XIV. Nr. 41. 



\aturwissenschaftlichc Wochensehrift. 



519 



made out for ascertaining' more accurately than we do 

 at the present the population of countries where it is not 

 possible to take a regular census. The subject is of 

 interrest to geographers as well as to statisticians. The, 

 subject is dealt with under the following heads: - 



1. What we ought to aim at, 



2. By what methods we may hope to attain our end, 

 I!. By whom the work is to lie done, and 



1. How In provide for the expenses and the admini- 

 stration. 



I. A. The aim should be to ascertain, as accurately 

 as possible, the population of all the countries, regions 

 and cities of the world, whose population is not at pre- 

 sent known. It is not necessary to point out how deficient 

 our present knowledge is in this respect. It is only 

 necessary to state that the estimate of the total number 

 of the inhabitants of the globe can only be given within 

 10U or perhaps even 200 millions. Still more vague of 

 course, are our notions concerning the population of cer- 

 tain divisions covering large regions of the earth, parti- 

 culary in Asia and Africa. If we were to draw a map 

 showing, by means of different shades of colour, the coun- 

 tries and regions where the population is more or less 

 known, such a map would show a vast number of regions 

 completely blank, and others where the shade of colour, 

 corresponding to our present degree of knowledge, would 

 be very, very light. 



B. Another no less important object, at which we 

 ought to aim in this respect, in the question of the com- 

 position of the population of the different nations and 

 tribes, according- to sex, age, place of birth, occupation, 

 language, religion, in short, the general demographieal 

 features, which in civilised countries are investigated by 

 the aid of censuses. 



II. A. As to the method by which the number of 

 the inhabitants should be ascertained, the best would of 

 course be a regular census such a has been suceesfully 

 worked in British India and in the vast Russian Empire. 

 It seems only reasonable to expect that those nations 

 which have colonies or other foreign possessions will be 

 interested in taking censuses as far as possible in their 

 respective possessions: The examples quoted above will 

 afford very useful hints as to the manner in which such 

 censuses may be worked. But it is evident, that for a 

 long time regular censuses will be impossible in large 

 regions, even those which are under the direct admini- 

 stration of european powers, not to speak of those, which 

 are considered as only under the "influence" of the respec- 

 tive powers. 



The next best method would probably be to count 

 the number of houses, huts and other dwellings hi the 

 entire region and to combine this investigation with 

 partial representative investigations concerning the ave- 

 rage number of inhabitants per house, distinguishing 

 between rural districts of varied geographical character 

 and towns. In regions where such a method might be 

 applied, it would appear possible to calculate the popu- 

 lation tolerably well. 



But in many and vast regions even this method will, 

 for the present, be impracticable and an examination 

 ought therefore to be made as to which method or 

 methods ought otherwise to be applied in order to make 

 a provisional estimate of the population. 



It seems that an estimate of the population, where 

 a complete numbering would be inpracticable, could be 

 approximatively made by counting in representative re- 

 gions of varied geographical character the number of 

 houses, huts and other dwellings, combined with investi- 

 gations as to the mean population of each dwelling. Such 



investigations should be made separately and with a 

 distinction between towns and country districts, for each 

 of the following divisions: coasts islands, valleys, plains, 

 marshes, swampy regions, plateaus, prairies, stepps, de- 

 serts, bushlund, woodland, forest, moorland and moun- 

 tainous districts. 



The area of each of these divisions in square kilo- 

 metres, and its mean population per square kilometre 

 should be as far as possible, ascertained and, in case of 

 the impossibility of accurately investigating the whole 

 region, partial representative investigations should lie made 

 as a provisional measure. 



If detailed investigations such as these were arranged 

 as far as possible in broad bands stretched over the whole 

 unknown country or continent, then we should obtain two 

 important advantages, viz. 1 st that the regions covered 

 by the bands would be tolerably well examined (and in 

 the course of time such bands might be extended side by 

 side, until they at last covered the entire surface); 2 1 " 1 

 that it would perhaps be possible, by aid of the detailed 

 results of the careful examination of the aforesaid regions 

 covered by the bands (showing the average population 

 per square kilometre under various geographical conditions), 

 to draw conclusions as to the probable density of popu- 

 lation in corresponding unknown regions situated outside 

 the bands, but surrounded by bauds of known districts. 



While admitting the importance of applying a uni- 

 form method (either such as described above or another 

 better one) everywhere, the importance must also be re- 

 cognised of providing supplementary information by the 

 aid of other methods which might used with avantage, 

 according to the different circumstances in each indivi- 

 dual case. 



B. The demographical details mentioned unter I. B. 

 might be obtained by filling up, in representative towns 

 and country regions, individual schedules of about the 

 same form and contents as the ordinary census schedules 

 for a certain number of households or families, selected 

 so that together they might be considered as fairly re- 

 presenting the population of a certain region. The more 

 numerous the schedules are, the greater will be their 

 value. 



But it will be of still greater importance in this 

 respect to take precautions for recuring their representative 

 character. 



III. As to the persons, by whom the investigations 

 should be executed, the question is not the same for the 

 processions of civilised powers - - including the sphere 

 of their influence -- as for the countries outside their 

 influence. 



In the former of these, there will be an administration 

 which can superintend the operation. The experiment 

 made in British India and in the asiatic possessions of 

 Russia shows that enumerators can be got. With regard 

 to the regions which arc not under regular control, but 

 belong to the sphere of influence of any power, the 

 government of the latter will probably be able to lind 

 enumerators or agents. And, as has been the case in 

 India, natives will, to a great extent, be found useful for 

 the purpose. 



Even those who cannot read or write, will be able 

 to assist by counting the houses and the number of in- 

 habitants in them, and might also in other ways assist 

 in the work. 



In independent native countries, if sufficient pecuniary 

 means are forthcoming, agents might be found among 

 missionaries, exploring or commercial travellers or others, 

 and native assistants might be used in many eases. 



IV. The expenses incurred by the operations here 



