(1) All fleas and bugs which have sucked the blood of animals dying from plague contain plague1 microbes.

(2) Fleas and bugs which have sucked the blood of animals which are suffering from plague only contain plague microbes when the bites have been inflicted from 12 to 26 hours before the death of the animals, that is, during that period of their illness when their blood contains plague bacilli.

(3) The vitality and virulence of the plague microbes are preserved in these insects.

(4) Plague bacilli may be found in fleas from four to six days after they have sucked the blood of an animal dying with plague. In bugs, not previously starved or starved only for a short time (one to seven days), the plague microbes disappear on the third day; in those that have been starved for 4 to 4½ months, after eight or nine days.

(5) The numbers of plague microbes in the infected fleas and bugs increase during the first few days.

(6) The faeces of infected fleas and bugs contain virulent plague microbes as long as they persist in the alimentary canal of these insects.

(7) Animals could not be infected by the bites of fleas and bugs which had been infected by animals whose own infection had been occasioned by a culture of small virulence, notwithstanding the fact that the insects may be found to contain abundant plague microbes.

(8) Fleas and bugs that have fed upon animals which have been infected by cultures of high virulence convey infection by means of bites, and the more certainly so the more virulent the culture with which the first animal was inoculated.

(9) The local inflammatory reaction in animals which have died from plague occasioned by the bites of infected insects is either very slight or absent. In the latter case it is only by the situation of the primary bubo that one can approximately identify the area through which the plague infection entered the organism.

(10) Infected fleas communicate the disease to healthy animals for three days after infection. Infected bugs have the power of doing so for five days.

(11) It was not found possible for more than two animals to be infected by the bites of the same bugs.

(12) The crushing of infected bugs in situ during the process of biting, occasioned in the majority of cases the infection of the healthy animal with plague.

(13) The injury to the skin occasioned by the bite of bugs or fleas offers a channel through which plague microbes can easily enter the body and occasion death from plague.

(14) Crushed infected bugs and fleas and their faeces, like other plague material, can infect through the small punctures of the skin caused by the bites of bugs and fleas, but only for a short time after the infliction of these bites.

(15) In the case of linen and other fabrics soiled by crushing infected fleas and bugs on them, or by the faeces of these insects, the plague microbes can under favourable conditions remain alive and virulent during more than five months.

(16) Chemical disinfectants do not in the ordinary course of application kill plague microbes in infected fleas and bugs.

(17) The rat flea Typhlopsylla musculi does not bite human beings.

(18) Human fleas do bite rats.

(19) Fleas found on dogs and cats bite both human beings and rats.

(20) Human fleas and fleas found on cats and dogs can live on rats as casual parasites, and therefore can under certain conditions play a part in the transmission of plague from rats to human beings, and vice versa.

In a report on plague in Hongkong submitted to the Secretary of State for the Colonies in December 1902, Professor Simpson 2 says (p. 8) “The experiments undertaken demonstrate that pigs, calves, buffaloes, sheep, hens, ducks, geese, turkeys and pigeons are, in addition to rats, susceptible to plague, and particularly so when fed with plague material.” He points out that these experiments have a twofold importance for the Health Officer, viz.

In a previous paper (This Journal, vol. VI. p. 524) we have given an account of the quantitative estimation of plague bacilli in the blood of 16 septicaemic cases. It remains now to supplement this number with a further series and to add some observations on the infectivity of the urine and faeces.

Fleas (Aphaniptera, Siphonaptera) are wingless insects which run or hop about. They are for the most part parasitic in their habits, being found only on the homoiothermic vertebrates, especially on those which build nests or have more or less fixed haunts or homes. Like the majority of other parasites they have the body flattened, but differ from most parasitic insects in that this flattening is in a lateral direction, so that the transverse diameter of the body is small and the vertical diameter great. Attention has already been drawn to certain other morphological peculiarities of these insects (vol. VII. p. 446).

Fleas that have fed on septicaemic blood are capable of conveying infection to fresh animals on which they feed. This capability is associated with the presence of plague bacilli in the stomach, intestine and faeces.

1. Both in Bombay city and in the Amritsar district of the Punjab there is a distinct seasonal variation in the number of rat fleas.

2. In these places this seasonal variation of rat fleas corresponds directly in a general way with the plague mortality. During the season when plague is epidemic the average number of fleas per rat is above the mean, while during the non-epidemic season it is below the mean. The height of the epidemic corresponds fairly closely with the season of maximum flea prevalence.

We have no desire to claim for the tests described above that they will be found applicable to every strain of organism belonging to any particular group. Our experiments were necesarily limited to a few races of each—many of these being probably avirulent from long subcultivation on artificial media. We merely wish to indicate the general methods we would adopt in any case of difficult bacteriological diagnosis, although as a matter of fact the only tests we found it necessary to use in the course of our work were the method of cutaneous and subcutaneous inoculation into animals and the stalactite test.

The tests referred to in this paper may be summarised thus:

1. The plague bacillus gives a fairly characteristic type of colony on neutral agar; it forms typical stalactites in neutral broth; it gives certain definite fermentation reactions.

2. B. pseudotuberculosis resembles B. pestis more closely than any other organism. The animal test on white rats is probably the best for its differentiation.

3. Bacilli of the haemorrhagic septicaemia group appear to be inhibited in their growth in media containing sodium taurocholate.

4. Bacilli of the B. enteritidis (Gaertner) group are the most readily distinguished of all; the appearance of agar cultures, the negative stalactite test, and their fairly definite fermentation sugar reactions are sufficient to mark them off from the plague bacillus.