Changes in the optical density of the microtitre wells were measured over a set of period, recorded automatically and plotted graphically to give a growth curve for each well (fig. 1).

Wells containing vitamin B12 showed no change in absorbance over a short period of time at the start of the test (a). This initial lag phase was followed by a rapid increase in absorbance with time (b) and then a levelling off as no further increase in absorbance was recorded (c). This maximum absorbance level was dependent on the amount of vitamin B12 present in the sample, those with a high concentration of vitamin giving a higher maximum absorbance than those containing low levels. When no vitamin B12 was added to the media only a very gradual increase in absorbance was recorded, giving a very low maximum absorbance. Wells containing neither vitamin B12 nor bacterial cells also showed a very gradual increase in absorbance. In an effort to obtain a flat base-line from wells containing no vitamin B12, cells were incubated in a candle jar for 18 h at 37 oC in 5 ml of assay medium after centrifugation but before adding to the test wells. This should have depleted all B12 in and around the cells. Results were similar to those for cells incubated for 4 h at room temperature. For each vitamin assay, results for 10 replica wells for each vitamin concentration were averaged. Calibration graphs were constructed comparing the concentration of vitamin B12 to:

(a) Area under the growth curve (fig. 2)

(b) Maximum absorbance (max) (fig. 3)

(c) Maximum speed of mass growth (m.g.) (fig. 4)

Each calibration gave a very good correlation and showed a linear relationship between vitamin concentration and each of the three absorption parameters (a, b or c) at concentrations in Bioscreen wells of 0-0.04 ng ml-1 vitamin B12.

Wells containing 0.000 and 0.002 ng ml-1 showed very little difference between results (fig. 5).

At concentrations of 0.05-0.1 ng ml-1 only small increases in measured absorbance parameters were recorded with increasing concentrations of vitamin over measurement periods of up to 20 h (fig. 6, 7 and 8).

When the measurement period was extended to 38 h, there was no further increase in maximum absorbance (fig. 9) and results for area under the curve did not improve (fig. 10). There was a decrease in maximum speed of growth as vitamin B12 concentration was increased from 0.06-0.10 ng ml-1 (fig. 11).