H,— 34,

(i) Contamination from Milk-strainers.—ln order to determine the extent of contamination of one supplier's milk from that of another by the strainers used over the weighing-vat 111 cheese-factory practice a number of trials were made at the Institute factory using both cloth and metal strainers. The results showed that if good-quality milk followed inferior milk through a strainer the amount of standards at present in force. Buttekmaking Investigations. la) Butter-flavour —Flavour in butter is considered to be due mainly to the presence of traces of diacetyl derived from the natural oxidation of a flavourless precurser-carbmol-which is produced by the starto organisms. Starters in use in New Zealand dairy factories have been found to be similar to those used in the United States in regard to their production of these substances. The amount of carbinol and diacetyl found in butter is dependent 011 the amounts added 111 the staiter, as "tHe "Si" j«™lop e d in «he 0re.,,, .taring fc - TSTS S5 S this countrv Diacetyl tends to disappear from butter held m cold storage at 150 land most o the loss occurs during the early part of storage. The precursor (carbinol) content remains unchanged during storage. , arbino l-diacetyl substances is dependent on the degree of development of aciditv bv starters they are not acidic substances. It has therefore been found possible to produce butters containing relatively large amounts of carbinol, with othem of » l-int+er made from " low acid " cream. The flavour, however, is still mild owing to tiie tact mai but little of the carbinol has changed to the flavour-substance, diacetyl. Present work is mainly directed towards the possibility of inducing some of the carbinol present 111 cream or batter to change over to the diacetyl form. If this is found to be practicable it might be possible to manufac ure highly flavoured butters of low acidity and consequent good-keeping qua 1 y. (b) Comvosition of Butter.-Butterfats from Friesian and Jersey cows were collected at intervals 1 ' +1 um seasons and some of their chemical characteristics have now been determined. There is also a difference in composition from day to day which may be due to either weather COnd Srwork e is d preliminary to a more extended investigation which is to be made next season into the variations in Se composition of butterfat in New Zealand 111 relation to feed, breed and period of lactation It is considered that the keeping-quality, spreadability, and vitamin-content are related to the composition of the fat, but little is known with regard to the relation of these factors to Neutralization of Cream and Flavour of Butter.-In 1924 attention was drawn by Mr. Valentine siTS % 1 ;r v t "«W &TZ tataeu developed, 'f, » wrlfta. pltaH ££ Sta U P—,« to tile industry e.rly in the COmU (d) f Timber-taint —A previous small-scale investigation on the treatment of Pmus radiata for' the out both tawa and rimu boxes. Two different systems of treatment were used . (1) The Wiley double-spray treatment as used in Australia. (2) A proprietary single-spray treatment, developed m New Zealan . The results may be stated briefly as follows. (1) The taint from Finns radiata boxes was not adequately controlled by either system and boxes of this timber, even after the application of the surface-treatment, were not suitable containers fOT boxes after treatment still imparted to the surface of the butter a certain distinctive flaVo (3) the miller in the Manawatu district experienced coneffect on the h.vonr of the of the butter 111 this respect they were a uite the e,u.ls of the .h.te-pme boxes used » controls. They 6ioW (t KSd''oK.'SSl" ta StiUt T ".s P.mt is no, being further tested ° Ut " Butter packed in boxes treated according to either of the above systems was contaminated with'LX offc sulco, but only to . very slight extent. It is possible th.t kiln-drymg of the shooks after treatment, as is practised in Australia, would eliminate this trouble.

19