REPORT ON THE CRINOIDEA. 01 



tridisticnate species than occur in Antedon. I do not know any species of the latter 

 genus with articulated radials in which there is a fourth post-radial axillary, such as 

 occurs in Actinometra alternans, Actinometra variabilis, Actinometra magnified, and 

 Actinometra bennetti ; and it is decidedly rare to find a third axillary ; while the 

 singular variations presented by Actinometra belli and Actinometra nobilis are 

 altogether unknown in Antedon. 



The above tables show that it is possible to make a preliminary classification of the 

 species of Comatuke by using the characters of their successive arm-divisions. 



But how are we to deal with the seventy odd species of Antedon which have only 

 ten arms, or with the thirty more which have bidistichate primary arms ? The 

 characters of systematic value which may be employed for this further classification are 

 those of the cirri, arms, and pinnules. The number of the cirri themselves and also that 

 of their component joints are very useful characters within certain limits. Antedon 

 valida and Antedon parvipinna, which are both figured on PI. XV., are obviously cmite 

 distinct specific types; and the same maybe said of Antedon alternata and Antedon 

 incerta, represented on PI. XVIII. , not only as regards the cirrus-characters, but with 

 respect to the pinnules also. 



The shape and the relative sizes of these latter organs, especially at the bases of the 

 arms, often afford characters of much systematic value, as will be seen by comparing the 

 flagellate lower pinnules of Antedon quadrata and Antedon australis (PI. XXVII. 

 figs. 8-16) with the stiffer ones of Antedon occulta and Antedon variipinna (PI. XLVIII. 

 figs. 2, 3) ; while those of Antedon valida, Antedon incerta, and Antedon macronema 

 (PI. XV. figs. 5, 6; PI. XVIII. fig. 5; PI. XXXVIII. fig. 4) are of an altogether 

 different type from either of the others just mentioned. 



Another very useful character for systematic purposes is to be found in the shape of 

 the arm-joints. In one large group of Antedon-species the radial axillaries and the next 

 few joints beyond them have their apposed sides much flattened, as is well seen in 

 PI. XV. fig. 6. In the absence of this very striking peculiarity, the shape of the arm- 

 joints, as seen from the dorsal side, is often of much use in classifying species. Thus, for 

 example, the elongated joints of Antedon phalangium (PI. XXVIII. fig. l), the short 

 compressed triangular joints of Antedon patula (PI. XLIIL), and the rounded joints of 

 Antedon variipinna (PI. XLIX. fig. 1) all afford good specific characters ; while in the 

 genus Actinometra the contrast is strong between the short discoidal joints of Actino- 

 metra Jimbriata and the triangular ones of Actinometra elongata (PI. LVII. fig. 2; 

 PL LXII. fig. 3). 



The condition of the ambulacra in the arms and pinnules is also of much use in 

 classification. Thus, for example, Antedon accela, Antedon incerta, Antedon insequalis, 

 and other forms have both side plates and covering plates on the pinnule-ambulacra, 

 which are often better defined than in the Pentacrinida) ; while in other species, such as 



