` In every known case of “crazing” the surface crack is situated in the centre of a small ridge or corrugation (which may be easily felt even in the case of minute “hair” cracks by sliding the finger over the surface at right angles to the direction of the crack)

` These ridges or “crater” phenomenon allow for immediate recognition of “crazing as opposed to ordinary cracking - it is, without exception always present in the former but never in the latter.

`There appear to be no doubt, also, from experimental work carried out in the laboratory, that “crazing” is accelerated by the  action of ultra violet light. The latter probably has an oxidising action on the bitumen present in this surface skin, which again weakens it mechanically”

2. Reduction in thickness of the bituminous skin.

` From the previous section it will be gathered that the phenomiona of “crazing” is essentially and fundamentally dpendant for its origin and existence on the highly bituminous, unstable, and unprotected surface skin on the asphalt. It logically follows, therefore, that if the surface layer can be dispersed and effectively protected from sun irradiation, progress toward the elimination of this defect will naturally result. Both from a practical and theoretical standpoint, its is essential that this bituminous skin should be reduced to a mnimum if “crazing” of the material is to be avoided. The dimensions and bituminous nature of the surface skin will be largely dependant on the bitumen content of the asphalt and, in practise, it is a fact that, other important factors being equal, the extent to which “crazing” takes place, is proportionate to the bitumen content of the material. An improvement undoubtedly results when the bitumen content is cut down, but there is naturally a very definite limit below which it is impossible to go without seriously affecting the consistency of the asphalt. This is especially true where a natural rock aggregate is employed.

` An important point arising here with regard to previous satisfactory examples of natural rock mastic, is that the asphalt was invariably heavily gritted both in the “vertical” and  “flat”.

3. Protection of the bituminous surface layer

` It has been stated that it is not possible or expedient to eliminate entirely the bituminous surface skin, and it is therefore becomes advisable to protect, as far as possible, the remainder of the bituminous surface layer from the effects of solar radiation. This is effectively done by “sanding” or “gritting”  the surface with a suitable material, “rubbing” material at present in use vary considerably. Sands (of various grading), granite dust, whin and limestone dusts and even cement have been used at various times in different parts of the country. . It can be stated definitely that limestone dust and cement when used as “rubbing” materials are quite fatal from the point of view of crazing. They do not form an effective protective dressing owing, probably, to their absorbent nature and the fact that they tend to chemical instability when directly exposed to weather conditions. There is no doubt that the best material for this purpose is a hard, absolutely non-absorbent and chemically inert substance such as silica (sands) or granite dust.

A light coloured material of possibly a quartz nature would appear to be ideal fro a point of view off efficient diffusion of the incidence of sunlight and the breaking up of the surface skin. The grading of this material, within limits, does not appear to be of vital importance. It should not be fine, otherwise a too intimate physical combination with the bituminous surface layer results, and the latter is “dried out” excessively. A relatively high proportion of coarse material can, in many cases, be considered distinctly beneficial - especially when applied in a rather special manner. In this connection it is interesting to quote the method used some years ago, which gave roofs, which are in extremely good condition; here a comparatively large quantity of coarse, hard grit was beaten into the surface both on the vertical and the flat sections. The result is a stable and an efficient protective dressing.

4. Incorporation of grit in the asphalt

` The incorporation of grit in the asphalt for both “vertical” and “flat” work is also to be highly recommended as this helps to stabilize the material mechanically and has the important further effect of reducing the bitumen/aggregate ratio, without affecting the consistency. By the incorporation of 15 t0 20% of a coarse grit, such as Bridport grit used in the London area, the soluble bitumen figure of the finished mastic is reduced 1 to 1.5% without any harmful effects as regard the consistency.  

It is an interesting fact that the “beaten” asphalt previously referred to, was found to contain the quite remarkably high figure of nearly 40% grit (“10mesh” and “retained on 10 mesh”) - this is irrespective of the material beaten into the surface.

5. Use of suitable asphaltic cement.

In the light of previous work carried out on this subject it may appear extraordinary that more prominence had not been given in this report to a discussion of the suitability of standard asphaltic cements in relation to the crazing problem. Although this matter is, of course, significant, a practical and theoretical examination of the problem shows that it is not of vital importance - provided always that the asphalt cement is up to the standard normally required for mastic asphalt manufacture.

“ In the case of roofing mastic asphalt it would certainly appear to be advantageous to reduce the penetration of the asphaltic cement to a minimum consistent with local conditions and always provided that this reduction in penetration is not accompanied by a disproportionate loss of ductility.”

Summary

The investigation for the elimination n of crazing in roofing mastic asphalt can be summarized as follows:

1. Reduction in thickness of the bituminous skin by the use of   the optimum soluble bitumen content.

2. Protection of the bituminous skin.

3. Use of suitable aggregate.

4. Incorporation of grit in the aggregate.

5 . Use of suitable asphaltic cement.

Although tabulated in an appropriate order of importance it must be emphasized that these recommendations are largely interdependent and if the risk of crazing is to be entirely eliminated, due consideration must be given to each.

Of necessity much has been omitted from this report, but from these extracts the cause of crazing is clearly shown and simple practical recommendations for its elimination are tabulated.

Since this report was published in 1926, experience has shown two modifications of the views expressed.

The first is that the elimination of the skin that may float to the surface on mastic asphalt is desirable, rather than its reduction, provided that it can be done by simple practical means.

The second point is that the use of coarse, sharp rubbing materials is the most effective in reducing or eliminating the bituminous skin. This may not produce such an attractive finish as the finer types of sand but there is little merit in having a pleasant looking surface, which is likely to be marred by crazing soon after it's completion.