Estimation of Self‐Compatibility in Tetraploid Alsike Clover, Trifolium hybridum L 1

FREQUENTLY, sample size, and method of obtaining the sample become problems in the collection of experimental data. Self-compatibility in tetraploid alsike clover ranged from a single seed per several hundred florets to 95% of florets setting seed in a study by Townsend (3). The distribution of compatibility was positively skewed in that a high frequency of plants was highly self-incompatible. Variation in self-compatibility is due to genetic effects, to environmental conditions, and to unknown sources. It is thus necessary to obtain an adequate sub-sample to separate genetic variability from that due to other causes while avoiding erroneous decisions caused by extreme precision of very large sub-samples. Self-compatibility studies in tetraploid alsike clover are very time-consuming, and progeny size must remain small. ]f only a few heads are needed per plant, more plants can be evaluated. These studies were conducted to determine the roles of method of floret manipulation, number of manipulated heads per genotype, and number of florets per manipulated head in estimating selfcompatibility in tetraploid alsike clover. These studies were conducted in a screened greenhouse at Fort Collins, Colo., during the spring and winter seasons of 1961 and 1962, respectively. Day length was extended to 16 to 18 hours by incandescent lighting. Plants 1-4A, 1-5B, 1-6B, 1-9B, and 1-10B, came from Weibull original tetraploid; plant 4-1A from Otofte 4nll; 2-7A, 6-8A, T-2A, and T-8A from ’Tetra Alsike’ lot-1 from Weibull; and plant 3-7B from ’Tetra Alsike’ lot-2, also from Weibull. These plants were selected to represent the observed range of selfcompatibility. The plants were clonally propagated but all genotypes were not included in each study. The two methods of manipulation were: (a) tripping the florets with a folded piece of paper that had been trimmed to a point on one end (conventional); and (b) rolling the florets between the thumb and fingers (roll). Four heads per propagule per replication for a total of 16 heads per genotype were manipulated by each method. Old and immature florets were removed prior to manipulation. The hands of the operator were washed in absolute ethyl alcohol, rinsed in water, and dried with paper towelling between manipulations when the roll method was used. The operator’s hands do not need to be washed between manipulations with the conventional method because the operator’s fingers do not come in contact with the stigmas of the self-pollinated florets. Results from many self-pollinations of highly self-incompatible plants showed that pollen contamination was not a probIem. Data were analyzed with and without transformation to degrees of an angle. Since both methods gave the same results, the data presented herein are nontrans[ormed.