with 2.0 ml. of sterile physiological saline onto a large agar surface such as that provided by a Roux bottle containing 300 ml. of agar. Spread the suspension of organisms over the entire agar surface with the aid of sterile glass beads. Incubate 24 hours at 37° C. and store for 24 hours at room temperature. Wash the resulting growth from the agar surface with about 50 ml. of sterile physiological saline. Standardize this suspension by determining the dilution which will permit 20% light transmission through a filter at 6500 Ångstrom units in a photoelectric colorimeter. Add 1.5 to 2.0 ml. of this resulting dilution to each 100 ml. of agar which has been melted and cooled to 48° C. to prepare the inoculum for the plates. The suspension may be used for one week.

(f) Assay. Use four plates for each sample. Fill one cylinder on each plate with a 1.0 unit per ml. dilution, and one with a 0.25 unit per ml. dilution, of the working standard. Add the estimated dilutions of 1.0 unit per ml. and 0.25 unit per ml. of the sample under test to the remaining 2 cylinders on each plate. Carefully place the plates in racks and incubate 16 to 18 hours at 37° C. After incubation measure the diameter of each circle of inhibition to the nearest 0.5 mm. using a colony counter with a mm. scale etched into the supporting glass over the light source. Other measuring devices equal accuracy may be used.

of

where s and si are the diameters of the zones of inhibition in mm. of the 1.0 unit and 0.25 unit dilutions of the standard, respectively, and u and ur refer similarly to the corresponding dilutions of the sample under test. The value V is the sum of the v values for all plates and W is the sum of the w values for all plates. To estimate the potency locate the point on the chart corresponding to the values of V and W, and the potency can be read from the radial lines on the chart.

(2) The error of the assay is estimated by using the nomograph which requires five values, namely, the potency, V, W, Rv, and Rw. Rv (the range of the v's) is the highest value of v minus the lowest value of v obtained from the individual plates. Similarly, Rw is the difference between the highest and lowest w values. After obtaining these five values, connect with a straight edge the points corresponding to v and w on the respective scales on the right side of the nomograph. Mark with a pin or sharp-pointed pencil the intersection of the straightedge and the diagonal line of the nomograph. Move the straightedge so that it connects the value of Rw on its scale and the diagonal line at the point of the pin. The value for Q is thus determined by the scale value where the straightedge crosses the line labeled "Q". T is obtained by adding the squares of Q and Rv. On the left side of the chart connect the values of T and W with the straightedge and read the value of the ratio (error of assay-potency) where the straightedge intersects the scale of values for the ratio. This value multiplied by the potency equals the percentage error of the assay. The error of the assay calculated here estimates only how closely one assayist can check himself on any given set of dilutions of unknown and standard. It does not include any errors of weighing or errors due to variations in materials or subdivisions of a lot of penicillin.

The chart for determining potency should not be used for determinations of potency lower than 50% or higher than 150% of the standard. If the potency lies outside these limits, the assay should be repeated using a higher or lower dilution. The radial lines on the chart beyond these limits permit a rough estimation of potency from as low as 5% to as high as 1,000% when low values of W are found. If the value of V or W falls outside the limits of the chart, divide both V and W by the same proper number to bring them into the range of the chart and read the potency from the radial lines as before. If 11.4 Rw is greater than W, the slope of the assay does not differ significantly from zero and the assay is invalid. (The figure 11.4 was obtained by use of Student's "t" test for determining the significance of a slope.)

In certain laboratories it has been noted that with the 4 to 1 ratio, involving concentrations of 0.25 unit for

the low dose, the zone of inhibition given by this dose may either be too small for accurate reading or have edges which are poorly defined. In order to permit the use of a higher concentration of penicillin for the low dose the third of the attached charts (Chart 3) may be used in assays in which the ratio of doses is 2 to 1, i. e., the high dose (sH) is twice the low dose (SL). As in the preceding chart (Chart 1), if the potency lies outside the limits of 50% to 150% the assay should be repeated, using a lower or higher dilution. The potencies beyond these limits are to be used for rough estimation purposes only. These extensions can also be used for four (or more) plate assays if both V and W are divided by the same proper number to bring them into the range of the chart. The error of the assay using the ratio of doses 2 to 1 is estimated by using the nomograph (Chart 2) in the same manner as described for the 4 to 1 ratio of doses. However, the resultant error of the assay derived in this manner must be divided by 2 to give the correct error of the assay for the 2 to 1 ratio of doses.

(h) Standard curve technique. The potency of a sample may also be determined by the standard curve technique using a single dose of standard and unknown.

Dilute the sample to be tested to 1.0 unit per ml. (estimated) in 1% phosphate buffer pH 6.0. Place six cylinders on the inoculated agar surface so that they are at approximately 60° intervals on a 2.8 cm. radius. Use three plates for each sample. Fill 3 cylinders on each plate with the 1.0 unit/ml. standard and 3 cylinders with the 1.0 unit/ml. (estimated) sample, alternating standard and sample. Incubate the plates for 16 to 18 hours at 37° C. and measure the diameter of each circle of inhibition. At the same time prepare a standard curve using concentrations of the standard of 0.6, 0.7, 0.8, 0.9, 1.0, 1.1, 1.2, 1.3, 1.4, and 1.5 units/ml. in sterile 1% phosphate buffer pH 6.0. Use three plates for the determination of each point on the curve, a total of 27 plates. On each of three plates fill 3 cylinders with the 1.0 unit/ml. standard and the other 3 cylinders with the concentration under test. Thus there will be 81 one unit determinations and 9 determinations for each of the other points on the curve. After the plates have incubated read the diameters of the circles of inhibition. Average the readings of 1.0 unit/ml. concen

tration and the readings of the point tested for each set of 3 plates and average also all 81 readings of the 1.0 unit/ml. concentration. The average of the 81 readings of the 1.0 unit/ml. concentration is the correction point for the curve. Correct the average value obtained for each point to the figure it would be if the 1.0 unit/ml. reading for that set of three plates were the same as the correction point. Thus, if in correcting the 0.8 unit concentration, the average of the 81 readings of the 1.0 unit concentration is 20.0 mm., and the average of the 1.0 unit concentration of this set of 3 plates is 19.8 mm., the correction is 0.2 mm. If the average reading of the 0.8 unit concentration of these same 3 plates is 19.0 mm. the corrected value is then 19.2 mm. Plot these corrected values including the average of the 1.0 unit/ml. concentration on 2 cycle semilog paper using the concentration in units per ml. as the ordinate (the logarithmic scale) and the diameter of the zone of inhibition as the abscissa. Draw the standard curve through these points. The 10 points selected to determine the curve are arbitrary and should be so chosen that the limits of the curve will fill the needs of the laboratory. However the potency of the sample under test should fall in the interval of from 60% to 150% of the correction point of the standard curve.

To estimate the potency of the sample average the zone readings of the standard and the zone readings of the sample on the three plates used. If the sample gives a larger average zone size than the average of the standard, add the difference between them to the 1.0 unit zone on the standard curve. If the average sample value is lower than the standard value, subtract the difference between them from the 1.0 unit value on the curve. From the curve read the potencies corresponding to these corrected values of zone sizes.

(i) Potency. The potency of sodium penicillin, calcium penicillin, and potassium penicillin is satisfactory when assayed by the methods described in this section if the immediate containers are represented to contain:

(1) 200,000 units or less and contain 85% or more of the number of units so represented;

(2) More than 200,000 units and contain 90% or more of the units so represented.