Skip to main content
. 2016 Jul 15;67(4):421–430. doi: 10.1111/ejss.12359

Figure 2.

EJSS-12359-FIG-0002-b

(a) Soil organic carbon (C), (b) nitrogen (N) and (c) organic C:N ratio under long‐term grass (black circles, solid line), arable (1949–; grey triangles, long‐dashed line) and fallow (1959–; white squares, short‐dashed line) treatments in 2008. The data points show the treatment mean and standard error of the mean (n = 3) and the lines derive from residual maximum likelihood (REML) analysis with spline or linear models where appropriate (models (4), (5), (6)). The separate bar shows the average standard error of a difference of means (SED) for the model (see Tables 2 and 3 for statistical analysis).