Response to Four Cycles of Selection for Low Initial Rate of Digestion in Alfalfa.
B. E. Coulman, C. Duncan and B.P. Goplen
AAFC, Saskatoon Research Centre, Saskatoon, SK. S7N 0X2
Between 1979 and 1971 four cycles of recurrent selection for low initial rate of digestion (LIRD) were carried out
in a population of alfalfa, composed of the cultivars Beaver, Kane, Vernal and Anchor. Digestions involved a 4
hr nylon bag digestion of fresh chopped forage sampled at a vegetative stage in the rumens of fistulated steers.
Between 2 and 5 % of the least digested plants were selected in each generation. Cycles 1 and 2 took two years to
complete; cycles 3 and 4 took four years each as progeny testing was done. In grazing trials, the cycle-4
population (LIRD-4) had an initial rate of selection approximately 85% of that of unselected Beaver alfalfa and a
mean reduced bloat incidence of 62% (1, 2). The goal of the present study was to compare the four parental
cultivars (cycle-0) and the four selection cycle populations in a single experiment to determine the response to
selection for LIRD and associated changes in forage quality and plant development. Populations were established
in 10 replicate nurseries in 1994, with each plot consisting of a 5 m row of 15 plants. Plants were sampled in
1995 and 1996 at a vegetative anda mid-bloom stage. Characteristics measured were intitial rate of digestion (4 h
nylon bag digestion), 24 and 72 hour in vitro digestions, acid detergent fibre (ADF), neutral detergent fibre
(NDF), crude protein (CP), lignin, plant height, stem thickness, leaf percentage and maturity (mean stage by
count). Not all characters were measured at both growth stages. Analyses were done over years as there were no
year*cycle interactions for any character.
For IRD, there was a sequential decline over cycles for the vegetative sampling, with cycle-4 having an IRD 89%
of that of cycle-0. No such sequential decline was found at the mid-bloom stage. Differences in IRD were not
significant (p=0.05) due to the high variability of this biological measurement. For other characters, significant
differences were found for ADF and NDF for both the vegetative and mid-bloom samples and for leaf percentage
and plant height, which were only recorded on mid-bloom stage samples. For NDF and ADF in vegetative stage
samples, there were linear increases over cycles, with cycle-4 being 11 gm kg-1 and 8 gm kg -1 higher than cycle
0. This amounted to a 6% increase in both NDF and ADF. For mid-bloom stage samples, increases in ADF and
NDF were more curvilinear. The lack of differences in lignin concentration among cycles suggests that increases
in fibre concentration have been due to increased cellulose and hemicellulose. This corresponds well to the
reported (2) increases in leaf epidermis and mesophyll wall thickness in the cycles of selection for low IRD, and
these cells are not highly lignified. No significant differences, nor obvious trends, were found in 24 and 72 h
digestions and crude protein concentrations. There was a trend towards increased plant height for all selected
cycles and all selected populations had a lower leaf percentage than the source population. There were no
differences in stem width or maturity.
References
Coulman, B., W. Majak, T. McAllister, B. Berg, D. McCartney, K.-J. Cheng, J. Hall and B. Goplen. 1996.
Reduced bloat incidence in grazing trials of alfalfa selected for low initial rate of digestion (LIRD). Report of the
35th North American Alfalfa Improvement Conference, Oklahoma City.
Goplen, B.P., R.E. Howarth, and G.L. Lees. 1993. Selection of alfalfa for a lower initial rate of digestion and
corresponding changes in epidermal and mesophyll cell wall thickness. Can. J. Plant Sci. 73: 111-122.
Previous Page