There are various means at our command for improving the ripening process. Perfect cleanliness in the entire manipulation necessary in milking and dairying, combined with freedom from disease in the milch cows, will carry us a long way on the road towards a good cream-ripening. Recently, however, a new method has been introduced, largely through the work and influence of Professor Storch in Denmark, which is based upon our new knowledge respecting bacterial action in cream-ripening. We refer to the artificial processes of ripening set up by the addition of _pure cultures of favourable germs_.[66] If a culture of organisms possessing the faculty of producing in cream a good flavour be added to the sweet cream, it is clear that advantage will accrue. This simple plan of _starting_ any special or desired flavour by introducing the specific micro-organism of that flavour may be adopted in two or three different ways. If cream be inoculated with a large, pure culture of some particular kind of bacteria, this species will frequently grow so well and so rapidly that it will check the growth of the other bacteria which were present in the cream at the commencement and before the starter was added. That is, perhaps, the simplest method of adding an artificial culture. But secondly, it will be apparent to those who have followed us thus far, that if the cream is previously _pasteurised_ at 70° C. these competing bacteria will have been mostly or entirely destroyed, and the pure culture, or starter, will have the field to itself. There is a third modification, which is sometimes termed ripening by _natural starters_. A natural starter is a certain small quantity of cream taken from a favourable ripening--from a clean dairy or a good herd--and placed aside to sour for two days until it is heavily impregnated with the specific organism which was present in the whole favourable stock of which the natural starter is but a part. It is then added to the new cream the favourable ripening of which is desired. Of the species which produce good flavours in butter the majority are found to be members of the acid-producing class; but probably the flavour is not dependent upon the acid. Moreover, the aroma of good ripening is also probably independent of the acid production. Of all the methods of ripening--natural ripening, the addition of natural starters, the addition of pure cultures with or without pasteurisation--there can be no doubt that pure culture after pasteurisation is the most accurate and dependable. The use of natural starters is a method in the right direction; yet it is, after all, a _mixed_ culture, and therefore not uniform in action. In order to obtain the best results with the addition of pure cultures, Professor Russell has made the following recommendations: 1. The dry powder of the pure culture must be added to a small amount of milk that has been first pasteurised, in order to develop an active growth from the dried material.