MICHIGAN ACADEMY OF SCIENCE. 75 



selection of the species for individual centers these factors have a vary- 

 ing importance, their influence being further modified by edaphic, geo- 

 graphic and historical considerations. 

 Alma College, March 15, 1905. 



THE NERVE-LAYER IN THE CORAL COENOPSAMMIA.* 



BY J. E. DUEUDEN AND S. A. AYRES. 



In 1879 the brothers O. and R. Hertwig (''Die Actinien," Jenaische 

 Zeitschrift, XIII) published their important results upon the nervous 

 system of the Actiniaria, By means of sections and macerations they 

 proved the existence of a definite nervous system, diffuse in character. 

 l)istinct multipolar nerve cells and epithelial >sensory cells with nerve 

 fibrils were demonstrated, and in certain regions the nerve fibrils were 

 found to constitute a distinct nerve layer (Nervenschicht). In the re- 

 ports on the "Challenger" Actiniaria (Vols. VI and XXVI) R. Hertwig 

 was the first to use anatomical characters for taxonomic purposes, a 

 method followed by all subsequent workers on the group. 



Dr. Carlgren in 1900 (''Ostafrikanische Actinien,"' Mitt. a. d. Nat. Mus., 

 Hamburg, XVII) introduced a classification of the Actiniaria in which 

 he made the presence of a columnar nerve layer one of the fundamental 

 features. Carlgren attempted to subordinate extelmal and anatomical 

 characters to histological details, tlie principal of which is the presence 

 of a nerve layer associated with a longitudinal ectodermal musculature 

 in the column wall and stomodaeum ; with these are often associated the 

 absence of gonidial grooves from the stomodaeum and ciliated streaks 

 from the mesenterial filaments. In Carlgren's opinion Actiniaria were 

 primitively provided with a nerve layer and an ectodermal longitudinal 

 musculature in the column wall and oesophagus. He would divide the 

 Actiniaria into two main tribes — the Protantheae, actinians retaining 

 their primitive histological and anatomical structures, and the Ni/7iatJieae, 

 those forms more highly developed in their histology, anatomy and ex- 

 ternal characters. 



Carlgren's classification has not been generally accepted. Writers such 

 as McMurrich, van Beneden and Duerden would be prepared to look upon 

 the ectodermal columnar nerve and muscle layers as ancestral and the 

 forms in which they occur as the lowest members of their group. But 

 they regard the details rather as characters which may turn up in forms 

 now very widely separated, showing that the species j)0ssessing them do 

 not represent a homogeneous group. 



Recent studies have shown that the madreporarian polyps are closely 

 allied to the Actiniaria, but no ectodermal columnar nerve or muscle 

 layer has hitherto been found, yet they have many of the characters 

 claimed to be ancestral, particularly the absence of gonidial grooves from 

 the stamodaeum and ciliated streaks on the mesenterial filaments. In 

 examining polyps of Coenopsammia manni Verrill, collected by one of 

 us from the Hawaiian Islands, a well defined ectodermal nerve layer was 



* Contributions from the Zoological Laboratory of the University of Michigan, No.- — The work 

 has been carried out by the aid of an appropriation from tlie Carnegie Institution. 



