H—29

During the winter a series of investigations was carried out of various methods of moisture estimations in milk powder to find out the limitations of certain types of ovens, &c., so that alternatives would be available if required. An important result of this, work was that a method was devised whereby a sample of milk powder could be heated for a short time with dry sand in a specially constructed heating bath and the moistureresult obtained within twenty minutes. This has provided a very useful method for factory control purposes. By finding the moisture content of powder soon after manufacture adjustments can be made to the machine to effect improvements needed. Pending the provision of the facilities needed at Wellington and New Plymouth to do the analytical work needed on dried-milk powder being graded for export, samples have been sent for analysis to Wallaceville. Although other work has had to be set aside to enable these samples to be dealt with, some useful experience has been obtained of the value of different methods and of apparatus used for analyses. The range of results has provided a useful indication of several quality problems which need more detailed investigation. Although good progress has been made, there is a need for further investigations of methods for fat and moisture estimations when suitable glassware and other apparatus are available. Besides these basic features of control work, there is also a need for somedetailed investigations of other properties of dried-milk powder which affect quality,, particularly solubility, alkalinity, copper and iron content, and bacterial content. When more staff becomes available it is hoped to be able to study these properties in detail to safeguard future quality requirements. Dairy Laboratory, Auckland. This year, with the commencement of milk drying by 17 factories in the Auckland area, much time has had to be devoted to the analyses of buttermilk, whole-milk, and. skim-milk powders. In the development of new techniques for dried-milk analysis, very little of which had beien done in New Zealand, much of the routine work has had to be reduced. Thisapplies particularly to the bacteriological examination of butter and cream and to chemical analysis of farm and factory water supplies. An attempt was made to analyse one sample representing each twenty-five bags of* powder received. In spite of shortage of equipment, delivery of which is very slow, and a shortage of laboratory space which was not overcome until about half-way through the season, it was found possible to keep abreast of the work offering. Delays in completing the analysis of some consignments did occur, but as experience was gained these have become less frequent. Considerable success has also been achieved with Jefcoate's modification of the Gerber test for fat in milk powders. This has been further modified, with excellent results, and a similar technique which has been applied to the Babcock test has now been brought to the stage when it should be suitable for use in dairy factories. Comparisons with the gravimetric methods show that the accuracy of both Babcock and Gerber tests is satisfactory for grading purposes. Up to 31st March approximately 1,200 samples of skim-milk, 3,900 of buttermilk,, and 500 of whole-milk powders have been analysed for moisture and titratable acidity,, and 1,200 samples of skim-milk, 1,700 of buttermilk, and 500 of whole milk have been examined for fat content. Kegular checks on the bacteriological quality and the acidity of butter passing; through the port have been made on approximately 2,400 samples. This is a slight decrease from last year's figures. Particular attention was paid to shipments of unsalted butter, and as much co-operation as possible was extended to Instructors and to factory managers in their endeavours to reduce the incidence of coliform contamination.

82