Weekend departures were uncertain in four US Arnold Red honeysuckle bloom samples in 1968-2008

We tested whether weekend departures in recorded first-bloom dates repeated in Arnold Red honeysuckle across two US networks and two periods. Four samples used 3,190 records from 1968 to 2008. Weekend departures from the 2/7 reference were -1.59, -2.74, -0.02 and -5.19 percentage points. All four state and year cluster intervals included zero after correction for four tests. Annual calendar-phase checks and planned weighting and flag checks also gave uncertain results. One state omission gave a clear late western deficit, but it was one of many sensitivity tests. The eastern samples were small. Unknown visits, selected long records, shared monitoring systems and calendar assumptions limit inference. This result does not establish unbiased phenology or a climate or pollution effect. The package preserves code, 4,804 eligible records, source hashes, all estimates, a figure and independent checks.

phenology · honeysuckle · observation bias · weekend · replication · null result