FACTORS WHICH INFLUENCE SPONTANEOUS SELF-FERTILIZATION IN SWEET CLOVER (MELILOTUS)

• Published in: Canadian Journal of Research
• Main Authors: Kirk, L. E., Stevenson, T. M.
• Format: Article in Journal/Newspaper
• Language: English
• Published: Canadian Science Publishing 1931
• Subjects: Pharmacology (medical); Complementary and alternative medicine; Pharmaceutical Science; Arctic
• Online Access: http://dx.doi.org/10.1139/cjr31-071; http://www.nrcresearchpress.com/doi/pdf/10.1139/cjr31-071

Seven factors have been described which seem to influence the tendency of sweet clover plants to produce seed by spontaneous self-fertilization. These are: length of stamens; stage of flower development when pollen is liberated from the pollen sacs; distribution of free pollen within the flower; size of cavity in the upper part of the keel; amount of pollen; condition of the pollen; and receptivity of the stigma.Spontaneous self-fertilization is the normal condition in certain plants of white blossom sweet clover, i.e., M. alba. This was not found to occur in any variety of yellow blossom sweet clover with the exception of "Redfield Yellow". True breeding lines of M. alba have been isolated consisting of naturally self-fertilized plants and other lines also which will not produce seed unless the flowers are manipulated. M. alba consists of a mixture of normally self-fertilized plants and plants that are random pollinated. This is especially true of certain varieties such as "Arctic". In some plants of M. alba an abundance of germinating pollen grains were found adhering to the stigmas before the flowers opened, thus precluding the possibility of natural crossing. Practically 100% of the flowers on such a plant produced selfed seed. The character of spontaneous self-fertilization exhibited clear cut segregation in a selfed line of M. alba. The importance of this character in relation to breeding improved varieties and pure seed production is emphasized.The efficiency of the "suction" method of emasculating sweet clover flowers depends on the peculiarities of the latter. This method is quite reliable with many plants but very unsatisfactory with others, depending upon the particular type of flower structure and behavior.