Discussion

With the exception of a myxosporidian (Henneguya vitiensis Laird, 1950), which was recorded from 12 of the 40 examples of Leiognathus fasciatus (Lacépède) examined, Haemogregarina salariasi n.sp. was the only protozoan found in the blood of marine fish during the present survey. The former parasite not being haematozoan (sensu stricto), it will not be considered further here. Thus, from the fact that only one (0.5 per cent.) of the 205 examples of 24 species studied was parasitized, the haematozoan fauna of Fijian marine fish appears to be a singularly poor one. This is emphasized by the results (as yet unpublished) of a recent survey of the haematozoa of New Zealand marine fish, in the course of which blood smears from 365 fish belonging to 45 species were studied. Fifty-six (15.5 per cent.) of these fish of 10 (22 per cent.) of the species concerned were infected with trypanosomes, and 35 (9.5 per cent.) of 8 (18 per cent.) of them were infected with haemogregarines. The overall total of the 365 fish parasitized by haematozoa was 81 (22 per cent.), 12 (26.5 per cent.) of the 45 species dealt with acting as hosts. However, the great majority of the New Zealand marine fish studied were of oceanic occurrence, and it was from species in this category that most of the blood parasites were described; while most of the Fijian fish from which smears were taken were reef- or shore-dwelling forms. Thus, although none of the few oceanic fish examined in Fiji were infected with haematozoa, perhaps a more comprehensive study might reveal the presence of these parasites.

page 13

Most marine fish from which haematozoa have been described dwell outside the tropics. There thus appears to be a possibility that the haematozoan fauna of marine fish is richer in temperate than in tropical waters. With the qualification mentioned above, the evidence from my surveys in New Zealand and Fiji supports this hypothesis. Nevertheless, in view of our present very patchy knowledge of this interesting group of parasites, and our almost complete ignorance of their vectors, it is too early to generalize. A map purporting to show the geographical distribution of marine fish haematozoa, constructed from the information at present available, would more truly depict the distribution of parasitologists interested in this group rather than that of the group itself.

Bufo marinus Linnaeus, a native of Central and South America, was the only amphibian studied. All the specimens handled, 59 examples of all ages, proved negative for haematozoa. In French Guiana this toad is parasitized by Haemogregarina cayennensis Leger (1918), H. darlingi Leger (1918a), and Lankesterclla minima (Chaussat) (Leger, 1918a). In infections with any of these haematozoans, the natural parasite level is usually very low. Being aware of this, I examined the whole area of each of my B. marinus smears, using a x5 ocular and x97 oil-immersion objective. Had parasites been present, even in very scanty numbers, they would certainly have been detected. B. marinus has been introduced into many of the islands of the Pacific because of its value as an agent in the control of insects of economic importance. The stock originally imported into Fiji came not from America, but from Hawaii. It would be of considerable interest to learn whether or not any of the blood parasites which have been recorded from this toad in America occur today in Hawaii; for somewhere along the line of introductions the Fijian stock of B. marinus would appear to have become freed from haematozoans, possibly as a consequence of the absence of suitable invertebrate hosts from the new habitats.

No parasites were found in the blood of any of the 28 reptiles belonging to 7 species which were examined. Although this is rather surprising in view of the general prevalence of saurian haematozoa, particularly haemogregarines, in the tropics, too few reptiles were dealt with for any significant conclusions to be drawn.

Although the only example of the fruit bat Pteropus nawaiensis (Gray) studied was heavily infected with Plasmodium pteropi Breinl, 135 examples of the long-tailed fruit bat Notopteris macdonaldii Gray proved uniformly negative for this or any other blood parasite. Plasmodium pteropi characteristically has a high natural infection rate—at least, among adult fruit bats. Thus Bearup and Lawrence (1946) found a 100 per cent. infection rate for 25 adults of Pteropus gouldi, a 91 per cent. rate for 11 adults of P. scapulatus, and an 83 per cent. rate for 18 adults of P. conspicillatus in Queensland, Australia; while McGhee (1949) found a 100 per cent. infection rate for 92 examples of P. geddiei and P. eotinus at Espiritu Santo, New Hebrides. Twenty of the examples of Notopteris macdonaldii from which smears were examined were juveniles, while the remaining 115 were adults. Some at least of such a number might be expected to be parasitized were this species indeed a host for haematozoa. In contrast with his findings for fruit bats, McGhee (1949) failed to find any haematozoa in 100 examples of Hipposideros cervinus (Rhinolophidae). Whether this apparent species immunity in bats dwelling in close association with species heavily infected with Plasmodium has a biological or physiological explanation, poses an interesting problem for future investigation.

page 14

In the case of Pteropus nawaiensis and Notopteris macdonaldii, a possible explanation might lie in the predominantly forest-roosting habit of the former species as opposed to the cave-dwelling habit of N. macdonaldii. Should the vector of Plasmodium pteropi prove to be a day-biting forest mosquito, it would obviously not have access to Notopteris macdonaldii in the normal course of events.