CELL-FORM ATION. 127 



tents and the wall of the parent-cell, in some places be- 

 tween the mucilaginous contents and the wall of the 

 simultaneously-produced fellow secondary cell. Here 

 again no other hypothesis is possible but that of the 

 origin of the membrane through secretion from the muci- 

 laginous contents. The surface of the contents secretes 

 layers of thickening up to the moment when cell-forma- 

 tion begins. The layer secreted at this moment is the 

 rudiment of the membrane of the new cell. 



In free normal cell-formation, lastly, the phenomena, 

 if they do not exclude every other theory, are not un- 

 favorable to the hypothesis founded on analogy. The 

 most indubitable case is that in Achlya, where the spo- 

 rangium-cell sometimes originates by parietal, sometimes 

 by free cell-formation. Since here, in the one case (in 

 the parietal mode of origin) the membrane is produced 

 through secretion from the contents, it is certain that the 

 same happens in the other case. In the formation of the 

 germ-cells of Algae, Lichens, and Fungi, and in that of 

 the endosperm-cells in the embryo-sac, the phenomena 

 admit of a twofold explanation : either that the membrane 

 is a secretion from the portion of contents becoming 

 isolated, or that it is a deposit from the fluid of the cell 

 surrounding the isolated portion. Schleiden and Schwann 

 have, as is well known, maintained the latter explanation. 

 They found an analogy for it in the inorganic crystalliza- 

 tion, but I believe that there is no analogy to it in the 

 region of organic processes. On the other hand, the 

 former explanation finds its certain analogy in lignifica- 

 tion, in abnormal free and parietal, and in normal parietal 

 cell-formation. Thus there exists no reason why we 

 should not assume the origin of the membrane in conse- 

 quence of secretion from the contents, to be certainly 

 established also in free normal cell-formation. 



The examples of the formation of membrane, in the 

 manner we have just seen, fall into two categories : 

 1, Those in which the phenomena directly require the 

 hypothesis that the membrane is formed through secre- 



