THE AROMA OF BUTTER 181 



demonstrated by the researches reported by P. SCHUPPAN (I.); and BENNO 

 MABTINY (II.) has drawn attention to the hygienic advantages attending such a 

 method of working. Finally, it was proved by POPP and BECKEK (I.) in 1893 

 that butter prepared from sterilised cream keeps better than that from Pas- 

 teurised cream, and far better than that from cream which has not been heated 

 at all. 



142. The Aroma of Butter. 



The results of Weigmann's researches up to the present (according to a 

 private communication) indicate that probably only a single species though 

 appearing as numerous varieties of bacterium sets up the lactic fermentation 

 now in question. This organism is a coccus (described by W. STORCH (T.) ), 

 measuring about i p. in diameter, and uniting to form chains. The varieties 

 (also called races] of this coccus, which, from the point of view of the systematic 

 botanist, do not differ sufficiently to be classified as separate species, generally 

 exhibit one or other of the following good qualities : they either give rise to a 

 powerful aroma, which imparts a very fine flavour to the butter, or else the 

 product, without exhibiting any marked excellence of flavour, is endowed with 

 good keeping qualities. Consequently such races or varieties should be used in 

 the manufacture of butters for export. Whether there are races in which both 

 these advantages are combined cannot yet be definitely asserted. The aroma 

 produced by the bacteria cannot have originated in the volatile acids of the 

 butter, since it is also developed in cultures free from fat, and containing no 

 nitrogenous nutriment other than peptone. Comparative experiments on the 

 flavour of butters prepared by the aid of different varieties of lactic acid bacteria 

 were made by ADAMETZ and WILCKENS (I.) in 1892. Owing to a noteworthy 

 discovery effected by H. W. CONN (I.) in 1895, the question has latterly taken a 

 new turn. He succeeded in isolating from a sample of South American milk 

 a fission fungus named Bacillus No. 41, which is not one of the acid bacteria, 

 but produces in milk and cream a fine aroma, identical with the highly-prized 

 flavour known in North America as " grass flavour " or " June flavour," because 

 it is produced only in the month of June, at a time when the cows are foddered 

 on tender grass rich with blossom and perfume. Cream inoculated with this 

 bacillus yielded butter endowed with this fine grassy aroma. By means of this 

 process butter ran now be produced with a uniform degree of excellence and 

 marketable value. The mode of working with this bacillus is simple. A culture 

 of the organism in milk is procured, the usual volume being about ^ litre (nearly 

 half a pint), and is poured into about 6 litres (1.32 gall.) of Pasteurised and re- 

 cooled cream. After a lapse of a couple of days the whole is transferred to the 

 bulk of the (fresh) cream, which is left for twenty-four hours and will then be 

 ripe for churning. About 6 litres of this ripe cream are reserved for inoculating 

 the next batch in the same manner as before. It will be noticed that the bulk 

 of the cream is not heated, and consequently the lactic acid bacteria therein will 

 be still living and capable of souring the cream, whilst the Bacilh's No. 41 acts 

 concurrently and develops the aroma. The latter organism, however, retains 

 the upper hand, having been initially present in excess. This flavour-developer 

 has now been tested and proved in more than a hundred North American dairies, 

 so that its employment can be recommended. Naturally, fresh cultures must be 

 introduced into the dairy at intervals (two to three months), since otherwise the 

 bacterium gradually loses its powers. 



The discoverer of the organism attributes to it the additional faculty of 

 remedying defective butters, but on this point the data at hand are insufficiently 

 conclusive. It should be mentioned that the microbe appears in the form of 

 non-motile short rods, 0.7 /x broad and i.i (* long, generally united in pairs, but 



