2 CEOWN-GALL AND SAECOMA. 



(2) The relation of the origin of the secondary tumors to the 

 primary tumor, i. e., whether the new growths result from the inde- 

 pendent migration of bacteria from the primary tumor to the site 

 of the secondary tumor or whether the tumor cell itself is the migrant. 

 That the latter was believed to be the case may be seen from the 

 second paragraph on page 164. 



(3) A comparison of the cell structure of the secondary tumor 

 with that of the primary tumor. 



NEW FACTS. 



Having obtained much favorable material for study from inocu- 

 lations made on the daisy in January of this year, using for inocula- 

 tion pure cultures of the schizomycete plated from tumors on the 

 daisy, the answers to these questions are at hand and are confirmatory 

 of the hypothesis made in the bulletin, viz, that the study of crown- 

 gall is calculated to throw some light on the origin of malignant 

 animal tumors. 



This study has yielded the following results. 



BACTERIA IN SECONDARY TUMORS. 



The bacterium causing the primary tumor occurs also in the sec- 

 ondary tumors. It is not plentiful, but, so far as we have plated, 

 occurs in about the same numbers as in the primary tumor. Xu- 

 merous poured-plate isolations and several successful reinoculations 

 have put beyond question its normal occurrence in these secondary 

 tumors. It occurs also sparingly in the tissue between tumors. In 

 other words, we do not have in this disease a cell stimulus begun in 

 the primary tumors by a parasite and then able to propagate itself 

 independently of the inciting organism. On the contrary, the 

 bacteria and the tumor cells occur together both in the primary 

 tumor and in the secondary tumor mid the one inside the other. 

 Owing to our inability thus far to stain the causal organism inside 

 the cells without at the same time staining many confusing granules 

 or to see it therein under the microscope so as to determine its pres- 

 ence in unstained sections with certainty, the evidence for this last 

 statement is indirect, but none the less conclusive. This evidence is 

 twofold: (a) By the poured-plate method, from the interior of 

 tumor tissue after surface sterilization we obtain the tumor-produc- 

 ing schizomycete, and, therefore, it occurs in these tissues; (b) when 

 such tissues are sectioned and studied under the microscope the ves- 

 sels and the intercellular spaces, so far as we have been able to ob- 

 serve, are free of bacteria. Numerous observations made during the 

 last three months confirm many earlier ones. If bacteria were com- 

 monly present in the vessels or spaces between the cells (as happens 



[Cir. 85] 



