U.S. Army Medical Department, Office of Medical History
Skip Navigation, go to content







AMEDD MEDAL OF HONOR RECIPIENTS External Link, Opens in New Window






Chapter XIII



Fort Bragg Fever

Worth B. Daniels, M.D.

The recognition of Fort Bragg fever as a specific new disease entity and the ultimate proof of its etiology is a contribution of the U.S. Army Medical Department to the science of medicine. The disease was described by Army clinicians, studied by Army medical personnel with the assistance of Army-consigned consultants, transmitted to animals by an Army research worker, and finally proved as to etiology by an Army veterinarian and others working at the Medical Department Professional Service Schools, Army Medical Center, Washington, D.C.

The story of Fort Bragg fever indicates, too, the superior opportunities which were available to military medical personnel for clinical research. Larger groups of patients with the same disease are more often available for study in military installations than in civilian institutions. It is doubtful whether this disease would have been recognized as an entity had the outbreak occurred in an urban civilian community; there, each patient might have been cared for by a different physician and treated in a different hospital, whereas, on an Army post all were concentrated in one hospital under the care of a closely knit medical service.


Late in July and early in August 1942, an unusual febrile illness occurred in a group of soldiers at Fort Bragg, N.C.1 It shortly became apparent not only that these men had identical symptoms but also that they all came from a few organizations quartered near one another in a limited area of the reservation. Between 29 July and 11 September, 40 patients with this illness were admitted to the hospital. The history was one of relatively sudden onset characterized by malaise, mild general aching, lumbar pain, severe frontal headaches, and postorbital pain. On the first or second day of symptoms, mild respiratory manifestations consisting of coryza, sore throat, pain and soreness in the chest, and cough occurred in 30 percent of the patients. The respiratory symptoms were not persistent and were never suggestive of primary respiratory involvement such as is seen in influenza. In about one-fourth of the cases, nausea and vomiting occurred, rarely accompanied by abdominal pain. Shaking chills or chilliness and fever developed. The fever was consistently spiking and frequently showed two or more peaks each

1Daniels, W. B., and Grennan, H. A.: Pretibial Fever; An Obscure Disease. J.A.M.A. 122: 361-365, 5 June 1943.


day. Recurrent chills often accompanied the elevations. During the periods of temperature elevation, severe accentuation of the frontal and postorbital aching was experienced, but during the periods of lower temperature, the patients felt relatively well. The fever persisted for 2 to 8 days-averaging 5.4 days-with maximum elevations ranging from 99.8 to 105.6 F. In five patients, a transient elevation of temperature, sometimes as high as 101.4 F., occurred from 2 to 7 days after the original febrile period. Stiffness of the neck accompanied headache in three patients, but examination of the cerebrospinal fluid revealed it to be normal; there was no noticeable relief of headache following lumbar puncture. Adenopathy was not remarkable. A firm spleen was palpable early in the disease in 95 percent of the patients. Splenomegaly persisted in some patients for as little as 5 days; in others, there was still noticeable enlargement after 2 weeks.

The most distinctive feature of the disease, however, was the appearance of an unusual rash on or about the fourth day of illness. In 60 percent (24) of the patients, this was bilaterally symmetrical and limited in distribution to the pretibial areas; in an additional 20 percent (8 patients), the pretibial areas were the primary site of the rash, and a few lesions were scattered elsewhere. Two patients had splotchy, generalized cutaneous manifestations including the anterior surface of both legs. One had a single lesion on the hand. In five cases, typical in all other respects, no rash was observed. Individual lesions consisted of an erythematous localized blush of irregular outline with ill-defined borders fading into the surrounding skin. These were often from 2 to 5 cm. in their largest diameter, gradually coalescing with adjacent lesions. The lesions were raised, warmer than the surrounding skin, and sometimes slightly tender to touch. In some patients, the lesions vaguely resembled erythema nodosum. In two patients, the rash became diffusely distributed over the entire body, and in a few it appeared urticarial. Following the generalized type of rash there was a residual pigmentation which persisted for about 2 weeks. None of the lesions were purpuric. In most instances, the cutaneous manifestations lasted 2 days, but they persisted longer in a few patients. Figures 57 and 58 illustrate the pretibial and the generalized forms of the rash.

Biopsies of cutaneous lesions were performed in six typical cases. They showed diffuse edema and a slight to moderate perivascular infiltration with small round cells and macrophages. It is clear that drugs played no part in the development of this rash because only one patient had received sulfonamide therapy and no other patient received any drug except acetylsalicyclic acid and codeine. Leukopenia was noted sufficiently often to constitute a typical feature; it was present, in all except five patients, at some time during the acute illness. It developed most often between the third and the fifth day of illness. At the termination of the febrile period, the leukocytes again rose to normal, and in 14 patients a slight leukocytosis occurred. The number of leukocytes ranged from 2,800 to 14,000 per cubic millimeter. Dif-


FIGURE 57.-Erythematous skin lesions over the pretibial regions.

ferential counts usually showed no deviation from normal, although in a few cases a moderate relative lymphocytosis was present.

Cultures of the blood showed no growth. Agglutinations for Proteus OX-19, Brucella melitensis, the typhoid-paratyphoid group, and heterophile antibodies were negative. No patient had been in an area where coccidioidomycosis was endemic. Skin tests using coccidioidin antigen were negative in two patients after their recovery. Roentgenograms of the chest were made in many patients, but none showed any abnormality.

At the termination of the fever, the patients rapidly returned to their original healthy state. There was no postfebrile depletion or depression, and there were no complications.


Case 1-A white soldier, aged 25, admitted to the hospital on 7 August 1942, complained of fever and headache. He had been well until 2 days previously, when he felt feverish and chilly. Severe frontal headache, postorbital pain, and severe backache developed. The next day he felt slightly better, but on the day of admission the symptoms were all accentuated. The temperature was 101 F., the pulse rate 90, and the respiratory rate 20; the leukocytes numbered 8,150 per cu. mm. Nothing unusual was found on physical examination. On the fifth day of the disease, the spleen was firm and easily palpable. Over the pretibial surfaces of both legs, irregular, erythematous lesions were noted; these were faintly tender, warm to touch, and very slightly raised (fig. 57). The patient's temperature was swinging in character and ranged from 99.8 to 104.4 F. without a corresponding rapidity of pulse. The leukocytes numbered 4,900


CHART 24.-Temperature, pulse, and leukocyte count of patient

per cu. mm., with 65 percent polymorphonuclear cells. On the seventh day of illness, the rash had entirely disappeared, the temperature was normal, and the patient felt well. The spleen was still palpable at discharge from the hospital.

Case 2-A white soldier, aged 21, admitted to the hospital on 9 August 1942-the third day of disease-had as initial symptoms frontal headache and pain in the feet and legs which later involved various joints. At the onset there was nausea without vomiting. Shaking chills had occurred on two occasions in the 36 hours prior to admission. Physical examination revealed nasal congestion, pharyngitis, signs of generalized bronchitis, a temperature of 104 F., a pulse rate of 92, and a respiratory rate of 20 (chart 24). The leukocytes numbered 9,450 per cu. mm. Two hours after admission a rash appeared on the trunk, neck, and extremities. This consisted of discrete, brilliantly erythematous, slightly raised, warm plaques varying in size from 1 to 8 cm. in their largest diameter. The rash was generalized (fig. 58) and, as in the other cases, the anterior aspects of both legs were involved. During the next few days the temperature ranged from 99 to 104.6 F., with very little acceleration of the pulse. The spleen became palpable. The rash remained erythematous for 3 days, faded gradually, and left a definite purplish pigmentation that diminished during the succeeding 2 weeks. The leukocytes were reduced to 4,350 on the sixth day and to 3,700 per cu. mm. on the seventh day, with no abnormality in the differential count. Convalescence was uneventful.


FIGURE 58.-Generalized form of skin eruption.


A number of different diseases were considered in differential diagnosis. In the early cases, influenza was suspected, but the transient character and mildness of the respiratory symptoms, persistence of fever, presence of a firm palpable spleen, and the unusual rash eliminated this possibility. Endemic typhus seemed to be excluded by the short duration of the illness, by the entirely different type of rash, and by the consistently negative Weil-Felix reactions with OX-19 at different periods of the illness. Rocky Mountain spotted fever was similarly dismissed from consideration. There was no history of bites by ticks in this group. When these patients were studied, Bullis fever2 had not yet been described, although cases were being recognized in Texas. Again, absence of bites by ticks and difference in symptomatology indicated that the cases here described were not the same as those seen at Camp Bullis, Tex. In no cases did adenopathy or the characteristic hematological changes of infectious mononucleosis develop. The duration of the disease, negative cultures of blood and stool, and negative agglutination re-

2Woodland, J. C., McDowell, M. M., and Richards, J. T.: Bullis Fever (Lone Star Fever-Tick Fever): An Endemic Disease Observed at Brooke General Hospital, Fort Sam Houston, Texas. J.A.M.A. 122: 1156-1160, 21 Aug. 1943.


actions eliminated the typhoid-paratyphoid group. The parasites of malaria and relapsing fever were sought, but not found. Dengue was at first thought to be the probable diagnosis. However, as new cases were observed, certain prominent features of that disease were consistently absent. No patient had a "camel back" type of fever curve or any postfebrile depletion or depression. The usual vector of dengue, Aedes aegypti, was not found in the vicinity. It would be unusual for an outbreak of dengue to be limited to such a small group. The type and the distribution of the rash were not similar to that seen in dengue. Other denguelike diseases were considered; however, the clinical features of this disease showed definite dissimilarity to these in one or more important respects. Search of the literature did not reveal a description of a clinical entity into which this group of cases would fit.


The organizations from which these soldiers came were quartered in the northern third of the populated area of the reservation, near a small stream and its tributaries. Other areas of the post furnished no patients ill with this disease; therefore, local environmental factors were thought to be of etiological importance. There was no single locality off the reservation to which these men went during the month previous to illness, and they had no common meeting place on the reservation. Some of the men had done no swimming, while others swam in several different ponds. Although there was no swimming place common to all, more men swam in "McFadgen's Pond" than in any other place-on or off the reservation.

The factors responsible for the outbreak could not be determined. The medical inspector of the post made a search for insect vectors that might be suspected. Bedbugs, stableflies, chiggers, ticks, and horseflies were found but for various reasons were considered unlikely vectors.

During the period of this outbreak, the population of mosquitoes on the post was very low, but various members of the Culex species were found with Culex quinquefasciatus appearing most frequently. This mosquito had been disproved as a carrier of dengue. No representative of the dengue vector, A. aegypti, was found, but Aedes atlanticus, Aedes canadensis, and Aedes vexans were. Their presence was thought to be of doubtful importance.

After a decrease in the number of admissions to the hospital in mid-August, the incidence of the disease again increased sharply. The clinical picture, the long period of incubation, and the leukopenia suggested that the disease was probably due to a virus. The localization of the cases in one area of the post made it seem probable that it was transmitted by an insect vector. The local medical personnel were unable to carry out studies as to etiology, so The Surgeon General, U.S. Army, was requested to send especially trained individuals to study the outbreak.



On 3 September 1942, Dr. Norman H. Topping, Passed Assistant Surgeon, U.S. Public Health Service, Maj. (later Lt. Col.) Cornelius B. Philip, SnC, and Dr. John R. Paul, Professor of Preventive Medicine, Yale University School of Medicine, New Haven, Conn., and Director, Commission on Neurotropic Virus Diseases, Board for the Investigation and Control of Influenza and Other Epidemic Diseases in the Army, arrived at Fort Bragg to study the disease. The Commission reviewed the records of the case previously studied and examined the patients remaining in the hospital. They agreed that the soldiers were ill with a disease unknown to them, but Dr. Topping remembered that an outbreak of a disease with similar features had been encountered in the town of Wrens, Ga., in 1940. This illness was locally known as Brushy Creek fever. Inquiry later brought information from Dr. Charles D. Bowdoin,3 who had observed these patients and reported them at a later date, that about 35 cases had occurred. The clinical course and manifestations seemed to be quite identical with those at Fort Bragg. A laboratory technician who had made some of the cultures of feces was reported as having developed the disease.

The Commission began investigations in regard to the epidemiology and etiology of the disease.4 They showed that, although only 34 patients with a rash had been considered to have this disease by medical officers of the hospital, the number of cases-in one of the regiments involved-without a rash might have exceeded those with a rash by as much as 50 percent. The Commission confirmed the fact that the disease was not spread evenly throughout the camp population; but, of the 34 cases (with a rash) diagnosed between 27 July and 19 September, 75 percent had occurred in 3 regimental units. Practically all of the cases (almost 97 percent with a rash) occurred in troops quartered in the northern third of the developed area. A canvass was made of four artillery regiments quartered in the southern section of the post, and no "missed cases" were found among 4,800 troops.

These features, plus the fact that the rate of attack was particularly high in two units, led the Commission to suspect that some local, although unrecognized, environmental feature was responsible. Particular attention was paid, therefore, to two units. In each of these, an epidemic amounting almost to an explosive outbreak had occurred. These outbreaks were separated by about 3 weeks and were limited, to some extent (in the early cases), to a single company.

These explosive epidemics became apparent when a survey of the records of illness of one unit in particular, namely, the 88th Airborne Infantry,

3Bowdoin, C. D.: A New Disease Entity (?). J.M.A. Georgia 31: 437-438, 442, December 1942. 
4Preliminary Report of the Commission for the Study of an Unidentified Disease at Fort Bragg, N.C., 3-11 Sept. 1942, by Dr. Norman H. Topping, P.A. Surgeon, U.S. Public Health Service, Maj. Cornelius B. Philip, SnC, U.S. Army, and Dr. John A. Paul, Yale University School of Medicine. Submitted to The Surgeon General, U.S. Army, 15 Oct. 1942.


revealed that there had been a sharp increase of cases of "influenza,""nasopharyngitis," or "fever of undetermined origin" in this regiment during August. Of the 33 cases of "influenza," and other diseases, which were reported in August, 9 developed a rash after they had been hospitalized and were ultimately diagnosed as having Fort Bragg fever. In 14 others, whose hospital records were available for study, the picture was compatible with that of Fort Bragg fever, although the rash was lacking.

It was concluded, therefore, that a nonexanthematous form of this disease existed and (if all forms of the disease were considered) it could appear in explosive epidemic form resulting in almost 10 percent of the personnel in a given company being infected within the period of 2 weeks. Factors responsible for this type of explosive outbreak were not determined. 

Certain information about the period of incubation was obtained. Five men probably developed the disease while they were away from the post and several developed it within a few days after their return, following an absence of 10 days on maneuvers on Chesapeake Bay. If it is assumed that the disease was not spread by direct contact, the data accumulated would indicate that the incubation period was from 10 to 15 days or longer.

The Commission's entomological observations indicated that if parasitic arthropods were vectors of the disease, there were only three that were found or reported in sufficient numbers and of sufficient distribution to be considered. These were stableflies (Stomoxys calcitrans), mosquitoes, and chiggers (Trombicula irritans). These parasites had also been reported to have bitten men stationed in units in the southern part of the post, where none of the cases occurred.

As to mosquitoes, it was found that a ditched but swampy arm of Tank Creek separating the two most heavily infected units could have provided the most available source of breeding, but reports of adjacent troops and of sanitary personnel indicated insufficient incidence of mosquitoes in both July and early August to account for the outbreak. The evidence of transmission by mosquitoes was not considered strong.

Having collected and frozen material for further study, the Commission departed on 11 September to carry out investigations of etiology in their various laboratories.

Search for an etiological agent was carried out by members of the Commission at the National Institute of Health, U.S. Public Health Service, at the Yale University School of Medicine, and at the Medical Department Professional Service Schools.5

The material obtained for study consisting of blood from acutely sick patients and a variety of insects (including Stomoxys calcitrans) had been frozen and kept on Dry Ice. It was inoculated into embryonated eggs, guinea

5Final Report of the Commission for the Study of an Unidentified Disease at Fort Bragg, N.C., 3-11 Sept. 1942, by Dr. Norman H. Topping, P.A. Surgeon, U.S. Public Health Service. Maj. Cornelius B. Philip, SnC, U.S. Army, and Dr. John R. Paul, Yale University School of Medicine. Submitted to The Surgeon General, U.S. Army, 31 Mar. 1943.


pigs, mice, monkeys, and one chimpanzee. Human transmission experiments were tried by inoculating frozen whole blood and by allowing A. aegypti and Aedes albopictus mosquitoes previously fed on a patient ill with the disease to feed on volunteers. Each one of these attempts to transmit the disease failed. In retrospect, however, the negative findings with frozen human material were explained by an important discovery made about 2 years later, when it was shown that the causative agent of this disease is destroyed by freezing. It is small wonder, therefore, that the experiments of 1942 were unsuccessful. It was not until 1st Lt. (later Capt.) Hugh Tatlock, MC, succeeded in performing direct transmission experiments with fresh blood in guinea pigs in 1944 that the disease was successfully transmitted.

Outbreaks similar to that of 1942 occurred during the summers of 19436 and 19447 among soldiers quartered in the same area of the post. In 1943, a rickettsia-like organism was recovered by Tatlock8 from three of five guinea pigs injected with fresh blood from a patient ill with the disease. Attempts were made to show that this agent was concerned in the etiology of the disease. However, subsequent studies indicated that this organism was undoubtedly of guinea pig origin and not concerned in the etiology of Fort Bragg fever.

In the summer of 1944, Tatlock injected two guinea pigs and two Syrian hamsters intraperitoneally with blood freshly drawn from a patient ill for 4 days with the disease.9 Both guinea pigs developed fever (105 to 106 F.) 8 days after inoculation. Both guinea pigs were found dead on the 10th day. No passage was attempted from the hamsters. Serial transmission of the fever-producing agent in guinea pigs was accomplished by the intraperitoneal injection of citrated blood on the first day of fever. The incubation period was 4 to 8 days and fever persisted from 2 to 4 days. More than 70 passages produced no increase in virulence of the agent. No gross abnormalities were found in sacrificed guinea pigs. Histological sections of the liver showed areas of focal necrosis. No inclusion bodies or Rickettsia were seen in sections.

Tatlock found the agent capable of infecting guinea pigs, rabbits, and Syrian hamsters by both intraperitoneal and intracerebral inoculation. The disease transmitted was uniformly fatal for hamsters. Intravenous inoculation of 11-day chick embryos was accomplished. From these infected eggs, the agent was propagated by the yolk-sac route, and fresh tissue suspensions were infective for hamsters in 1 to 100,000 dilution. The agent passed a Corning fritted glass filter but failed to pass a single Seitz pad. Tissue emulsions in sterile skim milk, frozen rapidly and maintained at -70 C.,

6Personal observation of the author.
7Personal communication, J. M. Kinsman, to the author.
8Tatlock, H.: A Rickettsia-Like Organism Recovered From Guinea Pigs. Proc. Soc. Exper. Biol. & Med. 57: 95-99, October 1944.
9Tatlock, H.: Studies on a Virus From a Patient With Fort Bragg Fever (Pretibial Fever). J. Clin. Investigation 26: 287-297, March 1947.


retained viability. However, blood, whole tissues, or tissues suspended in broth lost infectivity when stored for 24 hours at 20 C., 4 C., or at -70 C. in sealed glass ampules.

Sera from five patients with Fort Bragg fever were used in protection tests. Neutralization was obtained in two patients with convalescent, but not with acute, sera. Comparable results were obtained with two sets of sera from infected guinea pigs.

By early 1945, Captain Tatlock had brought his studies to a point where he and others interested in the disease were confident that Fort Bragg fever was due to a well-characterized filtrable virus unrelated to any other known virus. At this juncture, he was ordered to Walter Reed General Hospital, Army Medical Center, for clinical duties but maintained the agent there by passages and by storage, frozen in an emulsion of milk.

During the summer months of 1945, no patients with this disease were admitted to the hospital at Fort Bragg. It is of interest that a large part of the post area in which patients had previously seemed to acquire the disease was unoccupied except for a period of "a week or two."10

Isolated reports appeared of single cases of Fort Bragg fever in Connecticut,11 New York,12 and California.13 From the descriptions given, these patients may have had this disease, although it would seem difficult to make an absolute diagnosis in the absence of an outbreak.


In the spring of 1946, Captain Tatlock was ordered on detached service to Cincinnati, Ohio, where, through the cooperation of Dr. Albert B. Sabin and of the Longview State Hospital, he began studies in transmission among a group of patients undergoing fever therapy. He inoculated three patients intramuscularly and intracutaneously with a 10-percent suspension of infected, embryonated chick liver in saline. This material had been through 80 passages in guinea pigs and 23 passages14 in embryonated eggs. After an incubation period of 8 to 9 days all three patients developed a short febrile illness. Successful inoculation of three other patients, using freshly defibrinated pooled blood from the previous patients, was accomplished, and a third passage was successful in seven of eight other individuals. All but one of the latter patients, after an incubation period comparable to that of the natural disease (8 to 14 days), developed fever, and the majority exhibited the clinical picture of Fort Bragg fever with rash and leukopenia. For the third passage, two patients immune to "New Guinea C" strain of dengue fever virus, two immune to the "Hawaiian" strain of virus, and two immune

10Personal communication, J. H. Dingle, to the author.
11Lipscomb, L. L., and McMahon, J. L.: Pretibial Fever. J.A.M.A. 128: 90-91, 12 May 1945.
12Greenfield, I.: Pretibial Fever; A New Disease Entity. Urol. & Cutan. Rev. 47: 435-436, July 1943.
13Personal communication, G. Cheney, to the author. 
14See footnote 9, p. 445.


to the "Middle East-Sicilian" type of sandfly fever virus were used. All were found susceptible to the "virus" of Fort Bragg fever.

"Viremia" was demonstrated in these patients by the immediate inoculation of fresh, defibrinated blood into young hamsters. The "virus" appeared in the blood of these patients shortly before the onset of fever and disappeared rapidly thereafter.

It now appeared clear to all interested in the disease that the "virus" isolated by Tatlock from a case of Fort Bragg fever in 1944 was a new and distinct agent capable of producing the classical manifestations of the natural disease when inoculated into man.

In 1947, in New Haven, Melnick and Paul15 inoculated four chimpanzees intracutaneously and subcutaneously, using the "virus" isolated by Captain Tatlock. These animals developed a disease similar in many respects to the experimental disease transmitted to humans by Tatlock, and subsequent bleedings showed that they had developed neutralizing antibodies. When "virus" inactivated by freezing was inoculated into five chimpanzees, no disease or neutralizing antibodies developed. Two of these animals were later given active "virus" and both developed antibodies. Two animals were studied for "viremia" during the period of fever. This was demonstrated in one of the animals by transmission to a chimpanzee and in the other by transmission to young hamsters.

During the ensuing years, the "virus" of Fort Bragg fever was maintained in several laboratories. The possibility that this agent might be a Leptospira was considered several times by Gochenour, Smadel, and others at the Army Medical Service Graduate School, Walter Reed Army Medical Center.16 However, results obtained with methods then employed failed to support this idea.

Later, a large collection of leptospiral strains was assembled by this group. In the fall of 1951, Gochenour, an Army veterinarian-with Smadel and other coworkers-found that with antigens prepared from these strains convalescent and immune sera of victims of Fort Bragg fever gave a high titer of agglutinating antibodies against Leptospira autumnalis. With this lead, a leptospiral organism was recovered by culture from the 365th passage of the Fort Bragg agent since its original isolation. Cross-agglutination tests showed that the Fort Bragg agent was essentially indistinguishable from Lept. autumnalis Akiyami A, the cause of autumnal fever in Japan and other areas of the Far East. There was then no doubt that Lept. autumnalis was being carried in this laboratory as the "virus" of Fort Bragg fever. Obviously, it was of the utmost importance to learn whether this was a contaminant or whether it really originated from patients with the disease at Fort Bragg.

15Melnick, J. L., and Paul, J. R.: Experimental Fort Bragg Fever (Pretibial Fever) in Chimpanzees. Proc. Soc. Exper. Biol. & Med. 67: 263-268, March 1948.
16Gochenour, W. S., Jr., Smadel, J. E., Jackson, E. B., Evans, L. B., and Yager, R. H.: Leptospiral Etiology of Fort Bragg Fever. Pub. Health Rep. 67: 811-813, August 1952.


Paired sera from Melnick and Paul's four euphoniously named chimpanzees-Rosebud, Mary Lou, Hickory, and Catawba-showed high titers of agglutinins or neutralizing antibodies against Lept. autumnalis in the convalescent, but none in the acute, sera. Paired sera from six of Captain Tatlock's patients infected in Cincinnati gave the same results.

Fortunately, paired sera from five of the original soldiers who had been bled by Tatlock in 1944 were still safe in the Deepfreeze. Three of these patients were found to have developed neutralizing antibodies and leptospiral agglutinins. Two who failed to develop neutralizing antibodies in Captain Tatlock's experiments also failed to agglutinate Lept. autumnalis.

It is of great interest that three lots of hyperimmune rabbit sera-one prepared in 1947 from the hamster line, another in 1951 from cultured Leptospirae of the Fort Bragg agent, and a third from cultures of Lept. autumnalis Akiyami A-protected hamsters against 100,000 LD50 of the Fort Bragg Leptospira.

Stored sera from 45 Fort Bragg patients ill in 1943 or 1944, though not paired, were studied. Sixteen of these gave clear-cut serological evidence of infection with Fort Bragg Leptospira.

Fort Bragg fever then is caused by a member of the Lept. autumnalis group. No member of this group was previously known to exist in the United States. This disease takes its place with Weil's disease, swineherd's disease, canicola fever, and other leptospiral diseases and is a contribution of Army medical investigation to medicine.