Space Invaders - Banana Cultures - page 39

tion of professional banana breeders undertook their experiments pos-
sessing very limited knowledge about the cytology, genetics, and taxon-
omy of the genus Musa. 90 However, they were aware of one critical trait
shared by the Gros Michel and most other banana cultivars: the plants
were parthenocarpic, meaning that they did not have to be fertilized by
pollen in order to produce fruit. As a result, their fruit tended to be seed-
less, a characteristic for which human cultivators selected over the course
of centuries if not millennia.
Parthenocarpy posed a majorchallenge for plant breeders who strug-
gled to obtain seeds and pollen from the highly infertile Gros Michel.
In fact, crosses with Gros Michel were possible only because the plants
could be induced to set seeds in small numbers when pollinated by ''wild''
(seeded) bananas. But fertility rates were extremely low. In one breed-
ing experiment conducted by British researchers in Trinidad during the
1920s, pollination trials were carried out with six varieties of bananas, in-
cluding Gros Michel. About 20,000 pollinated flowers yielded fewer than
two hundred seeds, fifty of which were empty. The remaining seeds came
from just two varieties, Gros Michel and Silk; all had been pollinated by
one of the seeded varieties. Only seventeen seeds germinated, and a mere
five survived to the fruit-bearing stage. The sheer size of banana plants
helped to make breeding trials a costly endeavor—20,000 banana rhi-
zomes planted at conventional densities would take up some 25-30 hect-
aresof land. 91 Furthermore,theinroadsof Panamadiseasemadeitdicult
to find healthy Gros Michel plants for breeding.Thework of early banana
breeding, therefore, was tedious, full of uncertainty, and posed a number
of logistical problems.
Sensing that successful breeding would require a wide range of germ
plasm (i.e., genetic diversity), United Fruit's research director, Dr. Otto A.
Reinking,assembledacollectionofsome150Musa accessions from Asia,
Cuba, and Central America. 92 Between 1925 and 1928, United Fruit scien-
tists inChanguinola, Panama, crossedvarieties withdistinct chromosome
counts on the assumption that some of the offspring would possess the
same numberof chromosomes as commercial cultivars.The trials yielded
fourteen sterile hybrids with edible, seedless fruit pulp, but United Fruit's
J. H. Permar noted that they had little economic value since ''in no case is
their qualityequal to the fruits that are generally recognized by the public
as 'bananas.''' 93
In 1930 United Fruit terminated its banana breeding experiments in
Panama. Workers transferred some 130-odd varieties to Lancetilla, the
company'sexperimentalgardenontheoutskirtsof Tela,wheretheywould