Structure of Heteropora neozelanica, Bash. 335 



section of the tubes in their final and most fully developed 

 condition ; secondly, one at right angles to the branch (a 

 transverse section) ; and, thirdly, a longitudinal section, 

 dividing the branch vertically through its median plane. 

 The following are the principal points brought to light by an 

 examination of these three sets of sections in H. neozelanica : — 

 (a) Tangential sections. — As just remarked, there are no 

 sections which yield more interesting and valuable results 

 than those which intersect the skeleton tangentially, just 

 below the surface upon which the tubes open. When we 

 examine such a section (fig. 2, A) we observe that it is gene- 

 rally quite possible to distinguish the proper zooecia from the 

 interstitial tubes by their size, but that these two sets of 

 tubes are in no way distinguished from one another in point of 

 structure, while their dimensions are often very nearly the 

 same. The tubes are all rounded ; and their walls are very 

 thick, and are composed of delicate calcareous laminas ar- 

 ranged concentrically around the cavity, and not showing 

 any line of demarcation between each other. In this respect 

 the walls have the structure of such species of Monticulipora 

 as M. ramosa, 31. Jamesi, &c, and differ altogether from 

 such other species as M. pulchella. The most remarkable 

 feature, however, in the structure of the wall consists in the 

 presence of numerous canaliculi, which pass transversely 

 across the thickened wall (fig. 2, A) and open at both ends 

 by wide funnel-shaped apertures into the cavities of the tubes. 

 In this way the zooecia and interstitial tubes are placed in 

 direct and free communication with one another throughout 

 the entire colony. Sections of this nature also prove with 

 absolute certainty that these canaliculi are strictly confined 

 to the icalls of the tubes — a point upon which, as will be seen, 

 longitudinal sections might leave us in some doubt. 



Another very interesting and important point brought out 

 by tangential sections is that both the zooecia and cancelli are 

 provided in this part of their course with numerous delicate 

 radiating spines, which spring from the wall (fig. 2, A) and 

 are directed inwards for a longer or shorter distance, usually 

 falling short of the centre. I am not aware that the presence 

 of these radiating spinules has hitherto been recognized as 

 occurring in the recent Heteropora?, or in the extinct forms, 

 except in H. clavata, Goldf. (according to Busk), and then 

 only at the mouths of the interstitial tubes. In H. neozelanica 

 they are very slender and delicate, and often break up in thin 

 sections, so that they may appear to be wanting in a greater 

 or less number of the tubes ; but I have never failed to recog- 

 nize their existence in some part or another of tangential 



