Figure C does show an inverse relationship. This is not changed by it being wrong for the situation and wrong at the axis.
As for A, it has a characteristic shaped curve, much like radioactive decay which also never reaches zero. It is not correct to say the value at zero is undefined. The relationship approaches zero but never actually reaches zero. There is no value at zero defined or otherwise.
Apologies to the collective for continuing the maths thread, but I will push further. And no, this is not objecting for objecting's sake.
Loosely in not-so-strict speech, if one quantity goes up whilst another goes down, people tend to say this is an "inverse" trend. For example, people may say that the price of an article will vary inversely with its supply to the market (i.e. low supply, higher price, and vv.), though in practice this is not necessarily in perfect proportion.
However, at least in strict mathematical speak, and more often than not in scientific circles, there is one definition of 'inverse relationship', which is as I described and is demonstrated in Figure A. This has nothing to do with the specific situation in the question - this is a general definition. Inverse relationships (or strictly "inverse proportion") are not defined ("no value") when either variable is equal to 0.
Figure C is a linear trend; because it has a negative gradient, it is called a negative linear trend.
Figure A does look a lot like the negative exponential trend/curve. That said, one subtle difference between the two curves in question is that a negative exponential
does have a defined value when the independent variable is equal to 0. This makes sense, e.g. in your example of radioactive decay, if the radioactive concentration is on the vertical (dependent) axis and time is on the horizontal, then when time = 0, there is a value for radioactive concentration (i.e. the initial concentration). If we attempted to describe this using an inverse relationship, this would be wrong, because there is no defined value at time = 0. The shapes of each of these curves also generally look quite different until the value of the dependent variable approaches 0.
If you get out a graphing program and plot the graphs of
y = 1/x (inverse relationship) and
y = exp(-
x) or
y = e[sup]x[/sup] (negative exponential), you'll see that Figure A looks closer to the inverse relationship. The negative linear of Figure C looks more like
y = -x, although to produce the graph in Figure C more closely, you would need a vertical shift, i.e.
y = -x + k for some positive value of
k.
In maths, "not defined" and "no value" are pretty much synonymous. For some reason, however, it is proper to use the term "not defined" for an answer which, for all intents and purposes, doesn't exist and we can derive no useful value from, e.g. the result of division by zero. "Not defined" does not mean that there
is an answer and we just don't know / can't agree what it is.
Phew! Maths lesson over (for now

). Anyone with high school kids, feel free to refer this discussion on. And yes, I used to be a high school maths tutor.
For anyone who was absolutely bored out of their minds, just remember that when there's discussion about anatomy on this board (there are many doctors!), even though I know that is more applicable to us, I still have not much idea what you're talking about most of the time
