M. Koidzumi 
241 
there occur only two species, viz. Leucotermes (Reticulitermes) speratus and 
Coptolermes formosanus. Though the termites of Formosa are so rich in 
species, the great majority of these species are not of common occurrence, 
but are rare or locally distributed, and only a few species are found commonly 
throughout the island. The species most commonly found in Formosa are 
Coptotermes formosanus and Odontotermes formosanus. A species of Leuco¬ 
termes is also more or less frequently found in Formosa, and referred to 
different species by different authors. M. Oshima distinguished it from 
Leucotermes speratus , saying that although it is scarcely possible to draw a 
sharp line of distinction between the imagines, yet the soldiers display some 
distinct differences which justify their separation. S. Yano, S. Hozawa, and 
Holmgren, on the other hand, persist in the belief that they should be united 
into one. Not being an entomologist, I myself will not express an opinion 
on this species question. As regards their protozoa, however, I find that 
distinct differences are recognizable between those of Leucotermes speratus of 
Japan proper and those of the disputed species of Formosa. Thus I feel it 
convenient for my description to let the Formosan species have its own name, 
and I prefer to call it by the name given by M. Oshima, Leucotermes (Reticuli¬ 
termes) jlaviceps. 
In carrying out my studies both in Tokyo and Formosa, I was fortunate 
in obtaining an ample supply of living specimens of four species of the host 
insects. While at Tokyo, my studies were done on the forms harboured by 
Leucotermes speratus, collected at several localities in Japan proper. On going 
afterwards to Formosa, three different species, Coptotermes formosanus, Leuco¬ 
termes fiaviceps, and Odontotermes formosanus, came into my hands. Moreover, 
I have had an opportunity of examining another species of rather rare 
occurrence, Capritermes nitobei. In Odontotermes formosanus and Capritermes 
nitobei, however, none of the protozoa under discussion were found. While 
working at Tokyo, I was unable to obtain any specimens of Coptotermes 
formosanus ; but shortly before finishing this study I made a trip to Kyushu, 
for the purpose of studying Coptotermes formosanus and Leucotermes speratus 
in that southern district of Japan proper. 
In all the three species of termites studied by me (Leucotermes speratus, 
L. fiaviceps, and Coptotermes formosanus), the protozoa are found in equally 
large numbers, and almost the entire cavity of the intestine of every individual 
is found filled with them, excepting the oesophagus and the proventriculus. 
The Malpighian tubules also were found frequently invaded by some of the 
forms with slender bodies. Young termites, lately hatched, have their in¬ 
testines free from protozoa; but as they begin to take food, the protozoa 
seem quickly to invade them, for individuals measuring only three milli¬ 
meters in length are usually found rather heavily infected. 
In the workers the intestine is more developed than in the other casts, 
and is found almost filled with protozoa, food substances being rather small 
in volume as compared with the contained organisms. In the soldier and 
Parasitology xiii 1G 
