[bowman] fundamental PROCESSES IN HISTORICAL SCIENCE 527 



station at a subsequent hour, the train will have spent substantially 

 that interval of time in actually covering the distance between these 

 two points. The punctuality or unpunctuality of the railroad in 

 question and the season of the year are not in evidence, hence the 

 probability is fixed here by the average punctuality of all railroads at 

 all seasons. On some railroads the trains are nearly always substan- 

 tially on time, while on others there is some unpunctuality, especially 

 at certain seasons; but taking all stations, trains, roads and seasons 

 together, it is a conservative estimate that there are five arrivals made 

 at stations by trains substantially on time to one such arrival made 

 substantially late. In the 6th conclusion the probability of 10/11 

 or 10:1 depends on the chances that a train already half an hour late 

 would lose additional time through being held involuntarily at stations, 

 especially at a border and custom's point where such additional de- 

 lays are not infrequent, as opposed to the chances that it would incur 

 further delay by reducing its speed, or otherwise, between stations. 



Case U. 



Narration . 



(The narrative of necessary conclusions in this case consists again 

 simply of the successive parts of evidence read uninterruptedly, with- 

 out regard to the intervening probable conclusions.) 



(Part 1): "On the whole one finds throughout the world an 

 approximate equilibrium of the two sexes." 



(No. 1): "From this fact it is a strong probability that there is a similar 

 approximate equilibrium in ever>' country." {High). 



(Part 2) : "In some countries, however, there is a pretty constant, 

 though in itself moderate, relative excess of female persons; while 

 in others on the contrary there is a similar excess of males. Thus 

 it has been found that, for every 1,000 males, the number of females 

 in Norway was 1,075, in Sweden 1,065, in Great Britain and Ireland 

 1,060, in Switzerland 1,056, in Denmark 1,051, in Austria 1,044, in 

 the German Empire and in Spain 1,040, in Holland 1,024, in Hungary 

 1,015, in France 1,007, and in Belgium 1,005. A deficiency of females, 

 on the contrary, was found, among other countries, in Italy with 

 995 females per 1,000 of male population, in Japan with 980, in Bul- 

 garia with 965, in Roumania and Canada with 964, in the United 

 States with 953, and in Serbia with 948." 



(No. 2): "Probably the number of male and female births varies in the 

 respective countries, the female births preponderating where the female popu- 

 lation preponderates, and the male births preponderating where the preponder- 

 ance in population is with the males." {High). 



