A novel mapping procedure is presented for organelle genes or any other genetic system exhibiting a measurable frequency of exchanges occurring at a constant rate over a measurable time interval. For a set of markers in a multiply-marked cross, the exchange rates measure relative map distances from a centromere-like attachment point. With this method, we present mapping data and a linear map of genes in the chlcroplast genome of Chlamydomonas. The data are plotted as log (percent remaining heterozygotes) against time and map distances are taken as proportional to slope. A statistical method which is an adaptation of jackknife methodology to a regression problem was developed to estimate slope values. A single line is fitted to pooled data for each marker from several crosses, and then lines are re-fit to a series of pooled data sets in each of which the observations from a single cross have been omitted. From these data sets a final summary slope is computed as well as a statement of its variability. The relative positions of new markers present in single crosses can then be estimated utilizing data from many crosses. The method does not distinguish between one-armed and two-armed linear or circular maps. However, evaluation of this map in conjunction with cosegregation frequency data (Sager and Ramanis 1976b) provides unambiguous evidence of the genetic circularity of the Chlamydomonas chloroplast genome.