11.15 Performance of the lemma awn isogenic lines of Atlas in the Canadian prairies.
S. Jana. Crop Science Department University of Saskatchewan, Saskatoon, Canada.
The performance of the four backcross-derived near-isogenic lines of Atlas barley developed by Briggs and Schaller - Full-awned (AABB), Half-awned (AAbb), Quarter-awned (aaBB) and Awnless (aabb) - has drawn attention to their value in elucidating the role of lemma awn in kernel productivity (see Schaller and Qualset, 1975 for a recent review). Faris (1974) grew these isogenic lines at Beaverlodge, Canada for three years and found that the Half-awned line with about 55% shorter awn than Full-awned consistently out-yielded it. The yield disadvantage of Full-awned was believed to be due to a greater stress within the spike at the time of awn formation under the long photoperiodic conditions at Beaverlodge (55°12'N). It was also believed that the yield advantage of Full-awned Atlas disappeared in environments conducive to low grain yield (Schaller et al., 1972; Faris, 1974).
The four 'green' isogenic lines (homozygous recessive at Sp/sp), which were kindly made available to us by Professor C.W. Schaller of the University of California, Davis, have been in use at Saskatoon, Canada (52°ll'N) since 1971 in a series of quantitative genetic experiments (BGN 4: 37-40). One of these experiments consists of growing the four isogenic lines in a randomized complete block design with six replications at a commercial seeding rate, giving about 270 adult plants per m2 (Experiment D), Both 'green' and 'spotted' (homozygous SpSp) were included in a large spaced plant experiment since 1971 (except 1973), in which the entries were grown in a twice replicated field trial (Experiment S). In each entry in a replication, plants were widely spaced (on 45 cm centers) to avoid interplant and intergenotype competition, and thus provide a reliable measure of mean performance of each genotype under conditions optimum for lemma awn development. Methods of planting and measurement for Experiment S are similar to those given in Jana (1973).
Experiment D: In this densely seeded experiment, an analysis of variance of randomized complete block design was first calculated for each year separately using the fixed effects model. Entry differences were significant (p < .05) for each year except 1975. For the years 1971-1974, the sum of squares for entries was subdivided into three a priori orthogonal comparisons as given below:
The data from 1971-74 were then pooled in a combined analysis of variance over four years (Table 1).
Experiment S. The method of analysis of data was similar to that described for Experiment D above, except that average single plant yield was used in Experiment S. The results of the analysis of variance, calculated over a three-year period (1971-74, except 1973) are given in Table 1.
The estimates of mean yields of the four isogenic lines are given in Table 2. It is evident from Table 1 that the yield difference between Full-awned and Half-awned is not significant (C1). Although the comparison between Quarter-awned and Awnless was significant in the densely seeded experiment, the yield advantage of the Quarter-awned lines seems to disappear when interplant competititon is minimized (C2). On the other hand, the comparison, C3 is highly significant at both spacings. The relative magnitudes of the mean squares due to the three orthogonal comparisons reveal that the bulk of the variation in kernel yield among the isogenic lines is ascribable to the reduced yield of the short-awned lines (Quarter-awned and Awnless). Thus, the lack of a significant difference between the long-awned types (Full-awned and Halfawned) and the highly significant difference between the long-awned and shortawned types indicate the greater yielding potential of the long lemma awn. Even this substantial difference between long-awned and short-awned isogenic lines disappeared (P:.1-.25) in the rather good yielding summer season of 1975.
Contrary to the earlier reports (Qualset et al., 1965; Faris, 1974), results at Saskatoon show no superiority in yield of the Half-awned genotype. At the commercial seeding rate, a much larger difference is noticeable between Quarter-awned and Awnless than between Full-awned and Half-awned. This is also reflected in the considerably larger mean square estimate for C2 than for C1 in Experiment D. In this respect our results also differ from those of Schaller and Qualset (1975), who found a significant yield advantage of Full-awned at Davis, California, over a three-year period.
Saskatoon is located in the semi-arid Canadian prairies. It is reasonable to assume that the moisture stress at Saskatoon, during 1971-75, for which kernel yield data are given here, was at least as great as at Beaverlodge during 1967-69. In addition, long daylength, which hastens development and presumably causes a greater nutritional stress within the spike at the time of awn formation, is prevalent at Saskatoon. Therefore, our results cannot be interpreted in terms of competition for nutrients between the developing awn and florets during the early autogeny of the spike. However, the diversity of results leads support to the postulate of Schaller et al. (1972) that the response of the four awn types to environmental conditions is very complex. A generalized conclusion regarding the relationship between lemma awn length and kernel productivity is bound to meet with many exceptions. However, consideration of this relationship should be very important in a barley breeding programme, particularly in the semi-arid prairies. In the case of a choice between a long-awned or a short-awned barley, perhaps the choice is obvious. The choice is less obvious between the Full-awned and Half-awned types. We have initiated an experiment to quantitatively evaluate the contribution of the lemma awn in the improvement of kernel size and yield under artificial and field level stress conditions commonly encountered on the prairies. It is hoped that, when concluded, the experiment would furnish additional criteria toward understanding the role of lemma awn length in barley improvement.
References:
Faris, D.G. 1974. Yield component development in four isogenic barley lines differing in awn length. Can. J. Plant Sci. 54: 315-322.
Jana, S. 1973. Factorial genetic analysis of a two-locus system in barley. Can. J. Genet. Cytol. 15: 473-482.
Qualset, C.O., Schaller, C.W. and Williams, J,C, 1965. Performance of isogenic lines of barley as influenced by awn length, linkage blocks, and environment. Crop Sci. 5: 489-494.
Schaller, C.W,, Qualset, C.O. and Rutger, J.N. 1972. Isogenic analysis of the effects of the awn on productivity of barley. Crop Sci. 12: 531-535.
Schaller, C.W., and Qualset, C.O. 1975. Isogenic analysis of productivity barley: Interaction of genes affecting awn length and leaf-spotting. Crop Sci. 15: 378-382.
