The most constant and dominating symptom in the clinical picture of malaria is the fever. The great attention always bestowed on this symptom and the readiness by which it could be proved resulted in its being made the base of classification of the different clinical forms.

Yet the course of the fever in malaria shows such wide deviations -from curves of mathematic regularity to curves of complete irregularity-and sudden changes that the classification could be only a forced one. Moreover, it may not at all correspond to the general severity of the attack.

The association of the fever with the other symptoms in this classification gradually produced numerous forms and categories, until Dutrouleau's expression, "C'est le chaos," seemed entirely justified.

In our opinion, neither the fever nor any other one symptom can be made the base of a classification of the malarial diseases, for the simple reason that the form of the fever and the severity of the infection do not correspond. As an illustration, let us take quotidian fever. It would be a great error to put two such cases under one head, on account of the similarity of the fever; for one case might be caused by a plural infection with quartan parasites, the other by an infection with the small pernicious parasites. The whole consideration of these two cases from an epidemiologic, clinical, prognostic, and even therapeutic standpoint would, therefore, be different. In all confidence, therefore, we plead for a new rational classification on the basis of the species of parasite which causes the infection. How this is to be carried out will be seen in the special part of this treatise.

Returning now to the symptom of fever, we may say, first, that it is extremely exceptional for a malarial infection to run its course from beginning to end without any elevation of temperature, though, on the other hand, during the course of the infection, long apyretic intervals may occur, so that it is not justifiable to conclude that the disease has terminated on account of the temporary absence of pyrexia.

Observing the forms of fever, we find that all known types occur: remittent, intermittent, and continued. Moreover, these types may occasionally alternate in the course of the same disease, and an intermittent may prove regular or irregular.

The regular intermittent fever occurs so rarely in connection with other diseases, and is so common to certain malarial infections, that it has stamped itself on malaria and lent it its name, intermittent fever.

We distinguish a quotidian fever when a paroxysm occurs daily, a tertian fever when there is an interval of one day between two fever days, and a quartan fever when there is an interval of two days. In addition to these common types there are others, as the double quartan, with a paroxysm on two days following each other, then a day interval; triple quartan, double tertian with daily paroxysms; duplicate quotidian fever * with two paroxysms on one day, etc. These designations take into consideration the causal factors of the fevers, which, from a general pathologic standpoint, would be regarded only as quotidian fevers.

Intermittent malarial fevers with long intervals, as, for instance,, septan fever, have been observed by Kelsch and Kiener in Algiers, by Borius, Thaly, and Mahe in Senegal, by Laure in Guiana, by Gelineau on Mayotta Island, and others. As will be shown in the section on Relapses, these fevers are not to be regarded as particular types caused by specific species of parasites, but as relapses occurring with a certain regularity.

The intermittent types may be regular and show corresponding curves, or they may manifest certain irregularities. Among these irregularities we may mention, first, anticipation and postponement of the paroxysm, in that this may occur a few hours before twenty four (or forty eight or seventy two) hours, or a short time after the regular interval. The former is much more frequently observed. The postponing type is usually the result of quinin treatment and precedes recovery.

Another form of irregularity consists in the change of type; for instance, a quotidian after several tertian paroxysms, or vice versa.

* We differentiate, with Sauvage, a febris duplicata or triplicata from a febris duplex or triplex; by the former we understand the occurrence of two or three paroxysms on one and the same day; by the latter, the occurrence of two or three paroxysms on different days. Further details will be found in the Special Part.

The quartan type may also alternate with the quotidian. A change from tertian to quartan by postponement, as described (doubtless theoretically) by older writers, does not occur; at least, I have never succeeded in finding a convincing example. The explanation of this limitation of the change of type will be found under Special Remarks.

Finally, there is an irregular intermittent fever, occurring especially in connection with quotidian and tertian, in which the well known characteristics can no longer be recognized through the irregularities, and an apparently arbitrary form is assumed. By remittent fever we understand a subclassification of continued. The difference is that in remittent the temperature now and then falls almost to normal, while in continued it remains at approximately the same level.

It is customary, too, to designate the remittent of malaria as sub continued or subintrant (or subingrediens). The remissions may show a regularity by recurring at definite intervals and to the same degree. This form is not uncommon, and may be produced by a prolongation of the individual paroxysms, by anticipation, or by increase of the paroxysms within a definite interval. Several factors often act together. In these cases, therefore, the second paroxysm sets in before the first one has fully terminated.