168 THE MICHIGAN ACADEMY OF SCIENCE. 



No. 4 was inoculated with bacillus coli ; 



No. 5 was inoculated with bacillus coli and treated with 100 mg. of 

 morphine; 



No. 6 was treated with 200 mg. of morphine and inoculated with 

 bacillus coli. 



All of these flasks were Ihen placed in the incubator at 37°, where 

 they were kept for two weeks. At the expiration of this time they were 

 all autoclaved at 120° for forty-five minutes. A 2 per cent, solution of 

 tannic acid in glycerine was poured into each flask in an amount equal 

 to the mass therein contained. The flasks were then placed in an incu- 

 bator and kept for forty-eight hours at 40°. Then the flasks were re- 

 moved from the incubator, the contents strained through muslin, and 

 the filtrate heated on a water bath at 60° for one-half hour, in order to 

 bring down the coagulable proteids. After filti-ation through paper the 

 filtrates were shaken with twice their volumes of petroleum ether. After 

 separation of the ether, the material was heated in order to drive off 

 traces of the solvent, and then rendered slightly alkaline with sodium 

 hydrate and shaken with chloroform. The chloroform extract from each 

 and every flask promptly reduced iodic acid and potassium peruiianganate. 

 Like results were obtained amjiic alcohol extracts from an alkaline 

 mixture. None of these residues gave either the ferric chloride or the 

 P^roehde test for morphine in a satisfactory manner. Great ditliculty 

 was experienced in obtaining clean pure residues. This was true with 

 both .the chloroform and amylic alcohol extracts, and it was found to be 

 necessary to repeatedly take up the residues with water slightly acidu- 

 lated with acetic acid, and again render alkaline and shake with the 

 solvent. Crystalline residues were finally obtained from only two flasks, 

 one of which contained morphine, while the other did not. The crystals 

 in both of these cases seemed identical microscopically with those ob- 

 tained from an alkaline solution of morphine shaken with chloroform, 

 and the chloroform extract evaporated. The crystalline residue was so 

 small in amount that the possibility of dete^'mining the melting point 

 was excluded. It must be evident from these results that a satisfactory 

 method of extracting morphine in medico-legal examinations is not fur- 

 nished by Kippenberger, and vrhile the work herewith reported is nega- 

 tive in character, it seems justifiable to assume that until we have more 

 definite konwledge of the putrefactive bodies giving morphine-like re- 

 actions, it will be impossible in most cases to determine the presence 

 of morphine by chemical means. 

 M. W. Clipt, 



Hygienic Laboratory, University of Michigan, 

 Ann Arbor. 



