THE ANALYSIS OF TETRAD DATA

Abstract

The theory of four-strand crossing over was founded by JANSSENS T (1909). H e arrived at his theory in an attempt to explain tetrads in which all four cells are different. None such had been observed at that time but JANSSENS believed, by a happy metaphysical insight, that since germ cells were always produced in groups of four, the four must be aH qfferent. It is curious that when tetrads were isolated and four-type tetrads obtained, their significance was often missed. A useful summary of the types of tetrad produced #by various crossovers has been given by WHITEHOUSE (1942, 1949). Haploid characters among the bryophytes, green algae, and fungi have been used almost exclusively in tetrad analyses although flowering plants would seem to offer excellent material. In many flowering plants, including nearly all the Ericaceae, pollen is shed in tetrads. If individual tetrads were placed on styles there should develop, with or without hormone treatment, four seeds corresponding to the four pollen grains of the tetrad. The purpose of the present paper is to show what information can be obtained from tetrad analysis and to analyze some recent published data. Tetrads can be classified into ordered and unot;dered kinds. In ordered tetrads the pairs of cells resulting from the first nidotic division can be distinguished and hence the second division segregation frequency for any gene pair estimated. In organisms which form spores in a linear order such as many ascomycetes and the smuts, they are represented by the top two or four spores and the bottom two or four. In organisms in which the tetrads form a square and in which the second meiotic division spindles form a cross, they are represented by spore pairs along the diagonals, which is probably the case in chiastobasidial basidiomycetes. Another type of ordered tetrad is represented by Saccharotnycoides ludm$ii. Here the second meiotic division spindles are parallel and at about the same level so that one daughter cell from each goes to the top and one daughter cell from each goes to the bottom (WINGE 1947). The top and bottom pairs of cells are not sister cells hut are. so to speak, cousin cells. In this case the second division segregation frequency for a locus is twice the frequency of tetrads in which the top pair and bottom pair of cells have similar alleles. In unordered tetrads spore pairs cannot be distinguished by their spacial position. In some cases they can be distinguished by other means. In organ