91 



second new genus, Ceratella, in a family called Ceratellidfe ; and 

 while strongly inclining to locate this family among the true 

 horny sponges, Dr. Gray at the same time calls attention to tha 

 fact that many of the characteristics of the keratose sponges are 

 not to be met with in the dry horny skeletons of the two species 

 described. 



On treating a small portion forsome(thirty-six)hourswithcaustic 

 potash, the only effect observed was a greater transparency of the 

 keratose fibres, and perhaps a greater flexibility of the entire 

 mass. There was no tendency, however, of the frame or network 

 to break up into detached pieces, such as might occur if it were 

 formed of a series of horny spicules united the one to the other; 

 indeed, the skeleton must be looked upon as continuous. 



On placing another portion in some weak nitric acid, effer- 

 vescence at once occurred, and the coloured horny material pretty 

 speedily disappeared, leaving, however, behind, a semi-transparent 

 basis, which in great measure preserved the form of the original 

 little twig, and which had all the appearance at first sight of 

 being siliceous. On being placed on a glass slide, and covered 

 with a piece of thin glass, it yielded to a slight pressure, and when 

 examined under the microscope showed a gelatinous basis, in 

 which were entangled a few biacerate siliceous sponge spicules, 

 and a few diatoms {J^avicuJa, Pi7innlaria, Coscinodiscus, Amplii- 

 tetras), the former, without doubt, just as much foreign to the 

 Dehitella as the latter. 



On examining the structure after it has been gently boiled in 

 distilled water for a few moments, it will be found to have 

 absorbed a certain quantity of the water, so that even after it has 

 been tightly pressed and flattened it will, on the pressure being 

 removed, soon recover its shape. It would appear, however, that 

 the so-called tufts of horny spines met with on the sides are 

 really not so much spiiies as sharj)-ending prolongations of the 

 common skeleton ; indeed, they cannot be called tufts of " spi- 

 cules" in the ordinary meaning of this word, and the arrangement 

 of the network will be best learned from the accompanying wood- 

 cut (fig. 1). On making a transverse section of the stem it is 

 apparent that the main fibres are continuous, and that thus there 

 is a series of canals permeating the entire mass (fig. 2). The skele- 



fif.i 



ton is to a certain extent regular ; that is, it is made up of a series 

 of long horny strands, which run almost parallel to one another, 

 either terminating in a loop or in an obtuse point. The horny 



