Order 1. The petitioner, who has been convicted for selling adulterated baffalo's milk, is the revision petitioner after having lost in appeal before the Sessions Court. 2. The Food Inspector, P.W.1, purchased from the petitioner of 16.2.1979 675 ml. of baffalo's milk. The milk was analysed and Ex. P-8 is the report of the analyst. The result of the analysis and his opinion as shown in Ex. P-8 read thus: “Specific grabity at 30°PC - 1.025 Milk fat - 5.3 percent Milk Solids not fat - 7.0 percent Starch and Sugar - Absent and am of the opinion that the said sample does not conform to the standards prescribed for Buffalo Milk under the Prevention of Food Adulteration Rules, 1955 and is therefore adulterated. I am further of opinion that the same contains not less that twenty two percent (22.0%) of added water as calculated from the Milk Solids not fat content. No decomposition had taken place in the sample that would interfere with the analysis.” 3. Under Appendix 11-01-011, baffalo milk should contain 5.0 percent of milk fat 9.0 percent of milk solids-not-fat. The analyst found that the sample did not conform to the standards. 4. Learned Counsel for the Petitioner raised two contentions. He contended that the milk is a primary food and the minor variation from the prescribed limits so far as milk solids-not-fat and milk fat are concerned cannot be attributed to any human agency and could only be due to natural causes and thus he was entitled to the protection of the proviso to Section 2(m) of the Prevention of Food Adulteration Act. 5. The second contention is that the Analyst, without conducting the necessary test, has found that there was added water only on the basis of his calculation of the milk solid-not-fat contained as seen from Ex. P-8. This cannot be accepted. 6. I think both the contentions will have to be accepted. 7. 5. The second contention is that the Analyst, without conducting the necessary test, has found that there was added water only on the basis of his calculation of the milk solid-not-fat contained as seen from Ex. P-8. This cannot be accepted. 6. I think both the contentions will have to be accepted. 7. Section 2(m) with the proviso omitting the Explanation of the Act reads thus: “2(m): If the quality or purity of the article falls below the prescribed standard or its constituents are present in quantities now within the prescribed limits of variability but which does not render it injurious to health: Provided that, where the quality or purity of the article, being primary food, has fallen below the prescribed standards or its constituents are present in quantities not within the prescribed limits of variability. In either case, solely due to natural cases and beyond the control of human agency, then such articles shall not be deemed to be adultered within the meaning of this sub-clause.” 8. It is clear that the onus to prove that the proviso applies is on the accused. But in a case of this type where the report shows that milk contains milk solds-not-fat and milk fat either slightly above or below the standard prescribed under the Act, it has to be inferred that it was due only to natural causes and was beyond the control of human agency. The proviso, therefore, applies and the benefit has to go to the accused. The milk sold to P.W.1 cannot therefore the deemed to be adulterated on the ground that it contained milk fat and milk solids-not-fat in quantities not within the prescribed limits of variability. 9. The report Ex.P-8 shows that the conclusion that sample contains not less than 22 percent of added water was arrived at solely on the basis of the quantity of the milk solids-not-fat content. This Court in the decision reported in FoodInspector, Cochin Corporation v. Kumaran Rajan FoodInspector, Cochin Corporation v. Kumaran Rajan (1976) K.L.T. 74 has held that the Public Analyst should have adopted freezing point test which has been accepted as the more dependable method for finding out whether there is added water in the particular sample of milk. This Court in the decision reported in FoodInspector, Cochin Corporation v. Kumaran Rajan FoodInspector, Cochin Corporation v. Kumaran Rajan (1976) K.L.T. 74 has held that the Public Analyst should have adopted freezing point test which has been accepted as the more dependable method for finding out whether there is added water in the particular sample of milk. This decision refers to the earlier decisions of this Court in Gopinathan Nair v. Palani Gopinathan Nair v. Palani (1971) K.L.T. 248 and in State of Kerala v. Vasudevan Nair State of Kerala v. Vasudevan Nair (1974) K.L.T. 617. If inspite of these decisions the Analyst does not take care to conduct the freezing point test to ascertain whether the sample contain added water or not, the prosecution cannot still succeed and then claim a conviction. It is un understandable as to why the Analyst does not conduct the freezing point test inspite of the decision referred to. It is necessary to bring this to the notice of the Director of Prosecutions for such action as he may deem fit to avoid the recurrence of such mistakes hereafter. 10. In reversal of the judgment of the two Courts the accused is to be acquitted. Conviction and sentence are thus set aside and the revision is allowed. Revision allowed.