genus, from the same inland waters. Topsent * and 

 IvUndbeck^ refer to M etschnikowia Carter's Isodictya spinispiculum 

 and also the species originally described by Topsent himself as 

 Reniera fdholi ; both these sponges being found in the Atlantic. 

 But it does not appear that either author had had an opportunity 

 of examining material from the type-locality of M. tuberculata , 

 the tj'pe-species of the genus, and I would prefer to compare 

 Caspian specimens with true marine ones before expressing an 

 opinion on this point. Dybowski's figures of the skeleton of M . 

 tuherculata and M. flava (1879 ; pi. iii,figs. 5 and 6) are detailed and 

 clear, but I have seen no similar figures of that of the marine species 

 placed in the genus by the two authors just named. In any case, 

 Dybowski's figures show that there is a somewhat thin and irregu- 

 lar fibre sheath present in the sponges he illustrated, and that this 

 sheath is strictly comparable to that of some species of Reniera. 

 His figure of M. fiava {op. cit. pi. i, fig. 8) proves the existence 

 in that species of well-defined subdermal exhalent channels. 



On the whole, keeping in view the close similarity between 

 some species of Stratospongilla and some of Nudospongilla, and 

 also the biological conditions in which those of the latter genus 

 are found, I am inclined to regard the indubitable resemblance 

 between Nudospongilla and Baikalospongia as due to convergence, 

 but to accept as probable the view that Baikalospongia is closely 

 related to M etschnikowia. Until, however, the larval history of 

 the different species assigned to all these genera is more fully 

 known, it is impossible to express with any confidence a dogmatic 

 opinion as to their mutual relationships. 



Of the nominal species assigned to Baikalospongia on a pre- 

 ceding page I have examined only two, B. hacillifera and B. inter- 



' Evans, Quart. Journ. Micro. Sci. XLI, p. 425, pi. 38, figs. 6-8 (1899) has, 

 however, described a well-defined fibre-sheath in one species (A'', moorei) which 

 I assign provisionally to Nudospongilla. The systematic position of this sponge 

 is problematical. 



2 Mem. Soc. zool. France XI, p. 226 (1898) and Res. Camp. Sci. Monaco 

 XXV (Sponge. A9ores), p. 243 (1904). 



3 ■ Ingolf' Exp. \l (1), p. 52 (1902). 



