They selected white rats for their experiments, for a number of reasons. Rats are small, easy to handle and multiply rapidly. They usually live not more than three years and "280 days suffices to complete the entire period from growth to maturity"; which means that several life cycles can be watched. Not only does the rat come under observation during a period which would correspond to about 60 years in the life of man, but the effects of the diet on the offspring may also be noted. In fact, several generations are included in such a survey.

Now what was to constitute health and what ill-health? In the infant a steady increase in weight is taken as the best criterion of normal development. Of course we cannot always be sure that an increase in weight means just this, but in a large percentage of cases it does; and it becomes a simple way we have of measuring progress. A certain sign of the sick man returning to a state of normality is when he gains weight. We say he "puts on flesh." Under normal conditions the adult neither gains nor loses.

How To "Plot" Curves

Drs. Osborne and Mendel adopted this standard of measuring progress by noting the increase in weight of their rats.

To present their results graphically in a way that a mass of figures never could, they plotted curves in much the way that nurses in hospitals and the "modern" mother do when they wish to keep an easily presentable record of their infants' gain in weight from week to week. Such a nurse or mother first records in her notebook the various dates and corresponding weights of the child. Thus, to take an example, we may find some such record as this:

And so on. Now she takes her square-lined paper and marks off suitable horizontal spaces to indicate time and vertical spaces to indicate weight. A convenient way she finds is to call every seven squares in a horizontal direction one week (7 days), the number "1" at the end of the first seven squares indicating one week, "2" at the end of the first fourteen squares indicating two weeks, and so on. Similarly along the vertical line, since she deals with pounds and ounces, she selects sixteen squares to represent one pound; so that at the end of the first sixteen squares she puts the figure "1," representing one pound; at the end of thirty-two squares she puts "2," representing two pounds; and so on.

At birth (zero week) the child weighed 6 pounds 1 ounce. You travel vertically until you reach the figure 6 and then you go one square more. You make a cross at that point. At the end of the first week the child weighs 6 pounds 4 ounces. You travel horizontally until you reach "1" (one week) and vertically until you reach "6," and then go four squares above that (to represent 4 ounces). You make another cross here. And you continue that from week to week. Then you draw your line through the intersection of your crosses and you get a curve such as is shown in figure 1.

Figure 1. - Growth of an Infant

For an explanation of this chart, see page 60 of text.

What do you gather from such a chart? So long as the curve is up an incline, so long as you are traveling up-hill, the child shows a constant increase in weight, and therefore represents a probable normal development. The steeper the incline the more rapid the increase in weight. But where the slope reverses and goes down, instead of uphill, the curve represents a loss in weight and the child is therefore not developing as well as it should be.

From the curve under discussion (figure 1) you will notice that aside from a loss in weight during the first few days (a regular occurrence), the curve slopes steadily in an up direction until the tenth week is reached. Then there is a decline with the eleventh week. The doctor diagnosed the child's loss of weight as probably being due to insufficient nourishment, or perhaps to unwholesome nourishment, - because the mother was nervous. He suggested a supplementary milk diet. Notice how from the thirteenth week the curve starts up-hill again.

If I have dwelt for some time on this simple example, it is because once the reader understands the significance of such a curve as is represented in figure 1, he will have no difficulty whatsoever in understanding the subsequent charts. And as has already been stated, this graphical method of representation has the advantage over mere figures in that you can take in an entire experiment at a glance.