DESCRIPTIONS OF SPECIES. 163 



made on the minute structure of two species of Balanophyllia, viz, B. 

 irrorata (Conrad) and B. elongata sp. nov. For purposes of comparison 

 attention is called to I\Iiss Og-ilvie's work on the structure of Eupsammia 

 trochiformis} 



Balanophyllia irkorata (Courad). 



The wall. — Distally the septa are joined by synapticulse. True theca 

 was seen in one part of the section, i. e., a calcification line was seen between 

 the ends of two septa, but this might be a rudimentary septum. In another 

 place a very rudimentary septum was inserted between two older ; the ends 

 were fused so as to form a pseudotheca. 



The septa are not directly continuous to the outside, as in the so-called 

 Astrseids, Fungids, Turbinolids, etc. For a portion of the length a septum 

 is continuous, straight or slightly curved. As it approaches the exterior it 

 becomes perforate, and a spongy mass of synapticulfe and disconnected 

 portions of septa prevail. The main axis of the septal plane does not 

 coincide with the central vertical axis of a costa, but corresponds to one 

 side of it. A short disconnected rudimentary septum corresponds to the 

 other side of the costa. In places the apex of a costa seems formed by a 

 keystone piece set in between the ends of a large and rudimentary septum. 

 In other places the small septa may each have a con-esponding fine costa, 

 making a set of compound costa?. Very often it is extremely difficult, 

 because of perforations and synapticulse, to trace the course of a septum, 

 the whole being resolved into a spongy mass. But in general the costse 

 correspond in position to the septal (speaking of the developed septa) 

 planes projected through the wall. I do not know of any exception to this. 



The septal trabeculae, as judged by the granulations on the surface of 

 the septa, pass upward and inward. The course of a trabecula may con- 

 tinue uninterruptedly across several growth segments, new trabecula? being 

 introduced between the old, or, as Miss OgiKne has noted, the new 

 trabecular courses may alternate in position with the old trabeculse. Both 

 conditions occur, i. e., the trabecule are not always continuous across all 

 growth segments. In the spongy zone quite near the periphery of the 

 corallum is a line of divergence, outside of which the ti'abecula? pass 

 upwai-d and outward. 



'Microscopic and Systematic Study of Madreporarian Types of Corals, pp. 193-201. 



