1. The locomotor rhythm of Carcinus maenas was shown to be relatively similar at 10°, 15°, 20° and 25° C., but it was shown to be modified by changes in temperature.

  2. A fall of about 5 − 10° C. just before the onset of activity usually advanced and enhanced the initial peak of activity, whilst a rise of the same magnitude usually delayed and partly suppressed the initial peak, each without affecting the basic rhythm.

  3. A fall in temperature of about 10° C. or more between activity peaks usually initiated transient outbursts of activity, again without affecting the normal rhythm.

  4. Chilling to 4° C. for more than 6 hr., irrespective of the time of onset of chilling, resulted in a rephasing of the rhythm with the initial outburst occurring immediately on return to room temperature.

  5. In Carcinus from non-tidal docks chilling resulted in a change from the normal approximately nocturnal rhythm to an overt tidal rhythm, but in Carcinus from the virtually tideless Mediterranean the character of the normal nocturnal rhythm was unaffected by chilling.

The so-called temperature independence of many biological rhythms is of functional importance in maintaining the correct phasing of physiological and behavioural activities despite changes in environmental temperatures. Nevertheless, the rhythms can often be partly modified by changes of temperature and it is necessary to consider the extent of this temperature-dependence when speculating about the possible nature of biological clocks. Perhaps the most striking effects are produced by short periods of exposure to high or low temperatures, though the precise effects of such treatment may vary in different organisms. In some a rhythm may be delayed for the length of period of chilling (Brown & Webb, 1948; Harker, 1960; Sweeney & Hastings, 1960), but in others the delay varies according to the precise time of onset of the period of chilling (Stephens, 1957; Biinning, 1960; Sweeney & Hastings, 1960). The present paper describes a quite different effect of chilling on the locomotor rhythms of Carcinus maenas (L.). It also compares the effects of chilling on the tidal rhythms of Carcinus taken from the shore with the strikingly different effects of chilling on the predominantly nocturnal rhythms of Carcinus from non-tidal docks and of Carcinus from the relatively tideless Mediterranean Sea.

Experiments were carried out on Carcinus maenas (L.) collected from between tidemarks and from non-tidal docks in South Wales, and also on C. mediterraneas Czerniavsky at the Stazione Zoologica, Naples. Locomotor activity was recorded by aktographs described previously (Naylor, 1958) which were usually kept in constant-temperature rooms lit continuously by a dim red light. ‘Shore’ crabs were kept at 15°C. whilst locomotor activity was recorded, but ‘dock’ crabs were kept at higher temperatures than this (usually 22°C.) since the dock from which they were collected was warmed by a power station effluent. ‘Mediterranean’ crabs were usually kept at 24° C.

‘Shore’ crabs

Fig. 1 shows the average hourly activity of four groups of three crabs kept for three days at 10°, 15°, 20°, and 25° C., respectively. Clearly the phase and frequency of the rhythm were generally similar at each temperature, though there were some differences, particularly in the relative heights of the initial peaks in each experiment. Thus at 10°C. the first activity peak was particularly high and was in advance of the time of high tide, whilst at 25° C., the initial peak was small and occurred after the time of high tide. These differences were presumably related to the changes in temperature experienced by the crabs on being transferred to the constant-temperature rooms from the shore where sea temperatures would have been about 15°C. In other experiments, too, a rise in temperature resulted in a delayed and partially suppressed initial peak, whilst a fall in temperature was associated with the advance and enhancement of the initial outburst of activity. These and other transient effects of temperature change are more clearly illustrated in Fig. 2, in which it can be seen that a fall of about 10°C. at the time of onset of activity clearly enhanced the next peak (Fig. 2a, b), whilst a rise in temperature of the same order virtually suppressed the next peak (Fig. 2b). Moreover, a fall in temperature of 10°C. or more, but not a fall of 5°C., resulted in a transient outburst of activity if it occurred at a time when the crab would not normally have been active (Fig. 2 c, d). In addition there were considerable bursts of activity for up to 24 hr. after the fall in temperature, which might suggest that the lowered temperature initiated a transient rhythm and not merely a transient outburst of activity. The subsequent tidal rhythm was much less apparent than normal, often with every alternate tidal peak being more pronounced. However, a fall in temperature of even as much as 14°C. appeared to have no lasting effect on the phase of the rhythm, after the initial transitory effects.

Further experiments were carried out in which crabs were chilled for several hours. Fig. 3 a, b shows the effects of 24 hr. exposures to temperatures of 10° and 5·5°C. on two groups of four crabs collected when sea temperatures were about 12° −13°C. and kept at 15°C. after chilling. Exposure to 10° C. clearly had no effect upon the phase of the rhythm but chilling to 5·5°C. resulted in a rephasing of the tidal rhythm, starting with an initial outburst of activity immediately on return to the warm room. Similar results, obtained when crabs were chilled at 4°C. for 8 hr. and 11 hr., respectively, together with the results of control experiments, are illustrated in Figs. 3,c, d, e and f. In these experiments and in others of a similar nature (see Fig. 6 below) the character of the new rhythm was similar to that of the controls, but each started with a peak immediately after chilling, whether chilling began at the time of high tide, when crabs would normally have been active (Fig. 3,d), or between the times of two successive high tides, when crabs were normally inactive (Fig. 3 f). The rephased rhythm persisted for up to 5 days, the duration of the longest experiment, and the minimum period of chilling before the rhythm was reset seemed to be between 6 and 8 hr. When crabs were chilled for 2, 4 and 6 hr. the normal phasing of the tidal rhythm was unaffected.

Crabs did not survive periods of prolonged exposure to high temperatures, but warming to 30°C. for 2, 4 and 7 hr. seemed to advance the phase of the rhythm by about 2 hr. in each case (Fig. 4). Thus warming appeared to shorten the duration of one cycle of the rhythm by a maximum of about 2 hr.

‘Dock’ crabs

Chilling of Carcinus maenas collected from non-tidal docks, which show predominantly nocturnal outbursts of activity (Naylor, 1960), produced particularly striking results. Exposure to 4°C. for more than about 6 hr. usually resulted in the appearance of a distinct rhythm of tidal periodicity (Fig. 5,b, c and d), in contrast to the normal rhythm of approximately nocturnal periodicity (Fig. 5,a). The new rhythm was very similar to the normal rhythm of crabs collected between tidemarks, and again the first outburst of activity coincided with the time of return to the warm room. In addition, as in ‘shore’ crabs the phasing of the new rhythm was the same whether chilling began during periods of relative inactivity (Fig. 5,b and d) or at the time of onset of an activity peak (Fig. 5 c).

To compare the effects of chilling on ‘dock’ crabs with those on ‘shore’ crabs the results of several experiments, involving a total of 53 crabs, are pooled in Fig. 6. This shows the total hourly activity of 27 ‘dock’ crabs and 26 ‘shore’ crabs, each chilled for between 6 and 24 hr. starting with the time when the chilled animals were returned to the warm room. The approximately tidal (12·4 hr.) rhythm, indicated by open circles in Fig. 6, was clearly evident in chilled ‘dock’ crabs for at least 84 hr. and in ‘shore’ crabs for about 48 hr. Moreover, these results indicate that there is also a twice-tidal (24·8 hr.) component in the rhythms, marked by closed circles in Fig. 6. This was evident in both groups of chilled crabs during most of the 84 hr. period illustrated and it seemed to dominate the rhythm of chilled ‘shore’ crabs after about 48 hr. at normal room temperature. Thus there seemed to be some fundamental similarity between the two sets of results, particularly since each group of crabs showed minimal activity after about 18−20, 44−45 and 68−70 hr., i.e. at intervals of just over 24 hr.

‘Mediterranean’ crabs

Since chilling seemed to initiate a latent tidal rhythm in C. maenas from non-tidal docks it is worthwhile to consider some preliminary experiments at Naples on chilling C. mediterraneus, which habitually experiences virtually non-tidal conditions. Like ‘dock’ C. maenas, C. mediterraneus is also predominantly nocturnal in its normal habitat, but in the latter species the character of the normal rhythm is unaffected by chilling for several hours. In C. mediterraneus, chilling for between 7 and 15 hr. during normally active periods or during quiescent periods produced only slight delays in the normal 24 hr. rhythm (Fig. 7). A slightly greater delay was apparent in Fig. 7,f than in Fig. 7 e even though the former animal was chilled for a shorter period of time. However, it is not yet clear whether some stages of the rhythm of C. mediterraneos are more sensitive to temperature change than others, since chilling began in each case at about the time of onset of activity. The main point emerging from these results is that chilling did not affect the character of the rhythm as it did in C. maenas from non-tidal conditions.

The observations on ‘shore’ crabs seem to suggest that there are at least two physiological mechanisms involved in the control of locomotor rhythms in Carcinus maenas. One of these appears to be fairly deep-seated; it is unaffected by chilling for up to 6 hr. and seems to maintain the basic locomotor rhythm despite irregular fluctuations of environmental temperatures. Chilling for more than 6 hr. however, seems to cause the mechanism to break down, though it is reset on return to normal temperatures (Fig. 3). Warming to 30°C. for 2−7 hr. appears to affect the same mechanism by shortening one cycle of the locomotor rhythm and thus advancing the phase by about 2 hr. (Fig. 4). A second mechanism, on the other hand, is much more sensitive to temperature change, for changes in temperature of as little as 5°C. often produced transient effects which were temporarily superimposed upon the basic tidal periodicity. The temperature-sensitive mechanism, by responding to tidal and daily changes of environmental temperature, might serve to make continual adjustments of the phase of the more basic rhythm (see below).

At least two controlling mechanisms with differing degrees of flexibility have also been postulated for rhythms of colour change in Uca (Brown & Stephens, 1951) and Carcinus (Powell, 1962), for locomotor rhythms in Periplaneta (Harker, 1960) and perhaps also in the intertidal amphipod Synchelidium (Enright, 1963). There appear, however, to be no reports in the literature to compare with the observations that an apparently basic tidal rhythm appears in both ‘shore’ crabs and ‘dock’ crabs which have been chilled for more than 6 hr., irrespective of the time of onset of chilling. These results seem to confirm the hypothesis suggested earlier (Naylor, 1960) that Carcinus rnaenas retains the ability to exhibit persistent rhythms of tidal and twice-tidal periodicity even when kept in non-tidal conditions. This hypothesis is also supported by analyses of the activity rhythm of Carcinus kept in cycles of light and darkness which were different from the normal 24 hr. periodicity (Bünning & Müller, 1961; Blume, Bünning & Müller, 1962). Moreover, there is indirect evidence to support the hypothesis in the results of experiments on ‘Mediterranean’ Carcinus which habitually live in non-tidal conditions, for these forms show a predominantly 24 hr. rhythm (Naylor, 1961) the character of which is relatively unaffected by chilling (see p. 676).

The complex responses to changes in temperature observed in Carcinus are of interest in relation to the fact that in nature the species is frequently exposed to fluctuating temperatures, associated with periodic immersion and exposures to air in the intertidal zone. There are no doubt limitations to the view that temperature changes serve to phase a tidal rhythm (see Enright, 1963), but regular changes in temperature associated with tidal rise and fall might be expected to be at least partly involved in the phasing of the tidal rhythm of activity, perhaps through effects upon the temperaturesensitive mechanism which may in turn act upon the more basic controlling centre. On the other hand, spurious temperature changes, though perhaps resulting in transient responses, would not affect the basic rhythm since they would not be repeated at regular intervals. Such a mechanism might also explain the behaviour of crabs in winter, when they show little evidence of a spontaneous locomotor rhythm (Naylor, unpublished observations) and when larger crabs, at least, usually remain at or below low water mark (Naylor, 1962). At these times air temperatures tend to be lower than sea temperatures, so exposure to cold air on a falling tide might be expected to result in increased running activity which perhaps keeps the crabs in water. On the other hand, in spring, as air temperatures increase, crabs would be expected to remain inactive on sudden exposure to air and would be more likely to be stranded at low tide. From March to about August average sea temperatures are usually lower than average air temperatures, so each period of immersion might tend to result in increased activity associated with the fall in temperature. Apart from the fact that the crabs might be expected to be more active anyway when in water, frequent and rhythmic repetition of a fall in temperature might be expected to act as a ‘zeitgeber’ (see Aschoff, 1960) to phase the more deep-seated controlling centre and thus initiate the ‘endogenous’ tidal rhythm. There is some preliminary indirect evidence to support the view that temperature changes are important in phasing the rhythm, for crabs which showed an imprecise rhythm after several days in aktographs have been shown to revert to a distinct rhythm after a period of chilling (Naylor, unpublished observation). Incidentally, temperature changes may also account for the fact that in summer crabs are often active on the shore at night even when the tide is out (see Naylor, 1958). The absence of light is clearly important (see Naylor, 1960) but it may also be significant that in summer air temperatures at night may be lower than sea temperatures.

The responses to temperature change by Carcinus are also of interest when considering some general characteristics of endogenous components of biological rhyth-micity. First, the effects of increased activity and advancement of the initial peak in crabs subjected to a drop in temperature, with the opposite effects of decreased activity and delayed initial peak when subjected to a rise in temperature (Fig. 1), are in agreement with the generalization that those environmental factors which cause an increase in activity in a given organism also cause a shortening of the period of the rhythm, and vice versa (Aschoff, 1960). Enright (1963) has shown that observations on the intertidal amphipod Synchelidium also agree with this generalization, though in that species lowered temperatures resulted in decreased activity and delayed initial peaks. Lastly, changes in the amount of activity in response to changes of temperature, and which have also been reported in millipedes (Cloudsley-Thompson, 1951a, b), slugs (Dainton, 1954), cockroaches (Cloudsley-Thompson, 1957) and also in isolated nerve ganglia of slugs, crayfish and cockroaches (Kerkut & Taylor, 1956), indicate that many poikilothermal species have temperature-compensatory mechanisms. Such mechanisms may help to account for the ‘temperature-independence’ of biological rhythms (see Cloudsley-Thompson, 1961).

The experiments on Mediterranean Carcinus were carried out during a visit to the Stazione Zoológica, Naples, made possible through the kindness of Dr Peter Dohrn and through a generous grant from the Browne Fund of the Royal Society. I am grateful to Prof. E. W. Knight-Jones for much helpful discussion.

Aschoff
,
J.
(
1960
).
Exogenous and endogenous components in circadian rhythms
.
In Biological Clocks. Cold Spr. Harb. Symp. quant. Biol
.
25
,
11
28
.
Blume
,
J.
,
Bünning
,
E.
&
Müller
,
D.
(
1962
).
Periodenanalyse von Aktivitatsrhythmen bei Carcinus maenas
.
Biol. Zbl
.
81
, (
5
),
569
573
.
Brown
,
F. A.
Jr
., &
Webb
,
H. M.
(
1948
).
Temperature relations of an endogenous daily rhythmicity in the fiddler crab, Uca
.
Physiol. Zool
.
21
,
371
381
.
Brown
,
F. A.
, Jr
. &
Stephens
,
G. C.
(
1951
).
Studies of the daily rhythmicity of the fiddler crab, Uca. Modifications by photoperiod
.
Biol. Bull. Woods Hole
,
101
,
71
83
.
Bunning
,
E.
(
1960
).
Opening address in Biological Clocks
.
Cold Spr. Harb. Symp. quant. Biol
.,
25
,
1
9
.
Bünning
,
E.
&
Müller
,
D.
(
1961
).
Wie messen Organismen lunare Zyklen?
Z. Natur
.
16b
,
391
5
.
Cloudsley-Thompson
,
J. L.
(
1951a
).
On the responses to environmental stimuli and the sensory physiology of millipedes (Diplopoda)
.
Proc. zool. Soc. Lond
.
121
,
253
77
.
Cloudsley-Thompson
,
J. L.
(
1951b
).
Studies in diurnal rhythms. I. Rhythmic behaviour in millipedes
.
J. Exp. Biol
.
28
,
165
72
.
Cloudsley-Thompson
,
J. L.
(
1957
).
Studies in diurnal rhythms V. Nocturnal ecology and water relations of the British cribellate spiders of the genus Ciniflo Bl
.
J. Linn. Soc. (Zool.)
,
43
,
134
52
.
Cloudsley-Thompson
,
J. L.
(
1961
).
Rhythmic Activity in Animal Physiology and Behaviour
,
236
pp.
Academic Press : London
.
Dainton
,
B. H.
(
1954
).
The activity of slugs. I. The induction of activity by changing temperature
.
J. Exp. Biol
.
31
,
165
87
.
Enright
,
J. T.
(
1963
).
The tidal rhythm of activity of a sand-beach amphipod
.
Z. vergl. Physiol
.
46
,
276
313
.
Harker
,
J. E.
(
1960
).
The effect of perturbations in the environmental cycle on the diurnal rhythm of Periplaneta americana L
.
J. Exp. Biol
.
37
,
154
63
.
Kerkut
,
G. A.
&
Taylor
,
B. J. R.
(
1956
).
Effect of temperature on the spontaneous activity from the isolated ganglia of the slug, cockroach and crayfish
.
Nature, Lond
.,
178
,
426
.
Naylor
,
E.
(
1958
).
Tidal and diurnal rhythms of locomotory activity in Carcinus rnaenas (L.)
.
J. Exp. Biol
.
35
,
602
10
.
Naylor
,
E.
(
1960
).
Locomotory rhythms in Carcinus rnaenas (L.) from non-tidal conditions
.
J. Exp. Biol
.
37
,
481
8
.
Naylor
,
E.
(
1961
).
Spontaneous locomotor rhythms in Mediterranean Carcinus
.
Pubbl. Staz. Zool. Napoli
,
32
,
58
63
.
Naylor
,
E.
(
1962
).
Seasonal changes in a population of Carcinus rnaenas (L.) in the littoral zone
.
J. Anim. Ecol
.
31
,
601
9
.
Powell
,
B. L.
(
1962
).
Types, distribution and rhythmical behaviour of the chromatophores of juvenile Carcinus rnaenas (L.)
.
J. Anim. Ecol
.
31
,
251
61
.
Stephens
,
G. C.
(
1957
).
Influence of temperature fluctuations on the diurnal melanophore rhythms of the fiddler crab Uca
.
Physiol. Zool
.
30
,
55
69
.
Sweeney
,
B. M.
&
Hastings
,
J. W.
(
1960
).
Effects of temperature upon diurnal rhythms
.
In Biological Clocks. Cold Spr. Harb. Symp. quant. Biol
.
25
,
87
104
.