Content uploaded by Matthew Schrager
Author content
All content in this area was uploaded by Matthew Schrager
Content may be subject to copyright.
Eur J Appl Physiol (1995) 70:367-372 © Springer-Verlag 1995
Carl Foster • Lisa L. Hector • Ralph Welsh
Mathew Schrager • Megan A. Green • Ann C. Snyder
Effects of specific versus cross-training on running performance
Accepted: 7 August 1994
Abstract The cross-training (XT) hypothesis suggests
that despite the principle of specificity of training,
athletes may improve performance in one mode of
exercise by training using another mode. To test this
hypothesis we studied 30 well-trained individuals (10
men, 20 women) in a randomized longitudinal trail.
Subjects were evaluated before and after 8 weeks of
enhanced training (+10%/week), accomplished by
adding either running (R) or swimming (XT) to baseline
running, versus continued baseline running (C). Both
R (- 26.4s) and XT (-13.2s) improved time trial
(3.2 km) performance, whereas C did not (-5.4s).
There were no significant changes during treadmill
running in maximum oxygen uptake (VO2peak; -- 0.2,
--
6.0, and + 2.7%), steady state submaximal lzO2 at
2.68m's -1 ( - 1.2, - 3.3 and + 0.2 ml'kg-l-min-1),
velocity
at gO2pea k (
+ 0.05, + 0.25 and
+ 0.09 m' s -a) or accumulated 02 deficit (+ 11.2,
- 6.1 and + 9.4%) in the R, XT or C groups, respec-
tively. There was a significant increase in velocity asso-
ciated with a blood lactate concentration of
4 mmol- 1-1 in R but not in XT or C ( + 0.32, + 0.07
and + 0.08 m" s- 2). There were significant changes in
arm crank l/O2ve~k (+ 5%) and arm crank 1)O2 at
4 mmol" 1-1 ( + 6.4%) in XT. There was no significant
changes in arm
crank gO2pea k ( -~-
1.3 and - 7.7%) or
arm crank 1/O2 at 4 mmol" 1-1 ( + 0.8 and + 0.4%) in
R or C, respectively. The data suggest that muscularly
non-similar XT may contribute to improved running
performance but not to the same degree as increased
specific tranining.
Key words Cross-training • l/O2veak " Running •
Lactate. Oxygen deficit
C. Foster (N~) - L.L. Hector. R. Welsh. M. Schrager
M.A. Green ' A.C. Snyder
Human Performance Laboratory, Milwaukee Heart Institute,
P. O. Box 342, Milwaukee, WI 53201-0342, USA
Introduction
Systematic exercise training is associated with en-
hanced exercise performance in both previously seden-
tary individuals and athletes. Improved performance is
usually thought to be related to the magnitude of the
training stimulus and to be fairly specific to the type of
training undertaken.
Beginning in the early 1980s, with the emergence of
the triathlon, many single-sport athletes noted that
they performed remarkably well in their previous
specialty events despite reduced specific training to
allow for multi-event training. The concept emerged
that the non-specific training provided a "crossover
benefit" relative to general fitness. At about the same
time, recognition of the importance of both aerobic and
resistance training to the fitness participant, and an
increased number of middle-aged participants desiring
to minimize orthopedic consequences of single-mode
training, created interest in multi-event training. Ac-
cordingly the concept of "cross-training" emerged. It is
now fairly well accepted that multi-mode training has
a role for not only the general fitness participant but
also in maintaining general conditioning in athletes
during reductions of training associated with injury or
regeneration cycles.
Despite studies demonstrating the value of non-spe-
cific training in previously sedentary individuals (Lewis
et al. 1980; McArdle et al. 1978; Pollock et al. 1975)
there are comparatively few data regarding the value of
non-specific training in athletes. Loy et al. (1993) and
Mutton et al. (1993) have demonstrated significantly
improved running performance from either stair-
climbing exercise or combined run/cycle training that
were generally comparable to run-only training. Thus,
there appear to be data to support the concept that
non-specific, but muscularly similar, training may con-
tribute to enhanced running performance. At this time
there are virtually no data concerning the responses to
368
non-specific, and muscularly dissimilar, training (e.g.
swimming for runners). The concept of the "lactate
sink" (Stainsby and Brooks 1990) would suggest that
endurance training of muscle fibers not active during
running might reduce the magnitude of lactate accumu-
lation during running and, accordingly, contribute to
enhanced performance (Sjodin et al. 1982). Accord-
ingly, the purpose of this study was to evaluate the
effect of muscularly dissimilar non-specific training
(swimming) on running performance, and its physiolo-
gical correlates.
Methods
The subjects for this study were 30 volunteers (10 men, 20 women).
All provided informed consent prior to participation. Mean (SD)
characteristics of the subjects at the onset of training are presented in
Table 1. Although all of the subjects were previously well trained
and most had some experience in local running events, none were
serious competitive runners. During the 2 months prior to being
recruited for the study, the amount of running training ranged from
15 to 80 km week- 1.
Following recruitment, the subjects engaged in 8 weeks of baseline
training consisting of 30 min running at a moderate pace 5 days
weekly (25-30 km week- 1). This period was intended to ensure that
differences in training prior to recruiting the subjects were mini-
mized and that the level of training was consistent in all subjects
prior to application of the experimental intervention. After this
8-week baseline period, the subjects performed the first set of tests
and were randomized to one of three intervention groups. After
8 weeks of intervention, the subjects performed the second set of
tests.
Evaluation consisted of a 3.18-km (2-mile) running time trial held
on an indoor 201.2 m track, laboratory tests of peak oxygen uptake
(1202p~ak) and blood lactate accumulation, designed to elicit the
running velocity associated with a blood lactate concentration of
4 mmol" 1 1 (v4 mmol" 1-1), accumulated oxygen deficit (~O2 defi-
cit), running economy (at 2.68 m" s-1), and the velocity requiring
a 1202 equal
to VO2peak(/)VO2peak)
during treadmill running. Addi-
tionally 12Ozp~a k and
V02
at 4 mmol" 1-1 during arm crank exercise
were measured.
In order to minimize the influence of competitive experience, each
subject performed two time trials during each evaluation period.
Training during the day preceding each time trial was subjectively
easy. Each time trial was conducted without competitive assistance
from other runners. Lap times were provided throughout each trial.
The faster time trial during each evaluation period was accepted as
the criterion measure for that subject.
Blood lactate accumulation was evaluated using an enzyme elec-
trode system in capillary blood obtained from a warm fingertip
following steady velocity runs (4-6 rain) on the treadmill at 5%
elevation. The velocity during these runs was 2.2 m' s- 1 during the
first run and was increased by 0.2-0.3 m.s -1 during successive
stages. The duration of each ru n (beyond 4 min) was determined by
observing on-line plots of 1)'O2 and continuing the stage until a clear
steady state of 12Oz was observed. In order to facilitate blood
sampling, successive stages were separated by 1 min of walking at
1.3 m' s-
1 0% grade. Successive stages were performed at least until
the rating of perceived exertion (RPE) for that stage equaled 5 on the
category ratio Borg scale (Borg et al. 1987). This scale was slightly
modified by changing the verbal anchors into idiomatic American
English. For example the term "severe," which is the verbal anchor
for a rating of 5, was changed to "hard." We have found that this
slight modification of the scale greatly improves its clinical utility
(Foster 1988). v4mmol.1-1 was accepted as the index of blood
lactate accumulation. These steady-state runs were also used to
calculate the 1202 requirement of running to allow subsequent
calculation of running economy (the 1202 at 2.68 m" s- 1) and the
ZO2 deficit (Medbo et al. 1988). Following a 15-min recovery period
of slow walking/stretching after the last submaximal run, l?O2pe, k
was measured as the highest full minute 1202poa k during an uphill
(5% grade) running bout designed to allow measurement of the ZO2
deficit. The velocity during this bout was set at 0.9 m" s- 1 faster than
the velocity associated with an RPE of 5 (hard) during the submaxi-
mal runs. Pilot studies in our laboratory demonstrated that this
scheme for choosing velocity consistently caused the subjects to
become fatigued within 2-5 min. The ~O2 deficit was calculated
according to the procedure outlined by Medbo et al. (1988). The
1702 requirement was calculated from the submaximal steady-state
runs designed to measure v4 mm and running economy, extrapo-
lated to the velocity used during the maximal run. The ~O2 con-
sumed during this run was subtracted from ZO2 requirement to
compute ZO2 deficit. The vl2Ozp~, k was calculated by extrapolating
the submaximal velocity-l?Oz relationship to the l)'O2p~a k. Although
an independent measurement of 1202po, k using a conventional in-
cremental protocol and leveling off criteria were not used, the
protocol used in this study resulted in measured 1202 that was
consistently less than predicted from the submaximal studies, in
a respiratory exchange ratio greater than 1.10, heart rates within 5 %
of age-predicted maximal, and post-exercise blood lactate concen-
trations in excess of 10 mmol'l-1. On the basis of studies in our
laboratory demonstrating the behaviour of 1202poa k during non-
incremental protocols (Foster et al. 1993), we felt that the l?Ozpo, k
value thus achieved was likely to be comparable with that obtained
with a conventional exercise protocol.
Blood lactate accumulation during arm ergometry was evaluated
from blood lactate measurements obtained during arm crank exer-
cise with progressive 4-min stages. A 30-s break between stages was
allowed to facilitate blood sampling. During this bout, resistance on
the flywheel of the ergometer was adjusted to 10 N (males) and
5 N (females) on the basis of pilot studies. Power output during
successive stages was varied by changing cranking frequency. Arm "
~ZOzpea k was the highest
VO 2
during a bout where the subject
attempted to complete 200 revolutions of the arm crank as rapidly as
Table 1 Mean (SD) characteristics of the subjects of baseline
Age Height Weight % Fat
(years) (m) (kg)
TREADMILL ARM Crank
VO2pea k V4 mmol.1
1 ~Ozpea k
VO 2 at mmol'l- 1
(ml • kg- 1) (m. s- 1) (min- 1)
Men 34.2 1.77 76.3 15.9
(11.9) (0.05) (6.7) (4.7)
Women 24.6 1.67 58.8 17.9
(5.0) (0.07) (7.2) (4.2)
54.9 2.66 2.78 1.44
(6.5) (0.29) (0.38) (0.22)
52.4 2.52 1.91 1.21
(5.0) (0.28) (0.27) (0.23)
possible. The subject was instructed to start cranking as rapidly as
possible and to complete 200 revolutions in minimal time. At the
level of flywheel resistance used in this study, 200 revolutions at
maximal effort usually required about 2 min. We have previously
demonstrated that self-paced trials of this nature are very effective in
eliciting peak physiological responses in athletes (Foster et al. 1993).
The three randomly assigned intervention groups included a con-
trol group (n = 11), an enhanced running group (n = 9) and a cross-
training group (n = 10). The control group continued running
30 rain daily at a moderate pace, 5 days weekly. The enhanced
running group increased their running training by about 10% per
week. They completed eight training sessions during 6 days per week
including both interval and continuous running. The cross-training
group continued running 30 min daily at a moderate pace, 5 days
weekly. However, they increased their training frequency to eight
sessions during 6 days per week by adding three swimming sessions
per week to their running program, with the intent of having the
same total training load as the enhanced running group.
Training was quantitated by a modification of the method de-
scribed by Banister et al. (1986), which multiplies training intensity
by duration to create a training impulse score for each training
session• Because of practical difficulties in obtaining heart rate for
every training session in a large group of subjects and because of
concerns about the ability of heart rate to represent the training load
during interval exercise, we used the perceived exertion scale of Borg
et al. (1987) and asked the subject to rate their effort for the entire
training session. This is an extrapolation of the wide clinical use of
the perceived exertion scale as an adjunct to heart rate. The "ses-
sion" RPE was multiplied by the total duration of training (in
minutes) to create a training impulse score, which we refer to as
100 .....
t e/ r=0.65
901 .. /
=
t
"-/'
° 01
¢¢
:]/..
• e
60 . ~ . i . i , D - i . i • i • J • i •
0 1 2 3 4 5 6 7 8 9 10
Session RPE
Fig. 1 Comparison of the rating of perceived exertion
(RPE)
for
a 30-min steady-state running training session and the mean per-
centage heart rate reserve during that training session, r = 0.65
369
training load. In pitot studies there was a moderate correspondence
between average percentage heart rate reserve during 30-min
steady-state runs and the "session" RPE (Fig. 1). Following sugges-
tions that indices of lactate accumulation (and associated heart
rates) could be used to control training intensity (Coen et al. 1991;
Gihnan and Wells 1993; McLellan and Skinner 1981), we also
evaluated the relationship of the "session" RPE to the percentage of
time spent below, between and above commonly used blood lactate
transition zones (2,5 and 4.0retool.1-1) during 30-rain training
sessions (Fig• 2). These sessions included both interval and continu-
ous exercise. There was a good correspondence between the training
"session" RPE and the behavior of heart rate in relation to the
common blood lactate transition zones. Accordingly, we felt that the
training "session" RPE method provided approximately the same
information regarding the relative training intensity as the method
of Banister et al. (1986) which relies on continuous measures of heart
rate.
Statistical comparisons were made amongst the intervention
groups using repeated measures ANOVA. A Scheffe test was used
for post hoc comparisons. A P value of < 0.05 was accepted as
statistically significant.
Results
There were no significant differences amongst the three
groups prior to intervention in any of the primary
outcome measures. As per the experimental design, the
mean (SD) 8-week training load increased significantly
and similarly in the enhanced running [509 (24) to 695
(108)-week -t) and cross-training [519 (21) to 835
(126)" weak-t) groups and did not change in the con-
trol group [576 (14) to 549 (21) • week-1). The overall
increase in load in the enhanced running and cross-
training groups represented 7.1% and 10.3 % increases
per week of intervention, respectively. The increase in
training load was dominantly by an increase in training
duration in both the enhanced running and cross-train-
ing groups. There was a non-significant trend for in-
creases in training volume to predominate in the cross-
training group and training intensity to predominate in
the enhanced running group (Fig. 3).
Time trial performance improved significantly in
the enhanced running group (- 26.4 s) and in the
IOQ
90"
80"
¢¢ 7Q"
,J
A: Blood
Lactate < 2.5 rnmol*l-1
/ : ~ : : : •
1 2 3 4 5 8 7 8 0 10
Sesaton RPE
B: Blood Lactate
Between 2.5
& 4.0
mmoPl-1
100
80 •
8O
68 | •
50 •
40 •
30
20
10 • •
0 1 2 3 4 5 6 7 8 9
Session
RPE
100
90
80
¢ 70,
0 80
so.
2(1"
10"
10
C: Blood Lactate
> 4,0 mmot'l-1
II •
1 2 3 4 5 6 7 8 0 10
Seaslorl RPE
Fig. 2A-C Comparison of RPE for a 30-min running training session (including both steady state and interval sessions) versus: A the
percentage of time the heart rate is below that associated with a blood lactate concentration of 2.5 mmol. 1-1; B the percentage of time the
heart rate is between that associated with blood lactate concentrations of 2.5 and 4.0 mmol. 1 - 1, and C the percentage of time the heart rate is
above that associated with a blood lacatate concentration of 4.0 mmol. 1-1
370
1500
1000
.J
500
0
0
TRAINING LOAD
ENHANCED RUNNING
---II--- CROSS TRAIN
--'O-- CONTROL
Baseline
4
intervention
8 12 16
WEEK
,K
,.=
E
O
G
C~
_=
._=
300 •
250 -
200
150
Training Duration Training Intensity
o.
E 4
O
2'
E
.E
. . , • . . , • . . , . . . ~_ ~ ,
100 0
i
1 p M L q M
4 8 12 16 0 4 12 16
WEEK WEEK
Fig. 3A-C Schematic representations of the average training load duirng baseline and intervantion periods. Training Load (dimensionless
units) was computed as the weekly summation of daily duration (in minutes) multiplied by the session RPE for each day's training session.
Although the cross-training group (--M--) had somewhat greater training duration during the intervention period than the enhanced
running group (re--), training load was counterbalanced by a somewhat greater training intensity in the enhanced running group. There
were no statistically significant differences in training load between the enhanced running and cross-training groups. ~O--) Control
group
cross-training group (- 13.2 s). There was no signifi-
cant change in time trial performance in the control
group (-5.4 s) (Table 2). The change in time trial
performance was significantly greater in the enhanced
running group than in the cross-training group. Com-
pared to pre-randomization values, post-training time
trial performance decreased 3.2%, 1.4% and 0.6% in
enhanced running, cross-training and control groups,
respectively (Fig. 4).
v4 mmol' 1-1 increased significantly in the enhanced
running group but did not change in the cross-training
or control groups (Table 2). There was no significant
change in treadmill running 1202pe,k, running econ-
omy, V1202peak, or ~O2 deficit in the enhanced running,
cross-training and control groups, respectively.
There was no significant change in l)O2p~ak during
arm crank exercise in the enhanced running or control
groups. VO2wak during arm crank exercise increased
significantly in the cross-training group. There was no
significant change in the 1202 at 4 mmol-1-1 during
arm crank exercise in the enhanced running or control
groups. The 1202 at 4 mmol'1-1 during arm crank
exercise increased significantly in the cross-training
group.
Discussion
The main finding of this study is that muscularly dis-
similar non-specific training (swimming) significantly
improves running performance in well-trained recre-
ational runners. The improved running performance
with cross-training was significantly less, however, than
with increased running training. Our results are consis-
tent with previous reports of improved running perfor-
mance associated with muscularly similar non-specific
training, e.g. stair-climbing, cycling (Loy et al. 1993;
Mutton et al. 1993).
Despite significant changes in physiological re-
sponses during arm crank exercise consistent with an
upper extremity training response, there were only
371
Table 2 Mean (SD) pre and post-intervention results
(TDM 120=,, v4
mmol' 1-1 running velocity associated with blood lactate concen-
tration at 4 mlnol. 1-1, 1202 oxygen uptake, 20~
deficit
accumulated oxygen deficit,
v gO2pea k
velocity requiring peak [] 1202;
AC (/02peak,
AC
1202 4 mmol ' 1- ~.
Enhanced running Cross-training Control
Pre Post Pre Post Pre Post
Time trial 13.88 (1.50) 13.44 (1.4) am 14.77 (0.88) 14.56 (1.29) a 14.69 (1.59) 14.60 (1.52)
(rain)
TDM 1202~eak 58.0 (6.3) 57.9 (5.6) 53.6 (50.4) 50.4 (5.9) 53.5 (6.2) 55.1 (5.6)
(ml" min- 1 .kg- 1)
v4 mmol" 1-1 2.68 (0.26) 2.90 (0.28) a'b 2.49 (0.31) 2.56 (0.21) 2.56 (0.29) 2.65 (0.34)
(m" s -1)
1)'02 at 2.68 re's-1 45.4 (4.0) 44.3 (2.9) 45.5 (2.8) 42.2 (3.0) 44.3 (2.8) 44.5 (3.2)
(ml- min- 1. kg- 1)
202 deficit 44.8 (11.5) 49.8 (11.9) 42.5 (5.3) 40.8 (10.8) 44.7 (17.9) 48.9 (18.4)
(ml. kg- 1)
vl)Ozpeak 3.40 (0.45) 3.45 (0.41) 3.08 (0.33) 3.33 (0.29) 3.09 (0.37) 3.18 (0.37)
(m-s -1)
AC
[zO2peak 2.20 (0.62) 2.23 (0.66) 2.20 (0.55) 2.31 (0.48) a'b 2.22 (0.59) 2.05 (0.67)
1' rain- 1)
AC 1702 4 mmol.l- ~ 1.26 (0.32) 1.27 (0.26) 1.25 (0.19) 1.33
(0.32) a'b
1.28 (0.26) 1.29 (0.32)
(1. rain- 1)
a p < 0.05 pre vs post; up < 0.05 vs other intervention group
LU I05
Z
_0
I-- 100
Z
I.U
w
t-
z 95
e.e
e,i
kl.
o
N 90
TIME TRIAL
PERFORMANCE
I Run I
X
Train
|
El cont~o~ /
Fig. 4 Normalized time trial performance following intervention in
the enhanced running, cross-training and control groups. The time is
presented as a percentage (SE) of the pre-intervention score for that
group (I running, [] cross-training, [] control)
modest changes in the physiological responses during
running in the cross-training group that could explain
how cross-training enhanced running performance. As
expected the enhanced running group improved
v4 mmol' 1-1, supporting earlier suggestions we have
made regarding the mechanism of improvement in run-
ning performance in already trained runners (Daniels
et al. 1978). Non-active muscles are of established im-
portance relative to the uptake of lactate during exer-
cise (Stainsby and Brooks 1990). Lactate removal is
well established as a central factor in the control of
blood lactate accumulation during exercise (MacRae
et al. 1992; Mazzeo et al. 1986). However, there was
no change in v4 mm in the cross-training group that
would support the concept that a lactate "sink" in the
endurance trained upper extremity muscles contributed
to improved lactate metabolism during running. Hof-
fmann et al. (1993) have reported a dissociation be-
tween changes
in
gO2pea k and the ventilatory threshold
in response to either specific or non-specific training,
with the ventilatory threshold being less responsive to
non-specific training. Our results are also consistent
with findings of no change in the "local"RPE after
non-specific training despite increases
in
gO2pea k
(Lewis et al. 1980).
Medbo and Burgers (1990) have reported significant
increases in the Y, O2 deficit in response to high-inten-
sity training, although the precise nature of high-inten-
sity training did not seem to greatly influence the mag-
nitude of increase. Changes in YO2 deficit were, how-
ever, less in female subjects than in males. Given the
high percentage of female subjects in our sample and
the primarily moderate intensity of the training in this
study, large changes in the 5~O2 deficit were not to be
expected. Because our subjects were considerably less
than elite athletes and we wanted to be able to measure
steady-state 1/O2 over a reasonable range of running
velocities, we performed our treadmill runs with the
treadmill belt set at a slope of 5% compared to the
10-15% elevation recommended by Medbo et al.
(1988) and Olesen (1992). Olesen (1989, 1992) has sug-
gested that the measurement of the ~O2 deficit is very
dependent upon the testing and calculation procedures
used and has questioned the validity of the Y, O2 deficit
as a measure of anaerobic capacity. The longitudinal
nature of the present study should eliminate the magni-
tude of the treadmill slope as an issue since the slope
was equal both before and after intervention. We
372
(Foley et al. 1991) have demonstrated a comparatively
large day-to-day variability in the measurement of YO2
deficit which could contribute to the lack of systematic
change in this study. In any case, anaerobic capacity as
estimated by the FOg deficit did not seem to be a major
contributor to performance either before or after cha-
nges in training load.
An alternative explanation for the phenomena first
noted by single sport athletes training for triathlons is
that to compensate for a reduction in training volume
the athletes may have trained at a higher intensity.
Thus, higher-intensity single-sport training, rather than
a metabolic "crossover" effect, could be postulated at
the cause of the maintained single sport performance
capacity. Given the established importance of training
intensity to athletic performance (Foster 1983; Hickson
et al. 1985; Lehmann et al. 1992), there is a reasonable
rationale for such a suggestion. However, since the
increase in training load in the cross-training group
was achieved without a notable increase in training
intensity (Fig. 3), our data would not support this
suggestion. Even in the enhanced running group, in-
creases in training load seem mostly to be associated
with increases in training duration rather than train-
ing intensity. Certainly, the effects of independent
manipulation of the details of single-sport training
on subsequent performance is a high priority for future
studies.
In summary, the results of the present study indicate
that muscularly dissimilar exercise (swimming) contrib-
utes to improved running performance in individuals
training for running, although not to the same degree
as similar increases in muscularly specific training load.
Thus, muscularly dissimilar "cross-training" appears
to be useful relative to enhancing specific athletic
performance if the specific training load cannot be
increased to optimal levels. This effect is in addition to
the potential for athletes using non-specific training
during periods when ge.neral fitness is the primary
training goal.
Acknowledgement This study was supported by a research grant
from the Athlete Performance Division of the United States Olym-
pic Committee.
References
Banister EW, Good P, Holman G, Hamilton CL (1986) Modeling
the training response in athletes. In: Landers DM (ed), Sport and
elite performers. Human Kinetics Champaign, pp 7-23
Borg G, Hassmen P, Lagerstrom M (1987) Perceived exertion re-
lated to heart rate and blood lactate during arm and leg exercise.
Eur J Appl Physiol 65:679-685
Coen B, Schwarz L, Urhausen A, Kinderman W (1991) Control of
training in middle and long distance running by means of the
individual anaerobic threshold. Int J Sports Med 12:519-524
Daniels JT, Yarbrough RA, Foster C (1978) Changes in 1202Veak and
running performance with training. Eur J Appl Physiol 39:
249-254
Foley M J, McDonald KS, Green MA, Schrager M, Snyder AC,
Foster C (1991) Comparison of methods for estimation of anaer-
obic capacity (abstract) Med Sci. Sports Exerc 23:$34
Foster C (1983) lzO2 and training indices as determinants of com-
petitive running performance. J. Sports Sci 1:13-22
Foster C (1988) Translation of exercise test response to exercise
prescription. In: Oldridge NB, Foster C, Schmidt DH (ed) Clini-
cal exercise programs Mouvement Publications, Ithaca, N.Y., pp
169-178
Foster C, Green MA, Snyder AC, Thompson NN (1993) Physiolo-
gical responses during simulated competition. Med Sci Sports
Exerc 25:877-882
Gilman MB, Wells CL (1993) The use of heart rates to monitor
exercise intensity in relation to metabolic variables. Int J Sports
Med 14:339 344
Hickson RC, Foster C, Pollock ML, Galassi TM, Rich S (1985)
Reduced training intensityies and loss of aerobic power, endur-
ance and cardiac growth. J. Appl Physiol 58:492 499
Hoffmann JJ, Loy SF, Shapiro BL, Holland GJ, Vincent WJ, Shaw
S, Thompson DL (1993) Specificity effects of run versus cycle
training on ventilatory threshold. Eur J Appl Physiol 67:43 47
Lehmann M, Baumgartl P, Wiesneck C (1992) Training-overtrain-
ing: Influence of a defined increase in training volume vs training
intensity on performance, catecholamines and some metabolic
parameters in experienced middle and long distance runners.
Eur. J Appl Physiol 64:169-177
Lewis S, Thompson P, Areskog N-H, Vodak P, Marconyak M,
DeBusk R, Mellen S, Haskell W (1980) Transfer effects of endur-
ance training to exercise with untrained limbs. Eur. J Appl
Physiol 44:25-34
Loy SF, Holland GJ, Mutton DL, Snow J, Vincent WJ, Hoffmann
JJ, Shaw S (1993) Effects of stairclimbing vs run training on
treadmill and track running performance. Med Sci Sports Exerc
25:1275 1278
MacRae HSH, Dennis SC, Bosch AN, Noakes TD (1992) Effects of
training on lactate production and removal during progressive
exercise in humans. J Appl Physiol 72:1649 1656
Mazzeo RS, Brooks GA, Schoeller DA, Budinger TF (1986) Dis-
posal of blood lactate in humans during rest and exercise. J. Appl
Physiol 60:232-241
McArdle WD, Magel JR, Delio DJ, Tonier M, Chase JM (1978)
Specificity of run training on gO2max and heart rate changes
during running and swimming. Med Sci Sports Exerc 10:16-19
McLellan TM, Skinner JS (1981) The use of the aerobic threshold as
a basis for training. Can J Appl Sport Sci 6:197 201
Medbo JI, Burgers S (1990) Effect of training on the anaerobic
capacity. Med Sci Sports Exerc 22:501-507
Medbo JI, Mohn A-C, Tabata I, Bahr R, Vaage O, Sejersted OM
(1988) Anaerobic capacity determined by maximal accumulated
02 deficit. J. Appl Physiol 64:50-60
Mutton DL, Loy SF, Perry DM, Holland GJ, Vincent WJ, Heng
M (1993) Effect of run vs combined cycle/run training on aerobic
capacity and running performance. Med Sci Sports Exerc 25:
1393-1397
Olesen HL (1989) Anaerobic capacity; uphill running-a clue to its
accurate determination (abstract) Acta Physiol Scand 140:33
Olesen HL (1992) Accumulated oxygen deficit increases with inclina-
tion of uphill running. J Appl Physiol 73:1130 1134
Pollock ML, Dimmick J, Miller HS, Kendrick Z, Linnerud AC
(1975) Effects of mode of training on cardiovascular function and
body composition of middle aged men. Med Sci Sports 7:
139-145
Sjodin B, Jacobs I, Svedenhag J (1982) Changes in onset of blood
lactate accumulation (OBLA) and muscle enzymes after training
at OBLA. Eur J Appl Physiol 49:45 57
Stainby WN, Brooks GA (1990) Control of lactic acid metabolism in
contracting muscles and during exercise. In: Pandolf KB, Hol-
loszy JO (eds) Exercise and sports sciences reviews, vol 18.
Wiliams and Wilkins, Baltimore, pp 29-64
