In the light of the evidence it was agreed by all the parties concerned in the Inquiry that the point of origin of the explosion was at the retreating longwall face. There were other places, such as the junction of No. 22 Duckbill heading with the return, at which the lines of force indicated a possible origin, but in none of these, in fact in no place other than the longwall face, was any possible igniting source found.

Although electricity was so widely used, the evidence showed that it was not the means of ignition. Though some damage and defects were discovered in the electrical plant, it was shown either that they had been caused by the explosion itself, or that they were in such a position that they could not have given rise to the explosion.

In a careful examination no contraband whatever was found.

All mechanical appliances such as conveyors were examined but no signs of frictional heating sufficient to have served as a source of ignition were found.

The position of the shot firing cables and exploders showed that no shots were being fired at the time of the explosion. Evidence was given that shots had been fired in the roof at the longwall face to release chocks on some previous shifts and that on Sunday morning, 27th May, a shot hole was seen to have been bored for such a purpose, though it had not then been charged. Mr. Emery, inspecting about 8.0 a.m. on the following day, formed the impression that a shot had been fired near a chock. It seems to me unlikely, however, that if firedamp had been ignited by any such shot it would have continued to burn unseen when the heavy falls of roof occurred in the waste during the drawing-off operations on the evening prior to the explosion.

The safety lamps belonging to the men who were at or near the face when the explosion happened were found in a damaged condition, and there is of course no means of knowing what they were like beforehand. But as they were all damaged, and as the general standard of maintenance of the lamps was very high, it is reasonable to assume that the damage had been caused by the explosion.

There is the possibility that gas was ignited by sparks caused by rock falling in the waste, but no such sparks had ever been reported and, in my view, it is improbable that this did in fact happen.

It was proved that sparking did sometimes occur during the operation of the chock releases and a number of chock releases were found in the released state after the explosion. However, in tests made some years ago by the Safety in Mines Research Establishment on releasing chocks under load in an explosive atmosphere, and recently repeated under much bigger loads using a pair of the chock releases discovered to have operated spontaneously on the longwall face, no cases of ignition of firedamp occurred.

On the other hand, the evidence is that the coal cutter picks were moving at the time of the explosion, that they were then cutting through pyrites, and that sparks from this pyrites proved, under test by the Safety in Mines Research Establishment, readily capable of igniting mixtures of firedamp and air. Moreover, sparking had frequently been seen in other districts of the seam when a cutter jib was exposed during jibbing in, and on one occasion sparks had been seen in the undercut on this particular face.

I have no hesitation, therefore, in finding that the ignition was due to sparks caused by the cutter picks striking pyrites - a conclusion which was accepted by all the parties represented at the Inquiry.

From the position of the igniting source as well as on general grounds, it was agreed by all parties that the initial explosion was one of firedamp.

(a) Source of Firedamp

In considering the source of the firedamp, it is necessary to refer briefly to the roof conditions on this retreating face since it was started on 12th April.

After the face had travelled some 15 yards, that is to about 20 yards from the back rib side, some three to four feet thickness of roof fell in large pieces in the waste. On 27th April a heavier weight occurred causing a further fall of large stones in the waste and the breaking of many props at the face. After this, cutter breaks usually formed in the roof along the successive face lines and the withdrawal of the supports became difficult. On 11th May, when it had moved 32 yards, the whole of the face except near the ends was affected by a much heavier weight and ten yards of it was closed. In the waste a considerable thickness of roof fell again, mostly as large stones. The face was not fully re-opened until 21st May. From then up to the time of the explosion it advanced 4½ yards.

During coal filling operations on Monday, 28th May, breaks were observed in the roof at the face on the length of about 20 yards between the intake roadhead and a slip plane. During the day a break displacing the roof downwards on the goaf side developed along the face on the return side of this slip. On the back-shift, after the loading of the web of coal had been completed, the chocks were advanced and props and bars withdrawn. This was followed by falls of roof in the waste and by further displacement of roof along the break near the coal face, but about 8.0 p.m. the roof steadied and shortly afterwards the last of the back-shift workmen left the face feeling that it was quite secure.

There is indirect evidence, however, that further trouble occurred during the night-shift. Usually cutting was completed between 2.0 a.m. and 3.0 a.m., but when the explosion occurred at 4.35 a.m. 12 yards remained to be cut. Moreover, though men normally working in the duckbill headings were sometimes sent to the face to load cuttings, the presence at or near the face of four stone-men and the master shifter, together with the fact that the fore-shift workmen had not proceeded to the face, provides strong circumstantial evidence that the roof had again been "on weight".

Considering the reasons for the occurrence of this weight, it seems to me that as the face advanced, the roof over the waste, being without direct support, would be carried on abutments in the solid coal sides, in the coal left behind, and in the coal ahead of the face. With the continuing advance the span over the goaf and consequently the abutment pressures would increase and some roof beds would lower and fracture. An attempt to show the trend of these occurrences, though not the actual happenings, is made in Diagram I.

Although the immediate roof behind the waste edge supports collapsed in layers, the stone did not break up much in falling, and Mr. Emery, Mr. Hopkins and Mr. Morgan agreed, when questioned by Mr. Jones, that the conditions which developed were not those in which caving was likely to be successful.

As the seam was nearly seven feet thick it seems evident that the heaps of debris cannot have been high enough to give support to the unbroken beds above them. I envisage, therefore, that even after the weighting of 11th May, a considerable cavity in the goaf area was spanned by the higher beds.

It is likely therefore that during the night of the disaster the higher beds, overstrained by their increasing span over the goaf and affected perhaps by the comparatively slow advance since 11th May, and under the influence of the cut being made at the time, lowered gradually and finally fractured. Many props broke under the extra pressure and falls occurred, but most of the chocks remained effective and limited the extent of the falls. That the weight continued after the explosion is shown by the fact that some of the surfaces of the chock blocks exposed by fracturing were not coated with dust.

(b) Emission of Firedamp

Whilst all parties agreed that the firedamp came partly from the waste and partly from the higher beds, there was a wide difference of opinion as to the contribution made by each source. Mr. W. H. N. Carter, H.M. Senior Inspector for Special Duties, thought that the firedamp had in the main come from cavities in the waste and that the emission was on the whole gradual. On the other hand, Mr. H. E. Collins, the Production Director of the Durham Division of the National Coal Board, and Mr. H. F. Wilson, H.M. Senior District Inspector of Mines in charge of the South Durham District, thought there had been a sudden and very large outburst from the roof strata.

Considering the possibility of emission of firedamp from the waste, the space created by the removal of the coal was roughly 300 feet in length, 100 feet in width and six feet in height, that is about 180,000 cubic feet. Except in so far as this cavity had been reduced by the lowering of the roof strata massively, as opposed to the falling of the immediate roof which would merely transfer the space slightly upwards, it would remain open, a potential reservoir for firedamp, with a ready path to the air coursing the face.

The only evidence of the emission of firedamp on the face before the explosion was a sample, containing 0.08 per cent, of methane, taken on 6th May ten yards from the return side of the longwall face, and the testimony of one man that one of the deputies lost in the disaster had found it at the edge of a break on an occasion during the week beginning 21st May. Coal, however, usually emits firedamp most readily when subjected to extra pressure, and it was to be expected that the "make" would increase as the face advanced, and the extra pressure on the coal forming the margins of the excavation became of increasing consequence. On the First South longwall face firedamp had sometimes been found after a fall in the waste and it is probable that similar occurrences would have been observed on the Third South had it travelled a little farther.

In any event, on retreating faces with solid sides there is no circulation of air across the waste area, which, as shown in Diagram II, is ventilated only at the edge adjoining the face working, so that even a small flow of firedamp into the waste leads in time to a rich mixture.

If cavities are created in the waste area, a large accumulation may collect there, but, apart from an issue, roughly corresponding to the make, at the return end of the face, the firedamp may appear only as a result of ground movement within the waste area or of rapid changes of barometric pressure.

On the assumption that the cavity in fact did contain a rich firedamp mixture, the effect of the lowering of the upper beds, which almost certainly took place during the night of 28/29 May, would be to cause the emission of a stream of firedamp near roof level at the return end of the face. As a consequence of the occasional falls that were likely to be taking place in the waste area, there would also be surges of firedamp. To these may well have been added some extra make from the coal forming the face as it came under increased pressure; part of this would find its way into the air current via the undercut.

It is quite probable that the movements and collapses in the waste would also create channels whereby firedamp from the thin seams 136-185 feet above the Five Quarter Seam may well have made a contribution to the emission. I reject, however, the view that a large outburst occurred suddenly, partly because a gradual issue from the waste is more in accord with general experience, but more for reasons I advance in the following section.

Two widely differing views were put forward as to the reasons for the spread of the explosion outbye. Mr. Collins contended that the large volume of firedamp under high pressure, which he envisaged to have been released when the waste collapsed, was practically simultaneously ignited and that it carried the explosion through the whole of the affected area, with coal dust playing but a minor part. On the other hand, Dr. Tideswell envisaged a gradual emission of firedamp causing a considerable length of the return airway to contain an explosive mixture; when this was ignited it burst through into the First West intake roads and there initiated an explosion of coal dust which continued through the rest of the area.

Mr. Collins based his opinion on five main factors :-

I see nothing in the first three factors at all peculiar to explosions of firedamp; similar effects have been noted in previous explosions which after full investigations have been accepted as explosions of coal dust.

The uniformity with which the flame traversed the various roadways is, I think, what might well be expected, bearing in mind the very numerous connections between intakes and returns and that, almost without exception, the stoppings in these connections were destroyed.

The fifth factor refers to an attempt on the part of Dr. Jones, the Divisional Chief Scientific Adviser, to determine the nature of the fuel giving rise to the explosion from the composition of the residual gases. The method depends on the fact that when a fuel burns, the hydrogen combines with the oxygen in the air to form water, which condenses, so that it is possible from a comparison between the gaseous products of combustion and the amount of oxygen consumed to arrive at the proportions of hydrogen and carbon in the fuel.

In the laboratory this is a well established method of distinguishing between different hydrocarbons, but it has not previously been used as a means of determining the nature of the fuel taking part in an underground explosion. The method has possibilities, but as yet little is known about the variations resulting from the known instability of the oxides of carbon in a mine and from the absorption of oxygen by coal. Moreover, Dr. Jones himself made it quite clear that any conclusions based on the analyses of residual gases were valid only for the area in which the samples were taken, that is inbye of the First South headings.

Mr. Collins, however, suggested that as the conditions outbye of First South differed in no material sense from those inbye it was reasonable to assume that there was no difference in the explosive medium. But in fact the conditions differed very significantly in that in all the stentons inbye of No. 18 the indications of direction of force pointed from the return towards the intake, whereas in the stentons outbye from there the force had travelled from the intake towards the return. Mr. Collins thought that the explanation of this very important difference was partly that as the explosion travelled along two intake roads as compared with one return road, the force towards the return would be greater than that away from it, and partly that the explosive violence which occurred at the junction of the return with No. 22 stenton would tend to retard the explosion wave in its passage farther outbye along the return. He said, however, at another point, that he was "definitely of the opinion that the initial explosion caused the falls in the face return, in stenton 28, and that it blew the stopping in stenton 27, the stopping in stenton 26, the doors in the extension of the Belt Road towards the return, and finally passed along the return towards 22’s heading ". If, as Mr. Collins believed, the explosion was propagated by a fast moving column of firedamp with a burning front it is surely to be expected that when the explosion blew through into the intake at No. 26 stenton and at the belt extension, the postulated effect of the spread of the explosion along the twin intakes, as well as the "further development of explosive violence ", should have become evident there and the subsequent direction of force outbye of this point should have been towards the return. But in fact the direction of force in the stentons inbye of No. 18 was from return to intake and the signs of explosive violence were at the junction of No. 22 and the return. I therefore reject his explanation of the change in direction of the blast.

Finally, I find it quite impossible to accept the mechanics of a theory by which in effect an almost solid body of firedamp with a burning front propagated the explosion through the whole of the affected area by picking up oxygen as it advanced. This view is perhaps strengthened by Mr. Collins’ replies to questions which I put to him :-

In my opinion even if such a process could continue, it would be only by fits and starts, with interruptions whilst admixture with fresh air was taking place. There was, however, no suggestion that the men in the First West intakes and on and about the drift were taken other than unawares by the explosion and it is reasonable therefore to infer that the explosion swept through the workings without significant halt.

On Dr. Tideswell’s estimate that the flame reached the West level in the Main Coal Seam within 30 to 60 (or at most 90) seconds of being ignited, the outburst of gas postulated would have to be almost instantaneous and Mr. Collins himself estimated that it would be complete within a very few minutes. Assuming for the sake of argument, however, that it lasted five minutes and that the volume of air passing along the face was 5,000 cubic feet per minute, the atmosphere would contain 80 per cent. of firedamp. Such a mixture could only ignite in the very small fraction of time taken for the thin envelope of mixture containing between five and 1 5 per cent. of methane, at the front edge of the rapidly advancing gas, to pass over the source of ignition. I am very doubtful if it could do so, but again assuming that, as Mr. Collins suggested, the fringe did manage to ignite, I cannot believe that it would continue to burn. For at least the first 300 feet along the return this 80 per cent. mixture would have had to travel along a road in solid coal like a piston in a cylinder. As pure firedamp must be mixed with six times its own volume of air before it will burn, an 80 per cent, mixture would need between four and five times its own volume. Thus for every unit of the mixture burned there would be nearly six units of incombustible products and, being hot, they would occupy at least 20 times the volume of the original firedamp. There being no means of escape, a rapidly increasing wad of incombustible gases would gather between the advancing firedamp and the only supply of fresh air. Very soon, indeed almost immediately, it would become impossible for enough gas or air to pass across this wad to provide an inflammable mixture and the flame would die for lack of oxygen.

This analysis is, of course, quite unduly favourable to the hypothesis, for an emission of gas of 20,000 cubic feet per minute would have stopped the flow of air on to the face almost at once. In all probability the air would have been reversed and a part of the firedamp would have been forced into the intake roads where on the hypothesis it should have mixed with the air and burnt. In fact, as was brought out by Dr. Jones and Dr. Tideswell independently and supported by medical evidence, there was little, if any, combustion on the Third South intake roads near the face, and Dr. Jones gave evidence that little residual methane was found in these roadways as compared with elsewhere.

It is now necessary to consider whether the facts can be explained on the alternative theory that an initial gas explosion burst through to the intake and caused an explosion of coal dust.

Significant features of the initial explosion were the absence of flame in the intake in the neighbourhood of the longwall face, the development of violence at No. 28 stenton, the direction of force being from return to intake as far as No. 1 8 stenton, and that the stopping in the extension of the return to the North Haulage Road was not destroyed.

Dr. Tideswell held that, to account for these features, at the time of the ignition fouling of air must have extended well along the return but not beyond No. I 8 stenton, and that near the source of ignition it must have been a slow burning mixture. Estimating that the volume of the air passing along the face was of the order of 5,000 cubic feet per minute, he calculated that to produce such an inflammable mixture firedamp must have been issuing from the waste for a period of from ten to 20 minutes, or more, and that the volume of firedamp needed would be between 5,000 and 20,000 cubic feet depending on whether or not coal dust played any part in this phase of the explosion. For convenience in this calculation Dr. Tideswell assumed a uniform emission. It is, however, unlikely that the emission was uniform and, as diffusion would be slow in the current moving slowly along the unobstructed return, I think the gas would tend to persist in layers and that the concentration would vary from place to place. Moreover, as the path of the explosion crossed the conveyors in Nos. 24, 22 and 21 headings, it is reasonable to assume that coal dust did play some part in this phase. For these reasons, though the waste may well have contained a much larger volume of gas, I believe the effects observed may have been produced by a quantity approximating to the lower limit suggested.

Against this it was argued that such a large volume of firedamp could not have issued for so long without the three experienced machinemen working on the face becoming aware of it and taking some action. I believe the explanation is that :-

The difficulty, one might say the impossibility, of accounting for the fact that the force of the explosion had travelled from the return towards the intake inbye of No. 18 and from the intake towards the return from there onwards, on the theory that firedamp was the sole explosive medium, has already been discussed. But if, as Dr. Tideswell suggested, a coal dust explosion travelling along the intake was initiated by the firedamp explosion when it blew through at No. 26 stenton and the extension of the West Belt Road, the difficulty disappears.

The question then arises as to whether there was any evidence to support the theory of a coal dust explosion.

Dr. Tideswell estimated that there was a deposit of explosion dust - i.e. dust that had been raised into the air by the blast and subsequently deposited - about one-fiftieth of an inch in thickness corresponding to a cloud density of about 0.5 ounce per cubic foot throughout the area. As tests by the Safety in Mines Research Establishment showed that coal dust from the Five Quarter Seam required 60-65 per cent. of inert matter to render it non-inflammable, whereas most of the samples of the explosion dust contained only 35-50 per cent. of such matter, and as an explosion can be propagated by a cloud density of as little as 0.05 ounce of coal dust per cubic foot, he expressed the opinion that there was between two and three times as much explosive dust as was necessary to have carried on an explosion.

It is true, as Mr. Collins pointed out, that the samples taken by the management before the explosion and most of those taken by Mr. Badger afterwards contained considerably more than 65 per cent. of inert matter, and therefore it might seem that the dust should not have been explosive. These samples however were taken in the manner prescribed by the Regulations to determine the mean composition of the dust over the area of road sampled, whereas Dr. Tideswell’s survey showed that the composition of the explosion dust which, it must be. remembered, was the dust which had formed the cloud through which the explosion had passed, was much closer to that of samples taken deliberately from the more coaly deposits. This is quite consistent with the work described in Safety in Mines Research Board Paper No. 105* which shows that mine dusts are not uniform and that the mean composition of non-uniform deposits seriously under-estimates the inflammability of the deposits. In short, a far higher proportion of inert matter is needed to render dust non-inflammable when the coal dust and the inert dust are not intimately mixed.

Again, it was argued on behalf of the National Coal Board that, because of the work done in cleaning up, there were good grounds for thinking that there would not be enough dust available to propagate an explosion. Having regard, however, to Dr. Tideswell’s estimate of the amount of explosion dust, and to the evidence of Dr. Tideswell, Mr. D. E. Edwards, H.M. District Inspector of Mines, and Professor Granville Poole, lately Head of the Mining Department, King’s College, Durham University, who was called as an expert witness by the National Union of Mineworkers, on the presence, particularly round transfer points, of deposits of coaly dust after the explosion, I am unable to accept this view. Despite the clearing up and stone dusting which had been done, the blast would raise coal-rich dust from any coal or cuttings in transit on the conveyors, from the conveyor structures - which are notoriously difficult to treat or clean - from the belts themselves and from inaccessible places behind supports, etc.

It was said also that the amount of coking found in the area was small, and certainly much less than had been ex perienced at the Louisa Colliery explosion in 1947, and that this indicated that coal dust had played little part. Deposits of coked dust were, however, found widely even though lightly distributed inbye the top of the drift from the Main Coal. Dr. Tideswell held that the amount of plastered coke was about that usually found and that the amount and general appearance of the coked dust was consistent with an explosion travelling through a dilute and only moderately inflammable cloud of mixed coal and stone dust in air. Against this Mr. Collins contended, that the coking nature of the Five Quarter Seam would account for the presence of signs of coking in the explosion dust even if the explosion was a pure firedamp explosion, i.e. that the propagation was not due to the inflammation of the dust itself. The fact is that it is not possible to say from a superficial examination of an affected roadway whether an explosion had been one of coal dust or firedamp. Explosions of either give rise to widely differing degrees of violence; a coal dust explosion may leave much or little coked dust; and a firedamp explosion may produce coked dust that has played little part in the propagation of the explosion. One has to take into consideration the degree of probability of the presence of firedamp or of coal dust. In this case, while I see no likelihood of the presence of firedamp in these intake roads, I am entirely satisfied that there was enough inflammable dust to propagate an explosion.

Due to the thickness of the seam and to the fact that the roof caved in very large blocks, cavities of considerable volume existed behind the longwall face. As the waste was not ventilated, a large volume of firedamp slowly accumulated. When the roof weight, itself a consequence of the incompleteness of the caving, began during the night of 28-29th May, firedamp in the waste area was pushed nearer to the face and began to be emitted into the air current at roof level near the return end of the face. Roof fractures extending upwards led to influx of firedamp from the upper strata. At least to the point where disturbance was created by the air leaking through the stentons, the firedamp largely remained as a layer near the roof. Probably within half an hour before the explosion the emission was intensified. Falls in the waste caused surges of firedamp which would travel along the return as gradually dispersing concentrations.

Then, possibly not for the first time during the cutting operation, the cutter picks struck pyrites. The firedamp ignited by the consequent sparks may have been part of a surge from the waste. I think it more probable that it was firedamp released into the undercut as a result of the weight.

I believe that flame would emerge from the cut behind the machine and would alarm the machinemen who would naturally make for the adjacent roadhead without waiting to stop the machine. The flames, however, would lick towards the roof and there meet the under-surface of the flow from the waste.

The inflammation probably proceeded quietly to No, 28 stenton where its contact with a more inflammable mixture created explosive violence. The explosion passed backwards towards the face and forwards along the return, where it blew through to the intake at No. 26 stenton and at the extension of the West Belt Road. It continued along the return bursting southward at successive stentons to about No. 18 stenton. At No. 22 junction extra violence developed, possibly because the current then passing that point was highly inflammable. Thus into the lengths of the West Belt and Materials Roads between the Second South and Third South roads, where there were two double transfer points, flame came in two directions. I conclude that the explosion then became one of coal dust and that it spread as such, obtaining fuel mainly from the conveyors, the conveyor structures and the stentons.

The reason for its providential cessation on the West Main Level is uncertain. Dr. Tideswell suggested that it was checked because the outbye length of the fall on to this road had occurred before the flame reached that point. It may be, however, that as this was not a conveyor road but was in regular use as a haulage road, the coal dust and stone dust were more intimately mixed and therefore the stone dusting measures proved effective.