The different varieties of diabetes may not only be separated from one another by the differences alluded to above in the composition of the urine and in the degree to which the urine responds to dietetic treatment, but they may also be differentiated by their morbid anatomy and also to some extent by their clinical course. Thus the more severe forms of the disease are commonly associated with lesions of the pancreas, and hence are spoken of as pancreatic diabetes, and such cases commonly run an unfavourable course accompanied by much emaciation. A totally different type is that seen in middle-aged obese patients where the quantity of sugar excreted is far less, and where some of the more characteristic features of the disease, such as great thirst, are absent. In these cases of fat diabetes the clinical course of the malady is often much more prolonged and less severe. Again, there is a very distinct type seen in middle-aged persons associated with more or less advanced arterio-sclerosis, where the amount of sugar excreted is usually small, but where in addition albumin is present, and the clinical course of this variety is very different to that of the other forms of the disease, and such patients are far more liable to complications dependent on the vascular degeneration that is present, e.g. cerebral haemorrhage, thrombosis, gangrene of the extremities, etc.

It is obvious that these different types of diabetes must necessarily require very different dietetic treatment; thus to take the arterio-sclerotic form of the malady, it is clear that a highly nitrogenous protein diet cannot be partaken of in such cases without very considerable risks being incurred.

In regulating the dietetic treatment of this malady many fac tors have therefore to be taken into consideration. First the type of the disease as determined by the age, duration and clinical course of the malady, as for instance on the one hand the severe type seen in young adults characterized by great wasting and the excretion of a very abundant urine containing a very high percentage of sugar, together with acetone and diacetic acid, and on the other hand, the type seen in middle-aged or elderly fat patients where the symptoms are slight and the amount of sugar excreted far less.

The second most important point is the degree to which the excretion of sugar can be affected by dietetic measures. In all cases of diabetes, before any treatment is undertaken, medicinal or dietetic, the patient should be put on an ordinary diet, and careful determinations made of the total quantity of sugar passed in the twenty-four hours, together with determinations of the presence or absence of diacetic acid, acetone, etc. When these facts have been ascertained the diet should be gradually restricted until all carbo-hydrates are eliminated from the food. During this period no medicines, such as opium, antipyrin, etc., should be given. When the patient has been given a rigid diet for several days a further analysis of the urine should be made and the quantities of sugar excreted on this rigid diet compared with those passed on an ordinary diet. It will be found that the cases fall into one of three groups : in the majority of instances the sugar will undergo a notable diminution and the daily excretion will fall to half or one-quarter of that previously existent. Together with this fall in the output of sugar there will be a notable improvement in many of the symptoms, especially the thirst. In a certain proportion of cases, more especially where the quantities of sugar voided on the full diet were not large, the rigid diet will lead to a complete disappearance of the glycosuria. Lastly, in a third group of cases the rigid diet produces comparatively little effect on the sugar output and the patient continues to pass large, and sometimes very large quantities of sugar on the most rigid diet that can be devised. These three groups of cases indicate different degrees of severity of the malady, and perhaps in some instances differences in the nature of the lesion. They at any rate indicate very clearly that different dietetic treatment is necessary.

Lastly, the third factor that has to be taken into consideration in the dietetic treatment of the disease is the influence of the diet on the body-weight. Observations should also be made on the temperature, as a subnormal temperature is always a danger signal.