568 ANNUAL. REPORT SMITHSONIAN INSTITUTION, 19 3 



such very diverse conditions of season, climate, race, occupation, soil, 

 mode of life, food, dwelling, etc., that they can not properly be com- 

 pared with one another. The prevalence of some special disease in 

 exceptionally virulent or widespread development will raise the death 

 rate of any one year far beyond its normal figure. Again, the 

 presence of some plant disease or pest which causes loss of crops, and 

 the resulting lowered vitality of the people in consequence of insuffi- 

 cient food, may easily swell the death rate. Nowhere can these 

 tropical death rates properly be compared with those noted under 

 different conditions in other latitudes. Most of the reliable statistics 

 of mortality relating to white men in the Tropics are for soldiers, that 

 is, men, in the prime of life, subjected to rigid discipline; provided, 

 when possible, with carefully selected food, proper living quarters, 

 and adequate medical and sanitary supervision. Obviously, death 

 rates for white soldiers in the Tropics are not directly comi^arable 

 with the death rates in our own country, which include all ages, both 

 sexes, and very diverse living conditions, as well as deaths by accident 

 and old age. So various and so complex are the controlling factors 

 that critical comparative study is not worth while. Some years ago 

 I was asked to present a paper on tropical death rates before a scien- 

 tific body, but I declined the invitation on the ground that such a 

 discussion was really not a profitable one. 



Tropical death rates are certainly high, but this fact should not be 

 attributed solely to the dangers of the climate. Bad sanitary con- 

 ditions and lack of medical attendance account for many, if not 

 most, of the high mortality rates among the natives; and an irra- 

 tional mode of life explains many deaths among persons coming from 

 cooler climates. Tropical death rates have been, and are being, re- 

 duced with remarkable rapidity in all countries which are wholly 

 or partly under white control, and especially among white troops in 

 the Tropics. This is the result of experience with tropical condi- 

 tions, and of the increased precautions that are now taken in select- 

 ing and caring for the men. 



The death rates among white people in the Tropics range from 

 the appalling figure of 483 per thousand among European troops on 

 the Gold Coast of Africa in 1829-1836 — almost every other man 

 died — down through steadily lower figures until we come to a 

 rate as low as 5 per thousand for the British troops in India in recent 

 years. A few illustrations of the change may be given. A death 

 rate of 115 in French Cochin China in the 1860's had been reduced 

 to about 20 per thousand in the later 1870's and early 1880's. Among 

 Dutch white soldiers in Java the rate of 170 in the second and third 

 decades of the nineteenth century had been lowered to less than 20 

 in the last decade. Very striking, also, has been a phenomenal re- 

 duction in the death rate from dysentery among the Dutch troops 



