1. This appeal is brought by way of case stated by the appellant, a superintendent of the Royal Ulster Constabulary, against a decision of a resident magistrate Mr CP McRandal, sitting in Omagh Magistrates’ Court on 2 June 1998, whereby he acquitted the respondent on a charge of driving a motor vehicle on 22 December 1996 after consuming so much alcohol that the proportion of it in his breath exceeded the prescribed limit, contrary to Article 16(1)(a) of the Road Traffic (Northern Ireland) Order 1995. The magistrate held that he was not satisfied that the readings of the Alcolmeter used to ascertain the quantity of alcohol contained in a sample of breath furnished by the respondent were reliable. The issue before the court was whether on the facts proved or admitted he was entitled to reach this conclusion.

2. The facts relating to the incident on 22 December 1996 were set out by the magistrate in the case stated. A police officer on mobile patrol stopped the respondent’s car in Omagh at approximately 6 am, having observed that he was not wearing a seat belt. She formed the view that he had alcohol in his body and asked him to take a preliminary breath test. When this gave a reading which indicated that the proportion of alcohol in his body was above the prescribed limit of 35 microgrammes of alcohol in 100 millilitres of breath, the respondent was arrested and taken to a police station. There an evidential sample of breath was taken at 6.46 am. It was not in dispute that the relevant statutory requirements were observed in the procedure adopted.

3. The two samples of breath taken from the respondent produced readings of 66 and 74 microgrammes of alcohol respectively. Taking the specimen with the lower reading, as must be done under Article 19(1) of the Road Traffic (Northern Ireland) Order 1995, the proportion of alcohol was well above the prescribed limit. The respondent was warned that the matter would be reported with a view to prosecution and released from custody. A summons charging him with the subject offence was issued on 7 May 1997 for hearing on 22 July 1997. The matter eventually came on for hearing on 2 June 1998, when the magistrate dismissed the summons after hearing the evidence. The appellant applied to him by requisition dated 8 June 1998 to state a case. The magistrate stated and signed a case on 1 October 1998 for the opinion of this court, the question posed being –

4. When the matter was first listed before us on 12 January 1999 we formed the view that the question posed did not quite reflect the true issue in the appeal, which was whether it was open to the magistrate on the facts to reach the conclusion which he did. We accordingly sent the case back to him, with a direction to amend it by including his findings on the evidence given by an expert witness from the Forensic Science Agency of Northern Ireland (FSANI) and the conclusions which he formed about it. The magistrate restated the case on 1 March 1999, with a brief addition relating to that evidence. For some reason which remains unexplained the restated case did not come to the attention of the court until much later and the appeal was not listed for substantive hearing until 23 June 2000, a delay which is regrettable.

5. The device which was used to take the evidential samples of breath was a Lion Alcolmeter AE-D3, serial number NI/0050-0161. It was issued to Omagh RUC station by FSANI on 12 November 1996. When the device is used a prescribed series of steps has to be taken:

6. In respect of each such device two logs are kept, one by FSANI and one by the police. The FSANI log records the movement of the device between FSANI and various RUC stations, together with details of all repairs and recalibrations carried out to it by FSANI personnel. The police log is a record of all usage of the machine, whether for purposes of calibration or for measuring a sample of breath furnished by a suspect. At the conclusion of each use of the device it is recorded whether it was reliable or unreliable. If a fault occurs, its nature is sometimes recorded in the log, but generally no description is given.

7. The police log, a copy of which was produced in evidence before us by agreement, shows that between July and October 1996 the machine was being used in Armagh, and that during this period it functioned reliably on almost all occasions. It was then returned for repair or recalibration and issued to Omagh, where its use was first recorded on 1 December 1996. On that date and on 4 December it was recorded as having operated reliably. On 22 December 1996, on the occasion on which the respondent was tested, the log shows the following information:

8. The device was recorded as ‘reliable’.

9. On the next occasion on which it was used on 24 December 1996, the first calibration check reading was 40, the test was aborted and the record read ‘Unreliable’. Later on the same day a reliability test was carried out on the machine. The first subject reading was 32 and the second 20, and the result was recorded as ‘Unreliable’. On 29 December 1996 it was again used on a subject, when the first subject reading was 84 and the second 115. The discrepancy between these readings was too wide, and the result was recorded as ‘Unreliable’.

10. Immediately below that entry there is the impression of a rubber stamp, which reads “FSANI, 24 MAR 1997, REPAIRED”. Below that again there is an entry bearing the date “30.12.96” and recording a reliability test at Newtownards, which was restricted to the first calibration check. The following entry on the line immediately below is dated “30.3.97” and the location is again given as Newtownards. The record reads only “Unreliable, not coming off standby”. There then appears another rubber stamp, reading “FSANI, 18 APRIL 1997, REPAIRED/CALIBRATED”. The following six entries relate to use at Dunmurry between 28 April and 30 August 1997, in respect of which the device functioned properly on each occasion. The resident magistrate comments in the case stated that the entry commencing “30.12.96” is incorrect, which it undoubtedly is. It appears fairly clear, however, that it must have been made after the FSANI stamp was imprinted on 24 March 1997 and that it is the date which has somehow been incorrectly recorded.

11. The relevant entries in the FSANI log read as follows:

12. The amended case stated did not set out the evidence given before the magistrate by Mr Ian William Fulton, a Senior Scientific Officer at FSANI, only his conclusions about that evidence. By agreement between the parties, however, Mr Fulton’s report was furnished to us, it being agreed that this represented correctly the evidence in chief which he gave. In the course of his report Mr Fulton stated:

13. He expressed his conclusion in the final paragraphs of his report as follows:

14. The arguments presented to the magistrate and his conclusions are set out in the case stated (which, regrettably, is not set out as prescribed by this court in Emerson v Hearty [1946] NI 35 at 36) as follows:

15. The passage in italics is the portion which was added by the magistrate when he restated the case.

16. The answer to the question as it is posed in the case is No. The evidence about the analysis of the specimens of breath furnished by the respondent was admissible and it would not be correct as a matter of law to exclude it. If the magistrate considered it unreliable, he could place as much or as little weight on it as he thought fit, even reject it and leave it out of account completely, and it appears that this is what he may have meant in saying that he excluded it. The substantial issue, however, is whether, as counsel for the appellant contended, that evidence should be regarded not as unreliable but on the contrary as so compelling that the only sustainable conclusion was to find the case proved.

17. A court is not obliged to regard the result of a breath test given by such a testing device as conclusive, and evidence may be admitted which tends to give rise to the reasonable inference that it was not reliable. The court may reject evidence relating to a specimen of breath if it comes to the conclusion that the print-out from the particular machine is unreliable: Cracknell v Willis [1988] AC 450. The magistrate correctly appreciated this, and the issue is whether he was entitled to do so on the evidence before him.

18. The critical part of the magistrate’s findings is that italicised by him on the last page of the amended case:

19. We have examined the evidence contained in the logs with minute care and we are unable to see any basis for these conclusions. On all three occasions on which the test is recorded as having been unreliable when the device was in use in Omagh the test cycle was aborted immediately after a false reading was produced. When used on a subject on 24 December 1996 its first calibration check reading was 40, outside the permitted tolerance, and the test was aborted. When tested for reliability later that day it went as far as “Subject 2 reading”, which at 20 was more than 20 per cent below the first subject reading. The zero check 3 reading was 000, then the test appears to stop. When in use on 29 December the first subject reading was 84, but the second was 115, again outside the 20 per cent variation. It is recorded that the display showed “E” and the entry ceases.

20. These results gave rise to the complaint recorded in the FSANI log that the readings were unreliable high and low. The evidence from the log is that on every occasion when the unreliable readings were produced the test was terminated. This, as we understand the evidence and counsel’s submissions and explanations, is precisely what the device is designed to do. If the magistrate thought that it would abort as soon as the fault manifested itself, without the incorrect reading appearing on the read-out, which appears to be the purport of the last two sentences which we have quoted from the case, we think that he must have misunderstood the operation of the machine.

21. We conclude that there is nothing in the evidence appearing in the logs which contradicts or throws doubt on the correctness of Mr Fulton’s evidence that the fact that the device successfully completes the test cycle proves the reliability of that test and the accuracy of the results in that test. No expert evidence was called to counter Mr Fulton’s view, nor is there any indication that it was significantly challenged at the hearing. The defendant did not give evidence and the magistrate did not have facts before him which might have tended to establish that he was not intoxicated and so throw doubt on the validity of the breath test results. There accordingly was no ground on which the magistrate could properly reject the evidence furnished by the test results. The magistrate did not refer to the statutory presumption contained in Article 18(2) of the Road Traffic Offenders (Northern Ireland) Order 1996 and its statutory predecessor, but it may be debatable whether that presumption extends to the reliability of the device. Be that as it may, the facts in our opinion led irresistibly to the inference or conclusion that the device gave a reliable and accurate result when the respondent was tested on 22 December 1996 and that the proportion of alcohol in his breath exceeded the prescribed limit.

22. We would reframe the question posed in the case stated in the following terms:

23. We answer that question in the negative, allow the appeal and remit the case to the resident magistrate with a direction to convict the respondent of the offence charged.