Study of Madreporarian Types of Corals. 223 



and these lie called " /)seM(7o-synapticul£e." The name of 

 " true synapticulae " he limited to such as were formed around 

 new centres of deposit out of the septal plane ; these he found 

 in Siderastra^a, but did not farther examine typical genera 

 belonging to the families Funginse and Lophoserinse. The 

 author's results are that pseudo-synapticulcB occur not only in 

 Thamnastrjeinffi, but also in the Funginas, Lophoserina3, and 

 occasionally in Astrfeidte and Eupsammidte ; while true 

 synopticulre occur chiefly in Funginse and Eupsammidsp, 

 rarely in Lophoserina^, and never in Astrteidse. The author 

 regards true synapticulfe as hasal structures representing 

 modified dissepiments. The advantage of synapticula3 to the 

 polyp is that they afford a basal support over which the fleshy 

 parts and mesenterial loculi may bend and be continued to 

 some depth. The author, in pointing this out, refers to the 

 analogy of the internal canaliculate visceral system thus pro- 

 duced with the external canaliculate system attained by a 

 porous coenenchyme in " Perforate " colonies. At the same 

 time she thoroughly disagrees with the prevailing opinion that 

 the synapticulate types have any nearer relationship with 

 Madrepora^ Porites^ &c., since the skeletal parts show many 

 important differences of structure ; neither is the " porous 

 coenenchyme " in any way homologous with the synapticulate 

 calyx. 



No essential difference is presented between septa, costse, 

 and wall in respect of their microscopic structure ; and the 

 author found it also for other reasons practical to distinguish 

 in her work the septa and costte under the inclusive name of 

 radial structures, the wall, on the other hand, as a tanyential 

 structure. Dissepiments, tabular, true synapticulae, and certain 

 kinds of columella are regarded as hasal structures. 



The microscoinc structure of dis^einments and tahulce is 

 demonstrated hy the author to he the same. Both are composed 

 of a series of calcareous growth-lamellse laid down from one 

 surface only of the aboral body-wall of the polyp. The fibro- 

 crystalline deposit is therefore perpendicular to the plane of 

 contact between polyp and skeleton. The distance from one 

 platform of dissepiments to the next above coincides in all 

 typical Astrjeids with the interval between two growth- lines 

 on the septal surface. It may be deduced from this that the 

 polyp lays down a new basal support for itself at the close of 

 each growth-period. The solid calcareous deposit (usually 

 called '' stereoplasm " or " endotheca ") at the base of the 

 short simple calyces of most Turbinolids has the same micro- 

 scopic structure as tabula3 or dissepiments, differing from 



