Why do calculated recombination frequencies between pairs of loci that are located far apart underestimate the true genetic distances between loci? A. Recombination frequencies cannot exceed 50% and asymptotically approach 50% as the distance between genes increases because only one crossover can occur per chromosome. B. As the distance between loci increases, some multiple crossovers go undetected such that the relationship between recombination frequency and map distance ceases to be linear. C. Loci that are far apart on a chromosome often recombine to produce combinations of alleles that result in a lethal phenotype and such recombinants are not recovered or identified. D. Loci that are far apart on a chromosome are often susceptible to a form of chromatic interference that results in reduced crossing over, and this interference does not occur for loci that are closer together. E. If two crossovers occur between two loci, then only nonrecombinant gametes are produced; therefore, recombination frequencies actually begin to decrease for more distant pairs of genes.

B. As the distance between loci increases, some multiple crossovers go undetected such that the relationship between recombination frequency and map distance ceases to be linear.  

Explanation:

When calculating recombination frequencies, and hence, map distances, we might notice that these distances are not completely additive. They might vary. For example, let us say that we have three genes, A, B, and C, in that order. We calculated that the distance between A and B equals 5.9 MU and that B and C are 19.5 MU apart. According to this, we might say that the total distance between A and C is 25.4 MU (5.9 + 19.5). However, after a two-point calculation between A and C, the value equals 23.7 MU.

The recombination frequency between these two genes located in the extremes and far apart underestimate the actual genetic distances between them because there might occur other crossing-overs that were not detected. This is when calculating the distance between A and C, we probably will not detect the occurrence of a double recombinant between them, and hence, we might sub-estimate the real distance.

The relationship between the actual map distance (number of crossing overs) and the recombination frequency between two loci, is not lineal. The farther apart are the two genes, the worse is the distance estimation.  

not too sure but maybe in the hot springs.