Ck. Revell et al., LONG-TERM SOFTENING OF SURFACE AND BURIED HARD SEEDS OF YELLOW SERRADELLA GROWN IN A RANGE OF ENVIRONMENTS, Australian Journal of Agricultural Research, 49(4), 1998, pp. 673-685
A 3-year field experiment was conducted to investigate seed softening
in yellow serradella (Ornithopus compressus L.) in a low rainfall Medi
terranean environment at Merredin, Western Australia. The study examin
ed seeds of 4 accessions of serradella from separate growing sites (Pi
ndar, Merredin, and Badgingarra), and included the effect of pod buria
l (only for accessions grown at Merredin). Pods were placed on the soi
l surface in December 1992 and sampled in March, June, and October for
the next 3 years. Burial treatments (2 and 6 cm beneath the surface)
commenced in June 1993 after pods had been on the soil surface for one
summer. Samples were collected in June for the next 2 years. Softenin
g of seeds over the first summer in the field was compared with that o
btained in the laboratory with 16 weeks at a diurnally alternating tem
perature of 60/15 degrees C. Few seeds of any accession softened (gene
rally <6%) at the soil surface during the first summer, but the rate o
f softening increased over the next 2 years. The highest annual rate o
f softening was about 55% in the third year in accessions GEH72-1A and
GEH72-2A. Accession of serradella had more influence on pattern of se
ed softening than site at which seeds were produced. Burial of pods at
2 cm markedly accelerated seed softening in all strains, particularly
GEH72-1A and cv. Madeira, in which over 95% of hard seeds softened du
ring the first year of burial. Softening at 6 cm was similar to that a
t the soil surface. Shallow burial of pods, as would occur during cere
al cropping in one year, could improve regeneration of serradella, but
reduce the longevity of its seed bank.Laboratory treatment at 60/15 d
egrees C generally over-estimated field softening during the first sum
mer. The spread of germination in time in laboratory tests differed be
tween accessions and was much wider in GEH72-2A than in others, extend
ing up to 35 days. Such behaviour could provide insurance against tota
l seedling loss following false breaks of season.