Discussion
Although the number of samples were limited and most had degraded since their discovery in 1993, the details revealed by microscopic examination and the conditions of their discovery leads to the conclusion that this was a very important find.   The two principle questions posed by this examination are: where did the iron and its oxides come from and where did the heat come from? 

There has been much speculation as to the source of the iron.   Let us examine some of these theories.

Local Contamination
As has already been stated, it was at first suspected that the farmer had emptied the tanks of his spraying equipment at that spot.   However, the farmer was adamant that he had not done so and could not account for the deposit.   Also, such an explanation could not account for the evidence of heating.
Other mundane explanations were also considered and over the following weeks and years, the area has been thoroughly examined. There were no signs of any ferrous deposits or the remains of rusting farm implements anywhere in the field.

The Perseids
Much had been made of the coincidence between the appearance of the H-glaze and the Perseid meteor shower (July 23 to Aug. 20, peaking at Aug. 12).   One widely held belief is that the glaze was formed from meteoric dust that somehow propelled itself down to earth by means of a ‘plasma vortex’.   This theory cannot be correct.

Firstly, meteor showers are created when the earth passes through an extended dust trail produced by a comet.  The Perseids are believed to be the result of a dust cloud produced by the comet Swift-Tuttle.   It is know that these cometary trails are composed mostly of dust, ice and gas.   Iron is not a significant constituent and, therefore, meteor showers cannot account for the iron in the H-glaze.

The other component of this theory, the plasma vortex, is frequently referred to in articles and lectures and, over the years, has acquired general unquestioned acceptance.  The phrase seems to have first appeared in the work of Terence Meaden who tried to attribute crop circles to natural meteorological activity.    It may be that he had something else in mind and used the term for want of something better but the truth is that gaseous plasma does not occur in the atmosphere at ground level except in extreme conditions, such as a lightning strike, arc welding or in vigorous combustion.   By whatever means it is induced, it cannot sustain itself without receiving considerable energy from somewhere.   The moment this energy is removed, the plasma extinguishes.   Plasma may have vortices within it and these can induce a ‘pinching’ effect on any electrical current distributions within it.  However, such ionic movement has negligible mechanical impetus and is quite incapable of flattening one single stem of corn, let alone levelling the swaths of crop attributed to it.     Whatever caused the swirls of crop found at the location of the H-glaze, it was not a plasma vortex.

Meteoritic Iron
In addition to meteor showers, there are also sporadic meteors that enter the atmosphere at random. Most are vaporized but some are large enough to reach the ground. These are called meteorites. Broadly, there are three types: stony, stony-iron and iron. Stony meteorites are the most abundant, constituting over 90% of all meteorite falls. Only a very small proportion are iron. Of these, the smallest weigh about 5-30gms and are aerodynamically shaped into bullet-like tektites. Nothing like this was present in the samples.
Meteoritic Iron reaches the ground either as discrete lumps or it is ablated from the surface of the incoming meteors and distributed across the upper atmosphere in molecular form to descend slowly in the air currents. It is distributed more or less evenly over the earth's surface and there is no mechanism whereby the ablated remains of a meteor could be focused onto one point on the earth's surface. Whatever the source of the iron in the H-glaze, it was not meteoritic iron.

The Heating
The first indication that the glaze had been subjected to heating was its cohesion - i.e. the way it was bonded together and to the substrate. There were only a limited number of ways that such adhesion could have been produced - the use of a bonding agent, chemical bonding or sintering.

No bonding agent was found, despite close inspection, and the analysis gave no indication of the presence of any medium.
Chemical bonding can take many forms. Basically, it involves molecular interaction at the points of contact of the constituents. This could have occurred, with time, between corroding iron and the substrate and, indeed, there was evidence that some of the glaze was the result of natural corrosion. This is not surprising as it is known that it had rained just prior to the discovery of the H-glaze. However, corrosion is a relatively slow process and the structure of the glaze on the first sample or the coating on the second would not have remained in place for the time it took for corrosion to occur. Also, it does not account for the differences in the samples or the fragments of the substrate on the surface of the first sample (see 'Other Factors' below). Superficially, the case for corrosion seems persuasive but it leaves many questions unanswered and without the corrosion argument, there is no case for chemical bonding as the known ingredients of the samples do not react together in any other way.
Sintering, or the coalescence of powder into a solid by heat, involves heating the constituents to a point were each particle forms a microscopic weld or fusion with its neighbouring particles at their points of contact. Such an explanation would account for the absence of bonding agent and explain the many other observed features such as the rounded inclusions of elemental iron (Fig 3), the contorted and cracked appearance of the thick glaze (Fig 1&2) and the apparent creation of vitrified features on and adjacent to the ferrous deposits (Fig 4,5&6). Interestingly, such a theory requires that the heating process should be very short - perhaps a few milliseconds - but very intense. The fact that the chalk substrate had not recrystallized and reports that the crop showed no signs of incineration indicates that the heat was localized to the glaze but must have had limited thermal capacity. It is also probable that the ground and crop were protected by the damp conditions after the rain reported by the farmer.

As we have seen, the heat appears to have been concentrated in the Iron and its oxides.   This is perhaps not surprising as it is the presence of iron that is unusual and it may well be that the process that created the iron also heated it.  However there is no convincing process that can exclusively heat the iron and not the surroundings.    For example, any form of electrical discharge or chemical reaction that could generate sufficient heat to melt the iron would inevitably incinerate the crop.    Another possibility considered was RF (radio frequency) induction heating – a process that induces circulating heating currents into metals.   However, it was quickly realized that such a process could not account for the evidence.   Firstly, the resistance of the H-glaze was too high for the required currents to be induced into it.   Secondly, it is only possible to induce sufficient current into a ‘circuit’ if it is larger than the wavelength of the RF energy.   Yet, we have seen from Fig 3 that the bead was about 10 microns in diameter.   As most of the elemental iron was in lumps of comparable size, the RF energy would have had to be in the far infrared.   Passive infrared detectors, for example, operate in the 5 to 14 micron range.   This would heat everything.  It would not exclusively heat the iron.

Other Factors
Another very curious, and potentially revealing, aspects of this case was the difference in appearance of the first two samples. Indeed, looking superficially at the two deposits, the first impression was that they were totally unrelated. One was a mottled accretion coated with fragments of the substrate and thick enough in places to have the appearance of having flowed over the chalk. The other was an uncontaminated coating of relatively uniform thickness and coloration that could have been sprayed onto the substrate. It was only the known provenance of these samples that meant one had to accept that they were from the same event. These differences appear not to have attracted the attention of previous investigators or those who found the deposit, yet they may well hold the key to explaining how the H-glaze came to be formed in the first place.
Unfortunately, we can do no more than speculate as to the cause of this because the essential survey work was not carried out at the time of the discovery. It may be that the first sample was taken from a swirled feature of the formation and that the second sample came from elsewhere. Another possibility is that differences in height could account for their dissimilarity. We cannot answer these basic and potentially very revealing questions without the results of an all important survey.