1^2 ECHINOIDEA. 1. 



The same form of globiferous pedicellariæ is foiind in Strongylocciitrotits ^ Aiitliocidaris^ and 

 Parasaloiia. The two former are distingnished by the globiferous pedicellariæ having a well devel- 

 oped neck, jDrovided with circular and longitudinal muscles — an otherwise unknown feature. These 

 three genera are likely to be rather nearly related; their spicnles, however, show that the relation 

 is not very close. In Stroiigylocciitrotus the spicnles are a little branched in the ends, but otherwise 

 the original form is bihamate; in some species only (?) common bihamate spicnles are found. In 

 Antliocidaris the spicnles are biacerate, pointed in both ends and with a branch in the middle. A 

 somewhat similar form of spicnles is found in Parasaleiiia; but in this genus the globiferous pedicel- 

 lariæ have no neck. Thus this latter .seems to form a special group; its obliqnity and the peculiar 

 anal piates indicate also that it must be interpreted as an aberrant form, of which the nearest, 

 although not ver\- near, relations are: ^Inthocidaris and Stroiigyloccntrotus. In the genera Hcliocidaris^ 

 Echiuostrephns^ Toxocidaris, Echinometra^ Hcferocciilrotus, and Colobocoitrottis there is a strong, unpaired 

 lateral tooth on the globiferous pedicellariæ, and they have all simple bihamate spicnles. Heliocidaris 

 occupies a somewhat isolated position; its globiferous pedicellariæ are not so much developed as those 

 of the other genera, it reminds to a rather high degree of Strrcchinus N^rniiinyrri^ but especially of 

 Psc7tdechiinis albocincfiis\ several things favonr the belief that Pseudechiiuis is really a transitional 

 form between Stcrechinus and Heliocidaris, and the latter leads on again to Toxocidaris, Echiiiomctra 

 etc. Thus we have here a very fine series of development where, together with the peculiar develop- 

 ment of the globiferous pedicellariæ, a marked tendency to obliqnity is seen, reaching the climax in 

 the genera Hcfcrocciifrofiis and Coloboccntrotus. There seems to be no occasion to separate these two 

 genera as a special group on account of their longitudinal axis not being placed in the same direc- 

 tion as in Ec/iinoiiicfra, because their pedicellariæ and spicules are exactly agreeing with those of 

 Echtnoinctra. It is constantly seen that spicnles and pedicellariæ are the most important systematic 

 characters, so that there is no reason for suddenh- following a new principle here. The %^\\^x2l Pseud- 

 eclinms, Heliocidaris, and Echinostrepluts must then be interpreted as more or less primitive oligopore 

 Echinometrids. 



Stoviopncustes occupies a quite isolated position; its globiferous jjedicellariæ and spicules are 

 so peculiar and so different from what is found in the other forms mentioned here, that there can be 

 no question of classing it with an> of theni; it forms a special group. 



The relation between these forms may most easih' be surveyed in the following diagram. For 

 safety's sake I shall expressly remark, however, that I do not mean it to be regarded as a phyloge- 

 netic one. I vvill in no way maintain that our Parcchimis is the ancestral form of Ecliiiins etc, but 

 onl)- express my opinion that it shows the simplest structure of the jorgans most important with regard 

 to classification. We may in the recent forms scarcely find more than an indication of the way the 

 development seems to have taken. Now there is unfortnnately only a small chance of finding these 

 fine strnctures in the fossil forms, so we shall hardly get so far as to be able with certainty to point 

 out the ancestral forms. Otherwise this survey of the relations of the forms shows clearh- that here is 

 everywhere a tendency to increase the number of tube feet, a development from oligopore to polypore 

 forms. The most original feature, no doubt, is that all the ambulacral piates are well developed with 

 primary .spine and three tube feet; then the primary spines disappear from every other ambulacral 



