THE BACTERIAL CONTENT OF MILK 



to burden these pages with many tables of the numbers of bacteria 

 in London milks, but we may insert here the results of an exami- 

 nation carried out by us in May 1903. Purchases were made in 

 nine shops, five situated in the borough of Finsbury, which is as near 

 the centre of the metropolis as possible, and four others from 

 shops in the four other central districts. In three cases small 

 poor shops were selected, but in the other six good class shops 

 were sampled. The sample was asked for in the ordinary way, 

 and it was not known that it was being taken for purposes of 

 examination. It was collected in sterilised bottles, and examined 

 within a few minutes of being purchased. Ordinary Petri dishes 

 were used, the medium was gelatine, and the dilution was -5-^. 

 The technique described on p. 49 was rigidly adopted. The 

 following table gives the results : — 



* These plates liquefied too rapidly for correct enumeration. 



It should be stated that the samples above recorded were 

 obtained in a haphazard manner, and do not necessarily represent^ 

 the usual quality of the milk in these several districts. Obviously, 

 a very large number of examinations would be necessary to form| 

 any opinion upon the comparative condition of the milk of different 

 districts, and other conditions would also have to be consideredj 

 The table must be taken for what it represents, namely, the 

 bacterial content of nine milks purchased by chance in nine 

 London shops. The figures were checked by three counters. 



It is, however, necessary to comment upon the first sample 

 recorded in the table. This was obtained in a poor class shop in 

 Finsbury, in which no precautions are taken to protect the milk 

 from dust, and yet the returns yielded are the lowest of the series, 

 namely, 340,000 bacteria per c.c, and none of which were liquefy- 

 ing organisms. We therefore suspected the addition of anti- 

 septics. By accident, antiseptics had not been estimated at the 



