Why large seeds with physical dormancy become nondormant earlier than small ones

Abstract

Under natural conditions, large seeds with physical dormancy (PY) may become water permeable earlier than small ones. However, the mechanism for this difference has not been elucidated. Thus, our aim was to evaluate the traits associated with PY in seeds of Senna multijuga (Fabaceae) and to propose a mechanism for earlier dormancy-break in large than in small seeds. Two seedlots were collected and each separated into large and small seeds. Seed dry mass, water content, thickness of palisade layer in the hilar and distal regions and the ratio between palisade layer thickness (P) in the lens fissure and seed mass (M) were evaluated. Further, the correlation between seed mass and seed dimensions was investigated. Large seeds had higher dry mass and water content than small seeds. The absolute thickness of the palisade layer in the different regions did not show any trend with seed size; however, large seeds had a lower P:M ratio than small seeds. Seed mass correlated positively with all seed dimensions, providing evidence for a substantially higher volume in large seeds. Since wet, but not dry, high temperatures break PY in sensitive seeds of S. multijuga, the data support our prediction that internal pressure potential in the seed and palisade layer thickness in the water gap (lens), which is related to seed mass (i.e. P:M ratio), act together to modulate the second step (dormancy break) of the two-stage sensitivity cycling model for PY break. In which case, large seeds are predetermined to become water-permeable earlier than small ones.

Fig 1. Large (L) and small (S) Senna multijuga seeds from collections S1 and S2.

Bars = 1 mm.

Fig 2. Sections of Senna multijuga seeds showing locations where palisade layer was measured.

(A) Cross-section in hilar region. (B) Detail of region indicated by rectangle in A. (C) Cross-section in distal region (lateral position of the median third). (D) Detail of the region indicated by the rectangle in C. M1–M5 = locations where measurements were made on each seed.

Fig 3

Mean (± s.e.) dry mass (A) and water content (B) of small and large S1 (seed collection 1) and S2 (seed collection 2) seeds. Different lowercase letters indicate significant differences between seed collections within a seed size and different uppercase letters significant differences between seed sizes within a collection, according to Fisher’s test (P≤0.05).

Fig 4. Thickness of palisade layer (mean ± s.e.) in different parts of S1 (seed collection 1) and S2 (seed collection 2) seeds.

Different lowercase letters indicate significant differences between seed sizes within a collection and different uppercase letters significant differences between seed collections within a seed size, according to Fisher’s test (P≤0.05). There was no interaction between seed size and seed collection in A and D.

Fig 5. Relationship between P:M ratio and seed size for S1 (seed collection 1) and S2 (seed collection 2) seeds (mean ± s.e.).

Different lowercase letters indicate significant differences between seed sizes within a collection and different uppercase letters significant differences between seed collections within a seed size, according to Fisher’s test (P≤0.05).

Fig 6

Relationship between seed mass and (A) seed length, (B) seed width and (C) seed thickness. (n = 100, all P < 0.0001).

Fig 7. Conceptual model of differences in water loss during onset of PY in relation to seed size.

Blue dotted arrows indicate path of water inside the seed towards the hilar region. Red scale represents the variation in water loss among the sizes of seeds. Blue scale represents the variation in water content among the sizes of seeds.