40 L. Rogers — Special Report on Fever in Dinojpnr Dist. [Supplfc. 



relationship of different parts of a village to sach a source of mosquitos 

 as a small stream may differ widely. Further, as already pointed out, 

 the conditions in the cold dry weather give but a faint and totally in- 

 adequate idea of the surroundings in the rainy fever season, as will be 

 further illustrated in the section on mosquitos. Nevertheless in a few 

 instances there seemed to be a definite relationship between certain con- 

 ditions and the fever death-rate which are worth mentioning. Thus 

 the Dinajpur circle comprised an area between two streams, one of 

 which was a clean sandy river bed with a good flow, while the other was 

 a stagnant weed covered and very sluggish one. Two villages on the 

 banks of the latter had a fever death-rate of 81 '8 per thousand, the 

 highest rate met with during the whole enquiry. The other villages did 

 not show such marked variations, but the death-rate of the whole area 

 was very high, as was the water-level and the malarial rate. In short, 

 it was a typical example of a waterlogged area in a " dying river " 

 district. 



Next we may take the Birganj circle, which furnished an instruc- 

 tive example. As already mentioned, the villages in this circle were 

 chosen partly on a good river and partly several miles from it in an area 

 quite dried up at the time of my visit. Nevertheless, the latter dry 

 area showed a death-rate of 54*3 per thousand against one of 41*1 in a 

 group of villages near the river and of 34-7 in a large village close to the 

 river, but on higher ground with a water-level 17 feet down. The 

 explanation of this difference was revealed when it was found that the 

 dry area had the high ground water-level of 9 feet. It was therefore less 

 well drained than the higher areas near the stream, and the floods of 

 the rainy season would be longer in subsiding at the end of the year, 

 and would thus form favourable mosquito-breeding grounds for a longer 

 time. In the Tliakurgaon circle there were four large villages of 

 scattered hamlets all under very similar conditions and with about the 

 same ground water-level (12 feet), and here the death-rates varied very 

 little, all being between 40 and 50 per thousand. In Ranisankail itself 

 the ground water was 12| feet down and the spleen-rate was 62' 7 per 

 cent., while the fever death-rate was 45'5 per thousand. In the villages 

 to the north and south the spleen-rate was 98 per cent, and the ground 

 water-level only from 5 to 11 feet down, while the fever death-rate was 

 74'3 per thousand, a very high figure. Here once more we find a rela- 

 tionship between a high ground water-level and a high spleen and fever 

 rate. It was here that numerous cases of chronic fever with very large 

 spleens were met with, the nature of which will be discussed in the 

 second part of this report. They are precisely those cases which have 

 always been known as " Malarial Cachexia " and which are responsible 



