KESEAECHES UPOF FEVER, 425 



than compensates for the loss of the food heat. In the dog at least, how- 

 ever, there are periods of high temperature in which less heat is being 

 produced than after a full meal in the normal animal. It is plain that 

 disturbance of the heat production is a factor of the febrile state; but 

 it would seem that there is also a disturbance of the heat evolution, else 

 more of the extra caloric would be gotten rid of. It has been shown 

 that the diurnal rhythmic changes of temperature are usually parallel 

 in health and fever; and it would seem that in regular fever there is a 

 moving up of the habitual plane of heat upon which the organism is 

 run. What has been made out as to the methods of heat ijroduction . 

 and evolution naturally suggests, as a plausible theory of fever, that the 

 febrile state is chiefly due to a loss of power of the upper centers, mus- 

 culo- vasomotor, or heat inhibiting, as the case may be, and to a less 

 extent of sensibility of the vasomotor centers which preside over heat 

 evolution. As the result of the weakening of the npiier centers, more 

 heat is developed; whilst the benumbed vasomotor centers require a 

 greater degree of heat to stimulate them to throw oft' the excessive 

 caloric produced. In investigating the truth of this theory, the first 

 efibrt was to determine how fever is caused. Many fevers are evidently 

 the result of the introduction of a poison in the blood ; but it has been 

 thought that inflammatory fever is produced by irritation of peripheral 

 nerve. If, however, all the nerves going to the hind leg of a dog be 

 divided, and after recovery has taken place the anaesthetic limb be 

 wounded, traumatic fever is developed as soon and as intensely as in 

 the normal animal similarly wounded. Such fever would seem to be 

 due to an absorption of a poison into the blood ; for fever is a general 

 pathological process, involving the whole system, and the blood in the 

 wounded animal, with divided nerves, is the one apparent channel of 

 communication between the injured part and the general system. It is 

 plain that a poison circulating in the blood may affect heat production, 

 either by acting upon the nerve centers or by acting on the tissues of the 

 body. There are certain clinical malarial phenomena which tell very 

 strongly against the idea of fever being due to an action upon the uni- 

 versal protoplasm. Such phenomena are the regular sequence of com- 

 plicated symptoms which usually make uj) the intermittent paroxysm j 

 the frequent regular interchange between this fever paroxysm and neu- 

 ralgic and other nerve storms, one replacing the other; the occasional 

 occurrence of local paroxysms of intermittent fever involving only or 

 chiefly certain regions of the body. To aid in elucidating the matter, 

 two series of experiments were performed: In the first, the ettect of 

 section of the cord was studied in animals suffering from fever. The 

 result was an exaggeration of the phenomena which occur after spinal 

 section in the uninjured animal, heat evolution being enormously in- 

 creased at first. This indicates that in fever the vasomotor nerves are 

 more active in restraining heat evolution than in health. The second 

 series of experiments consisted in irritating peripheral sensitive nerves 



