We typed five microsatellite loci in 52 landraces of Italian emmer wheat to determine if genetic analysis of cereals can provide information relevant to the spread of agriculture. Each of the five loci was polymorphic with 43 allele combinations identified in the 52 landraces. The allele combinations fell into two groups. Group 1 comprised 27 genotypes found in 42 landraces and Group 2 comprised 15 genotypes found in 10 landraces. The landraces with Group 1 genotypes showed a strong correlation between geographical and genetic distances (r = 0.601, p <0.001) but those with Group 2 genotypes did not (r = 0.116, p = 0.244). We inferred that the Group 1 landraces might therefore retain a phylogeographical structure that reflects ancient events. We present a phylogeographical model for the spread of agriculture that enables the point of origin of crop cultivation to be predicted by comparison between the genetic and geographical distances between landraces. We applied this model to the Group 1 landraces by positioning 131 hypothetical points of origin around the coastline and northern border of Italy. The highest correlation coefficients between genetic and geographical distances were seen for hypothetical points of origin located on the coast of northern Puglia. We repeated the analysis with 1040 hypothetical points of origin located within the Italian peninsula. Again, the highest correlation coefficients were located in northern Puglia. These predicted points of origin correspond with the location of the earliest agricultural sites in Italy. The results show that plant genetics can be used to study the spread of agriculture. © 2009 Elsevier Ltd. All rights reserved.
- Emmer wheat