Astylos])ongida3 and Anomocladina. 275 



spongia and Pulaowancn. In all essential cliavacters Melo- 

 nella and llastosia agree with Cylindroplnjma \ on the other 

 hand, Lecanella, which I fonnerly referred to the Anomo- 

 cladina, shows peculiarities which it is difficult to bring into 

 harmony with the other forms. 



In figs. 4 a-f I have had carefully represented some iso- 

 lated skeletal corpuscles of Lecanella which are remarkable 

 for their large size. These were obtained by the action of 

 hydrochloric acid from the still unique original specimen 

 from ISontheim. If the corroded parts of the sponge-body 

 are examined by direct light it is seen that the skeletal 

 elements are not firmly bound together, but that the furcate 

 ends are only loosely ajoplied to each other without being 

 amalgamated. By treatment with acid, therefore, we never 

 obtain connected parts, but only the isolated corpuscles shown 

 in fig. 4, which evidently represent the primary elements of 

 which the skeleton of Lecanella is built up. These, how- 

 ever, can hardly be likened to the simple rods, branched only 

 at the ends, of the typical Anomocladina, for the enlarged 

 nodes are here really the starting-points of 4, 5, 6, or more 

 arms, which are slightly branched at their extremities. The 

 genus Lecanella essentially influenced my former incorrect 

 definition of the Anomocladina ; if it be removed from that 

 group Ave find among the Megamorina the genera Doryderma 

 and Heterostinia^ or also Llolodictyon and Facliypoterion of 

 Hinde, the skeletal elements of which show an unmistakable 

 resemblance to those of Lecanella. 



I therefore do not hesitate to remove Lecanella to the 

 Megamorina, and the Silurian genus Ilindia^ Duncan, may 

 also find its right place in the same group. 



From the foregoing remarks it appears that, as supposed by 

 Hinde, the Astylospongida^, do not belong to the Ilexactinel- 

 lidaj, but to the Lithistidffi, and, indeed, to the family Anomo- 

 cladina. The definition of the latter will then have to be 

 modified as follows : — 



Family Anomocladina. 



Skeletal elements consisting of simple, generally straiglit, 

 but sometimes curved rods, more or less strongly branched 

 at the two extremities. The branched ends of several (4-9) 

 neighbouring rods meet together, and by their amalgamation 

 form nodes. In this way is produced a latticework re- 

 sembling that of certain liexactinellida?. 



The genera belonging to this family are : — - 



21* 



