6 I'l'of. J. Miiller on the Sti'ucture of the Echinoclerms. 



tangidce or of tlie Echinida, or that of the HolothuriadcB ? 1 

 imagine no one will be disposed to get rid of this difficulty by 

 supposing that the Holothuriadce are exceptions and creep upon 

 their backs. Let us conceive the typical form (Mittelform) of 

 the Echinoderms to be a sphere with radial arese, the question 

 is, whether this typical form passes from that of the Spatangus 

 into that of the Holothuria, by the body being rotated in the 

 plane of its middle longitudinal section, so that from resting 

 upon the inferior inter-radius it comes to lie upon the opposite 

 radius; or, whether this change is effected by the rotation of 

 the typical form round the axis of the radii, so as to bring the 

 globe from its position with an inter-radius inferior into that 

 with a radius inferior. In the former case, the conju ate sides 

 and the anal inter-radius remain constant, but the denomina- 

 tions of the conjugate sides as regards right and left become 

 changed wdth the rotation in the longitudinal plane. In the 

 latter case, the conjugate sides and the position of the anus would 

 shift, but the anterior and posterior ends would remain constant, 

 different radii and inter-radii becoming developed into the am- 

 bulatory disc, and the anal inter-radius undergoing a corre- 

 sponding alteration. 



If any such thing as a homology of the Echinoderms exists, 

 it can only be sought for in the former of these hypotheses. 

 By rotation round its axis the position of the Spatangus may 

 indeed approximatively be converted into that of the Holothuria, 

 but not wholly ; for even then to correct this position, and at the 

 same time to change a Spatangus into an Echinus, the rotation 

 in the opposite direction round a transverse axis, or in the lon- 

 gitudinal plane between the conjugate sides, would still be 

 required. 



If, therefore, we seek an ideal middle term among the various 

 actual forms of Echinoderms, this typical form will be found to 

 possess no constant anterior and posterior extremities ; it has the 

 mouth in the one pole of the ambulacra, but in passing into the 

 various forms of Echinoderms, it may turn sometimes one side 

 of its radial body forwards, sometimes exactly the opposite. 

 These anteriorly directed sides are, however, determinate in all 

 cases ; they are invariably cut by the same constant meridian of 

 the radial form ; i. e. the longitudinal plane of separation between 

 the symmetrical halves of the ideal typical form is in all cases 

 constant. 



Those terms are best which are derived from the intrinsic pe- 

 culiarities of a form and not from positions, which it may alter. 

 On the former principle we may bring the ambulacra of the 

 5-partite Echinoderms under two classes, the one of which con- 

 tains three, the other tw^o ambulacra ; we thus obtain a triradial 



