16 BULLETIN 581, U. S. DEPARTMENT OF AGRICULTURE. 
shown that high counts invariably indicate bad stock. From the 
chart it appears then that a count of 35 yeasts and spores in a unit 
volume may indicate 4 per cent or more of decayed material. By 
connecting the outlying points in the chart, the upper limits of the 
‘Zone of Possible Yeast and Spore Counts’’ are obtained. The 
limit of the zone shown beyond 12 per cent of rot has been based on 
the count for laboratory Sample 46. From such a chart it is possible 
to calculate the minimum percentage of rot which a certain count 
represents. | 
BACTERIAL COUNTS. 
Considering now the counts of bacteria in Tables 7 and 8, it is 
found that.in the samples with 1 per cent or less of rot the count on 
bacteria is relatively low. With greater amounts of rot more bac- 
teria may be found, till in Sample 45 a count of nearly a billion was 
obtained. In this case there was also a maximum mold count (100 
per cent). Comparing the results in Table 7 with those in Table 8 
it is noted that the mold count runs up more rapidly im the factory 
than in the laboratory series. With yeasts and spores and with bac- 
teria, however, the reverse is true, that is, the counts mount more 
rapidly in the laboratory series. This is explained by the fact that 
in the laboratory series the tomatoes had not been subjected to the 
mechanical or factory method of washing, which results in eliminating 
much of the soft rot wherein these organisms are most abundant. 
The decay which resists this washing process is more commonly due 
to molds. The other organisms when present are usually, though 
not always, the result of secondary infection, and produce some of 
the very soft types of decay. 
The results of bacterial counts on the laboratory and factory sam- 
ples, tabulated in Tables 7 and 8, have been plotted in Figure 4. 
Below 15,000,000 bacteria the results indicate little as to the amount 
of decay, but above this pomt, up to about 20 per cent of rot, the 
ratio of increase as shown on the chart is about 20,000,000 for each 
per cent of rot. Beyond 20 per cent of rot the upper limit of the 
zone is based on the count on Sample 45 (Table 7). As in each of 
the other charts, the rule holds true here that a low bacterial count 
does not necessarily indicate sound stock, but a high bacterial count 
always indicates bad stock. 
MICROSCOPIC COUNTS ON TOMATO SAUCES AND PASTES. 
Tomato sauce and paste are practically the same kind of product 
* as tomato pulp, but concentrated to a much greater degree. Hitherto 
no opinion has been published by the Bureau of Chemistry concern- 
ing the microscopic counts obtained on the more highly concentrated 
tomato products, when handled in accordance with good factory 
practice. During the season of 1912 a study of these products was 
