80 CELL HEREDITY 



the mechanism of exchange is not so simple and there may even be nega- 

 tive interference; this means that the occurrence of one exchange may 

 make it more likely that another will occur in its proximity. 



When evidence of interference is observed, the linear ordering of genes 

 by the additive method may fail. In this case, other topological methods 

 may be employed. For instance, in a three-factor cross, the middle gene 

 can be separated from its two outside neighbors by a double exchange, 

 one occurring between it and each of the terminal genes. Hence, the 

 middle gene can be identified as that which occurs among recombinants 

 least often in association with the other two from the opposite parent. 



Ordered tetrads, such as the meiotic products of Neurospora and some 

 other fungi and yeast, yield considerably more information than can be 

 gained from the use of gametes that have been collected at random. 

 Not only can independence in segregation or the degree of linkage be- 

 tween genes be derived from their analysis, but the 1 : 1 segregation of 

 alleles is repeatedly verified (or exceptions are found) and the centromere 

 is located and used as an additional marker. In three-point crosses not 

 only the occurrence of double crossing over but also the specific strands 

 involved in multiple exchanges may be observed directly. Unordered 

 tetrads are more common among the algae, yeast, and fungi; these too 

 have advantages, and special methods have been developed to exploit 

 them. In most organisms, however, the meiotic products are not found 

 in tetrads, so mapping is based upon the sampling of gametes at random. 



Some genes cannot be located on chromosome maps by the techniques 

 described; in fact, they do not show Mendelian segregation in crosses. 

 A case in point is a gene for resistance to a high level of streptomycin 

 in ChlamydomoTuis. It is not known to be linked to any other genes and 

 it passes from one parent through the zygote into all four gametes. 

 This and other cases that seem to involve nonchromosomal genes will 

 be discussed in Chapter 9. 



The genes to be discussed in Chapters 4-8 are located on chromo- 

 somes. Step by step their distribution along the whole length of a 

 chromosome can be approximated. Such genetic maps have been 

 checked cytologically by comparing the position of genes with the loca- 

 tion of chromosome breaks that lead to chromosome aberrations. This 

 can be done, for example, in the corn plant (Plate Vb) whose chromo- 

 somes show a pattern of enlargements (chromomeres), but much more 

 easily in Drosophila (Plate Villa), where the giant salivary chromosomes 

 appear cross-banded. The linear pattern of these bands can be com- 

 pared with the pattern of gene loci on the genetic map as shown in 

 Figure 3.7. The orders of genes are identical. The spacing does not 

 exactly correspond, either because of some nonrandomness of crossing 



