This section is from the book "Experimental Cookery From The Chemical And Physical Standpoint", by Belle Lowe. Also available from Amazon: Experimental cookery.
Custards. As can be deduced from the foregoing discussion, custards containing a high proportion of sugar may not thicken satisfactorily for serving purposes. A small amount of salt aids setting but too much increases the tendency to curdle.
Pie fillings. One question often asked is why butterscotch fillings for pies which are of a consistency for serving sometimes become thin and runny after standing for a short time. This of course is different from the instances in which thickening does not occur. There may be various factors that bring about this result, but the effect of the sugar in elevating the temperature at which egg coagulates is one explanation. The proportion of sugar in butterscotch fillings in different recipes varies from about 15 to 25 per cent, or 15 to 25 grams per 100 grams of filling. A usual procedure is to cook the sugar, cornstarch or flour, and scalded milk until thick. This mixture is often added to the beaten yolks by stirring the hot mixture into the yolks. Sometimes it is considered that the hot milk mixture will coagulate or cook the egg yolk sufficiently. However, if the temperature drops below 80°C. during this procedure and the mixture is not reheated the egg yolks will not be cooked sufficiently. The filling appears thick at the time, but on standing it becomes runny. If sugar is added to uncooked egg yolks, mixed, and left to stand for a few minutes the mixture appears thinner and more runny. If the yolks are not cooked sufficiently they act in much the same way, the sugar dissolving in the uncooked yolk and the filling becoming runny. This has never been observed by the author if the mixture is heated to a sufficiently high temperature after the egg yolks are added. The same thing may occur in chocolate and lemon pie fillings, although this in some instances is due to not using enough starch, the action of the acid on the smaller amount of starch lessening the stiffness of the filling. Another possible cause is the tannin of the brown sugar, because the tannin has a dehydrating effect on many sols. But to date it has not been possible to obtain thinning in the butterscotch filling by adding slightly more tannin than might be found in the sugar.