GENERIC CHARACTERS IN THE COCCACEAE 185 



groups of organisms. The first group, including the forms with 

 faint or meager surface development, corresponds to the genus 

 Streptococcus as defined above. It is sharply characterized by high 

 acid production and the absence of gelatin liquefaction or nitrate 

 reduction. Cocci of this type are characteristically parasitic, and 

 very rarely show the sarcina grouping. On the other hand, the 

 more vigorous forms are generally saprophytic, and frequently 

 show packets. They ferment sugar slightly or not at all, and often 

 reduce nitrates and liquefy gelatin. 



5. FERMENTATION or CARBOHYDRATES. 



In measuring the amount of acid produced in dextrose and lactose 

 broth, two check determinations were made in each case, and the 

 figure finally recorded was the average of these two determinations. 

 The correspondence between the two tubes was generally close. 

 From 150 cases for each sugar we have calculated the probable error 

 of a single observation, and find it only it o . 043 for dextrose and 

 o . 036 for lactose. Since our readings were only taken to loths 

 of a c.c., it is evident that even a single determination would be 

 sufficiently accurate for any long series. 



The general results of the titrations are shown in Table 25 and 

 in Fig. 2. It will be noticed that with both acids the organisms are 

 ranged with fair regularity about a single mode. The majority 

 of the cultures studied produce an acidity of 0.1-0.2 per cent 

 normal in dextrose, and fail to ferment lactose at all. In either 

 case a few cultures only show an alkaline reaction, and with lactose 

 less than half of the organisms form acid, giving the curve for that 

 sugar a very acute form. The curve for dextrose falls off much 

 more slowly, and shows slight secondary modes at acidities of 

 0.5-0.6 per cent normal and 1.3-1.4 per cent normal. Finally, 

 both curves show an extraordinary extension in the direction of 

 the higher acidities. It will be noticed that for each sugar several 

 of the highest reactions, ranging from 3 to nearly 10 per cent normal, 

 have been omitted from the chart. We may fairly consider the 

 fermentation of the two carbohydrates together, since, as shown 

 in Table 26, they are very closely correlated. The amount of acid 

 produced in lactose broth is almost always less than that in dextrose 

 broth, but the two vary together. 



