Separation of the inspiratory and expiratory reflex effects of alternate-breath oscillation of PaCO2 during hypoxia

Respiration Physiology (1977) 29, 379-390 @ Elsevier/North-Holland

SEPARATION EFFECTS

Biomedical Press

OF THE INSPIRATORY

AND EXPIRATORY

OF ALTERNATE-BREATH OSCILLATION HYPOXIA’

REFLEX

OF P&o2 DURING

SUSAN A. WARD2 and D. J. C. CUNNINGHAM iJniver.sity Laborat0r.v

qf Fhys~ology~ Oxford, United Kingdom

Abstract. Four healthy young men and women, selected for the responsiveness and steadiness of their breathing, were studied in rest and mild exercise (58 runs) while receiving alternate inspirates of low and high P,,, (0 and 8.6 kPa). PA,,, oscillated between ca. 6 and 7.5 kPa (45-55 torr); PA~~was held steady at more than one level between 6 and 9.6 kPa (45-72 torr). Using cross-correlation analysis, the phase relations were determined between the alternating PA~~, and the following reflex outputs: mean inspiratory and expiratory flows (tr and GE) and the reciprocal of the duration of expiration (~/TE), the two expiratory variables being lumped together for purposes’of expression, but not of calculation. TI, being relatively unaffected by alte~ating Pp~co~,was not re-studied (see companion paper). The common patterns of signi~~nt reflex alternation were: Gr alone, usually in phase (with PA~~J, 24 % ; GEalone, usually in phase, I7 76; both inspiratory and expiratory variables, in phase with CO, and each other, 1504; both inspiratory and expiratory variables, the expiratory being out of phase with both CO, and with the inspiratory, 23 %. Some runs showed a mixture of phase relations. In 71 %, end-expiratory lung volume (VL, E’ formerly called FRC) alternated significantly. It is concluded that expiratory events can be influenced by peripheral chemoreceptors independently of inspiration. Arterial chemoreceptors Breathing pattern

Carbon dioxide oscillations Control of expiration

In the accompanying paper we examined the extent to which the reflex responses to an experimental alternate-breath oscillation of alveolar Pcoz (PkoI) were dependent upon the presence of hypoxia (Ward and Cunningham, 1977). We noted that the patterning of the responses with respect to volumes, flows and durations was variable, and it is with this aspect that the present paper is concerned. Accepted for publication IS November 1976.

* This research was carried out with a grant from the Medical Research Council. 2 States of Guernsey Research Scholar; present address: The Physiological Laboratory, Liverpool, England. 379

University of

380

S. A.

WARD

AND

D. J. C. CUNNINGHAM

The form of an expiration (flow and duration) has been regarded by some authors as being more or less determined by the form of the preceding inspiration (Clark and Euler, 1972; Bradley et al., 1975). There is, however, evidence that besides the link with inspiration, expiration is considerably influenced by other factors including respiratory drive (Cunningham and Gardner, 1972; Cunningham and Gardner, 19’77: Gardner, 1975; Kay et al., 1975; Gautier et al., 1973; Bartoli et al., 1973; Bartoh et al., 1974). In some of the patterns of response to be described expiratory and inspiratory variables are more or less dissociated from each other, thus lending support to the view that expiration is to some extent independent of inspiration. Methods APPARATUS:

EXPERIMENTAL

PROCEDURE

A full description is to be found in Ward and Cunningham DATA

(1977)

ANALYSIS

Inspiratory and expiratory responses to the alternating CO, signal were characterised in terms of the corresponding mean flows, $1 and GE, and the times for which they occurred, Tr and TE expressed as their reciprocals, l/T1 and ~/TE, for reasons that will become apparent (see below). TI showed little tendency to alternate (Ward and Cunningham, 1977) and is not reported on here. Bivariate cross-correlation analysis (Sokal and Rohlf, 1969, p. 498) was used to examine the relationships between selected pairs of variables in each breath for whole runs and for subsections of runs: (1) a significant correlation between PA,,, and an output variable (Sokal and Rohlf, 1969, p. 516, P < 0.05) was taken to indicate the occurrence of a reflex alternation in that variable; (2) the relationships between inspiration and expiration were determined through inspection of the degree of correlation between inspiratory and expiratory output variables. The analysis was confined to runs conducted against a background of steady hypoxia (PACTc 9.7 kPa or 73 torr), both at rest and in mild exercise, as most of the reflex alternation occurred in this range of PAL*(Ward and Cunningham, 1977). RdtS PRESENCE

OF ALTERNATION

The data analysed here are those of the 58 hypoxic runs only. The incidences and absolute amplitudes of alternation that occurred in the output variables are to be found

BREATHING

IN MAN

381

in the companion paper (Ward and Cunningham, 1977); though the method of analysis used in the present paper was different, the incidences of significant alternation by the two methods were nearly identical. Change in end-expiratory volume* (VL, E’), obtained as the breath-by-breath difference between inspiratory and expiratory tidal volumes, alternated very often (column 8 of table 1). As reported earlier, the apparatus and the methods used for analysing its outputs recorded no appreciable alternation when the place usually occupied by the human subjects was taken by a reciprocating pump, the sinusoidal flow output of which covered a corresponding range of volume and period. The results reported here are thus not due to apparatus artefacts. The control determinations in which the alternation was between identical inspiratory gas mixtures are described in the companion paper.

PHASE RELATIONS

In table 1 cross-correlation coefficients that reached statistical significance (P < 0.05) are summarized subject by subject, with rest and exercise shown separately. The underlying patterns are reconstructed diagrammatically in figs. l-5; the magnitudes of the changes are much exaggerated in order that the different patterns should stand out. On the left of each figure is a hypothetical spirogram covering four respiratory cycles, i.e. two full cycles of alternation, constructed from the corresponding pattern of cross-correlation coefficients shown in the table. Inspiration is upwards (line lines) and expiration downwards (heavy lines). The high- and low-CO, breaths are drawn with full and broken lines respectively, and the slopes are notionally proportional to the mean flow rates. On the right of each figure the pattern is plotted diagrammatically as tidal volume, VT, us TI and TE (Clark and Euler, 1972; Cunningham and Gardner, 1972), the TE axis being reversed (Kay et al., 1975). The filled and open symbols correspond to the end-inspiratory and end-expiratory symbols in the spirogram (denoting high- and low-CO, breaths, respectively). In order to indicate directional changes in mean flows, the plotted points are joined to the origin, the lines following the conventions used in the spirograms. Columns 1,2 and 3 of table 1 show the occurrence or absence of significant cross correlation of the alternating ~~~~~ with inspiratory and expiratory variables. Absence of alternation (column 3) is illustrated in fig. 1, below, and the pattern of alternation to be expected from current theory is discussed later. Columns 4, 5, 6 and 7 of table 1 deal with phase relations. The headings to the columns are defined explicitly in the first four lines in terms of the signs of the cross-

* Commonly called Physiology.

FRC, a term now discouraged by the European Society for Clinical Respiratory

382

S. A. WARD

The occurrence

of the different

AND D. J. C. CUNNINGHAM

TABLE I of reflex alternation

patterns

Totals

of runs

of inspiratory

Insp. or expir. alternation

alternation inspiratory

no alternation

Gl

(3) -

(4)

(5)

+ 0 0 0

-

expiratory (2) ___-

(1)

I

I nsp. and expi tlternation

GE and:01 r

I ‘Tt

and expiratory

variables Alternation of

i n phase

out

piW

6)

17)

0

i

~ or +

+

+ _

[+I

c-

V1.t’

Signs of correlation coefficients CO, I’St-1 CO, I’,, c’l CO2 1.5 I Tr: 51 1‘5Tt, and,or

1 Tt.

or + 0

?I: t‘.>1 Tt

+ N

Subject

1 I

4:1

7

0 3[i]

71. 0

6

‘II’ 0 0 I

3

I 0

21’1 0 0 I[13 ?[{I

6

7

4&[21

511

21

VI ~X41

X[. 13?,/

41

409 R

Y

6

6

0

E

x

7

8

0

421 R

Y

4

8

0

I 0 1

4

5

0

1

2

I

0

7

I

0

2

4

0

1 3 8

3

E

6

425 R

x

/

1 I 5

E

6

4

392 R

7

6

0 4

E

8

5

2

All R

30

17

I8

E

28

20

I5

5

8

3

58

37

33

13

14

IO

All

In columns

6 and 7, figures

and expiratory nation

of CO,

in square

output

variables.

of end-expiratory

3 4 4 4 6

0

20

22191

cross correlation

7

brackets

variables

show runs in which the phase relations

were confirmed

These ligures

are included

lung volume

is also shown

by direct cross corrugation

in the unbracketed (column

indicated

by the

between inspiratory

vnalues. The occurrence

of alter-

8).

no response

7

Ts Fig.

1. Reconstruction

of pattern

of response.

L&I

: schematic

0 -

spirogram

Tt inspiration

upwards.

l;

in-

spiration; ‘\ expiration. Full lines, high-CO, breaths; broken lines. low-CO, breaths. Right: conventional VT, TI, TE relation with TE plotted leftward from origin. The thin dotted curve is the mean steadystate relation

between

VT and TE (Cunningham

n is number

of responses

and Gardner,

of the type shown.

3972). Lines and symbols

as in spirogram.

This figure shows no response.

BREATHING

383

IN MAN

correlation coefficients; 0 denotes a lack of significant correlation while blank spaces indicate that testing the correlation was considered unnecessary.

INSPIRATORY

OR EXPIRATORY

ALTERNATION

ALONE

Column 4 of table 1 and fig. 2 show the occurrence of correlation between P~co~ and GI unaccompanied by significant correlation with either of the expiratory variables : nearly all these correlation coefficients were positive. Figure 2 shows that alternation of end-expiratory volume (VL, E’) is inevitable with this pattern. Inspiratory alternation by itself was the commonest pattern observed in the series. Column 5 of table 1 shows the occurrence of correlation between P~co* and expiratory variables only; here the correlations were positive more often than not (table 1, subject 421). When GEand ~/TE alternated together they were always positively correlated with each other, that is to say, they were never systematically out of phase (see Discussion). Figure 3 shows, above, alternation of both CE and ~/TE, the ‘stimulated’ breath occurring concurrently with the high PA,,~. Below, only GE

inspiratory alternation only

Fig. 2. lnspiratory constant,

alternation

and both expiratory

only. Conventions

as in fig

slope, GE and TE are constant. monest

pattern

observed

I. The inspiratory

slope, ;I, alternates,

YL, E’ has to alternate.

Tr is

This was the com-

experimentally.

expiratory alternation only

T-‘-T E

I

Fig. 3. Two out of three possible types of expiratory alternation. Conventions as in fig. I. Above, both irk and TE alternate in phase. Below, only 6~ alternates, and forces a correspon$ing alternation of VL, E’.

384

S. A. WARD AND D. J. C. CUNNINGHAM

alternates, ~/TEbeing constant. In this case, too, the ‘stimulated’ breath is shown as concurrent with high PACES. Alternation of VL, E’ might occur with the upper pattern, but was inevitable with the lower. The third possibility, alternation of ~/TE at constant GE, was seen twice; it is illustrated in combination with inspiratory alternation in fig. 5, bottom (note that the expiratory slope is constant).

INSPIRATION AND EXPIRATION BOTH ALTERNATING

Columns 6 and 7 of table 1 record 22 runs in which both inspiratory ($I) and expiratory variables (GE and/or ~/TE) were significantly correlated with P~co~ within the same runs. More often than not, the correlation coefficients were of opposite sign, indicating that the two halves of the respiratory cycle alternated out of phase. Inspiratory alternation was correlated with expiratory in nine of these runs (figures in brackets: 4 direct, 5 inverse), indicating that for a substantial part of each run the alternations occurred simultaneously rather than in consecutive sections of the runs. When the correlations were inverse, increase of inspiratory activity (itself associated with the high-PA,oL breaths) was followed by decrease of expiratory activity within the same respiratory cycle and vice versa on the low-PAN,,, breaths. Figure 4 shows two of the three possible combinations of in-phase alternation. In one both GEand ~!TE alternated and in the other alternation was confined to GE; the third possibility ( I/TE alone) was not seen. By changing the relative magnitudes of the alternations of the different variables a stable VL, E’ could have been obtained with either pattern, but in fact this rarely occurred. Figure 5 shows two of the three possible combinations of out-of-phase alternation, the upper pattern being by far the commoner. The third combination of expiratory

inspiratory-expiratory

alternation in phase

Fig. 4. Two types of in-phase (i.e. synergistic) inspiratory-expiratory alternation. Above, both irk and TE alternated in phase with each other and with the alternating $1. Below, CE alternated in phase with ;I, TE being steady. Though neither pattern necessitated it, alternation of VL,E’ nearly always occurred.

BREATHING

385

IN MAN

inspiratory -expiratory alternation out of phase

Fig. 5. Two types of out-of-phase one of the commonest could

equally

patterns

have been drawn

(i.e. antagonistic)

inspiratory-expiratory

seen. The same combination as an expiratory inspection

pattern

alternation,

of cross-correlograms like the upper

the upper one being on the expiratory

side

one in fig. 4, as can be seen by

of the VT, TE plots.

was not seen in out-of-phase runs, but it occurred in other contexts (figs. 3 and 4, lower spirograms). The common pattern could conceivably be drawn to give a steady VL, E’, but in fact this was seldom if ever seen. In thirteen runs grouped in columns 6 and 7 of table 1 the cross-correlations between inspiratory and expiratory variables over the whole runs were not significant, although both inspiratory and expiratory activity were correlated with the alternating PAC02.Presumably in these ‘anomalous’ runs, although alternation occurred in both parts of the respiratory cycle it was often confined to only one or the other at any one time; such overlap as occurred was insufficient to emerge in the form of a significant inspiratory-expiratory correlation. The extent of the overlap was examined by dividing the runs into subsections arbitrarily chosen to be 20 breaths long, each of which was subjected to cross-correlation analysis, as described earlier for the full runs. As suspected, while these short subsections usually conformed significantly to one pattern of alternation or another, the patterns were not homogeneous throughout the whole runs. Three of the more inhomogeneous runs are reconstructed in fig. 6. In the top run, judging from the CO,-inspiratory and CO,-expiratory correlations, subsections 3, 5 and 6 show predominantly in-phase inspiratory alternation, subsection 2 shows out-of-phase alternation of 1/TE and subsections 1 and 4 show both. The inspiratory+xpiratory analysis shows three further correlations, those in periods 3 and 5 being qualitatively unexpected in that they are in-phase, unlike all the others on this subject. The interpretation we have adopted is of a tendency towards out-ofphase inspiratory*xpiratory alternation for the first 40 breaths, followed by ‘indecision’ as between in-phase and out-of-phase expiratory alternation for the remaining 80 breaths of the run. This run as a whole is scored as two separate halves in columns 6 and 7 of table 1. alternation

386

S. A. WARD

AND D. J. C. CUNNINGHAM

Total

+

0

+

+

+

+

0

+

1

2

3

4

.

Fig. 6. Three anomalous runs, analysed in subsections. Abbreviated patterns ofcross-correlation for each period as +, - or 0. For interpretation see text.

are shown

Figure 6B illustrates a similar analysis which led to the run as a whole being classed as in-phase inspiratory~xpiratory alternation (table 1, column 6); the expiratory component persisted throughout while the inspiratory component faded and reappeared. In the second subsection of fig. 6C the direct inspiratoryyexpiratory correlation was significant even though neither of its components was correlated with the alternating CO, signal. The remainder of this run was not unlike that in fig. 6B. Run A of fig. 6 and one other were the most difficult to classify of the twelve ‘anomalous’ ones in columns 6 and 7 of table I ; most were no more difficult than the lower ones in the figure.

DIFFERENCES BETWEEN SUBJECTS

The patterns of alternation varied considerably from subject to subject but were moderately consistent for any given subject, whether resting or exercising. Thus subject 409’s patterns showed predominantly out-of-phase inspiratory-expiratory alternation, while those of subject 421 were primarily expiratory, often with some in-phase inspiratory alternation superimposed. Subject 392’s patterns tended towards inspiratory-only alternation, though several runs showed none at all. Lack of any response was even more marked in subject 425, especially during rest.

BREATHING

IN MAN

387

Discussion

The four young people studied here have it in common that they all showed a reflex alternation that was dependent on the presence of hypoxia (Ward and Cunningham, 1977). Apart from that, their responses could scarcely have been more disparate. In what follows, attention has been concentrated on the expiratory response, particularly on its alternation simultaneously with inspiration. The runs that appear as bracketed figures in columns 6 and 7 of table 1 showed without doubt that such simultaneous alternation occurs and that the two halves of single respiratory cycles may be either out of, or in phase with one another. This conclusion was surprising, and led us to the more detailed analysis of the 13 anomalous runs by the method outlined in fig. 6. Viewed in this way, it emerged that even the most perplexing pattern seen so far (the top spirogram of fig. 6) could be regarded as being due to a slow shift of emphasis between the inspiratory and expiratory components of a mainly out-ofphase inspiratory-expiratory pattern. In more recent work (Drysdale and Ward, 1976) the out-of-phase inspiratory-expiratory responses predominated in two more individuals. Neither the present nor the more recent work allows any statement on the frequency with which such responses might be expected in the population of normal individuals, but there can be no doubt as to their reality.

THE INDEPENDENCE

OF EXPIRATION

According to a scheme for the steady-state control of the respiratory cycle (Bradley et al., 1975), inspiratory events in the main determine the characteristics of the immediately following expiratory half-cycles. Bradley et al. (1974), reporting on the anomalous behaviour of TI during transient responses to step changes of Pcoz in the cat, mention expiration briefly and then only to say that the ratio TE/TIis preserved during the transient (but see Gardner, 1974, and Widdicombe and Winning, 1974). The present results are expressed in terms of inspiratory and expiratory times and flows rather than volumes and so the details shown in the spirograms are not immediately comparable with those of Bradley et al. (1974; 1975). A comparison may, however, be made with the help of the accompanying VT, TI, TE diagrams. In general a model based on their views would be restricted in several respects : since expiratory variables should not be influenced except through inspiratory events (or by lung inflation during expiration, Knox, 1973, which is scarcely relevant here), and since TI should be unable to follow quick changes, the patterns of alternation would comprise only those shown in fig. 2 ($1 and VTI) and fig. 4, bottom, in which CE and VTE would alternate in phase with VTI. The following would be impossible: (1) any expiratory alternation unaccompanied by inspiratory alternation (fig. 3); (2) any alternation of TE at all (figs. 3 and 5); (3) any out-of-phase alternation (fig. 5). The results described above are often contrary to the hypothesis and lead to the conclusions that expiratory events are not wholly dependent upon inspiration; the

388

S. A.

WARD

AND D. J. C. CUNNINGHAM

alternating chemical signal therefore has access to them by some pathway not concerned with inspiration. Furthermore, although end-expiratory volume is generally considered to be relatively stable from breath to breath (but see Hlastala et al., 1974) under the highly artificial conditions of these experiments the termination of expiration and the initiation of the next inspiration are not related to the attainment of a precise end-expiratory volume. In other words. if the transition from expiration to inspiration is related to a deflation reflex, the reflex is either imprecise or is easily overridden by other factors. Indeed, the existence of an effective reflex of this kind would render impossible some of the patterns observed. The work of Cunningham and Gardner (1972), Gardner ( 1974). and Cunningham and Gardner (1977) on man, of Bartlett, Remmers and Gautier (1972), Gautier et al. (1973), and Widdicombe and Winning (1974) on anaesthetized and unanaesthetized cats, and of Bartoli et al. (1973; 1974) on dogs on cardiopqmonary bypass provides further evidence that an expiration is at least partly independent of the preceding inspiration. The present results also have points in common with the results and formulations of Koepchen (1974), who distinguished symmetrical and asymmetrical reflex patterns of inspiration and expiration. Gautier et al. (1973) and Bartlett, Remmers and Gautier (1973) described the control of expiration in terms of a tidal volume of gas to be emptied through an airway with upper (laryngeal) and lower resistances; the driving forces consist of passive recoil of the lungs and chest wall opposed or supplemented by more or less braking by inspiratory muscles and by less or more expiratory muscle activity, respectively. There appears to be intrinsic integration and control of these agents (e.g. Bartlett and Remmers, 1975 ; Bartoli et al., 1973), and chemical as well as mechanical influences from outside the system can influence them (Nadel and Widdicombe, 1962; Sterling, 1969; Bartoli et al., 1974). At present we have no means of knowing what combination of these effector systems is accessible to the alternating CO, signal. In the companion paper (Ward and Cunningham, 1977) we have given reasons for thinking that the arterial chemoreceptors are primarily involved in the detection of the alternating CO, signal, and recent measurements of reflex latency (Drysdale and Ward, 1976) are consistent with such a view. It is, however, conceivable that the full response develops as a result of some sort of interaction between the arterial chemoreceptors and thevagal inputs from lung mechanoreceptors sensitive to CO, (Mustafa and Furves, 1972; Bradley et al., 1976), such as has been demonstrated between chemoreceptors and lung irritant receptors (Glogowska et al., 1972). The variation in the pattern between runs may be attributed tentatively to the variations in the effectiveness of carotid chemoreceptor volleys according to their timing within the respiratory cycle (Black and Torrance, 1967; Band et al., 1970). In particular the detailed timing studies of Eldridge (1972a,b) show that the cat ehemoreflex pathway includes expiratory elements that appear to be independent of inspiration. Eldridge’s results, together with the small periodic variations that probably occur in awake individ~ls in the circulatory transit time between the lungs and the chemoreceptors (see Cunningham, 1975) could probably account between

BREATHING

389

IN MAN

them for all the phenomena described here, including the composite periodic effects analysed in fig. 6.

Acknowledgements

We thank Mr E. Aldsworth, Miss L. M. Castell, Messrs D. V. GoffandT. J. Meadows for technical and clerical assistance; SAW is grateful to Professor D. Whitteridge, F.R.S., for allowing her laboratory facilities.

References Band,

D. M., I. R. Cameron

carotid Bartlett,

artery

D., J. E. Remmers

and H. Gamier

physiologist

frequency. Bartlett,

and S. J. G. Semple (1970). The effect on respiration

pH and Pco2 in the cat. J. Physiol. (London)

of abrupt

changes

in

211: 479494.

(1972). The role of the larynx in the regulation

of respiratory

I5 : 79.

D., Jr., J. E. Remmers

and

H. Gautier

(1973).

Laryngeal

regulation

of respiratory

airflow.

Respir. Physiol. 18 : 194204. Bartlett,

D. and J. E. Remmers

(London) Bartoli,

A., E. Bystrzycka,

pulmonary

vagal

A., B. A. Cross,

of carbon (London] Black,

(London)

respiratory

rhythm

A. Guz, S. K. Jain,

in the absence

M. I. M. Noble

and duration.

J. PhJlsiol.

(1973). Studies ofthe

of breathing

movements.

and D. W. Trenchard

(1974). The effect

a vagal reflex studied in the dog. J. Physiol.

and alveoli on ventilation; (1967). Chemoreceptor

cycle. J. Physiol.

effects in the respiratory

189: 59AlP.

G. W., C. von Euler, 1. Marttila determining

and B. Roos (1974). Transient

G. W., C. von Euler, I. Marttila

and B. Roos (1975). A model of the central

in the cat. Biol. Cybernet.

effects of CO,

and reflex inhibition

19: 105-I 16.

G. W., M. I. M. Noble and D. Trenchard

F. J. and C. von Euler

and steady-state

Acfa physiol. scund. 92: 341-350.

rate and depth of breathing.

(1976). The direct effect of pulmonary

discharge produced by changing the CO, concentration , bypass. J. Physiol. (London) 256: 25S26P. Clark,

airflow

240: 91-109.

of inspiration Bradley,

of expiratory

230: 449465.

dioxide in the airways

on mechanisms Bradley,

of central

A. M. S. and R. W. Torrance

Bradley,

Reflex control

A. Guz, S. K. Jain, M. I. M. Noble and D. W. Trenchard

control

J. Physiol. (London) Bartoli,

(1975).

247: 22P.

(I 972). On the regulation

of the airways

stretch receptor

in dogs on cardio-pulmonary

of depth and rate of breathing.

J. Physiol. (London)

222: 261-295. Cunningham, expiratory Cunningham,

D. J. C. and W. N. Gardner times during

CO,

inhalation

D. J. C. (1975). A model illustrating

Naiure (London) Cunningham,

(1972). The relation

steady-state

tidal volume and inspiratory

the importance

and

227: 50-51P.

of timing in the regulation

of breathing.

253 : 440442.

D. J. C. and W. N. Gardner

during steady-state (submitted).

between

in man. J. Physiol. (London)

CO, inhalation

(1977). A quantitative

description

in man, with special emphasis

of the pattern

on expiration.

of breathing

J. Physiol. (London)

Drysdale, D. B. and S. A. Ward (1976). The reflex latency of inspiratory and expiratory responses alternate-breath oscillation of alveolar Pco 1 in man. J. Physiol. (London) 260 : 4142P.

to an

Eldridge, F. L. (1972a). The importance of timing on the respiratory nerve stimulation. J. Physiol. (London) 222: 297-318.

sinus

effects of intermittent

carotid

390

S. A. WARD

AND D. J. C. CUNNINGHAM

Eldridge, F. L. (1972b). The importance of timing on the respiratory effects of intermittent chemoreceptor stimulation. .I. Ph.vsiol. (London) 222: 319-333. Gardner,

W. N. (1974). The pattern

I London) Gardner.

H.. J. E. Remmers

Physiol. Glogowska.

step changes

of alveolar

P,o> in man. J. fhysiol.

Affecting

and D. Bartlett,

Respiration

in Man. Oxford

Jr. (1973). Control

University.

of the duration

D. Phil. thesis

of expiration,

Respir.

18: 20522 I M.. P. S. Richardson,

nerves, peripheral

Kay, J. D. S.. E. Strange of breathing

in resting

Petersen

in man during

pathways

(1974). Periodic

changes

in functional

man. Acra physiol. stand. 90: Suppl.

and H. Vejby-Christensen steady-state

C. K. (1973). Characteristics

(1972). The role of the vagus

in the genesis of augmented

breaths

16: 179-196.

and C. J. Lenfant

variables

and A. J. Winning

and other afferent

Respir. Physiol.

M. P., B. Wranne

other respiratory

J. G. Widdicombe

chemoreceptors

in cats and rabbits. Hlastala,

following

body

242: 75P.

W. N. (1975). Some Factors

Gamier,

Knox,

of breathing

carotid

exercise.

of inflation

residual

capacity

(1975). Mean and breath-by-breath

J. Physiol.

and deflation

/London) reflexes

and

396. 74~. pattern

251 : 657-669.

during

expiration

in the cat. J.

central

respiratory

drive and in-

Neuroph.tm’ol. 36 : 284295. Koepchen,

H. P. (1974). Concept

hibition.

In: Central

Hippokrates, Mustafa,

Sterling,

receptors

in the lungs of rabbits.

on tracheal

volume

and H. P. Koepchen.

by inhalation

of 7 “‘4 CO,.

S. A. and D. J. C. Cunningham

San Francisco,

of decreased

in blood

to airflow.

on the pattern

of breathing

from slowly-adapting

gas tensions

J. Physiol.

and carotid

/London)

sinus

163: 13-33.

Calif., Freeman.

specific airway

conductance

in man during

hyper-

Clin. Sci. 37: 539-548.

(1977). The relation

between

alternation of breathing in man. Respir. Physiol. 29: Widdicombe, J. G. and Winning, A. (1974). Effects of hypoxia, perature

Stuttgart,

16: 197-212.

(1962). Effect of changes

G. M. (1969). The mechanism caused

Respir. Physiol.

and total lung resistance

R. R. and F. J. Rohlf (1969). Biometry.

capnia Ward.

edited by W. Umbach

pp. 122-133.

J. A. and J. G. Widdicombe

pressure Sokal,

types of reflexogenic

and Regulation,

M. E. K. Y. and M. J. Purves (1972). The effect of CO, upon discharge

stretch Nadel,

of two general

Rhythmic

hypoxia

hypercapnia

in cats. Respir. Ph~siol. 21 : 203-222.

and CO,-induced and changes

reflex

in body tem-