Excise sampling and provisional levy principles sustain a demand where valid test reports and departmental tolerance are not disproved.

Refusal to retest a third sample did not breach natural justice where the samples were drawn in the assessee's presence and the earlier reports substantially matched. A valid sample drawn under the excise sampling regime could govern yarn cleared until the next sample date, absent proof of any material change in manufacture. The absence of an express testing rule did not defeat reliance on the reports, and a demand was sustainable under the provisional levy scheme even if it cited the wrong rules, so long as the correct statutory basis existed. The departmental 2.5% tolerance was also not shown to be arbitrary, and the levy challenge failed.

---

---

Issues: (i) Whether refusal to retest the third sample vitiated the levy for breach of natural justice; (ii) whether the test result of a sample drawn on one date could govern yarn manufactured and cleared up to the next sample date; (iii) whether the demand could be sustained in the absence of a specific statutory rule for testing samples and despite the levy being cited under wrong rules; and (iv) whether the tolerance of 2.5% adopted by the department was arbitrary.

Issue (i): Whether refusal to retest the third sample vitiated the levy for breach of natural justice.

Analysis: The samples were drawn in the presence of the petitioner and the first two test reports substantially agreed on the relevant count of yarn. The petitioner did not successfully challenge the sample-taking process or the earlier reports. In those circumstances, the further retest of the third sample was treated as unnecessary, and no denial of fair hearing was established.

Conclusion: The refusal to test the third sample did not violate natural justice.

Issue (ii): Whether the test result of a sample drawn on one date could govern yarn manufactured and cleared up to the next sample date.

Analysis: The department was not expected to test every bale or every day. The court accepted that periodic sampling is the practical method of regulation, and the test result of a validly drawn sample can govern production until the next sample, unless the assessee shows that the manufacturing process changed in the meantime. No such change was shown here.

Conclusion: The test reports could be applied to the entire quantity manufactured between the two sample dates.

Issue (iii): Whether the demand could be sustained in the absence of a specific statutory rule for testing samples and despite the levy being cited under wrong rules.

Analysis: The absence of an express rule prescribing the mode of testing did not prevent reliance on the test reports. The power to draw samples under the sampling rule was treated as sufficient to include the allied power to test them. The demand was also held sustainable under the provisional levy scheme, and a wrong reference to other rules did not invalidate the demand because an order may stand if supportable under the correct statutory provision.

Conclusion: The test reports were admissible and the demand was sustainable under the provisional levy provisions.

Issue (iv): Whether the tolerance of 2.5% adopted by the department was arbitrary.

Analysis: The court found no reason to interfere with the departmental tolerance standard, which had been accepted in earlier proceedings as based on experience and testing. The contrary notification relied on by the petitioner did not displace the departmental standard for excise assessment in the facts of this case.

Conclusion: The 2.5% tolerance was not shown to be arbitrary.

Final Conclusion: The challenge to the additional excise levy failed in all respects, and the levy as sustained on the remaining quantity was upheld.

Ratio Decidendi: A valid sample drawn under the excise sampling regime may govern clearances until the next sample, and the resulting differential levy can be sustained on the basis of provisional assessment even if the demand cites the wrong rule, provided the correct statutory basis supports it.

---