126 COLLECTIONS FKOM MELANESIA. 



8. Linckia panciforis. 

 Martens, Arch./. Nat. xxxii. (I860), p. 69. 



In only one example is one of the arms otherwise than perfect ; 

 but this does not seem to have been budded out in place of one cast 

 off, but to have been bitten or broken off not far from its tip. 



Three dry specimens from Bird Island, jN t .E. Australia; coral- 

 reef. 



9. Linckia megaloplax. 



Arms five. E=67, r=9, or li=7'5 r about. Adambulacral 

 spines flattened, one for each plate, with blunted end; externally 

 to and alternating with these are somewhat shorter papilliform 

 spines ; the general granulation of the actinal surface of the disk 

 extends between these latter; externally to them there is a row of 

 larger spines, the distribution of which is extremely irregular, for 

 while at some points they are almost as closely packed as those of 

 the more internal row, they are at others separated from one 

 another by the distance of three or four of the inner spines. The 

 rest of the abactinal surface is closely covered with subequal gra- 

 nules of some size. The abactinal surface is traversed very regu- 

 larly by six rows of poriferous spaces, which are comparatively 

 large and markedly rectangular ; the smallest spaces are found in 

 the lowest row on either side ; a short distance from the tip of each 

 arm the spaces completely disappear, and the granulation becomes 

 a little more prominent, and there is here, as in some allied species, 

 a large specially modified plate with one or more large tubercles 

 upon it. The disk itself is covered with large pore-areas, and the 

 only noticeable character is the large and distinct, though not pro- 

 jecting, raadreporic plate. The pore-areas are about 2 millim. wide, 

 and the length of the madreporic plate, around which the granules 

 are very distinct, may be 4 millim. The colour of the abactinal sur- 

 face is deep brown or black, whilst that of the actinal surface is 

 lighter. 



In a specimen smaller than that which has formed the chief basis 

 of this description the two surfaces are, towards the free end of the 

 arm, distinguished from one another by the development of some 

 short marginal spines. In another, dried, specimen, intermediate 

 in size between these two, the spines at the upper and lower mar- 

 gins of the sides of the arm, though insignificant, are both larger 

 and more numerous. 



The idea that the younger forms would have a larger supply of 

 spines is opposed by the fact that in a still smaller specimen these 

 spines are altogether absent*, while the skeletal plates are stouter 

 than in the specimens which bear the spines. The question now 

 arises as to whether there are three stages — («) one in which the 

 ossicles are so stout that no defensive spines are needed in addition ; 

 (b) one in which growth has proceeded so rapidly that the bars of 



* Cf. Ann. & Mag. N. H. (5) viii. p. 441. 



