acidic and established that this regimen made
them less able to use oxygen efficiently. In turn,
this led the researchers to infer that induced alka-
losis could have an opposite effect. Dill et al.
(1932) demonstrated that runners could have a
1% decrease in running times when alkalotic.
While it has been shown previously the muscle
cell membranes are impervious to HCO 3 - (Katz
et al.1984; Costill et al.1988), an increase in
extracellular HCO 3 - increases the pH gradient
between the intracellular and extracellular envi-
ronment. The effect of this increased pH gradient
is to facilitate the efflux of intracellular lactate
and H+, thus reducing the fall in intracellular pH
(Katzet al.1984; Costill et al.1988). Both lactate
and H+have been shown to follow a favourable
pH gradient (Roth & Brooks 1990). The time
course for the production of lactate has been
shown to vary from 5 s (Pernow & Wahren 1968;
Saltinet al.1971; Jacobs et al.1983) to several
minutes (Wilkes et al.1983). In an early, well-
conducted study, Osnes and Hermansen (1972)
measured postexercise blood lactate levels in
subjects who ran distances from 100 to 5000 m.
Lactate concentrations increased with increasing
distance up to 1500 m, after which there was no
further increase: pH and blood bicarbonate con-
centrations were lowest after the 1500-m run.
This would seem to suggest that acid–base
balance shifts occur most dramatically after exer-
cise lasting 4–5 min. It is reasonable to assume,
therefore, that if sodium bicarbonate were to be
effective as an ergogenic aid, it would be so over
a similar time span, since these time periods are
dependent upon high rates of energy production
from anaerobic glycolysis.
There has been some suggestion that the mech-
anisms whereby sodium bicabonate loading is
effective lie not with the bicarbonate ion but are
possibly due to the sodium load (Saltin 1964;
Kozak-Collinset al.1994). Sodium could change
intravascular volume, which in turn could alter
performance. Kozak-Collins et al.(1994) tested
this hypothesis with the ingestion of either
NaHCO 3 or NaCl, which both provided equimo-
lar amounts of sodium given prior to repeated
396 nutrition and exercise
bouts of 1-min exercise. Performance was not
enhanced in either condition but haematocrit
measures suggested that intravascular fluid
status remained similar. pH was significantly
raised in the bicarbonate trial when compared
with that in the NaCl trial. Further studies are
required to determine whether intravas-
cular volume is responsible for the increased
performance.
A greater understanding of acid–base balance
during rest and exercise can be gained by reading
Jackson (1990), Jones (1990), Lindinger and
Heigenhauser (1990) and Heigenhauser et al.
(1990).
The work of Jervell (1928), Dennig et al.(1931)
and Dill et al.(1932) was, by and large, forgotten
by the coaching and scientific communities. The
modern era of the study of acid–base balance
during exercise performance essentially began in
the 1970s, with publication of the work of Jones et
al.(1977). These workers studied five men who
acted as their own controls, through treatments
consisting of either a placebo (calcium carbon-
ate), 0.3 g ammonium chloride (acidic) per kilo-
gram of body mass, or sodium bicarbonate in
the same dosage. All doses were taken after an
overnight fast and over a 3-h time period. The
exercise consisted of cycle ergometry utilizing
three different protocols: 20 min at both 33% and
66% of previously determined maximum oxygen
uptake (V.
o2max.), followed by exercise to exhaus-
tion at 95% V.
o2max., without rest in between. Time
to exhaustion at the 95% V.
o2max.power output
level was approximately 4 min for the control
condition. In the bicarbonate treatment, exhaus-
tion time was approximately twice that of the
control, whilst in the acidic condition, it was
about half the control time. Blood lactate concen-
tration in the bicarbonate treatment was signifi-
cantly greater (P<0.01) than the control at both
the 66% power output level and at exhaustion. In
the ammonium chloride condition, blood lactate
levels were significantly lower in these two work
periods. Blood pH was consistently higher in
the bicarbonate treatment group and lower in
the ammonium chloride treatment than in the