574 EXPERIMENT STATION RECOED. 



under aseptic conditions until defecation toolc place. The freshly deposited 

 feces were emulsified in a drop of sterile water; from this emulsion slides were 

 prepared, and cultures were made on bile-salt neutral red lactose agar and in 

 litmus milk. When colonies of bacilli formed on the bile-salt medium they 

 were picked off and subcultures were made in litmus milk, gelatin, and peptone 

 solution and in fermentation tubes containing 1 per cent solutions of glucose, 

 lactose, and saccharose in peptone water. Incubation was conducted in all cases 

 at 37° C. 



" Cultures were made from 17 specimens ; of these, 5 produced colonies of 

 bacilli upon the bile-salt medium with subculture results as follows: Four 

 produced acidity and clotting of milk, acid and gas in glucose, lactose, and 

 saccharose, and production of indol ; the bacilli were Gram-negative, and did 

 not liquefy gelatin, thus corresponding in their reactions with BaciUns lactis 

 (fTOffcncs. One specimen produced gas in glucose and lactose, liquefieil gelatin, 

 and coagulated milk with subsequent peptonization, thus corresponding with 

 B cloavw. 



" In 5 cases after acidification had been brought about by fecal emulsion 

 inoculated into litmus milk, the development of greenish molds of the Asper- 

 gillus variety occurred, and in 1 case the spores of the fungus were present 

 in sufficient numbers to be easily identified by the microscope in the freshly 

 made fecal emulsion. 



" In order to prove the transmission of known organisms, I fed several 

 specimens upon the sputum from a tuberculous patient, upon pus containing 

 staphylococci, and upon spores from cultures of an Aspergillus fungus. In the 

 first case the insects were preserved until defecation took place, and after the 

 feces had been examined and proved free from acid-fast bacilli, the cockroaches 

 were fed upon the tuberculous sputum, which they all devoured voraciously. 



" The B. tuherculosis was present in feces passed within 24 hours of feeding 

 and was readily identified by the Ziehl-Neelseu staining reaction. In 1 case 

 an insect vomited salivary fiuid within a few minutes of feeding and died 

 within a few hours, but whether or not death was brought about by indigestion 

 of the morbid material I am quite unable to say. 



" Two specimens which fed upon staphylococcic pus showed staphylococci 

 in their feces, and cultures upon agar were obtained from these. Four speci- 

 mens fed upon morsels of bread dusted with the spores of Aspergillus fungus 

 showed the spores in great numbers in the feces some few hours later, and 

 cultures of the fungus grew luxuriantly upon milk which was inoculated with 

 these feces." 



These experiments indicate that cockroaches, and probably other insects, may 

 be active agents in souring milk and in disseminating molds and other patho- 

 genic bacteria in pantries and cellars. 



Dairy fundamentals, O. Erf and O. C Cunningham (Columbus, Ohio, 1911, 

 pp. 96). — A quiz compendium, accompanied by laboratoi-y exercises, on milk 

 and milk products, and testing milk, cream, butter, and cheese. 



The application of the theory of errors to investigations on milk, S. H. 

 Collins {Jour. Bd. Agr. {London], 1911, Hup. 7, pp. JfS-oo, fig. 1). — ^A demon- 

 stration of the value of the theory of errors in testing milk, and a discussion 

 of the relative ac-curacy of different methods of determining the specific gravity 

 in estimating the amount of fat and total solids in milk. 



Milk and other liquids were first tested with a 20 cc. plummet on a balance 

 turning to a tenth of a milligram, and then tested with the lactometers. The 

 error of the plnniuiet was about ±0.00003, a small amount compared with from 

 ±0.00018 to ±0.002 in the errors of the lactometers. 



