On the other hand, on the same river Taff, and in the tributary valley of the Rhondda, there is an immense population of miners, among whom the enteric fever death-rate will probably be found to have been higher in 1881-90 than in any other registration district. The most populous part of the district is the town of Ystradyfodwg, which had 44,046 inhabitants in 1881 and 68,720 in 1891, an increase of over fifty per cent., the highest urban rate of increase in the country. On the mean of the last three years, 1891-93, its enteric fever death-rate has been ·62 per 1000. There are several populous towns or townships in the mining districts of the north-east which have in like manner kept their high rate of typhoid mortality--Auckland, Easington, Bellington (Morpeth) and Middlesborough. It is held by many that enteric fever has been most characteristically a product of the modern system of closet-pipes and sewers. It is, of course, the defects of the system that are, in this hypothesis, to blame, including its partial adoption, the transition-state from the older system, the tardy extension to new streets, as well as cheap and faulty construction. All those things, together with the inherent difficulty of connecting with a main sewerage the irregular squattings of a mining community, are probably to be found in highest degree in those districts of Durham and South Wales that are most subject to enteric fever. While enteric fever is in some places steady or endemic from year to year, in others its force is felt mostly in great and sudden explosions. One such happened in the city and district of Bangor in the summer of 1882. The registration district had only 95 deaths from enteric fever in the ten years 1871-80, but in the single year 1882 it had 87 deaths registered under that name. Of 548 attacks (with 42 deaths) which were known from 22 May to 12 September, 407 fell in August and the first twelve days of September[407]. In the following year and throughout the rest of the decennium the district had its usual low average of enteric-fever deaths. One thing relevant to the explosion was probably the excessive rainfall of June and July (9·5 inches, as compared with 4·8 inches about London).