736 MR. p. H. CARPENTER ON THE [DeC. 19, 



division of 3 joints, the axillary with a syzygy. Subse- 



quent divisions of two joints united by syzygy 



,b 



novce-guinea. I (2)A'RDP^ 



b 



typica. lA'RD^. 



B. Second and third radials united by ligament. 



Many arms. First ray-division of three joints, the axillary 

 with a syzygy. 

 a. Rays divide three times. Subsequent divisions Uke the first. 



robustipinna. A'(D)P— '. 



japonica. v3A'R?' 



parvicirra. 3A'DP^. 



(i. Rays may divide five times or more. 



I. Third and fifth ray-divisions like the first. Second and 

 fourth divisions of two joints, the axillary without a syzygy. 



alternans. SA'RFP"?. 



II. All ray-divisions like the first. 



schlegeli. 3A'RDP^. 



>c 



bennetti. 3A'RDP^. 

 peronl SA'RDPj. 



Prof. Bell's fbrmulse do not give any thing like a proper idea of 

 the characters of Act. novcB-guinea and Act. typica, especially the 

 latter. Both species are among "those rare cases in which 

 divisions extend beyond the palmars;" and Prof. Bell should there- 

 fore have made use of his symbols P' and P". These two would 

 have sufficed for A. novcR-guineee, which has only two axillaries be- 

 yond the palmars. Strictly speaking, however, neither P, P', nor 

 P" have any proper place in the formula; for the palmar and sub- 

 sequent axillaries are not syzygial joints homologous with the dis- 

 tichal axillaries, any more than the radial axillary is, either in these 

 two species or in the Solaris group ; and as pointed out above, it is 

 equally incorrect, for morphological reasons, to describe the first 

 brachials as being syzygial joints homologous with the third brachials 



1 It is absurd to put the D within brackets in this formula, because the onlj- 

 specimen described has no syzygy iu the axillaries of three out of the nine 

 primary arms. I have described a specimen of Act. parvicirra in which five 

 out of the ten distiohal axillaries have no syzygy, and another iu which .there 

 are four axillaries with and four without a syzygy. Here therefore we have a 

 character which " frequently though not always obtains" just as in Act. robusti- 

 pinna. Why is the one case noted in the formula but not the other ? Prof. 

 BeU's experience of the variations in these characters must surely have taught 

 him that it is the exception and not the rule for all the distichal and palmar 

 series of any many -armed specimen to be exactly alike, and that a speciJSc dia- 

 gnosis must be based on the characters of the majority. When, however, some 

 specimens of any type have distichals or palmars, and others may be altogether 

 without them it is usefiU to put the D or P within brackets ; and this should 

 have been done iu Bell's formula for Act. parvicirra, as I shaU shortly point out. 



