CANCEROUS TUMORS. 19 



are accompanied by an abundant small-celled infiltration of 

 the surrounding connective tissue, the luxuriance of which 

 keeps pace with the degree to which the glands are altered. 

 (One photograph, No. 27, shown.) 



I trust the foregoing illustrations will enable you to follow 

 me when I say that in epithelial cancers we have to do, on the 

 one hand, with the ingrowing from previously existing epithe- 

 lial structures of branching epithelial cylinders ; on the other 

 hand, with a small-celled infiltration of the connective tissue, 

 the cells of which in all probability have migrated from the 

 blood-vessels. All that I have been able to see inclines me to 

 give my assent to the opinion that the cylinders lie in the lym- 

 phatic capillaries, or in distended Ij'^mph spaces ; the small cells 

 occupy the smaller lymph spaces of the connective tissue in 

 which they are found. The preponderance of the fully formed 

 cancer cylinders, or of the small-celled brood, determines con- 

 siderable diversity in the details of different growths, and this 

 is increased by the circumstance that in some cases the connec- 

 tive tissue papillae of the part involved grow outward, and 

 branch as in the first case I presented to you. I cannot pause 

 here even to sketch the varieties thus produced, but I wish to 

 add that I am not prepared to agree with Billroth that Thiersch 

 and Waldeyer were right in regarding it as fully established 

 that the cancer cylinders grow by a multiplication of their 

 epithelial elements by division. On the contrary I wish to 

 draw attention not merely to the fact of the swarm of small 

 cells about the terminal buds of the cylinders, but to the addi- 

 tional circumstance that, in almost every section of epithelial 

 cancer in the Museum collection, I find among the epithelial 

 cells numerous unmistakable wandering corpuscles, often fixed 

 by the reagents used, with their processes extended as in the 

 act of migration. 



If Biesiadecki* is right in believing that the ordinnry growth 



♦Rindfleiscli, op. cit., p. 101. 



