Only one of four English samples had a spring warmth interval above zero for relative Gatekeeper timing in 1985-2024

We tested whether March-May warmth changed the mean flight date of Gatekeeper relative to Meadow Brown and Ringlet at the same English sites. Four models used 9,690 site-years in northern and southern England in 1985-2004 and 2005-2024. Models controlled May-June warmth, site effects and time. All four slopes were positive. Only later North had a four-test interval above zero: +2.45 days per C, interval +0.39 to +4.51. Later South gave +2.13, interval -0.003 to +4.26. Both early intervals included zero. Removing later warmth control gave positive corrected intervals in both later samples. Common-site and year-intercept checks limit a stable response claim. Missing visits, count selection, shared weather and short series prevent causal interpretation. The study gives a signed test of a prior unsigned spread result on overlapping records. Original code, aggregate results, source hashes, a figure and independent numerical checks are included. External databases and record-level extracts are not bundled.

butterfly · phenology · temperature · relative timing · observation bias · regional replication