The First Government of Northern Ireland, Education Reform, and the Failure of Anti-populist Unionism, 1921–1925

This article examines education reform under the first government of Northern Ireland (1921–5). This embryonic period offered the Ulster Unionist leadership a chance to construct a more inclusive society, one that might diminish sectarian animosities, and thereby secure the fledgling state through cooperation rather than coercion. Such aspirations were severely tested by the ruling party’s need to secure the state against insurgency, and, more lastingly, to assuage the concerns of its historic constituency. The former led to a draconian security policy, the latter to a dependency on populist strategies and rhetoric. It is argued here, however, that this dependency was not absolute until July 1925. Before that, the Belfast government withstood growing pressure from populist agitators to reverse controversial aspects of its education reforms, only relenting when Protestant disaffection threatened the unity of the governing party and the existence of the state.

Defining Spatial and Temporal Hydromorphological Sampling Strategies for the Leigh Brook River Site

Detailed surveys of depth and velocity are undertaken to describe hydro-ecological status of rivers. Fieldwork for these surveys is time consuming and expensive. This paper aims to describe the methodology applied in order to determine the most suitable depth sampling strategy for effective field data collection and river representation in time and space at the Leigh Brook river site, Worcester, UK. The accuracy of three different sampling strategies for predicting depth at non-measured points has been compared and the mesohabitats that better characterise depth changes due to variations in discharge have been identified. The results show that depth changes due to discharge change are mainly located at shallow and deep glide mesohabitat types. The analysis for the comparison of sampling strategies indicates that grid sampling strategies give better results than regular transects. Since the results also show that higher errors in predictions are obtained in the deepest areas, higher sampling densities should be applied in these locations.