When writing a user defined equation, does the prefix <B> mean data set B or the second analyzed data set?

The latter. Here is a sample equation:

  <A>Y=1/(1+Ka*X^h)
  <C>Y=(Ka*X^h)/(1+Ka*X^h)

The equation above seems clear at first. The first line is for data set A, and the second line is one for data set C. But what if you asked Prism (via choices on the right half of the Analyze dialog) to only analyze data sets A and C, skipping B? Now it is a bit ambiguous. Does <C> mean data set C, or does it mean the third data set included in the analysis? Prism uses the second definition. So you need to enter the equation like this.

  <A>Y=1/(1+Ka*X^h)
  <B>Y=(Ka*X^h)/(1+Ka*X^h)

Now the second line is for the second data set, which in this example is data set C (since only A and C were selected in the Analyze dialog).