4t2 iSTAiTE BOARD OF AGRICULTURE 



day period, but a decrease OA^er the initial niimiber. At the end O'f two 

 weeks 4 out of 6 samples 'showed a decrease over the one week period while 

 2 showed an increaise over the one week period but a decrease over the one 

 day period. Only two samples w^ere analyzed at the end ol the three 

 week period. One showed an increase over the two week period but a 

 decrease over the initial count while the other showed a decrease over 

 both the two week period and the initial count. 



The general co.U'clusions that one is forced to draw from the data pre- 

 sented in Table VII are to tlie effect that the numbers of bacteria decrease 

 during the storage of ice cream. There may be fluctuations in individual 

 samiples during the process of storage, but the number of bacteria 

 in the end is less than at the beginning. Hammer (2) made deter- 

 minations on twelve samples that had been held from periods rang- 

 ing from 10 to 44 days and reached this conclusion: "From these 

 results it seems safe to conelude that during the period of sto'rage of ice 

 cream there is no increase in the number of bacteria capable of develop- 

 ing on agar." Esten and Mason (7) held several samples of ice cream 

 for about a month and concluded that there is no marked increase or 

 decrease in the number of bacteria. Ellenberger (8) who went into this 

 matter quite extensively reached the following coniclusion : "There is no 

 Tadical change in the total number of bacteria in ice cream during stor- 

 age. There seems, hoiwever, to be a tendency toiwards a slight decrease 

 during the first two to four days, with a more noticeable increase and a 

 corresponding decrease again between the fourth and twenty-first day, 

 after which time there is a very gradual falling off in numbers." 



GENERAL DISCUSSION OF RESULTS 



As was stated in the 'beginning out pla.n w^as to start with the mix and 

 trace it through each procesis in the factor}- until it came out the finished 

 product. The resuKs of the influence of each process on the bacterial 

 count are shown in Tables Villa, Vlllb and VIIIc. The data on the 

 bacterial counts of the mix were divided into three parts. Those data 

 tabulated in Table Villa represent the bacterial counts of those batches 

 whose bacterial count is below 500,000 bacteria per c.c. In Table Vlllb 

 are tabulated batches whose bacterial counts are between 500,000 and 

 990,990 and in Table VIIIc batches w^hose bacterial counts are over 1,000.- 

 000 per c.c. 



The 'Counts were then averaged and the averages plotted in the 

 form of a graph. The number of bacteria per c.c are represented as 

 ordinates, while the abscissae represent the manufacturing operations. 

 The straight line curve represents the results from Table Villa, the 

 broken line curve the results from Table ^^IIIb, and the dotted line curve 

 the results of Table ^aIlc. 



A study of this graph shows some very interesting things. It is inter- 

 esting to note that all three take the same general direction. The broken 

 line graph is veiy irregular. This is due to the fact that few samples 

 fell in this class so that the averages are erratic, which in turn make 

 the grapih irregular. The other two lines, however, show the same general 

 trend upw^ard. The course that these graphs take indicates that each 



