index of the decrease of the putrescent matter, and therefore of the efficiency of the filtration. Also the amount of nitric nitrogen in the filtered effluent indicates the extent of conversion of putrefactive nitrogenous matter in the tank effluent into non-putrefactive nitrogenous compounds, so that the higher this figure is the more active, as a general rule, is the purifying power of the filter.

The question of sewage disposal is a highly technical one, and only those who have had practical experience can imagine the difficulties which have to be overcome, and therefore Local Authorities and others who have this matter to deal with should be careful before committing themselves, unless advised by experienced engineers, as a method of disposal which may answer in one case may prove useless in another. It is important to remember that it is a chemical and bacteriological, as well as an engineering question, and the chemical composition of the sewage, which I have previously mentioned, varies greatly, and has to be taken into consideration in forming an opinion as to the method which is likely to prove most satisfactory.

In conclusion, I wish to say that the evidence given before the Royal Commission of Sewage Disposal has been of the greatest value to the Sanitary Engineer, but until the Commissioners' final report is published we are not in a position to say which of the bacteriological methods at present recommended is the best; and no doubt, as I mentioned above, this to a large extent is governed by the cir.cumstances in each case; neither can it be said that we have arrived at finality in our knowledge as to the best means of availing ourselves of Nature's processes, but that we are on the right lines is perfectly

certain.

FERRO-CONCRETE

BY H. KESTNER, C.E.

There can be no question that the use of reinforced concrete as a building material has made steady progress in England, America, and on the Continent during the last few years, so that one is justified in asking, Why has this building system been so much neglected in South Africa?

To arouse some interest and to draw the attention of engineers to this system of construction will be the object of this paper.

The practical use of ferro-concrete is of course known to every engineer, and it is only the theory which has not kept pace with the progress of this form of construction, about which there still exist very vague and erroneous views, especially with regard to the statical action of this method of construction. The reason that the theory has not kept pace is on account of the rapid development of ferro-concrete building.

Only ·

A method of calculation conformable for the purpose must, as far as possible, correspond with the existing statical action of both materials forming the ferro-concrete, i.e., concrete and iron. under these conditions can safety and economy be gained. Very much depends on the kind of calculation adopted, and as to whether the formulæ are very long and intricate, or simple and easy of comprehension. It must at once be confessed that the various theories about ferro-concrete in the vain hope of obtaining complete accuracy, make the calculations so intricate that they are completely useless for the purpose they are intended. Every engineer who is a friend of this new building system, and is desirous of assisting its progress, should avoid these faults and only choose the simplest formulæ.

The most accurate theories which always consider the elastic qualities of both constituent materials, concrete and iron, in every point, have their fundamental rules always connected with certain presuppositions, and this disqualifies them of the term accurate, but on account of the care which has been applied in the solution of the question they are the most meritorious theories at present we possess.

In the present state of the theoretical investigations, it is in my opinion more justifiable to adhere to a theory connected with presuppositions instead of adopting a of adopting a fallacious accuracy. It naturally depends entirely on the more or less careful investigation of the basis of this theory. This will be the object of my paper, and I will make use of the facts which experience has taught us.

CO-OPERATION OF CONCRETE AND IRON.

In the theoretical investigation of ferro-concrete it is generally accepted that the concrete is in a position to follow the iron in its changes of form through tension, compression or other strains. If we place instead of the modulus of elasticity of the iron the letter Ee, and of that of the concrete the letter Ec, we assume that the elastic strains of the reinforcement, and of the concrete,

which adjoins the reinforcement, are in the proportion of Ee to Ec. Most of the theorists have accepted this fundamental rule as a basis of their formulæ, without investigating the correctness. It is, however, of great value to examine it more closely in order to prove whether this fundamental rule can be accepted or whether it is possible to disregard it completely.

Experience has not given us many facts to answer these questions. Harel de la Nae, a French Engineer, assumes that in a ferro-concrete body under strain the cross-section shows a point of contraction, but only then when the limit of elasticity of the concrete has been reached. According to this engineer's assertions, the slipping of the reinforcement within the body can either be caused by the elongation of the iron when it reaches the limit of elasticity, or by shearing of the concrete. These assumptions are completely proved by loading tests which were carried out to the breaking of test pieces. Should the breaking of such a test piece occur in the centre, the adhesion of the concrete to the iron ceases through the elongation of the iron. Should, however, the test piece break near the supports, the reinforcement is dislocated by the shearing forces. This proves that the adhesion of the concrete to the iron is attacked by external forces which at a certain limit destroy, the adhesion. In any case it is most interesting to know whether these forces give such an elastic change to the form of the body before reaching the limit, that both materials cannot be considered as co-operating.

From the purely theoretical point of view it is evidently inaccurate to assume that the fibres of the concrete and iron lying close together are in the same way pulled or compressed.

Even when shearing stresses resulting from the external forces are not appearing, the diversity of the material must necessitate a particular connection between concrete and iron at the point where both materials come into contact with one another. It cannot be disregarded that the theory of elasticity can really only be applied to bodies of the same material. The theory of the strength of materials which does not permit of the consideration of all irregularities in the shape of the homogeneous bodies, is more difficult to apply to the elastic change of form of those bodies which are built up of irregular voids, cracks and widely differing materials. Differently composed bodies do not therefore allow a purely scientific investigation. If we now consider the shearing strains which under the influence of the load come into action at the point of contact of the iron with the enveloping concrete, then we find a fresh reason for the probable disconnection of the concrete and the iron. At another page of my paper I will refer to the fact that the crosssection of a homogeneous body under a bending moment must be subjected to changes of form through shearing forces. In the ferroconcrete body the shearing force, which the iron exercises on the concrete, must have a reaction of the same kind. Experience does not give us any facts regarding the value of the modulus of shearing elasticity, and on the other hand such elastic changes of form are not taken into consideration in the calculation of the homogeneous

bodies. It is, therefore, conformable for the purpose and in accordance with the usual stability calculations, to disregard the elastic change of form resulting from the shearing forces. One must therefore be satisfied to admit an eventual elastic sliding of the iron in the concrete without being able to take the change of form into consideration. All practical tests have proved that the iron takes an actual part in the elongations of the concrete.

UNIFORMITY OF FERRO-CONCRETE.

Uniformity in the ferro-concrete is the chief factor, and upon this the complete co-operation of the iron and the concrete entirely depends. There is, for instance, the Monier system; in this system a reinforcement of iron rods of small cross-sections are used, which are divided in the concrete in regular and short distances, and thereby both materials are as closely as possible connected with one another. For such a system it must be admitted that the smallest particles of the enveloping concrete take an active part in the bearing of the stresses, to which the reinforcement may be subjected. If, however, the reinforcement is limited to only a few iron bars of larger diameter, or if it consists of profilated iron embedded in greater distances from one another, the co-operation of both materials will not be the same as in the former system, and the actual bearing power will be less if these reinforcements are not interlocked by cross-connections.

This has been fully recognised by practical men, and most of our constructional systems are based upon the principle that it is better to divide the reinforcement uniformly, and to distribute iron rods of smaller cross-sections at shorter distances. A construction disregarding this principle cannot be acknowledged as a ferro-concrete building system, as the assumption of the co-operation of both materials is not applicable with such a form of construction, and economy cannot be effected as they require a greater factor of safety. This will be especially proved with ferro-concrete constructions under a bending load, which naturally causes shearing stresses. The disadvantage of a reinforcement of greater cross-sections will not be so apparent with parts under compression or with bridge arches. where shearing stresses of importance seldom appear.

IMMUTABILITY OF THE CRoss-Sections.

The common law of the strength of the building materials is based upon the following assumption :

Each cross-section of a prismatic body, which is subjected to stresses caused by external forces remains even during its change of form, and all scientists engaged in the investigation of ferro-concrete have accepted this law as a basis for their researches. It may, however, be possible that this assumption is not correct. I may mention that it has been proved through extensive tests, that in concrete bodies under tension, the tensile strains are not equally distributed over the cross-sections of the body. This fact proved

by the above-mentioned tests will doubtless exist in ferro-concrete structures under a bending or compression load. Supposing we had to investigate a column or a concrete wall, I am sure you will agree with me that the assumption of the immutability of the cross-sections cannot be correct, at least not at the point where the external force strains the structure, as the load is always distributed over a more or less extensive area of the surface of the structure. But the impossibility to estimate the manner of load distribution in structures makes every investigation in that direction dubious.

If we take the case of a body under a bending load, we will also find that the assumption in question does not possess a scientific basis. The law of bending, as taught to-day, shows us that the assumption of the immutability of the cross-sections was originated for the purpose of simplifying our calculations. In certain limits, however, it was possible to prove this assumption through tests which were made with bodies of rectangular cross-sections, but nothing has confirmed it with bodies of irregular shape, but on the other hand it is known that the immutability is incompatible with the existence of shearing stresses, because these cause the section to curve into an S-shaped form.

LAW OF THE ELASTIC POWER.

Through the two above-mentioned assumptions, the manner in which a body of ferro-concrete changes its form is defined, and it only remains to determine the law, which will define in the equilibrium actions the elastic power of the concrete in connection with the corresponding changes of form.

This question has troubled theorists very much, and has been the cause of long disputes..

To make the point in question as clear as possible, we will only consider the condition of prismatic bodies subjected to bending.