In the handling of the large starter, care should be used not to overripen, since the larger quantity is more likely to "whey off" than is the smaller starter. Skim milk rather than whole should be used for this. It should be selected with care and heated to 200° F. for thirty minutes. When it is impossible to secure fresh milk for starter making purposes, either condensed skim milk or milk powder may be used. The condensed milk is diluted with water until its volume is about the same as the milk before concentration; the mixture is then treated the same as fresh milk, being heated and cooled before inoculation. In the case of milk powder, one part of the powder is added to ten or twelve parts of water, allowed to dissolve as far as possible, and the mixture heated and cooled. Either of these liquids will give satisfactory starters; the cost however is high, and in most places milk can be obtained more cheaply. The inoculation and the temperature should be so controlled, as to ripen the starter at the time it is to be needed. These conditions must be determined by the maker for himself. It should be remembered that the bacteria grow much more rapidly, as the temperature is increased; and hence, the amount of inoculation is dependent on the temperature at which the starter is to be kept. When the starter is propagated under practical conditions, it sooner or later deteriorates, either in acid production, or in flavor, and a new pure culture must be procured from the manufacturer. It is impossible to give a hard and fast rule as to the length of time a starter can be kept in good condition. It will depend on how well the maker satisfies the conditions necessary for maintaining its purity and strength. The use of imperfectly sterilized milk, or dirty utensils soon contaminates it; overripening is likely to injure the flavor. One of the most frequent troubles encountered is the appearance of a slimy or ropy condition in the starter, although the acidity developed may be normal and the flavor desirable. It has been found that this condition is not necessarily due to contamination, as was considered true in the past, but rather to some change in the lactic bacteria themselves. If the propagation is continued, the slimy condition will often disappear.