Its chief characteristic, as some one has said, is its lack of characteristics. The rash usually begins about the fourth or fifth day, often first appearing upon the hands, forearms, and feet. It may remain confined to these parts or spread to the chest, forehead, and remainder of the body. The rash is often morbilliform in character and, as McCulloch says, often so like measles that there is no use trying to

FIG. 3.-Graphs of observed temperature and pulse rate in dengue fever case (typical textbook attack).

differentiate the two rashes. Perhaps almost as often, however, it is scarletiniform in character, and may cover the whole body. Typical urticarial rashes are occasionally described in dengue, and Phillips described one case with a petechial rash resembling typhus. King described a flea-bitten rash and thinks the rash is most pro

nounced where the pains are greatest. Hare states that a rash resembling prickly heat is common in the Tropics, but with the lesions less elevated. Agramonte, Von Dühring, and others state that the eruption may occasionally last only a few hours and may easily be overlooked by the physician or nurse. The rash usually lasts for three or four days, but in occasional instances has been recorded by Robertson et al., Von Dühring, and others, to have persisted for from two to four weeks. Itching and burning are not uncommon at the site of the eruption, especially of the soles and palms. The itching usually persists for only a day or two.

During convalescence a fine furfuraceous desquamation is quite common, which, at times, may amount to peeling, as noted by Kraus, Agramonte, Sandwith, Graham, and others; and it may be as pronounced as that seen in moderate cases of scarlatina. Goldsmid and Crosse have described a fine stippling of the soft palate as often the only rash seen at the first visit.

Gastro-intestinal symptoms.-The tongue is usually covered with a whitish yellow fur which gives way at the edges to a clean red mucous membrane. The appetite in practically every case is lost and may amount to a loathing of food. However, in Cleland's thirteen experimental cases only one showed anorexia. There may be a feeling of pain or discomfort in the epigastrium; and nausea, often accompanied by vomiting, is quite common. Jaundice is very rare in dengue, but when occasionally seen, as by Goldsmid and Crosse, is always described as mild and transient. Mild constipation is often present, but yields readily to laxatives.

The loss of from 7 to 14 pounds is usually suffered during an attack of dengue, probably a result of the fever, anorexia, etc.

Pulse. The pulse in dengue usually varies with the temperature and may be rapid (Chart II), but more often it is slower than would be expected in most fevers of corresponding degree.

Rush did not mention this feature at Philadelphia, 1780, but it has been observed in most epidemics since that time, the pulse in many cases not going above 100. During convalescence the pulse is also usually slow. Koizumi noted rates from 44 to 48 as common during this period. Faget's sign is very rare in dengue.

Genito-urinary symptoms.-The urine, as in most fevers, is reduced in amounts. In most epidemics, albumin is not described as being present, but is occasionally encountered. When present, it is usually transient and in small amounts, coming on about the end of the second day and disappearing with the fever. McCulloch found no albumin in his cases; Hanabusa noted small amounts in 8.3 per cent of his cases; Hare states he observed slight amounts commonly; Carpenter and Sutton observed albumin in 6 out of 122 men; Koizumi noted albumin in 15 per cent of his cases; Agramonte

in Habana, 1905, states that albumin was present in practically every one of his 154 cases and in many more seen in consultation, usually beginning on the second day and lasting throughout the disease. The same was true of 70 cases treated at the Los Animas Hospital in the same epidemic. These varied results may be due to the character of the tests employed.

A number of authors describe menstrual disturbances in dengue. Hare thinks that menstruation is increased and prolonged when present, or, if absent, its onset is often precipitated; and he considers this one of the most distinctive features of the disease. Dennis noted these features in three-fifths of his adult female cases, and Robertson states that they are the rule. Rice noted them in 75 per cent of his adult female cases. Craven states that the menstrual irregularities may persist for several years after an attack of dengue.

Glands. Glandular involvement is a feature wherein great variation is apparent in different epidemics. The writer saw no cases of glandular enlargement in Monroe, La, 1922; King, in Fort Worth, Tex., 1907-8, states that enlarged glands were found in a few cases and were without soreness; Goldberger and McCoy noted a few enlarged glands in Brownsville, Tex., 1907, but state that they were discrete and neither tender nor painful; Levy, at Galveston, Tex., 1918, states that a small proportion showed glandular enlargement. On the other hand, Pridmore, describing an outbreak in Burma, 1902, states that 75 per cent of cases showed enlargement of the cervical, axillary, inguinal, and supra-condylar glands; Castellani mentions enlarged and tender glands as present in from 30 to 70 per cent of cases, and states that they often remain for weeks after the acute attack is past. He also states that dengue was produced in two out of three cases by the injection of aspirated gland juice into healthy subjects. Lane, in the Virgin Islands, 1918, noted affection of all the large superficial glands in 47 out of 75 cases. There was no suppuration, but in many cases the glands were so painful as to require the application of ice.

McMullin describes enlargement of the spleen, coming on with the secondary rise of temperature, in about 50 per cent of cases; Ardate also mentions splenic enlargement as common.

Joints. Swelling of one or more joints is not uncommon in some epidemics, though almost absent in others. No swollen joints were encountered by the writer in Monroe, La. Skottowe, in the Fiji Islands, did not see a swollen joint, neither did Levy at Galveston. Cleland states that, with one doubtful exception, he saw only one joint affection. The swelling when present is described as being a sort of puffiness of the tissues about the joint and never leads to suppuration. Pridmore states that joint pains and swelling may persist in a few cases and cites one case in which they lasted for one

month. Castellani mentions the same condition, which, though rare, he states, may last seven to eight weeks.

Nervous symptoms.-During the first days of an attack there is usually great restlessness, with insomnia, lasting for three to four days. These features are less in children. The sense of taste is almost universally altered, and many patients complain of a bad taste, with a loathing of food. Photophobia is not uncommon during the earlier days of the ailment. Itching has already been mentioned, and paresthesia of the skin is not uncommon. All types of mild mental confusion may exist during the period of high fever, and occasionally a patient is seen who is drowsy, indifferent, responds with difficulty to questions, and rapidly sinks again into somnolence. The writer saw one such case in Monroe, La. Moulliac mentions such a case in which the symptoms persisted for two months, recovery being without sequelae. Giddiness is often complained of upon rising, and Goldsmid states that fainting was not uncommonly observed by him in elderly women. Couffon and Pagnier state that they observed nearly constant absence of knee jerks and pupillary responses to light, but that these symptoms appeared several days after the temperature reached normal. (From his description it may be questioned as to whether this outbreak was really dengue.)

The asthenia, mental depression, dejection of spirits, and irritability which often follow the attack are quite marked in dengue, even in a few cases leading to suicide, as noted by Love.

Blood findings.-The white-blood count is quite characteristic in dengue. Practically all clinicians who have studied this feature of the disease have noted marked leucopenia, with reduction of the percentage of neutrophiles, accompanied by a relative and absolute increase in the mononuclear elements, especially the lymphocytes. The percentages of large and small lymphocytes vary considerably in different cases and in the same case at different times. So, while they do behave in a somewhat characteristic fashion, they are of less importance from a diagnostic standpoint than is the leucopenia with reduction of the polymorphs.

Stitt, in the Philippines, studied 100 cases and found the average white count to be 3,200; the lowest, 1,700; the average percentage of polymorphs was 51 per cent; the lowest, 29 per cent.

Ashburn and Craig give the following percentages for the whiteblood cells in a case on different days of the disease:

Harnett, Stitt, Carpenter and Sutton, and others have observed an increase of the eosinophiles, beginning about the third to sixth day and continuing well into convalescence, and they attribute considerable diagnostic significance to it.

Red cells. Carpenter and Sutton, Stitt, Vedder, Graves, and others have noted no change in the red cells, either their number or hemoglobin content, whereas Eberle and Levy, on the other hand, report a reduction in red cells and hemoglobin as common.

Complications.

Complete recovery is the rule in dengue, but complications occasionally occur. It has been suggested that the marked leucopenia in the disease may leave the patient susceptible to various infections.

Boils or small abscesses are mentioned by Allen, Kennedy, Hare, Skottowe, and others. Orchitis, inflammation of Cowper's glands, pericarditis, catarrhal ophthalmia, and other pyogenic affections have been occasionally described.

Eye complications.-Barkan notes one case of paralysis of accommodation which was first noted two weeks after the onset of dengue. In this case there was no response to accommodation and but slight response to light. The Wassermann was negative. The patient gradually recovered. Van Milligen mentions two cases with normal vision who developed weakness of accommodation with presbyopia. Upon rest, both recovered. Gibson encountered a case of acute glaucoma in a lady of 58, which began on the second day of dengue. Hare mentions a case of acute inflammatory glaucoma following dengue. Barkan noted abducens paralysis coming on nine days from the onset of dengue and clearing up later. Gibson observed three cases of "keratitis dengue" and five cases of "keratitis postdengue." Archibald observed keratitis in one case.

Spadero mentions one case of choroiditis following dengue. The writer noted that his eyes tired easily for several weeks following the attack.