A total of 92 isolates of Vibrio harveyi (as identified by ribotyping)
were tested for siderophore activity on a modified chrome azurol sulp
hate agar. Whilst 4 isolates failed to grow, of the remaining 88 strai
ns, 78.4% were positive for the production of siderophores. There was
no correlation between siderophore production and length of time since
isolation. There were significant differences between the host groups
from which the bacteria were isolated (F = 3.23, df = 3.85; p < 0.05)
. The colony halos produced by environmental isolates (halo to colony
ratio = 2.7) and fish isolates (halo ratio = 2.4) were significantly l
arger than the halos from invertebrate isolates (halo ratio = 1.6). Pa
thogenicity in an Atlantic salmon bioassay was barely significantly re
lated to siderophore production (t = -1.87, df = 64; p < 0.1). The max
imal non-inhibitory levels of 2,2'-dipyridyl showed no association wit
h siderophores (t = -1.65, df = 30; p > 0.05) nor with pathogenicity (
t = 0.16, df = 87; p > 0.05). Siderophores must be considered a virule
nce factor in piscine infecting V. hanreyi but not in strains infectin
g invertebrates.