Using a historic collection of Australian lentil varieties released between 1988 and 2019, we show that genetic gain in yield (29 kg ha^-1 yr^-1) associates with faster reproductive development. A focus on five loci related to reproductive development, the lentil ELF3 ortholog Sn, two loci linked to clusters of lentil FT orthologs and two loci without proposed candidates in chromosomes 2 and 5, revealed the potential for introducing novel early flowering alleles into current breeding material for further phenological diversity with potential implications for yield.

We phenotyped 26 Australian lentil lines representing 32 years of breeding and selection for yield in 12 environments resulting from the combination of two seasons, three locations in South Australia, Western Australia, and Queensland, and two sowing dates. Rainfall is winter dominant in South Australia (SA) and Western Australia (WA) and summer dominant in Queensland (Qld); locations were Kapunda, SA (-34.40, 138.85) and Merredin, WA (-31.49, 118.27) in both seasons, and Gatton, Qld (-27.54, 152.33) in 2021 and Forest Hill, Qld in 2022 (-27.57, 152.36). Sowing dates were 5^th May and 29^th June 2021 at Kapunda; 19^th of May and 25^th of June 2021 at Merredin; 24^th May and 22^nd June 2021 at Gatton; 10^th May and 15^th July 2022 at Kapunda, 19^th of May and 23^rd of June 2022 at Merredin and 16^th June and 20^th July 2022 at Forest Hill. Treatments were arranged in a randomised complete block design with three replicates.

Phenology was monitored 2–3 times per week in the centre rows of each plot to visually determine the timing of 50% of plants reaching: emergence, flowering, podding, end of flowering and maturity. Chronological time was converted to thermal time (^oCd) based on cumulative daily average temperature minus a base temperature of  0 °C.