II30 CHEMOTHERAPY OF BACTERIAL DISEASES 



observations do not warrant the amount of clinical enthusiasm that has been aroused. 



Creosote, guaiacol, and their derivatives have not given much if any evidence of 

 specific bactericidal effects in vivo, and the results of animal tests have been generally 

 of a negative character. 



Cinnamic acid and the cinnamates, bile and the bile salts, fats and fatty acids, 

 and various hydrocarbons have generally yielded negative results in experimental in- 

 fections, although cod liver oil may owe its curative properties in human tuberculosis 

 to vitamines which bring about an accelerated rate of recalcification and an increase 

 of inorganic phosphorous in the blood in both rickets and tuberculosis in much the 

 same manner as heliotherapy, without involving the question of specific tuberculoci- 

 dal action from the oil, which is doubtless a very remote and improbable effect. 



COMPOUNDS OF ARSENIC, MERCURY, AND OTHER SUBSTANCES IN THE CHEMO- 

 THERAPY OF EXPERIMENTAL STREPTOCOCCUS, PNEUMOCOCCUS, 

 STAPHYLOCOCCUS, AND OTHER INFECTIONS 



Much less investigation has been devoted to the subject of the chemotherapy of 

 experimental pneumococcus, streptococcus, staphylococcus, and other bacterial infec- 

 tions with compounds of the metals, although this is a very important field worthy of 

 intensive study because of the frequency and highly fatal character of these diseases. 



Arsphenamin and neoarsphenamin may exert a feeble curative effect upon experi- 

 mental streptococcus infections, but these effects are not appreciable in the presence 

 of infections sufhciently virulent to maintain the organisms in the blood. Sodium 

 cacodylate has proved without appreciable effect in my experience in doses as high as 

 0.005 gn^- psr kilo (corresponding to 5 gr. per 60 kilos). 



Arsphenamin and neoarsphenamin, especially the former, possess in slight degree 

 an appreciable bactericidal activity for the anthrax bacillus in vivo, but the curative 

 effects are so feeble that it is doubtful if these compounds are capable of favorably 

 influencing anthrax bacteremias in man. 



I have also employed arsphenamin, neoarsphenamin, mercurophen, and mercuro- 

 chrome in experimental pneumococcus infections of mice and rabbits, but with nega- 

 tive results, although occasionally the lives of mice treated simultaneously with 

 inoculation or within two hours thereafter were prolonged from one to three days 

 beyond the controls. 



Better results, however, were observed by Harkins and myself with mice infected 

 by subcutaneous injections of hemolytic streptococci, according to Morgenroth's 

 method. As a general rule we inoculated the mice every three days and produced 

 thereby a streptococcic cellulitis with infection of the bloodstream lasting from seven 

 to ten days and more closely duplicating conditions as encountered among human 

 subjects with streptococcus infections. Mercurochromc was given by intraperitoneal 

 injection in doses of 0.005 Z^- P^i" kilo, at daily intervals and four hours after each 

 inoculation. Controls receiving the drug alone lived indefinitely; untreated controls 

 succumbed with severe streptococcus cellulitis and bacteremias whereas 30 per cent 

 and sometimes more of the treated animals lived indefinitely and showed no local 

 lesions at the site of inoculation. In my experience this technique has proved the most 

 satisfactory of all employed, and I believe that mercurochrome demonstrated unmis- 

 takably some degree of curative activity. 



