1. The alimentary tracts of 104 lice (Haematopinus spinulosus), which had been taken from rats infected with T. lewisi, were examined. In 51 lice no trace of T. lewisi could be found. In 53 lice T. lewisi, unchanged in every particular, was found in various parts of the gut. The trypanosome was seen to be quite unchanged in stained preparations.

2. The alimentary tracts of 45 fleas (Ctenophthalmus agyrtes), which had fed on infected rats, were examined. The T. lewisi could not be found at all in 43 fleas. They disappeared very rapidly from the ingested blood and could not be found in stained preparations. In 2 fleas trypanosomes were found, but they did not differ in form from those seen in the blood of the rat.

3. 263 lice were examined from rats which were apparently uninfected with T. lewisi. No form of T. lewisi was discovered in fresh or stained preparations.

4. 31 fleas from uninfected rats were dissected and their alimentary tracts examined. No trace of T. lewisi was found.

5. 15 lice were examined 18–48 hours after their removal from an infected rat. T. lewisi was found in 7 of these, but they shewed no developmental changes.

6. Observations made upon 36 lice from infected rats shewed that at the earlier stages of digestion T. lewisi was more often found than at the later stages. In some lice however trypanosomes could not be found at the early stages of digestion, although they were present at an advanced stage of digestion.

7. The examination of other organs than the alimentary canal in fleas and lice did not shew any form of T. lewisi.

8. Crithidia ctenophthalmi was found in 9 Ctenophihalmus agyrtes.

9. About 370 lice and 75 fleas were examined but no evidence was obtained of developmental changes of T. lewisi in either fleas or lice.

The morphology of the spirochaetes has proved in recent years a fruitful source of controversy. One has only to look through some of the vast literature on the subject to be convinced that the time has not yet come for drawing final conclusions, so varied and so contradictory are the observations by different well-accredited authors.

A large number of the trypanosomes and allied flagellates are now known to be the causal agents of some of the most fatal diseases of man and of the lower animals. The study of these parasites during the past fifteen years has truly made phenomenal strides, and at the present time it holds a position of commanding interest. Novy (1907) in a recent paper aptly states the reason for this active research when he says, “It is not merely because six or seven of these diseases have been recognised in domestic animals in Asia, Africa and South America. It is not wholly due to the discovery of a human form of the disease, but rather to the fact that the mastery by man of whole regions is called into question. The successful contest with malaria and yellow fever has now given place to the fight with trypanosomiasis and upon the issue of this struggle depends to a large extent the destiny of Equatorial Africa.”

In a recent paper Dr Sambon (1909) has made a violent attack on me for my gentle reminder to him that he has overstepped the boundary of established truths regarding the haemogregarines, and has let his imagination run wild. As he still persists in maintaining his position I feel I am bound to expose his methods, and in the present paper I propose examining in some detail his recent observations on the haemogregarines, and at the same time dealing with his reply to my paper.

It has been stated that in the stomach of the bed-bug the spirochaetes of relapsing-fever may frequently be observed to lose their normal regular curves and become more or less “worm-like.” This is of interest to those who oppose the view that spirochaetes are rigid spirals, a view which, after gradual abandonment by most observers, has lately been again advanced by Novy and Knapp in their study on S. recurrentis.

There seems little reason to doubt that most blood-sucking invertebrates harbour in their digestive tract natural flagellate parasites, which have erroneously been described by various authors as part of the life-cycles of blood-parasites from vertebrates. Among arthropods such natural flagellates have now been observed in ticks, lice, bugs, mosquitoes, biting and non-biting flies, and fleas; and the list of such hosts steadily grows.

1. We have discovered that Trypanblau and Trypanrot are highly efficient remedies in the treatment of canine piroplasmosis, no drug or mode of treatment having hitherto been found to exert any appreciable effect upon this very fatal disease.

2. The drugs exert a direct and observable effect upon the parasites (a) by causing the pyriform parasites to quickly disappear, and (b) in most cases, by causing the total disappearance of the parasites from microscopic observation in the peripheral blood.

3. The disappearance of the parasites from the blood is usually temporary, the parasites reappearing in small numbers after an interval of 9 to 12 days, but the dogs, as a rule, show no symptoms and gradually progress toward recovery. In our experience there is but slight loss of weight in treated animals, this being in marked contrast to what is usually observed in dogs which recover naturally. In the two dogs which died of a relapse the parasites reappeared after an interval of 4 to 5 days.

4. The experiments were conducted upon 21 dogs of all ages and of a variety of breeds. Many of the dogs were highly susceptible puppies. Twelve dogs were subjected to curative treatment, one dog received preventive treatment, and eight dogs served as controls.

5. Twelve dogs received treatment as follows:

(a) Two dogs were treated with Trypanrot and survived 111 days and 20 days respectively. In the latter case the dog was treated in an advanced stage of the disease, the parasites disappeared, and death appeared to be due to the after-effects of the malady.

Two control dogs died respectively on the 9th and 7th day after they were inoculated.

(b) Ten dogs were treated with Trypanblau and seven made a good recovery. Three of these dogs are still alive, having been respectively under observation for periods of 90, 83 and 65 days. Four of the recovered dogs died respectively of distemper on the 42nd and 43rd day, of severe mange on the 52nd day and of distemper and mange on the 69th day. We failed to cure the disease in 3 out of 10 cases treated: two small wormy ill-nourished puppies died from a relapse (on the 13th and 15th day) after they appeared to be on the road to recovery; one dog, treated only 3 hours prior to death, and when 50% of its corpuscles were infected, succumbed to the disease, but even in this case a visible effect was exerted upon the parasites by the drug.

Of the 7 control dogs, 6 died of piroplasmosis within 7–13 days after inoculation with virulent blood, and the remaining dog died of piroplasmosis on the 36th day after inoculation.

6. After the parasites disappear under the influence of the drug, the temperature usually falls to normal, but slight rises of temperature may at times occur, together with the subsequent reappearance of the parasites.

7. When the parasites reappear after treatment they occur in small numbers and can usually be detected only at the edge of the film. In such cases the parasites tend to occur in higher multiples [(PPPP) or more] within the individual corpuscles. The same appearance has been observed in one of the control dogs, which died of chronic piroplasmosis on the 36th day.

8. In the two puppies 10 and 11, in which death followed upon a relapse, the parasites reappeared after a shorter interval (4 to 5 days) than in any of the other dogs. The blood counts made during the fatal relapse are remarkable as showing a high percentage of dividing forms (D); that rapid invasion of fresh corpuscles was proceeding is also shown by the high percentage of single pyriform parasites (P) which occurred in the corpuscles.

9. One experiment is reported (Dog 13) in which Trypanblau was injected 24 hours after the animal was inoculated, with the result that no parasites appeared in the dog's blood up to the 65th day, and the dog remained perfectly well. The control dog died of piroplasmosis on the 7th day. (This experiment has since been successfully repeated.)

Further experiments of this character will be shortly reported upon.

10. Arsacetin and Soamin exert no curative effect upon canine piroplasmosis.

It would appear from the data presented that the favourite location for the common rat fleas of this vicinity, Ceratophyllus fasciatus, Bosc., and Loemopsylla cheopis, Roth., is about the hind quarters of the rat, while Ctenopsyllus musculi (Dugès), Wagner, prefers the region of the head and neck.

The same regional distribution of rat fleas was found in the case of the guinea-pigs.

Squirrel fleas are most numerous on the hind quarters.

These observations, while they are not to be insisted upon as throwing any special light upon the regional distribution of buboes in naturally infected plague rats, are of particular interest in showing the very constant preference of Ct. musculi for the region of the head and neck.

In regard to the other species of fleas it should be borne in mind that the skin areas are not of equal dimensions; that of the head and neck being smallest; that of the hind quarters being the largest; and the skin area of the fore quarters being between these in size.

The above-mentioned considerations bearing upon the influence of insect prevalence and the influence of meteorological conditions on the prevalence of trypanosome infections in rats may be summarized as follows:

(1) There are definite seasonal variations in the prevalence of trypanosome infections in M. rattus and in M. decumanus.

(2) No definite correlation is forthcoming between the seasonal prevalence of T. lewisi and the seasonal prevalence of rat fleas. The seasonal prevalence of the transmitting insects is apparently a factor of subsidiary importance in the causation of the trypanosome prevalence.

(3) The evidence adduced above indicates that the dominant factor determining the seasonal prevalence of trypanosome infections is the atmospheric temperature, the optimum temperature approximating to 79° F. It is probable that temperature operates by influencing a developmental cycle of the trypanosomes in the transmitting insects.

(4) It is possible that in addition to temperature the atmospheric humidity may play a part by influencing the direct or mechanical transference of the infection.