Whole-week leaf-to-bloom peaks were uncertain in four US plant samples in 1965-2008

We tested whether recorded intervals from first leaf to first bloom had local peaks at whole weeks. We used 13,900 paired plant-year records from historical US lilac and honeysuckle programs. Four separate program and period samples compared gaps of 14 to 56 days at seven-day steps with their immediate neighbors. The estimated excess ranged from 0.15 to 4.60 percentage points. All four state-plus-year cluster intervals included zero after correction for four tests. Equal site weights, flag removal, and alternative neighboring days did not establish a repeated peak. One later eastern lilac subgroup had a positive interval, but this did not repeat across samples. The records do not establish weekly visits or show that observation bias is absent. Missing visit histories, selected long records, few clusters, and the assumed local shape of biological gap frequencies limit the test. We provide an inspectable diagnostic, paired records, source hashes, code, and reproduction instructions. We do not estimate climate or pollution effects.

phenology · observation bias · weekly sampling · lilac · honeysuckle · null result