8 BULLETIN 581, U. S. DEPARTMENT OF AGRICULTURE. 
were used for other products. The followimg comparison of the 
counts on them is, therefore, of interest: 
Fields | Yeasts 
with and Bacteria. 
molds. | spores. 
Per 1/60 | Million 
Percent.| cmm., Per ce, 
Catsup made from trimmings of tomatoes peeled for 72d............ Ba od 90 143 3 
Pu!p from tzimmings of tomatoes peeled for 84d -_.-.............----5----- 109 63 €0 
Puip from trimmings of well-sorted tomatoes pee:ed for 73a...-.....-.---. 20 13 9 
The contrast between the character of the first two and the last 
is a good illustration of the counts obtained on such types of products 
handled as these were. 
RELATION BETWEEN PERCENTAGE OF ROT AND MICROSCOPICAL 
COUNTS. 
During the earlier investigations made upon tomato products the 
character of the stock was usually determined by means of visual 
inspection only. In later experiments the counts were made on 
products in which the amount of rot in the raw stock was determined 
by weight. These experiments were divided into two groups. One 
group of samples was prepared in the laboratory from stock of known 
character, while the other group consisted of inspected stock made 
in the factory. ‘‘Decayed”’ or ‘“‘rotten’’ stock is taken to mean 
-such portions as would appeal to the average intelligent housewife 
or consumer as being decayed or rotten and unfit or undesirable for 
food. For purposes of study of the molds the two groups are con- 
sidered separately. 
MOLD COUNTS. 
LABORATORY SAMPLES. 
In tests conducted in the laboratory the amount of rot in the whole 
tomato was determined by cutting out and weighing the decayed 
portion and calculating the percentage. These portions were then 
mixed and pulped by rubbing through a 20-mesh sieve. In a few 
cases this procedure was modified by pulping the good and bad por- 
tions separately and then mixing them in the proportions desired. 
Experience has shown, however, that such a pulp is usually not quite 
as satisfactory to work with as the products normally manufactured 
because it is somewhat difficult to produce in the laboratory a pulp 
of just the same texture as that made under good factory conditions. 
The experiments and counts are nevertheless incorporated in this 
bulletin, because when compared with the counts on pulp made under 
factory conditions they show that the same general relationship 
between the percentage of rot and counts is maintained. The 
