}forj)holo<j7j of the Madreporaria. 221) 



iiidivitluMlity of the septa is soniewliat uncertain owing to 

 possible loss of orientation. To remeily these defects a 

 method was devised by which, with suitable material, one 

 can follow step by sti-p all the develoj)nientaI stages from 

 beginning- to end without any uncertainty of orientation, and 

 secure drawings ot all desirable stages. The process involves 

 the grinding down of an individual fixed coralluni from one 

 end to the other, and the study and drawing of all the stages 

 as revealed. The broad end of a corallum is first ground 

 smooth, and fixed by Canada balsam to a glass slide, in the 

 manner usually followed by geologists in j)reparing micro- 

 scopic sections of rocks; grinding down with fine emery or 

 on a ground-glass plate is then commenced at the opposite 

 end, the narrow tip, and continued all the way. Willi 

 favourable specimens the septa can be most clearly recognized 

 under a low power of the microscope, and their arrangement 

 outlined by the aid of a camera lucida. The distinctness of 

 the septa can be often em))hasized by etching the exposed 

 surface with a little weak acid, and to secure the best 

 reflection of the light iVom the ground surface while drawing 

 tiie latter may be smeared with weak glycerine or balsam. 



By these devices the whole septal develo{)ment of a simple 

 corallum can be followed almost as satisfactorily as if one had 

 watched its actual growth day by day. It was from such a 

 series that the ten stages representing the septal development 

 of Streptelasma rectum given in the paper on tiie Fossubi 

 were secured, and also the series here reproduced (figs. 1-8), 

 depicting the septal development of Lophcphyilum yroliftrum , 

 and intended to replace those given in 1902. 



For purposes of the present paper it is not necessary to 

 describe the sections of Lophophyllum at greater length than 

 is given in the explanation to each. They reveal nothing 

 fundamental beyond what was brought forward in the first 

 paper, but a confusion in the latter of main and counter septa, 

 alluded to by Gordon, is corrected. In place of the older 

 terminology 1 think the time has come to adopt that founded 

 upon more modern knowledge of the relationships of the 

 group and accepted for the Anthozoa geneially. With the 

 exception of unimportant details, the septal sequence of 

 Lophophyllum here given bears the closest relationship to 

 that of btnptehisma rectum in the sixth part of these con- 

 tributions; moreover, it is that found to be characteristic of 

 a large number of other species of rugose corals which I have 

 studied by the same method (c/. figs. 9-12 ; 13-16). Septal 

 and mesenterial development is unquestionably one of the 

 most reliable means zoologists possess for determining the 



