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



The affinities of Lubomirskia are, in my opinion, with Pachy 

 chalina, Schmidt, from which the genus differs in its spiny 

 spicules and in the peculiar structure of the terminal part of the 

 vertical fibres, and, with Veluspa, Miklucho-Maclay, which has 

 smooth tylote spicules. The structure of the skeleton- fibre 

 appears to have been misunderstood by Dybowski and by most 

 subsequent writers owing to the facts that the section figured by 

 him (1880 ; pi. II, fig. 5) was too thin to show the real structure, and 

 that the precaution of staining preparations of this genus with 

 some reagent that would display the chitinoid sheath of the fibre 

 has not hitherto been adopted. 



The method I have m^^self used in making the preparations 

 of L. haicalensis figured on plate ix is a very simple one. After 

 cutting a thick hand-section of the dried sponge I dissected out a 

 few fibres with their attachments under a binocular microscope 

 and washed them in running water, brushing them at intervals 

 with a camel' s-hair brush, until the cellular debris was removed. 

 I then placed them for about ten minutes in a strong aqueous 

 solution of pyrogallic acid. This solution of course stained both 

 the sheath and any remains of cells that still adhered to it, but 

 the former were easily distinguished b}^ their apparently granular 

 nature and removed b}' further brushing in water. This method 

 is naturally applicable only to skeleton-fibres that have a definite 

 horny sheath. 



It will be noted that in fig. la on plate ix that the horny or 

 chitinoid substance is deposited in the interstices between the 

 smaller twigs of the fibres in concentric layers and that these 

 interstices are often almost completely filled up in this manner. 



Subfamily RENIERINAE. 

 Baikalospongia, gen. nov. 



Sponge massive or encrusting, resembling Lubomirskia in 

 general structure but friable (though hard) and not at all elastic. 

 A stout basal membrane of a horny nature is present. 



Skeleton superficially resembling that of Lubomirskia, except 

 that there is no horny sheath to the fibres and that the vertical fibres do 

 not form definite brush-like tufts at their distal extremity but are more 

 or less distinctly splayed out to form a horizontal skeletal reticulation 



Spicules. — There are no true microscleres . The skeleton-spicules 

 as a rule resemble those of Lubomirskia, but the spines at their 

 extremities are usually differentiated more distinctly. In one 

 species {B. irregularis (Swart.) ) the macroscleres are smooth and 

 blunt at both ends. 



Gemmules. — These bodies have been discovered as yet only 

 in one species {B. bacillifera) , in which they are ovoid or pear- 

 shaped structures with a simple horny covering which is distinctly 

 depressed in a crateriform manner at the narrower end (pi. ix, 

 fig. 3&). They lie in the stout basal membrane of the sponge 

 with their long axis parallel to it. 



