RECENT RESEARCHES OlS^ THE RADIOLARIA. 153 



may or may not be of similar length, but they always meet within 

 the capsule and consist of twenty acanthin spicula. 



They may merely meet together centrally, and be all alike, as 

 in Acantliometra, 



On the other hand, the differences in their lengthi and size may 

 be so great as to make the body conspicuously bipolar, as in Am- 

 philoncTie, where two spicula are greatly in excess of tlie others. 



Four spicula may be in excess, to a less degree, forming two 

 long axes at right angles, as in Acanthostauriis*' . 



Moreover the processes may give forth transverse processes, as 

 in Xiphacantha (fig. 8) ; and these processes may be very complex, 

 as in Lit]ioptera-\ . 



The spicula may not merely meet together, but they may anky- 

 lose together centrally, as in Astrolithium and Staurolithium. 



Two other very exceptional conditions may obtain : — (1) The 

 spicula may merely adjoin when they meet, but instead of regu- 

 larly radiating on all sides, may all be directed one way, diverging 

 irregularly from one polar axis, as in LitholophusX. 



(2) The spicula may be ten in number, and go right through 

 the central capsule in all directions, each spiculum perforating 

 the capsule's wall in two places, the spicula meeting but not in 

 any way uniting centrally. 



But radial elements may either join or both join and ankylose, 

 when one or more circumferential shells also exist, the radii per- 

 forating them and either simply joining or ankylosing together 

 in the centre, as in Dorataspis, Halliommatidium, and Aspidomma. 

 One of the most peculiar of all B-adiolarian shells is that of 

 Diploconus^, where we have two very large polar spicula (as in 

 Ainphilonche) united with ten short, cylindrical, radiating spicula 

 in the middle of the capsule by absolute ankylosis. To this 

 structure there is added two siliceous homogeneous cones, with 

 th.eir apices coinciding with the central mass and with serrated 

 margins at the open ends, from the middle of each of which pro- 

 trudes one of the large polar spicula. Each of these cones may 

 be conceived as formed of diverging spicula (like those of Litho- 

 loplms) united together ; and thus Diploconus would be like 

 LitholopJius, with a double set of opposite, modified, and fused- 

 together spicula, to which other ankylosed spicula, like those of 

 AmpMlonche, are added. 



* L. c. pi. xix. fig. 5. t L. c. pi. XX. fig. 1. 



X L. c. pi. xix. fig. 6. § L. c. pi. XX. fig. 7. 



LTN^T. JOTJRN. — ZOOLOGY, VOL. XIT. IL 



