THE LIMITING SURFACE OF PROTOPLASM 235 



cartilage cells, that a few filaments (so-called limiting filaments) 

 of the framework occur on the surface, but no continuous mem- 

 brane of condensed protoplasm. 



On the other hand, Heitzmann had from the first (1873), 

 although not persistently, represented the surface of the proto- 

 plasmic framework as shut off by a continuous membrane of the 

 skeletal substance. Later (1883) he discusses this point in more 

 detail, and derives the membrane, in the same manner as that of 

 the vacuoles, from the formation of a so-called " membranous 

 layer" (see above, p. 229) from the framework. 



On the botanical side the matter has received rather different 

 treatment. The idea had notoriously prevailed among botanists 

 for a long time that the protoplasm must be limited towards 

 the exterior by a special cMticular layer (Hautschicht), which was 

 distinguished by its properties from the rest of the protoplasm. 

 This cuticular layer has often been directly observed, i.e. the most 

 external non-granular hyaline protoplasm, which we usually term 

 ectoplasm in Protozoa, was termed the cuticular layer. The 

 assumption of an ever-present, even if not directly visible 

 cuticular layer, was, however, first founded upon the osmotic 

 experiments of Pfeffer and others, which in the former's opinion 

 could only be explained on the supposition of such a protoplasmic 

 membrane possessing special osmotic peculiarities. 



Schmitz, who was the first on the botanical side (1881) to 

 study accurately the reticular structure of protoplasm, thinks it 

 may be assumed that the cuticular layer is specially distinguished 

 by its more narrow meshed structure from the remaining proto- 

 plasm. Strassburger also (1882, Zellliilute) was inclined to agree 

 with this view, but is of opinion that the meshes of the cuticular 

 layer of protoplasm could also become completely obliterated 

 (p. 195). Upon this anomalous structure of the cuticular layer 

 its peculiar osmotic properties might well depend. On pp. 235, 

 236, he speaks quite definitely of the fact that he assumes the 

 " anatomical meshes," i.e. the meshes of the reticular structure, 

 to be quite closed in the cuticular layer, and that in it the so- 

 called molecular network is more stable, and the molecules have 

 a more definite arrangement. If the cuticular layer is wanting 

 on occasions, " then the granular protoplasm might contract the 

 meshes at its surface, and take on the function of the cuticular 

 layer." In 1884 Strassburger also explains the nuclear mem- 

 brane as a corresponding cuticular layer of the protoplasm, and 

 remarks upon this point (p. 104) that, "like every cuticular 

 layer " it is " produced by narrowing of the meshes " from the 

 network of the cytoplasm. 



Schneider (1891), a consistent representative of the theory 



