MATTHAI— RECENT COLONIAL ASTR^ID^ 25 



former I term " principal " couples and the latter " subsidiary " couples. In any species of 

 this group, the number of principal couples varies within narrow limits, while that of the 

 subsidiaries may vary considerably ; sometimes the latter may be entirely wanting between 

 two adjacent principal couples. Moreover, some of the principal couples may be incom- 

 plete. The subsidiary couples differ irregularly in their radial and vertical extent, some 

 of them being very narrow and short ; two mesenteries constituting a subsidiary couple 

 may also differ in their relative size. This variation is no doubt due to the differences in 

 the time of formation of the mesenteries. The number of subsidiary mesenteries in any 

 exocoele is also liable to great variation*. 



All the mesenteries are attached above to the oral-disc, and along their outer margins 

 to the corallum, by means of the wedge-shaped mesogla^al processes, but they vary con- 

 siderably in their depth, the primary and principal mesenteries reaching very nearly to the 

 base of the polyp. The arrangement and structure of the mesenteries, and the numbers 

 of cycles and of the couples composing them are of considerable value in the classification 

 of corals. Following Fowler's terminology (44, p. 578), the part of the gastro- vascular 

 cavity included within a couple is known as an "entocoele" ("loge" of French authors) 

 and that between two neighbouring couples as an " exocoele " (" interloge " of French 

 authors). 



As the two directive couples and grooves are situated at the ends of the longer 

 diameter of the stomodaeum and as the mesenteries are arranged on the hexameral plan, 

 the polyps of Group I are divisible into two similar halves along this plane, viz. the 

 sagittal ("plan commissural" of Faurot, p. 59, and "plan median" of van Beneden, p. 12); 

 they may also be divided into two similar halves along a plane at right angles to the 

 sagittal, viz. the transversal (" plan median " of Faurot). In some polyps the bilateral 

 symmetry is not quite perfect owing to the incompleteness of the last cycle of mesenteries. 

 Typically, however, Group I may be regarded as bilaterally symmetrical along two planes. 

 Their mesenterial couples are also arranged radially with reference to the oro-aboral axis 

 ("axe-vertical" of Faurot, "axe oro-aboral" or "axe du corps" of van Beneden), passing 

 through the centre of the mouth at equal distances from the two directive grooves. Like 

 the couples of mesenteries the disposition of the tentacles may also be assumed to be 

 symmetrical both bilaterally and radially, although from preserved polyps I have not been 

 able to settle this point. 



In Group II, on the other hand, the bilateral symmetry is wanting owing to the 

 absence of the directive couples of mesenteries and of the grooves ; a trace of the bilateral 

 symmetry is, however, left in the lateral compression of the stomodaeum. The radial 

 symmetry is interrupted by the indefinite number and the irregular width of the 

 subsidiary couples of mesenteries and by the incompleteness of some of the principal 

 couples. In Gyractis, Boveri (19) regarded the absence of directive mesenteries as only 

 apparent, their presence being disguised by the development of a mesentery on either side 

 of each of the two Edwardsian directive couples, but McMurrich (99) believed that the 

 disappearance of the directives may be due to " the mesenteries which really represent 



* Parker (110) has made similar observations on the numerical variation and on the irregular arrangement 

 of the incomplete mesenteries in the " monoglyphic " forms of Metridium marginatum. 



SECOND SERIES— ZOOLOGY, VOL. XVII. 4 



