33
 D400 group. This finding was consistent with the feral origin of the colonies,

 in which African mtDNA was found (Hall & Smith 1991).

 Some of the drones from southern Mexico were collected from managed

 colonies established from feral swarms, while the majority were collected from

 feral swarms (Hall & Muralidharan 1989). It was expected that both

 USA/European and South African markers would be found. The number of

 variants common to Mexico and South Africa was far greater than those

 common to the USA and Mexico, although there were no alleles common to the

 Mexican and South African sample populations. The variants and alleles unique

 to the Mexican samples resembled those in the South African samples and,

 apparently, represented the recent establishment of the feral African population.

 As in Honduras, most of the European-type variants and alleles common to the

 USA and Mexican populations were in the M300 group (Table 2, distributions

 2 and 3).

 The greater number of variants and alleles identified in the neotropical

samples relative to the USA reflected the presence of African as well as

European variants. Variants unique to the neotropical drones, but similar to

those found in USA bees, probably reflect regional differences in European

races. Five alleles detected only in Honduras and Mexico (Table 2, distribution

5) were composed of variants that either were found, or were of the same

variant-type as those found, in the South African drones (Table 2, distribution