270 THE EVOLUTION OF CONTINUITY 



its external aspect. The other outstanding point is that 

 the unexposed surface is always in contact with connective 

 tissue. The free epithelial surface is very liable — and 

 specially so in the case of skin or mucous membrane — to 

 be wounded, severed, or have its continuity broken by 

 many hostile agencies, and its normal functions may involve 

 a continual cell loss. On the other hand, on the connective 

 tissue side the blood and lymph may bring to the cells 

 toxic substances which can wound or destroy as thoroughly 

 as externally acting physical agencies. When the lesion 

 heals, it does so through the release of cells from arrest, 

 which, their purpose accomplished, become re-arrested ; 

 but it may happen that the arrest, once broken, is never 

 re-imposed and that tumour growth results. 



The theory of Waldeyer and Thiersch presumes that the 

 underlying connective tissue exercises normally a restraint 

 on its covering epithelium, and that malignant epithelial 

 tumours might arise through the removal of this restraint 

 with advancing age. Advancing age may well be only one 

 factor, but this apart, the influence of connective tissue 

 on its epithelium would seem to be proved by experiments 

 which have been performed in vitro. 



Those of Champy are very suggestive. 



He found that on growing in culture a piece of embryo 

 rabbit's kidney, tubules were soon produced of a primitive 

 type, and finally these merged into an unformed mass of 

 undifferentiated embryonic cells. Smooth muscle cultivated 

 outside the body also gave rise to embryonic cells, this being 

 specially noteworthy because smooth muscle cells do not 

 divide when in the body. Champy's most interesting 

 discovery was, however, that when a portion of attached 

 connective tissue was included in the culture the epithelial 

 cells multiplied typically. 



It would thus seem clear that if tissues are removed 

 from the body and grown in culture the amount of growth 

 is uncontrolled ; and this because the tissues in question 

 are no longer a part of the body's personality. And it is 

 clear, in addition, that in the case of epithelium if no attached 

 connective tissue be included, the character of the epithelial 

 growth is also uncontrolled. One cannot but conclude 

 that the connective tissue exerts the influence which keeps 



