22 
A. K. MORISON AND J. R. ANDERSON 
A: FEMALES 
8 
B: 
25 
□ STAGE 2 
□ STAGE 3 
■ STAGE 4 
25 
MALES 
50 
50 
S 
75 
75 
100 
125 
100 
125 
STANDARD LENGTH (mm) 
MARCH 
n = 12 
150 
175 
MAY 
n = 5 
i ' m 1 
150 
175 
] " 
MARCH 
- 
n =7 
■■HI 
175 
175 
Fig. 3. Length-frequency histograms showing maturity 
stage for (A) female and (B) male Galaxias brevipinnis 
caught in March and May 1990. 
turbidity made efficient electrofishing imposs¬ 
ible. 
Vertebral morphology 
An analysis of variance of the mean number of 
vertebrae of G. brevipinnis in our samples and in 
10 of those studied by McDowall & Frankenberg 
(1981) showed significant differences among the 
samples (F = 17.85, P < 0.001). However, a 
Newman-Keuls test could not separate samples 
into distinct groups. 
DISCUSSION 
Our survey results indicate that populations of 
G. brevipinnis are present at 11 sites in nine 
streams in catchments in north-eastern Victoria. 
Moreover the distribution of the 11 sites, the 
size range of G. brevipinnis we caught and their 
reproductive condition suggest that the species 
has been present for several years at least and has 
established breeding populations that spawn in 
April or early May. The populations reported 
here may be part of the natural range of the 
species, and perhaps remnants of a once wider 
distribution within the Murray-Darling Basin, 
or they may be recently established popu¬ 
lations. 
Pre-1989 surveys 
Fish surveys carried out prior to 1989 at 89 sites 
in the Upper Murray catchment and 29 sites in 
the Kiewa River catchment recorded no speci¬ 
mens of G. brevipinnis , suggesting that the 
species has recently extended its range into the 
Murray-Darling drainage basin. These surveys 
included multiple samples from all major tribu¬ 
taries and were widely distributed between their 
headwater and lowland sections. Sampling 
methods included dip-netting (27 sites), electro¬ 
fishing (4 sites), fyke nets (5 sites), and the use of 
the piscicide rotenone (77 sites). The surveys 
recorded Galaxias spp. at 35 sites, including 
both G. olidus (32 sites) and G. rostratus{ 3 sites 
near Albury). We expect that G. brevipinnis, if 
present in the area, would also have been caught 
even though none of the pre-1989 sampling sites 
coincided exactly with sites where we caught G. 
brevipinnis. 
It is possible that G. brevipinnis was caught 
during these earlier surveys but was not correctly 
identified. However, all the specimens in the 
Museum of Victoria have been recently exam¬ 
ined by T. Raadik (formerly of the Museum’s 
Department of Ichthyology) who found no G. 
brevipinnis among the many collections of G. oil- 
dus from north-eastern Victoria. Specimens of 
G. olidus reported from north-eastern Victoria 
by McDowall & Frankenberg (1981) were care¬ 
fully examined by those authors and they re¬ 
ported no G. brevipinnis. Species of Galaxias 
recorded during Fisheries Division surveys have 
not ahvays been unambiguously identified but 
no specimens of G. brevipinnis have been col¬ 
lected. Small specimens of Galaxias can be dif¬ 
ficult to identify in the field but. because G. 
brevipinnis grows much larger than G. olidus and 
G. rostratus (the only other galaxiids known to 
inhabit the same area), the larger specimens are 
readily distinguishable. Presumably G. brexi- 
pinnis would have been correctly identified 
among the larger specimens. 
M\urray-Darling populations 
The April-May spawning period indicated for 
G. brevipinnis in north-eastern Victoria is within 
the autumn or early winter spawning period pre¬ 
viously reported for G. brevipinnis in mainland 
Australia (Frankenberg 1969, Koehn & O’Con¬ 
nor 1990), Tasmania (Andrews 1976) and New 
