REPORT ON THE CRTNOIDEA. 343 



Challenger at Station 186, and in at least six of the seven from Samboangan. The usual 

 rule is that the tentaculiferous anterior arms have about twice as many joints as the 

 ungrooved hinder arms, which terminate definitely in a miniature axillary joint, bearing 

 a couple of pinnules ; while the anterior arms always seem to end in a growing point as 

 is the case with all the arms of Antedon. 



The problematical ovoid bodies, which occasionally appear as brown spots in the 

 centre of the dorsal surface of some of the segments of the pinnules on the ungrooved 

 arms, occur in single individuals of this species from Station 186, Banda, and 

 Samboangan, and also in five out of the eleven examples obtained by Semper in the 

 Philippines. 1 am at present quite unable to throw any light upon their character, 

 though I hope that the researches of Dr. 0. Hamann, in whose skilled hands I have 

 placed several of the pinnules containing them, may add considerably to our knowledge 

 of their nature and structure. They are not peculiar to Actinometra parvicirra, as they 

 also present themselves in Actinometra elongata from Banda, and in the Brazilian 

 Actin ometra me ridiona lis. 



The number of cirri which occur in Actinometra parvicirra seems to vary consider- 

 ably, though the number of joints remains fairly constant at ten to sixteen. In three 

 of the Philippine specimens the centro-dorsal is reduced to a thin disk bearing three or 

 four moderately developed cirri, with indications of other sockets which have been more 

 or less completely obliterated (PI. LXI. figs. 1,5); while in the individual from Station 

 174 the centro-dorsal is very irregularly shaped, and bears quite rudimentary cirri with 

 imperfect sockets for others (PI. LXI. fig. 3). Another Samboangan specimen has a 

 larger number of cirri, but they are all small and rudimentary on a very thin centro-dorsal 

 (PL LXI. fig. 4). As a general rule there are ten or a dozen cirri which are not unfre- 

 quently disposed in pairs, two at each angle, with a few others in intermediate positions 

 (PI. LXI. figs. 2, 6 ; PI. LXVII. fig. 3). But I have seen individuals, both from the 

 Philippines and from the Cape of Good Hope, with as many as twenty -five sockets on the 

 centro-dorsal, which almost entirely conceals the first radials, though of course they may 

 not all have borne cirri simultaneous^. There is also a good deal of variation in the 

 development of the spines on the later cirrus-joints, and in the characters of the terminal 

 comb on the lower pinnules. Three modifications of this comb in different individuals 

 from Samboangan are shown in figs. 8-10 on PL LXI. In the original of fig. 8, the comb 

 is so small that it might easily escape notice ; but the other two pinnules are more normal 

 in character. The number of pinnules which bear a comb is also very variable. I have 

 seen specimens both from Africa and from the Philippines, in which there is no comb 

 after the third brachial ; while in others from both localities it may be found on the 

 pinnule of the fifteenth brachial, and in some of the Philippine specimens the later 

 pinnules of the arms may have small combs. In like manner I have seen individuals 

 from the Cape and from the Philippines in which the basal joints of the genital pinnules 



