* Loc. cit., p. 48, 55.

I am convinced that the reverse is the case. The highly contagious influenza extends with great rapidity wherever an infected individual is found; the far less contagious, but markedly localized dengue fever remains confined to its locality (sea coasts, ships) and extends with difficulty into the interior, because the local conditions there for the exogenous propagation of the germs are ordinarily not so favorable as on the sea or river coasts.

Very interesting are the most recent reports of Skottowe on the dengue epidemic of 1885-1886 in the Fiji Islands. The introduction of the disease into Numea (New Caledonia) was due to the fact that an infected European was permitted to land. The time required for the disease to extend over all the various inhabited islands (about 80) was from May, 1885, to March, 1886. Everywhere the places on the coast were first affected; then the pestilence spread up along the river bank. The distribution into the interior was slow and sparse. Just before and just after the outbreak of dengue there was an influenza epidemic, and thus one could, states the observer, recognize the great difference between dengue and influenza. The latter became disseminated, like the wind, over all the islands,-"justifying the hypothesis of atmospheric transmission,"-while dengue, on the other hand, as a contagious disease, remained confined to the coasts and became distributed only moderately and slowly toward the interior. The comparison is interesting, but the explanation is incorrect, as we have shown above.

In explaining the different mode of distribution of dengue and of influenza, there is one factor which deserves greater recognition than has been accorded to it up to the present time. The dengue patient, even in the mild cases; on account of the articular pains, is generally immediately confined to the bed and to the house, while innumerable influenza patients continued their journeys to all quarters of the compass and thus carried the germs to all parts with great rapidity.

The difficulties in explaining the various epidemiologic relations in dengue are lessened and the contradictions resolved if we assmne

* Zulzer writes: "Dengue is preeminently characterized by its contagiousness; it is infectious in the "true sense of the word, and that apparently in an even higher degree than typhus or small pox. Influenza, on the other hand, is not contagious." Von During: "Dengue is exquisitely contagious but influenza not so." that dengue, although under certain conditions a disease directly contagious from person to person, nevertheless is also dependent upon conditions of time and place in so far as the development and increase of the contagion outside of the human organism are concerned. This process is at present elastically termed "miasmatic." According to this definition, dengue would, therefore, belong to the contagious miasmatic diseases. [Graham has described a parasite, resembling Pirosoma, in the blood of cases of dengue fever. According to him, the disease is transferred by the mosquito Culex fatigans, and he has been able thus experimentally to convey the disease. The parasite passes through a sporing stage in the mosquito in an analogous manner to the malarial parasite.-Ed.]

We will now consider some other important epidemiologic characteristics of dengue:

1. The contagium, the materies morbi, is unknown, but is undoubtedly of a microparasitic nature.

The meager bacteriologic investigations made up to the present time have yielded no result. The elements found in the blood by Cunningham and Charles (1873) and by Laughlin (1886) scarcely deserve mention. Experiments on transmission to animals were negative; only Vodermann claims to have obtained positive results in the ape. As of influenza, it has also been asserted of dengue, that it can be transmitted to dogs, cats, and other domestic animals. Especially in America whole herds of cattle are said to have contracted the disease (with paralysis of the posterior extremities). Our remarks regarding the supposed identity of influenza in man and the horse (p. 587) are applicable here too.

Nothing is known regarding the mode of entry or exit of the contagion nor of its mode of transmission. (Carried by air, perhaps also by water?)

2. There is no definite proof of the transmissibility of the germs of this disease by merchandise. The fact that dengue, a fever of short duration (three to seven days), was in earlier times, when ships were slow, nevertheless carried to some distances, from India to Aden, Suez, and Port Said, from the West Indies to Cadiz, when the ships' crews were undoubtedly free from infection after the lengthy voyage, is partly in favor of its transmissibility by merchandise, linen, etc.

Zulzer pointed out that the long continued, branny desquamations of dengue convalescents is to be kept in view as a factor in the transmission of the disease. In the opinion of several physicians in Smyrna, the disease was imported by rags, which had been imported by the Hebrew rag pickers yearly into Smyrna from Palestine, Syria, and Cyprus. The Greek government therefore forbade the importation of rags from

Smyrna into Greece. Von During quotes the following as an example: "A laundress, with the dengue eruption still on her, washed, ironed, and glossed the linen of five families. In each one of these families a member contracted the disease, and on each occasion that member who had used this clean linen; moreover, these were all individuals who, a few weeks previously, when the epidemic was in that quarter, had remained exempt.

3. The incubation period in dengue, as in influenza, is very brief, varying from twenty four hours to a few days.

The period of incubation in the epidemic of 1889 in Constantinople was generally forty eight hours. The shortest was twenty four hours, and the longest probably not more than four to five days (von During). This was also the case in Athens (Orenstein). In earlier epidemics a longer period of incubation was noted-generally four days (Cotholendy, 1873, at Reunion; Burnett, 1872, in Bombay). Skottowe and Corney (the Fiji Islands, 1885) found the period of incubation a little over twenty four hours; the observer of the epidemic in Hong-Kong (1895) mentions that the period of incubation was generally less than twenty four hours.