26 Coelenterata. 



form. In both of these cases the conditions of internal pressure were ap- 

 parently similar to those which exist at the oral end, but the less rapid 

 regeneration of the aboral tentacles indicates either a difference in structural 

 relations at the two ends or a difference in polarity . 



Child ( 2 ) states that regeneration in Cerianthus is not proportional to the 

 size of the piece. The smaller the piece, other conditions remaining the same, 

 the greater the relative amount of regeneration. As regards absolute amount 

 of regeneration the small and the large pieces are at first alike, but later in 

 the small piece regeneration is retarded and ceases sooner than in the large 

 piece, probably owing to lack of available energy. Cylindrical pieces usually 

 undergo a change of form, increase in length and decrease in diameter, during 

 or after regeneration. The change is slight during the earlier stages before 

 the oral structures are developed. It is probably the result, either directly 

 mechanical or reactive or both, of longitudinal tension upon the tissues, which 

 may in turn be due in part to internal circulatory currents and in part to 

 the habit of creeping over surfaces in the direction of the longitudinal axis. 

 In pieces maintained in the collapsed condition the length decreases and the 

 diameter increases, at least relatively. Forms with duplicated oral or aboral 

 ends may be produced by partial longitudinal splitting: in cases of oral dupli- 

 cation each disc is essentially a fractional structure, bearing approximately the 

 number of tentacles corresponding to the portion of the circumference which 

 it represents; at the aboral end a new aboral pore may be formed when the 

 cut does not pass through the old pore. No marked regulation, reduction or 

 absorption of these duplicated structures has been observed except occasionally 

 where the duplication included only the extreme terminal portions. Many 

 peculiar forms result from the union of whatever portions of the cut surfaces 

 come into contact, and may be more or less modified by the tensions and 

 pressures to which the tissues are subjected. 



Hammatt states that a fragment cut off the base of Metridium marginatum 

 curls until its extremities meet so that the mesenteries previously almost parallel 

 now become radial; it develops a spreading base, a column, 12 tentacles and 

 a mouth, the whole process being complete in three weeks. Serial sections 

 of a parent specimen show a complete or partial infolding of the body wall 

 prior to fragmental fission which evidently frequently occurs, as such frag- 

 ments are common in the shore pools. Previous to fission the body becomes 

 asymmetrical and the fragment is cut off on one side of the directive plane 

 of the parent. The infolding of the body wall proceeds gradually downwards 

 and inwards and the fragment separates from above downwards. 



Heath describes Zoanthella galapagoensis, a new species of Semper's larva 

 from the Galapagos Islands. The single specimen was pelagic, spindle-shaped 

 and 6 mm long. At the end of the slenderer portion is the mouth but no sign 

 of tentacles, at the opposite end there is no trace of aboral pore. A ciliated 

 fringe extends backwards along the median line to a point 1,75 mm from the 

 aboral pole. The strip of ectoderm on which these cilia are borne is much 

 wider on the side in contact with the mesoglcea than on the exposed side, 

 being almost fan-shaped in section. The individual cells of this strip are 

 slender, each with deeply staining nucleus and a distinct external cuticular 

 border surmounted by a large cilium. The cilia are not fused to form an 

 undulating membrane. The general ectoderm consists of the usual slender 

 cells with occasional more fusiform sensory cells, gland cells, stellate pigment 

 cells, and nematocysts of two kinds one abundant, relatively small and 

 slender, with one end level with the surface of the body, the second kind larger 



