Oxygen Consumption of the Terrestrial Mite Alaskozetes Antarcticus (Acari: Cryptostigmata)

The maritime Antarctic affords a unique opportunity for the environmental physiologist to study the effects of low temperatures on terrestrial arthropods. The terrestrial arthropods there consist almost entirely of Acari and Collembola, and although respiration rates have been measured for the collembolan, Cryptopygus antarcticus Willem (Tilbrook & Block, 1972; Block & Tilbrook, 1975), there is little information on the physiology of the terrestrial mites. Accordingly, the present study of oxygen consumption of a common cryptostigmatid mite was undertaken at Signy Island, South Orkney Islands, during the southern summer in 1971–2.

Alaskozetes antarcticus (Michael) of the family Podacaridae is endemic to the Antarctic and Sub-Antarctic regions, and is one of the most southern over-wintering land animals at almost 78–latitude (Wallwork, 1967). There are two distinct subspecies: A. antarcticus antarcticus and A. antarcticus intermedius. This study was concentrated on the former subspecies, which is widely distributed throughout the Antarctic Peninsula area. The habitats of A. antarcticus are varied, from inter-tidal debris, penguin guano, seal wallows, and bird nests to the undersides of loose rocks, bones, crustose lichens and algae, especially Prasiola crispa (Lightf.) Menegh. (Gressitt, 1967; Strong, 1967; Tilbrook, 1967; Wallwork, 1973). It occurs often in dense aggregations which may include all stages from egg to adult. Aggregations are present throughout the year, and development, moulting and maturation occur within them. A. antarcticus is a scavenger feeding on detritus, mostly of vertebrate origin. It is one of the largest Antarctic terrestrial arthropods, but the adult is only ca. 1 mm in body length, and all stages are slow moving.

The objectives of the present study were to investigate the influence of body weight and temperature upon the oxygen consumption of a range of individual mites representing the post-embryonic life stages of A. antarcticus. In addition, data were collected on the variations in metabolism due to sex and breeding condition of the adult. This project was complementary to a long-term ecosystem research programme of two terrestrial sites at Signy Island, which is being undertaken by the British Antarctic Survey (Tilbrook, 1973).

Great care was taken to ensure that collected animals were maintained and studied at temperatures representative of their summer habitat conditions (December-March).

Samples of live mites for respirometry were collected daily by hand from habitats close to the B.A.S. Research Station on Signy Island. Field temperatures were measured at the time of collection by a Grant Model S thermistor thermometer. The experimental temperatures were 0°, + 5° and +10 ° C, and as far as possible the measurements of oxygen uptake were made within 5° of the field temperature. Once collected, the mites were handled entirely in a cold room at the experimental temperature. The mites were sorted and identified to their various life stages under a × 16 stereomicroscope and then weighed individually on a Beckmann electromicrobalance (LM 500) sensitive to 0×1μ g. The micro-respirometer used was a Cartesian Diver (Linderstrøm-Lang, 1943; Holter, 1943), and stoppered divers were utilized (Zeuthen, 1964) with gas volumes in the range 1·90-33·25 μl. The respirometer had a capacity for 14 measurements at one time, and was also in the controlled temperature room. The respirometer water-bath temperature was controlled to ± 0·;01 ° C.

Mites were placed singly in the divers, and at least 30 min was allowed for equilibration after loading into the respirometer. Respiratory measurements were made over 5–6 h, and the individual activity of the mite was recorded. Following this the animals were removed, killed in 75 % ethyl alcohol, mounted in 80 % lactic acid on slides and examined. Confirmation of life stage by the setation of the anal and genital areas in the adults, and the genital papillae in the juveniles (Wallwork, 1962) was then made, along with observations on sex, egg number, and gut contents. Calculations of individual and weight-specific oxygen consumption rates of A. antarcticus were as described for C. antarcticus in Block & Tilbrook (1975), except that individual live-weight measurements, rather than estimated values, were used for the weightspecific rates.

The mean individual live weights of each life stage of A. antarcticus, which were calculated from the material used for the respirometric measurements, are given in Table 1. There was a steady live-weight increase from the larval to the deutonymphal stage, each being approximately double the weight of the previous stage. The largest weight increase (almost 3 × ) occurred between the deuto- and the tritonymph. The adult female was significantly heavier (mean live weight of both gravid and non-gravid individuals) than the tritonymph (P < 0·01).

Both gravid and non-gravid females separately were generally heavier than the males, and the mean weight of the combined gravid and non-gravid females was significantly different (P < 0·01) from the male. Within each life stage, there were no significant differences in live weight of material used at each temperature.

Log₁₀ oxygen consumption rate (× 10^{− 3}μl O₂ ind^{− 1} h^{− 1}) was plotted against log₁₀ live weight μg) for each animal measured at each of the three experimental temperatures (Fig. 1). Linear regressions were fitted to these data, and the resulting equations and correlation coefficients are shown in Table 2. There was a steady increase in respiration rate with increasing live weight at each temperature. The regression lines for + 5°; and +10 ° C were almost parallel, indicating that the relationship of log₁₀ oxygen consumption to log₁₀ live weight was similar at these two temperatures. At 0° C, however, the regression line diverged from those of +5° and +10 ° C over the upper portion.

To investigate this further, the homogeneity of the three regressions was examined using the method of Ostle (1963). This tests whether the combined data can be represented by a single regression line. The overall regression equation for respiration rate on live weight (W) for the three temperatures combined was log₁₀W− 0·7791. The analysis and variance ratio tests (Table 3) showed that the three lines were not homogeneous (P < 0·001) and could not be represented by a single regression. However, the slopes of the lines were not significantly different, and the failure of the overall regression equation resulted from between-temperature differences. Further, the mean respiration rate at each temperature (P < 0·01) and the within and between temperature slopes (P < 0·001) were significantly different. Therefore, the relationship of respiration to weight in A. antarcticus was similar at the three temperatures, but significant differences in respiration rate were detected between temperatures.

Examination of the individual weight-specific respiration rates compared to live weight shows that there was only a very slight negative correlation (Table 2, Fig. 2). The correlation coefficients for each temperature were not significantly different from zero. Therefore weight-specific respiration rate was not correlated with live weight at these temperatures in A. antarcticus.

These results contrast with those for the Antarctic collembolan, C. antarcticus, reported by Block & Tilbrook (1975). For this species, the +5 ° C regression line of individual respiration rate against live weight was significantly different from those of 0° and +10 ° C. It was concluded that the smaller, immature individuals had a significantly higher metabolic rate at + 5° C, which may be a seasonal adaptation to summer temperatures in the region of +3° to +5 ° C. A. antarcticus at 0 ° C exhibited a different metabolism-weight relationship from that at + 5° and +10 ° C, but this was not significant. The metabolism of A. antarcticus in respect to live weight was not affected by a rise in temperature to +10 ° C, which was above the summer norm.

Considering the weight exponent b in the relationship (where respiration rate, and W: live weight) for A. antarcticus (Table 2,) b varied from 0·830 to 0·976 with a mean of 0·927 over the 10 °C temperature range. These values are generally higher than those found for terrestrial mites. Berthet (1964) determined that b = 0·72 for 16 species of temperate oribatids, and gave a range of b values : 1·372 (0 ° C), 0·123 ( + 5 ° C), 0·459 ( + 10 ° C) and 0·722 ( +15 ° C) for the adult mite Steganacarus magnus Nicolet. The coefficients of the slope of the lines at +10 °C and +15 °C did not differ significantly, and both these groups taken together gave a value of 0-630. These results were partially explained by Berthet in that the correlation coefficient between log₁₀ oxygen consumption and log₁₀ live weight was highly significant at all temperatures except + 5 ° C, and also that the variability of oxygen uptake determinations was greatest at 0 ° C, presumably due to decreased sensitivity of the diver. For four species of phthiracarid mites, comprising 44 individuals (mostly adults), Wood & Lawton (1973) determined a weight exponent b of 0·539 at +10 ° C-In three oribatid species (all life stages) the value of b ranged from 0·511 to 0·697. For a further 12 species of Cryptostigmata covering 92 individual adults, b was 0·572 at the same temperature. They considered that, after weight, activity was the most important factor influencing mite respiration. However, the results for A. antarcticus were obtained on resting animals and the effect of activity on metabolism can largely be excluded. Webb (1975) calculated a mean b value of 0 ·686 for all life stages of 5. magnus at 4-18 °C. He concluded also that two regression lines adequately represented the relationship between respiratory rate and live weight: one for adults-(In In W − 1 ·819) and another for juveniles (In In W − 0-201). Weight therefore, has a major influence upon metabolism of the micro-arthropods studied, but it is not a constant effect either within or between species at differing temperatures within their normal range. This may be due to the varying degree of chitinization of the exoskeleton between individuals and species. Amongst the mites and Collembola, A. antarcticus has the largest weight exponents so far determined.

In order to examine differences in oxygen uptake between individual life stages of A. antarcticus more closely, the mean respiration rates for each stage at each temperature (Table 4) were calculated from the individual data in Fig. 1. Mean oxygen consumption per individual increased steadily with mean live weight of the juvenile stages at each temperature (Fig. 3 a). The mean adult male respiration rate decreased slightly below the tritonymphal rate at o °C but increased above it at + 5 ° C, and also at +10 ° C where it continued the juvenile trend. There was much more variability in the results for males at +10 ° C than at + 5 ° C and 0 ° C. A similar situation was seen for the mean female respiratory rate over the three experimental temperatures, except that at + 5 ° C the female rate continued the juvenile trend and did not increase as in the male.

Marked differences between the juvenile stages occurred in mean weight-specific respiration rates (Fig. 3 b), both within and between temperatures. At 0 ° C the mean protonymphal rate was depressed compared to both the larval and deutonymphal rates. Thereafter, there was a steady decrease in metabolic rate with increasing live weight, with the adult female having the lowest rate. At + 5 ° C there were no significant differences between the juvenile stages; but the adult male rate was distinctly increased compared to the female level. At +10 °C metabolic rate increased greatly from the larval to deutonymphal stage, thereafter a steady decline ensued, with no significant differences between the tritonymph and adult male and female. The results stress the considerable differences in oxygen consumption between the various life stages and these are further complicated by differing metabolic responses to the experimental temperatures.

There are few published data on respiration rates throughout development in terrestrial mites. In Nothrus silvestris Nicolet, the protonymphal stage had a high rate (213 ·1 μl O₂g^{− 1}h^{− 1}) compared with other juvenile stages and adults (range 119 ·2–·185·8 μl O₂ g^{− 1}h^{− 1}) at +10 ° C (Webb, 1969). The deutonymphs of Nothrus pahistris C. L. Koch and Parachipteria willmanni Hammer had significantly higher respiration rates than the protonymph at +10 ° C (Wood & Lawton, 1973). Again, the deutonymphal metabolic rate of S. magnus at +18 ° C was considerably higher than that of the other nymphal stages, but so also was larval metabolism (Webb, 1975). It seems that there are differences both between species and between temperatures in this respect for the five mite species for which there are data. In A. antarcticus the proto- and deutonymphs show the greatest metabolic response at +10 ° C.

The sex ratio of the weighed individuals (see Table 1) was 1 female: 1 ·21 male. There were no significant differences in oxygen uptake per individual and per g for male and female A. antarcticus at each of the three temperatures studied. Further comparisons of the mean respiration rates of male and gravid female did not reveal any differences. This was surprising in view of the fact that the female was heavier (Table 1) and a large proportion (77 %) of the females were carrying eggs. The mean number of eggs was 4 ·3 per gravid female, and it is interesting to note that gravid females only occurred in the samples from 27 January onwards. These results are similar to C. antarcticus (Block & Tilbrook, 1975), in which the effect of sex on respiratory rate was not detected at the same temperatures.

Table 5 shows a further breakdown of the respiration data for adult females into gravid and non-gravid components at each experimental temperature. No data were obtained for non-gravid female A. antarcticus at 4-10 ° C, and thus the comparison is restricted to the two lower temperatures. The gravid female rate was higher than that of the non-gravid female at +5 ° C on an individual basis, but a multiple regression analysis of oxygen consumption of the adult female on live weight and number of eggs and prelarvae failed to confirm this difference. For mean weight-specific oxygen uptake the gravid rate was lower than the non-gravid female at both o° and + 5 °C. Thus there was a slight reduction in metabolic rate due to an increase in the female weight component (Table 1).

In N. silvestris, Webb (1969) found that gravid females showed a 25 % increase in weight-specific oxygen uptake compared to non-breeding adults (non-gravid females and males). In S. magnus the egg content of the female was more important than live weight or number of prelarvae in its effect upon respiratory rate at +10 ° C (Webb, 1975). Wood & Lawton (1973) studying a range of oribatid mites, recorded that in five out of seven species, the gravid female was heavier than the non-breeding adults (i.e. males and non-gravid females) and usually exhibited higher respiratory rates than these at +10 ° C. However, significant differences were only established between individual rates for Ceratoppia bipilis (Hermann) and S. magnus, and between weightspecific rates for Damaeus onustus C. L. Koch and S. magnus. These trends are similar to those found in A. antarcticus.

Fig. 4(a) shows the mean individual oxygen consumption plotted against temperature for each life stage of A. antarcticus. For all stages there is a steady increase of respiration rate with temperature over the experimental range. There is a separation into two groups in this respect: one consisting of larva, protonymph and deutonymph, and the other of tritonymph, adult male and female. This may be a reflexion of an increased individual metabolism from early in the tritonymphal instar, which occurred at all three temperatures, due to an increase in activity. The range of oxygen consumption values over all life stages is small at 0 ° C, greatly increased at + 5 ° C, and largest at +10 ° C. At + 5 ° and +10 ° C the order of increasing metabolism follows the developmental cycle, but at 0 ° C, the tritonymph has a higher level than the adult male and female. At + 5 ° C the oxygen consumption of the male exceeds that of the female, whereas they are similar at 0 ° and +10 ° C. A more uniform pattern for all stages emerges when the weight-specific respiration rate and temperature relationship is examined (Fig. 4b). The larval stage showed the least increase of metabolism (g^{− 1} h^{− 1}) from 0 ° to + 10 ° C.

Q₁₀ values were calculated for each instar of A. antarcticus for the temperature ranges o° to +5 °C, 4·5° to +10 °C and 0° to +10 °C, using the mean respiratory rate animal^{− 1} (Table 6). Over the full o° to 10 °C range, there was only a small variation in temperature response from a Q₁₀ of 2 ·07 (deutonymph) to 3·83 (protonymph). If Q₁₀ values for the component temperature ranges are examined, more striking differences are apparent. Between o° and 4·5 °C, Q₁₀ varied greatly from 1·98 (tritonymph) to 7·23 (adult male), and between +5 °C and +10 °C it varied only slightly from 1·78 (larva) to 3·10 (protonymph). The deuto- and tritonymph stages exhibited least change in Q₁₀, whereas the adult male showed most change over the ranges examined. It is concluded that environmental temperature changes within the normal summer range for A. antarcticus at Signy Island elicit a complex series of metabolic response patterns, which may be associated with activity of the mites. Again, as for the collembolan, C. antarcticus (Block & Tilbrook, 1975), there are very marked changes in Q₁₀ above and below + 5 °C. Diurnal temperature fluctuations in summer in the habitats of A. antarcticus in the maritime Antarctic are probably large (− 5° to +16 °C ; Chambers, 1966), but at Signy Island the mean summer temperature is in the range − 0·5° to +9 °C. The total exposure time of the individual to temperatures around 4-10 °C in summer is probably small. This may account for the increased variability in respiration levels especially of the adults (Fig. 3 a) and juveniles (Fig. 3 b) at +10 °C. The results suggest, therefore, that there may be metabolic adaptation to summer temperatures in A. antarcticus.

Few calculations of Q₁₀ have been made for Acari. For 16 species of oribatid mites, Berthet (1964) found a Q₁₀ range of 3·5 ·5·7 with a mean of 4·0 from o° to +15 °C. A Q₁₀ of 2·65 was calculated for N. silvestris for the temperature range 4·10° to +20 °C (Webb, 1969), and similarly a Q₁₀ of 2·03 for S. magnus over the range 4·11° to 4·25 °C (Webb, 1975). These were all adults of temperate species, but they have some similarity with the Q₁₀ values for the various stages of A. antarcticus over o° to 4 ·10 °C (Table 6). The larva, protonymph, adult male and female of A. antarcticus have higher temperature coefficients than Webb’s species, but lower than Berthet’s species. The Q₁₀ values of these instars of A. antarcticus approach Berthet’s mean for adult mites over a similar span of temperature. Compared with a Q₁₀ range of 1·99–2·54 for C. antarcticus (Block & Tilbrook, 1975) over the same temperature interval, values for A. antarcticus are generally higher. It is expected that cold-adapted arthropods will exhibit a greater response to changing temperature than temperate species, and this is confirmed by A. antarcticus. This is also true for male blowflies (Tribe & Bowler, 1968) in which standard metabolism is temperature dependent over the range 10° –30 °C, but not for many marine invertebrates (Newell & Pye, 1971). In Littorina littorea (L.) standard respiratory rate is almost independent of temperature, but the active rate is markedly temperature dependent and the point beyond which a decline occurs varies seasonally. The mechanisms of metabolism-temperature interaction are little understood at present, and current research on A. antarcticus is directed towards this end.

The results of the present study showed that the relationship between oxygen consumption rate and live weight remained constant in A. antarcticus over the temperature range 0° to +10 °C. Significant differences in respiration level were detected between these temperatures. This is in contrast to a similar study of the collembolan C. antarcticus (Block & Tilbrook, 1975). Comparative data for other species suggest that weight is the major influence on metabolism in micro-arthropods, but this is not constant within or between species and it may vary with temperature. This is probably caused by the varying degree of sclerotization of the different life stages and species for which data exist.

The influence of temperature on oxygen consumption of A. antarcticus was also important. In terms of individual respiration and temperature, there were two lifestage groups, one consisting of larva, proto- and deutonymph with low rates, and the other composed of tritonymph, adult males and females with high rates. The early instars (proto- and deutonymph) exhibited the highest weight-specific respiration rates, particularly at +10 °C. Also, the various life stages showed markedly different temperature responses as indicated by Q₁₀. The effect of temperature changes, which are representative of those occurring in the natural environment, upon cold-adapted species such as A. antarcticus require further investigation.

Comparative respiration data for temperate mites have been recorded by Berthet (1964), Webb (1969,1970,1975) and Wood & Lawton (1973). At o °C, adults of four species of oribatids over a similar weight range (calculated from Berthet, 1964) had much lower (1·073 − 5·872 ×^{− 3}μO₂ ind^{− 1} h⁻) respiration rates than adult A. antarcticus (Table 4). Also, at +5 °C, adult A. antarcticus showed much higher levels of metabolism, compared to adults of 16 oribatid species (Berthet, 1964) but this difference was not so pronounced at +10 °C. Fig. 5 (a) is a double log₁₀ plot of all the available respiration data at +10 °C against live weight of adult Cryptostigmata. The equation for the fitted regression line (C) of respiration rate on live weight (W: μg) excluding A. antarcticus is (n = 48, r = +0·923, S.E._b) =0·044)-Published data for respiration rates at +10 °C of adults of 19 species of Mesostigmata (Webb, 1970; Wood & Lawton, 1973) and adults of five species of Prostigmata (Wood & Lawton, 1973) allow the following equations to be derived on the same basis :

Respiratory data covering all the post-embryonic life stages of individual mites are very limited. Fig. 5 (b) compares results for six species of Cryptostigmata including A. antarcticus for which life-stage data are available at +10 °C. Linear regressions have been fitted for each species, and the regression equations are given in Table 7.

Over the upper part of its weight range, individuals of all life stages of A. antarcticus have much higher respiration levels than other species, which suggests some degree of metabolic adaptation in this species. Variation in the regression coefficient occurs between species, and the b value for N. silvestris (0·930) approaches that of A. antarcticus ; the remaining species having lower weight exponents. A general relationship Unking oxygen consumption and live weight has been derived for the six oribatid species in Fig. 5(b): ⁠. A-similar relationship of has been derived from data covering adults and nymphs of several oribatid mites (Chapman & Webb, 1977). By contrast, analysis of life-stage respiration and weight data for four species of Mesostigmata (Wood & Lawton, 1973) gives the following relationship: ⁠, indicating that live weight has a greater influence on the measured respiration rate over the life cycle in these mites than in the Cryptostigmata.

The present study reports a b value ranging from + 0·830 (0 °C) to + 0·976 (4·10 °C) for A. antarcticus. This is generally higher than that of other Antarctic invertebrates which have been studied: +0-120 (0°C) to +0·570 (+10 °C) for cultured C. antarcticus (Tilbrook & Block, 1972), + 0·669 ( + 5 °C) to + 0·825 ( +10°Q for field C. antarcticus (Block & Tilbrook, 1975) and +0·51 ( + 5° and +10 °C) for the tardigrade Macrobiotus furciger J. Murray (Jennings, 1975). Individual respiration is almost directly dependent upon live weight rather than surface area in A. antarcticus. Zeuthen (1947) concluded that animals < 1 g in weight do not obey the surface law of ⁠, and this is confirmed by examination of the limited micro-arthropod data available. Live weight has been shown to be a major factor affecting individual respiration levels in such arthropods, but its effect varies within the species according to developmental stage and between species over their normal temperature range. Studies of respiration and growth rates together, over a range of field temperatures, in selected arthropods are required to clarify this.

Considering the effect of temperature alone on the respiration of A. antarcticus, a regression analysis of the individual data for the three experimental temperatures allows the following equation to be derived: ⁠, where⁠: respiration rate ( × 10^{− 3}μl O₂ ind^{− − 1} h^{− 1}) and T : temperature (°C). A mean Q₁₀ value may be calculated as the antilog of 10 × coefficient b, which is 3·31 and this corresponds to the average value for the range of temperature coefficients given in Table 6. However, the weak regression and correlation coefficients confirm that temperature alone does not influence respiration significantly in this species. Berthet (1964) transformed his individual weight data to a standard weight equal to the mean of all the mites used in his respiration experiments. For adult data from 16 species at +5°, +10° and +15 °C he calculated the relationship, ⁠. The regression coefficient for temperature is not significantly different from that of A. antarcticus, but further comparison is precluded because of life-stage differences in the data.

The effects of both live weight and temperature on individual respiration of A. antarcticus can be examined by means of a multiple regression, the equation being calculated, where respiration rate ( ×^{− 3} O₂ ind^{− 1} h^{− 1}), W: live weight (μg) and T: temperature (°K). This was further developed in respect of the absolute temperature by analogy with Arrhenius’ law to give ⁠, where ⁠: respiration rate (× 10 μl O₂ ind^{− 1} h^{− 1}), W: live weight μg) and T: 1 × 10⁴/ T absolute. This general relationship is representative of the active life stages of A. antarcticus over the temperature range 0° to +10 °C. This allows a comparison with Berthet’s (1964) generalization for adults of several species over the temperature interval + 5°to + i5°C,⁠. It can be seen that weight exerts a greater influence on respiration level in A. antarcticus than in the temperate species, and that temperature has a slightly reduced effect on respiration for the Antarctic mite.

The available data on metabolic rate (weight-specific oxygen consumption) and live weight, which have been reported for a range of Antarctic terrestrial invertebrates, are shown in Fig. 6. For A. antarcticus, only Marsh (1973) is comparable, and his average value lies within the range of results for the present study at + 5 °C. The collembolan C. antarcticus is generally smaller in size, and its metabolic rate (μl O₂ g^{− 1} h^{− 1}) is higher than that of A. antarcticus. The other collembolan Isotoma klovstadi Carpenter is larger than C. antarcticus, but higher metabolic rates have been measured for this species even at −4 °C (Strong, Dunkle & Dunn, 1970). Gamasellus racovitzai (Trouessart) (Goddard, 1976a), a mesostigmatid mite, although of similar weight to A. antarcticus, has a higher metabolism probably due to its greater locomotory activity at low temperatures associated with its predatory role. The Antarctic Prostigmata are amongst the smallest arthropods to be measured with a Cartesian diver microrespirometer, and several of them are very active species. Hence their higher metabolic rates (Goddard, 1976b; Block, 1976), compared with other mites and the Collembola.

Considering all the data for Antarctic terrestrial invertebrates, a relationship of metabolic rate to temperature over the range − 4 ° to +22 °C has been derived as: ⁠, where ⁠: metabolic rate (μl O₂ g^{− 1} h^{− 1}) and T: temperature (°C). The overall Q₁₀ calculated from the regression coefficient b is 3·04, which is slightly higher than the majority of values found for temperate micro-arthropods. Comparison of this relationship for Antarctic species with that derived from 109 species of temperate Acari over 0° to + 25 °C, where ⁠, indicates that there is no significant difference between the regression coefficients. Therefore it may be concluded that the metabolic response to temperature of the two groups of Antarctic and temperate terrestrial invertebrates is similar over their normal temperature ranges, but the Antarctic group has a generally elevated level of metabolism. This elevation is 3 to 5 times the metabolic rate of temperate Acari over the range o° to + 20 °C.

It is possible, knowing the respiratory rate of an animal and the calories lost for each unit of oxygen consumed, to calculate the minimum energy required to support metabolism. Using an oxy-calorific equivalent of 4·74 calories ml^{− 1} of oxygen consumed (Petrusewicz & Macfadyen, 1970) for an animal feeding on a mixed diet and with an RQ of 0-82, and the respiratory rates of A. antarcticus determined at Signy Island (Table 4), it was calculated that from 0·093 (larva at 0 °C) to 4·241 (adult male at +10 °C) μcal ind^{− 1}24 h^{-− 1} were required for maintenance metabolism. Marsh (1973) using a mean respiration value of 21·875 x 10^{− 3}μl O₂ ind^{− 1}h^{− 1} at +5 °C, an ingestion rate of 0·341 /4g dry weight of food 24 h^{− 1} and a calorific food equivalent fo 4·7 μcal μg^{− 1} determined that 78·2% of the energy assimilated was utilized in respiration for an individual A. antarcticus of 81·0μg weight. Marsh’s mean respiration figure at + 5 °C falls between the tritonymph and adult values in the present study (Fig. 3 a), but his mean weight corresponds to midway between the deuto- and tritonymphal stages (Table 1). Notwithstanding these differences, a similar calculation of the proportion of energy used by A. antarcticus in maintenance metabolism, based on the Signy Island respiration data at +5 °C and a mean weight of 81·0 μg, gave a value of 82·3 %. This is rather higher than Marsh’s estimate, but together the two values suggest that the majority of energy assimilated by A. antarcticus is used in respiratory metabolism. It is concluded that this is probably a feature of microarthropod energetics in habitats with low environmental temperatures, leaving only a small proportion of energy for production of tissues and young. Hence, these animals have slow growth rates and generally long life cycles compared to temperate species (Block, 1965).

There has been little attention paid to metabolic compensation to temperature amongst terrestrial poikilotherms of high latitudes. In studies of arctic insects, Scho- lander et a1.(1952) found scant evidence for a relative elevation in respiratory activity at low temperatures. This is in contrast to the results of the present study. Most cold-adapted terrestrial arthropods are probably dependent on exposure to relatively high environmental temperatures, for albeit short periods, for their activity and development. At Signy Island, habitat temperatures in the summer for A. antarcticus are mostly in the range − 5° to +9·5 °C, when growth and reproduction are maximal. As temperatures fall below o °C, a capacity to survive mechanical damage due to freezing rather than to compensate metabolically becomes more important. With the severe winter conditions of its habitat (minimum temperature − 20°to − 30°C), A. antarcticus must be able to withstand freezing in most of its life stages. Investigations are presently in progress to determine if the capacity to withstand freezing is linked to a facility to supercool, and to elucidate the possible mechanisms involved.

I thank the British Antarctic Survey for support throughout the 1971– 2 Antarctic summer season, the Leverhulme Trust for a Research Fellowship, the Royal Society for a travel grant and Leicester University for leave of absence, without which this research would not have been possible. Finally, I appreciate the criticism of draft manuscripts by Drs P. J. Tilbrook, R. R. Harris and N. R. Webb.