132 M. K. A. Zlttel on Fossil Lithistidce, 



variety prevails than among the true skeletal elements. They 

 are either uniaxial or quadriaxial siliceous structures. 



The former present no peculiarities worth notice. Bacillar 

 spicules from 0'5 to 10 and 20 millims. in length may be 

 observed in the most various modifications. They are usually 

 spindle-shaped and pointed at both ends, sometimes blunt at 

 one end and pointed at the other, or rounded off at both ends. 

 Most frequently they are straight ; but curved, S-shaped, and 

 sometimes even undulated spicules occur. Their surface is 

 smooth, rarely spinous. In an undescribed recent species, 

 which I have received from Dr. W. Marshall, there are undu- 

 lated spicules which are furnished at regular intervals with 

 pointed frill-like processes, and in their habit resemble the 

 form of spicule figured by Bowerbank in the Mon. Brit. Sp. 

 pi. i. fig. 14. 



The spicules of the quadriaxial type, to which it would be 

 better to give Carter's name " trifid " or " ternate," are much 

 more multifarious in form, as one axis is always differently 

 developed from the rest and appears sometimes as a long shaft, 

 sometimes as a short style, and sometimes only as a button- 

 like thickening. Apparently perfect equality of the four rays 

 never or extremely seldom occurs in the Lithistidge. At least, 

 I have never observed the so-called clievaux defrise either in 

 living or in fossil forms. 



Most frequently we find anchors with long simple shafts, 

 diminishing towards the free end. The three prongs at the 

 opposite end are rarely simple, and are then usually short 

 (patento-ternate, recurvo-tevnate, expando-ternate, incurvo- 

 porrecto-ternate spicules, &c. of Bowerbank, I. c. figs. 45-54, 

 128, 129) ; but generally they divide again into two (rarely 

 more) prongs, forming so-called double anchors (bifurcated 

 expando-ternate spicules, Bowerbank, I. c. pi. v. fig. 130, 

 and spiculated dichotomo-patento-teruate spicules, fig. 53, 

 &c.). 



In the simple anchors the three prongs either diverge ob- 

 liquely outwards at equal angles or they are bent back. This 

 is the case also in the forked anchors ; but in these the three 

 furcate prongs more frequently lie in the same plane, starting 

 at right angles from the shaft, or their ends may even bend a 

 little backwards. In my monograph of the genus Coelopty- 

 cJiiimi I have figured (Taf. vi. figs. 3-30, and Taf. vii. figs. 

 1-10) many such anchor-shaped structures, presumably all 

 derived from Lithistidge ; so that any further description ap- 

 pears superfluous. Among these figures there are some 

 (Taf. vii. figs. 9, 10) in which the arms of the forks are not 

 smooth and straight, but furnished on the outside with branch- 



