NEUROMUSCULAR MECHANISM IN METRIDIUM 445 



The sections were mounted in xylol balsam under a cover glass 

 and have proved permanent. The method is best carried out 

 with the least exposure of the material to light. It is also well 

 to use cold fluids, about the temperature of melting ice, up to 

 the change to silver solution for five days. This and the sub- 

 sequent changes to the parafiine were carried out at room tem- 

 peratures. As with most metal impregnations, the results of 

 this method are precarious, but in good preparations, the neuro- 

 fibrils appear as black or brownish-black lines in an almost trans- 

 parent field. In studying preparations one precaution is neces- 

 sary. The impregnation stains the filaments of the large nettle 

 capsules in the acontia very deeply and in many preparations 

 these are found to have been discharged and to have entered 

 such parts as the column wall and the mesenteries. Care must 

 therefore be taken not to confuse them with nerve fibrils ; but a 

 few moment's inspection is usually enough to make sure what a 

 given fibril is. 



The ectoderm of the tentacles in Metridium exhibits a neuro- 

 muscular structure much like that ascribed to actinians in general 

 by the Hertwigs. The epithelial sense-cells with their basal 

 ends branching into fine fibrillae, can be demonstrated by the 

 isolation method, the Golgi method, and that described in this 

 paper. In the actinians studied by Groselj ('09) these cells 

 were often stained by methylen blue. In Metridium the nerv- 

 ous layer to which the central processes of these sense-cells 

 give rise, is of considerable thickness and is directly next the 

 longitudinal muscle fibers. According to the Hertwigs ('79-80; 

 p. 488) this layer in the tentacles of many actinians contains 

 at most only a few scattered ganglion cells and in the forms 

 studied by Gro'elj ('09, p. 294) none whatever were found. 

 In the nervous layer of the tentacular ectoderm of Metridium, 

 Havet ('01, pp. 408, 411) was unable to identify ganglion cells 

 at all and we, too, have looked there in vain for these structures. 

 We have examined for this purpose certainly many thousand 

 sections of these organs but without being able to discover a 

 single well defined ganglion cell. Nowhere have we seen any- 

 thing comparable to the cells shown by Schneider ('02, p. 623, 



