The figures contained in these regulations were not intended, either literally or by implication, to be taken as standards for milk. A milk which contains less than 8·5 per cent. of solids‐not‐fat is not necessarily adulterated—one that contains 8·5 per cent. or more is not necessarily genuine. All that the regulations do is to move the onus of proof. In the case of the prosecution of a vendor of milk for a sample which contained 8·5 or more of solids‐not‐fat the Local Authority would have to prove that the sample was adulterated, in the case of a prosecution for a sample which contained less than 8·5 per cent. of solids‐not‐fat, the defendant, in order to escape conviction, would have to prove the milk to be genuine. The weight which has been given to this limit of 8·5 per cent. of solids‐not‐fat has varied considerably. There are those who appear to consider that it is almost an absolute minimum, and that any milk which contains less than this amount is almost certainly watered, whilst others attach little importance to this figure. It may be desirable to interpolate at this point the figures which have been obtained recently on the samples taken in the County of Lancaster. Since the beginning of the year 1930, 5,959 samples of milk have been examined, of this number 121, or 2·0 per cent., have contained less than 8·5 per cent. of solids‐not‐fat. By means of some of the methods which are described below each of these deficient samples has been examined for the presence of added water, and it has been found that 102 contained added water, whilst 19 were naturally poor. It follows, then, as far as these samples are concerned, that in the case of herds of cows, only 0·3 per cent. give milk containing less than 8·5 per cent. of solids‐not‐fat. From this it must of necessity follow that the limit of 8·5 is at least a very good sorting test. In fact it is far more likely to fail to detect slightly adulterated milks (containing, say, from 1 to 5 per cent. of added water) than it is to describe milks as adulterated which are in reality genuine but poor. Dr. J. F. Tocher, who holds the position of Public Analyst to many of the Scottish Counties, and who is a very outspoken critic of the methods adopted for the determination of added water, has written on this subject to a considerable extent. The following statement, which was made by him in his 1925 Report, has been brought to my notice § with the suggestion that it should be referred to in this report. Dr. Tocher writes:—“The general conclusion from these results is that it is quite unsound on the part of analysts to express a definite opinion that water has been added to milk, when a sample has been found to be below 8·5 per cent. in solids‐in‐fat.” If such a statement as this merely means that a milk is not necessarily watered if the percentage of solids‐not‐fat is below 8·5 it is, of course, not only correct, but absolutely unassailable; in fact, it is merely putting the limit of the Regulations into other words. To those, however, who are familiar with Dr. Tocher's other writings, it may appear that there is something more than this behind the words used. On many occasions in the past Dr. Tocher has stated categorically that it is not possible to prove by chemical or physical examination that a milk is or is not watered, and that all that an analyst can say is that the milk is below the limit, and leave the interpretation of the fact to others, the final evidence being obtained from those who have handled the milk. Apart from the fact that it is not usual to give undue weight to evidence obtained from a defendant it would be quite impossible to rely entirely on this source, for the reasons given in the following paragraph.