42 



is deserving of especial attention. But before this is considered it 

 will be well to discuss the process of normal cavity-formation to- 

 gether with its relation to primary and secondary growth. 



3. The formation of cavity by the destruction of cells and 

 its prevention by the use of gypsum casts. 



De Bary 1 in his Vergleicliende Anatomie has pointed out the 

 intimate relation between schizogenous and lysigenous modes of origin 

 of the cavity in tbe pith and mentioned the fact that one mode may 

 replace the other in closely allied plants. De Bary's statement may 

 be carried farther: One mode may replace the other in different 

 individuals of the same species and in different internodes of the 

 same plant. The fact is that there is no boundary between schizo- 

 genous and lysigenous. If for instance in Vicia faba one studies the 

 cavity-formation in the stem at the distance of several internodes 

 from the earth be finds the pith-cells forming a loose network which 

 as primary extension progresses suffers many breaks, the cells becom- 

 ing widely separated before any die. This beginning of cavity-for- 

 mation is schizogenous. If now one traces cavity-formation in the 

 epicotyl where the pith remains solid till primary extension has 

 ended, he will find the cells dying before separating from one another, 

 the mode being lysigenous. If the study is carried into the next 

 inter node above, where the cavity begins about the time that secon- 

 dary growth begins, one finds the pith-cells shortly before cavity-for- 

 mation with intercellular spaces larger than those appearing in the 

 internode below, smaller than those in the internode above. The cells 

 may die with or without any split appearing in the tissue, but if 

 such split appears before any cells die its extent is very small. Nam- 

 ing the beginning of cavity -formation in such an internode either 

 schizogenous or lysigenous is purely formal. 



TJrtiea dioica, Lamium garganicum, and Caltha palustris become 

 hollow as does Vicia faba, and the 2 former show in their lower 

 internodes the same variation as pointed out for Vicia. 



In Archangelica, Myrrhis, Dahlia, Melianthus and Ricinus , the 

 early stages of cavity-formation are almost alike in the higher inter- 

 nodes where the stem becomes hollow before elongation has ended 

 and in the lower internodes where the pith is solid for some time 

 after secondary growth has begun. In the former case the inter- 



i p. 223 et sq. 



